Upload
others
View
1
Download
0
Embed Size (px)
Citation preview
School of Health Sciences
Attention in Rodents:
Pharmacology and the Role of the
Frontoparietal Network
A thesis submitted to The University of Manchester for the
degree of Doctor of Philosophy in the Faculty of Biology,
Medicine and Health
Andrew James Hayward
2017
2
Table of Contents List of Figures ....................................................................................................................................... 5
List of Tables ........................................................................................................................................ 6
Abbreviations ....................................................................................................................................... 7
Thesis Abstract ..................................................................................................................................... 9
Declaration ........................................................................................................................................ 10
Copyright ........................................................................................................................................... 10
Acknowledgements ............................................................................................................................ 11
1 GENERAL INTRODUCTION .................................................................................................................. 12
1.1 CONTEXT OF THESIS ......................................................................................................................... 12
1.2 ATTENTION .................................................................................................................................... 12
1.2.1 History of attention ............................................................................................................. 12
1.2.2 How attention is used.......................................................................................................... 13
1.2.3 Neurobiology of attention ................................................................................................... 14
1.2.4 Neurotransmitters of attention ........................................................................................... 15
1.2.5 Pharmacology of attention .................................................................................................. 16
1.2.6 Neural Oscillations .............................................................................................................. 18
1.3 ANIMAL MODELLING ....................................................................................................................... 19
1.3.1 Abstract .............................................................................................................................. 21
1.3.2 Introduction ........................................................................................................................ 22 1.3.2.1 Value of Animal Models and the Need for Improved Translation..................................................... 24 1.3.2.2 Measuring Attention and Impulsive Action in Rodents .................................................................... 25
1.3.3 Methods of Separation ........................................................................................................ 26 1.3.3.1 Models of High Impulsive Action .................................................................................................... 26 1.3.3.2 Models of Low Attention ............................................................................................................... 28 1.3.3.3 Models Most Relevant to ADHD Symptomatology .......................................................................... 29
1.3.4 Neurobiology Identified Using Separation Methods ............................................................. 33 1.3.4.1 Using Impulsive Models ................................................................................................................. 33
1.3.4.1.1 Behaviour ................................................................................................................................ 33 1.3.4.2 Pharmacology/ Neurobiology......................................................................................................... 34 1.3.4.3 Using Models of Attention ............................................................................................................. 36 1.3.4.4 Using Models of ADHD................................................................................................................... 37 1.3.4.5 Using Models developed in 5C-CPT ................................................................................................ 37
1.3.5 Selection of the most appropriate method of separation ...................................................... 44
1.3.6 Conclusion .......................................................................................................................... 45
1.4 GENERAL AIMS OF THESIS ................................................................................................................. 46
2 GENERAL METHODS .......................................................................................................................... 48
2.1.1 Animals ............................................................................................................................... 48
2.1.2 Animal Usage ...................................................................................................................... 48
2.1.3 Apparatus ........................................................................................................................... 49
2.1.4 5C-CPT Task Structure ......................................................................................................... 50
2.1.5 5C-CPT Training................................................................................................................... 51
2.1.1 Signal Detection Theory....................................................................................................... 53
2.1.2 Variable Stimulus Duration .................................................................................................. 55
2.1.3 Division into subgroups ....................................................................................................... 56
3 PHARMACOLOGICAL VALIDATION OF A REFINED LOW ATTENTIVE MODEL USING METHYLPHENIDATE
AND ATOMOXETINE .................................................................................................................................. 57
3.1 CONTRIBUTIONS ............................................................................................................................. 57
3.2 CONTEXT ...................................................................................................................................... 58
3
3.3 ABSTRACT ..................................................................................................................................... 58
3.4 INTRODUCTION ............................................................................................................................... 59
3.5 METHODS ..................................................................................................................................... 60
3.5.1 Animals ............................................................................................................................... 60
3.5.2 Apparatus ........................................................................................................................... 61
3.5.3 5C-CPT Training Procedure .................................................................................................. 61
3.5.4 Testing Procedure ............................................................................................................... 61
3.5.5 Division into Subgroups ....................................................................................................... 61
3.5.6 Statistical Analysis............................................................................................................... 62
3.6 RESULTS ....................................................................................................................................... 63
3.7 METHYLPHENIDATE ......................................................................................................................... 63
3.7.1 Group Performance ............................................................................................................. 63
3.7.2 High/ Low Comparison ........................................................................................................ 64
3.8 ATOMOXETINE ............................................................................................................................... 67
3.8.1 Group Performance ............................................................................................................. 67
3.8.2 High/ Low Comparison ........................................................................................................ 68
3.9 DISCUSSION ................................................................................................................................... 71
4 NICOTINE AND CAFFEINE IN THE 5 CHOICE CONTINUOUS PERFORMANCE TASK ............................... 73
4.1 CONTRIBUTIONS ............................................................................................................................. 73
4.2 CONTEXT ...................................................................................................................................... 74
4.3 ABSTRACT ..................................................................................................................................... 74
4.4 INTRODUCTION ............................................................................................................................... 75
4.5 EXPERIMENTAL PROCEDURES ............................................................................................................. 76
4.6 ANIMALS....................................................................................................................................... 76
4.6.1 Apparatus ........................................................................................................................... 76
4.6.2 5C-CPT Training Procedure .................................................................................................. 76
4.6.3 Testing Procedure ............................................................................................................... 76
4.6.4 Division into Subgroups ....................................................................................................... 77
4.6.5 Statistical Analysis............................................................................................................... 77
4.7 RESULTS ....................................................................................................................................... 78
4.7.1 Whole Group Baseline ......................................................................................................... 78
4.7.2 High/Low Comparison ......................................................................................................... 80
4.8 DISCUSSION ................................................................................................................................... 85
5 PARTIAL AGONISM AT THE Α7 NICOTINIC ACETYLCHOLINE RECEPTOR IMPROVES ATTENTION,
IMPULSIVE ACTION AND VIGILANCE IN LOW ATTENTIVE RATS ................................................................. 88
5.1 CONTRIBUTIONS ............................................................................................................................. 88
5.2 CONTEXT ...................................................................................................................................... 89
5.3 ABSTRACT ..................................................................................................................................... 89
5.4 INTRODUCTION ............................................................................................................................... 90
5.5 EXPERIMENTAL PROCEDURES ............................................................................................................. 92
5.5.1 Animals ............................................................................................................................... 92
5.5.2 Apparatus ........................................................................................................................... 92
5.5.3 5C-CPT Training Procedure .................................................................................................. 92
5.5.4 Testing Procedure ............................................................................................................... 93
5.5.5 Division into Subgroups ....................................................................................................... 93
5.5.6 Statistical Analysis............................................................................................................... 93
5.6 RESULTS ....................................................................................................................................... 94
5.6.1 Group Performance ............................................................................................................. 94
5.6.2 High/ Low Comparison ........................................................................................................ 96
4
5.7 DISCUSSION ................................................................................................................................. 100
6 FRONTOPARIETAL NEURAL ACTIVATION CORRELATES WITH VIGILANCE IN THE 5C-CPT .................. 104
6.1 CONTRIBUTIONS ........................................................................................................................... 104
6.2 CONTEXT .................................................................................................................................... 105
6.3 ABSTRACT ................................................................................................................................... 105
6.4 INTRODUCTION ............................................................................................................................. 106
6.5 METHODS ................................................................................................................................... 107
6.5.1 Animals ............................................................................................................................. 107
6.5.2 5C-CPT Apparatus, Training and Testing ............................................................................ 107
6.5.3 Tissue Extraction ............................................................................................................... 108
6.5.4 Immunohistochemistry ...................................................................................................... 108
6.5.5 Quantification ................................................................................................................... 110
6.5.6 Statistical Analysis............................................................................................................. 110
6.6 RESULTS ..................................................................................................................................... 112
6.7 DISCUSSION ................................................................................................................................. 113
7 OSCILLATIONS IN COGNITION: AN ATTENTION-PROMOTING DOSE OF METHYLPHENIDATE
AUGMENTS ALPHA RHYTHM IN THE DORSAL ATTENTION NETWORK ..................................................... 116
7.1 CONTRIBUTIONS ........................................................................................................................... 116
7.2 CONTEXT .................................................................................................................................... 117
7.3 ABSTRACT ................................................................................................................................... 117
7.4 INTRODUCTION ............................................................................................................................. 118
7.5 METHODS ................................................................................................................................... 119
7.5.1 Animals ............................................................................................................................. 119
7.5.2 Surgery ............................................................................................................................. 119
7.5.3 Recording.......................................................................................................................... 119
7.5.4 Histology........................................................................................................................... 120
7.5.5 Analysis ............................................................................................................................ 120
7.6 RESULTS ..................................................................................................................................... 120
7.7 DISCUSSION ................................................................................................................................. 123
8 GENERAL DISCUSSION ..................................................................................................................... 126
8.1 AIM 1: TO FURTHER DEVELOP A METHOD OF MODELLING ATTENTION DEFICITS IN RATS. .................................. 126
8.2 AIM 2: TO USE THE HIGH AND LOW ATTENTIVE MODEL TO STUDY THE PHARMACOLOGY OF ATTENTION ................ 127
8.3 AIM 3: TO EXAMINE THE INVOLVEMENT OF THE FRONTOPARIETAL NETWORK IN PERFORMANCE OF THE 5C-CPT AND
EFFECTS OF ATTENTION PROMOTING AGENTS ................................................................................................... 128
8.4 LIMITATIONS AND FUTURE RESEARCH ................................................................................................. 129
8.5 CONCLUSION................................................................................................................................ 132
9 REFERENCES .................................................................................................................................... 133
5
Word Count: 39,692
List of Figures
Figure 1.1: Brain regions involved in alerting, orienting and executive networks. .......................... 15 Figure 1.2: The 5 Choice Serial Reaction Time Task vs 5 Choice Continuous Performance Task 26 Figure 1.3: Baseline separation of HI-filled circles/LI-empty circle animals using the Dalley et al.
(2007) method. ............................................................................................................................ 27 Figure 1.4: a dimensional analysis method of performance separation ......................................... 30 Figure 2.1: Animals used for 5C-CPT experiments and the order of use ...................................... 49 Figure 2.2: 5C-CPT Apparatus for training and pharmacological testing ....................................... 49 Figure 2.3: 5C-CPT Task Structure. ............................................................................................. 50 Figure 2.4: Representative progression of training. ...................................................................... 53 Figure 3.1: Methylphenidate increased accuracy at 0.5 and 1 mg/kg in LA animals. ..................... 65 Figure 3.2: Vigilance is increased by 0.5 and 1 mg/kg in LA animals. ........................................... 66 Figure 3.3: No significant differences were seen on the latency measures due to methylphenidate.
(A) No significant differences in correct latency due to methylphenidate. ...................................... 67 Figure 3.4: In LA animals Accuracy is increased and premature responses reduced by
atomoxetine. ............................................................................................................................... 69 Figure 3.5: Atomoxetine reduced d′ in HA animals, but increased it in LA animals ........................ 70 Figure 3.6: Atomoxetine had no significant effects on the latency measures in HA or LA animals. 71 Figure 4.1: Comparison of accuracy and d′ after vehicle treatment in HA and LA groups when
grouped by these parameters ...................................................................................................... 81 Figure 4.2: Performance measures of the go trials in 5C-CPT when treated with saline, nicotine
(0.1 and 0.2 mg/kg) and caffeine (2.5 and 5 mg/kg) ..................................................................... 82 Figure 4.3: Signal Detection Theory related measures of the 5C-CPT .......................................... 83 Figure 4.4: Latency measures of the 5C-CPT following treatment with vehicle, nicotine (0.1 and 0.2
mg/kg) or caffeine (2.5 and 5 mg/kg). .......................................................................................... 84 Figure 4.5: Number of trials completed per 20 minutes of the 5C-CPT. ........................................ 84 Figure 5.1: Group differences between high and low attentive groups (HA/LA) at increasing SDs. 97 Figure 5.2: Encenicline improved selective attention and vigilance in low attentive (LA) animals
only. ............................................................................................................................................ 97 Figure 5.3: Encenecline reduces impulsivity in LA animals at 0.75 s SD. In HA animals pFA (A)
trended to significance for 0.75 s SD and 0.1 mg/kg (p=0.081). ................................................... 99 Figure 5.4: Latency measures of the 5C-CPT showed no significant TREATMENT*GROUP
interaction. ................................................................................................................................ 100 Figure 6.1: Regions analysed for C-Fos quantification ............................................................... 109 Figure 6.2: Correlation between count order and the number of identified c-Fos-positive neurones.
................................................................................................................................................. 111 Figure 6.3: C-Fos-positive cell count was higher in PPC compared to PFC. ............................... 112 Figure 6.4: Vigilance (d′) correlates positively with the number of c-Fos-positive cells in mPFC and
PPC .......................................................................................................................................... 113 Figure 7.1: representative power spectral densitiy for saline recording. ...................................... 121 Figure 7.2: Methylphenidate takes 20-40 minutes to cause changes in the neural oscillations .... 122 Figure 7.3: Methylphenidate significantly increases power in the alpha band, particularly in the
mPFC, while reducing power at lower frequencies. .................................................................... 123
6
List of Tables
Table 1.1: Summary of published methods for modelling impulsivity, attention and combined
symptom subtypes using performance based separation techniques. .......................................... 31 Table 1.2: Summary of pharmacological effects in models based on behavioural separation........ 39 Table 2.1: Training stages for 5C-CPT. ........................................................................................ 52 Table 2.2: Summary of measures in the 5C-CPT ......................................................................... 54 Table 3.1: Group vehicle treated baseline before HA/LA grouping for methylphenidate ................ 63 Table 3.2: Group vehicle treated baseline before HA/LA grouping for atomoxetine....................... 67 Table 4.1: Whole cohort 5C-CPT performance following vehicle compared to nicotine and caffeine
................................................................................................................................................... 79 Table 5.1: Effect of encenicline on the whole cohort..................................................................... 95
7
Abbreviations
Abbreviation Meaning
5C-CPT 5 Choice Continuous Performance
5C-SRTT 5 Choice Serial Reaction Time Task
5-HT 5-Hydroxytriptamine
Acc Accuracy
ADHD Attention Deficit/Hyperactivity Disorder
ADHD-C Attention Deficit/Hyperactivity Disorder Combined Subtype
ADHD-HI Attention Deficit/Hyperactivity Disorder Hyperactive/Impulsive Subtype
ADHD-I Attention Deficit/Hyperactivity Disorder Inattentive Subtype
ANOVA Analysis of Covariance
BOLD Blood Oxygen Level Dependent
BSA Bovine Serum Albumin
Cg1 Cingulate Cortex 1
COMT Catechol-O-Methyl Transferase
CPT Continuous Performance Task
d′ D Prime
DAT Dopamine Transporter
DRD4 Dopamine D4 Receptor
E Efficent Performers
EEG Electroencephalography
fMRI Functional Magentic Resonance Imaging
GABA Gamma Aminobutyric Acid
HA High Attentive
HI High Impulsive
HP High Performance
IA Inattentive
IA-I Inattentive-Inpulsive
ITI Inter-Trial Interval
LA Low Attentive
LC Locus Coeruleus
LFP Local Field Potential
LH Limited Hold
LSD Lest Squared Difference
M Moderate Performers
MANCOVA Multivariate Analysis of Covariance
mPFC Medial Prefrontal Cortex
MPH Methylphenidate
MPtA Medial Parietal Associative Cortex
NAc Nucleus Accumbens
NAcC Nucleus Accumbens Core
nAChR Nicotinic Acetylcholine Receptor
NAcS Nucleus Accumbens Shell
NAT Noradrenaline Transporter
NICE National Institute of Clinical Excellence
OFC Orbital Frontal Cortex
PBS Phosphate Buffered Saline
8
PBS-T Phosphate Buffered Saline with Triton
pFA Probability of False Alarms
PFC Prefrontal Cortex
pHR Probability Hit Rate
PPC Posterior Parietal Cortex
PR Premature Responses
PrL Prelimbic Cortex
RDoC Research Domain Criteria
RI Responsivity Index
SD Stimulus Duration
SI Sensitivity Index
vITI Variable Inter-Trial Interval
vSD Variable Stimulus Duration
9
Thesis Abstract
The recent high profile Phase III failures of new drugs to treat psychiatric disorders demonstrated that there is a clear need to improve translation of pre-clinical findings. The Research Domain Criteria (RDoC) proposes that this can be achieved by focusing on trans-diagnostic criteria. A good example of this is inattention, which is prevalent in conditions such as schizophrenia, Alzheimer‟s disease and attention deficit hyperactivity disorder (ADHD). Validating and refining novel and existing models of inattention will facilitate development of new therapeutic strategies. The work presented in thesis refines a rat model of low attention to increase its translational potential for investigating attention promoting compounds. RDoC also suggests that research at multiple levels of analysis is essential. Therefore, this work also investigates involvement of the frontoparietal network in attention using electrophysiology and cFos techniques ex vivo. This thesis first refines a 5 choice-continuous performance task (5C-CPT) method to group rats based on attentional performance by using a more translational calculation of vigilance (d′) alongside stricter separation criteria. This method demonstrated pharmacological validity by replicating previous effects of methylphenidate and atomoxetine in low attentive (LA) rats. It also showed enhanced translational potential by revealing reduced omissions (increased sustained attention) in LA rats, which is an important clinical symptom that the previous method failed to uncover. Nicotine and caffeine, which improve sustained attention in humans, were assessed in high and low attentive animals. Both compounds increased impulsivity (probability of false alarms (pFA)), irrespective of attentive grouping. However, these impulsive symptoms may have been caused by off-target effects. This was controlled for by testing the selective α7 nicotinic acetylcholine receptor partial agonist encenicline. The rationale for this was that knockout of the α7 nicotinic acetylcholine receptor promotes an inattentive phenotype in mice; therefore, encenicline was studied for its potential to improve attention. Encenicline improved accuracy, d′ and pFA, in LA rats, showing the potential of this target to improve attention, vigilance and impulsivity, respectively. Therefore, it is proposed that partial agonists at the α7 nicotinic acetylcholine receptor warrant further investigation for inattention symptoms. Connections between the prefrontal cortex (PFC) and the posterior parietal cortex (PPC) form the frontoparietal network, activated in humans during tasks that require attention. Subsequent studies in this thesis investigated the involvement of this network in 5C-CPT performance by using the immediate early gene c-Fos, a marker for neuronal activation. The number of c-Fos-positive neurones in PFC and PPC positively correlated with task engagement/vigilance (d′). This finding further supports involvement of these regions in sustained attention, as well as the utility of the 5C-CPT as a highly translatable task. Electroencephalography in patients has shown that the attention-promoting dopamine re-uptake inhibitor methylphenidate (Ritalin), increases alpha oscillations in frontoparietal regions as well as increasing beta in frontal regions, which correlates with improved performance in attentive tasks. In the final study, local changes in low-frequency electrical oscillations of the PFC and PPC of the rat following an acute attention-promoting dose (1 mg/kg) of methylphenidate under urethane anaesthesia were analysed. Methylphenidate promoted alpha oscillations (8-12 Hz) in the PFC and high gamma oscillations (55-90 Hz) in the PPC and reduced delta (1-4 Hz), theta (4-8 Hz) and high gamma in the PFC. Alpha is linked to the suppression of irrelevant and selection of salient stimuli, which may be a mechanism by which methylphenidate promotes an attentive state. Clinical studies have positively correlated frontal theta to ADHD symptom severity and it is thought to be linked to underutilisation of attention. Therefore, reductions of theta in this experiment may also be associated with methylphenidate‟s ability to promote attention in the 5C-CPT. In conclusion, these studies support the use of a refined 5C-CPT as a translational task of relevance to conditions with inattention as a symptom, such as ADHD, schizophrenia and Alzheimer‟s disease. This thesis offers guidance for future work into how high and low attention in rodents interacts with pharmacological agents and the effects of encenicline show the potential of selective nicotinic α7 agonists to improve attention. The presented work also shows that methylphenidate improves inattention and promotes alpha band oscillations in the PFC, which is a biomarker in need of further study. Additional research is needed to link these oscillations in anaesthetized animals to specific behaviours, but the work outlined here provides important progress towards understanding the interaction between inattention, pharmacology and the frontoparietal network.
10
Declaration
No portion of the work referred to in the thesis has been submitted in support of an application for
another degree or qualification of this or any other university or other institute of learning
Copyright
i. The author of this thesis (including any appendices and/or schedules to this
thesis) owns certain copyright or related rights in it (the “Copyright”) and s/he has
given The University of Manchester certain rights to use such Copyright,
including for administrative purposes.
ii. Copies of this thesis, either in full or in extracts and whether in hard or
electronic copy, may be made only in accordance with the Copyright, Designs and
Patents Act 1988 (as amended) and regulations issued under it or, where
appropriate, in accordance with licensing agreements which the University has
from time to time.
iii. The ownership of certain Copyright, patents, designs, trademarks and other
intellectual property (the “Intellectual Property”) and any reproductions of
copyright works in the thesis, for example graphs and tables (“Reproductions”),
which may be described in this thesis, may not be owned by the author and may be
owned by third parties. Such Intellectual Property and Reproductions cannot and
must not be made available for use without the prior written permission of the
owner(s) of the relevant Intellectual Property and/or Reproductions.
iv. Further information on the conditions under which disclosure, publication and
commercialisation of this thesis, the Copyright and any Intellectual Property and/or
Reproductions described in it may take place is available in the University IP Policy
(see http://documents.manchester.ac.uk/DocuInfo.aspx?DocID=24420), in any
relevant Thesis restriction declarations deposited in the University Library, The
University Library‟s regulations (see
http://www.library.manchester.ac.uk/about/regulations/ ) and in The University‟s policy
on Presentation of Theses
11
Acknowledgements
Firstly I would like to thank my supervisors Joanna C. Neill, John Gigg and Michael Harte their
support and guidence has been invaluable throughout my PhD.
I would also like to thank members of Jo Neill, John Gigg and Michael Harte‟s labs including: Lisa
Adamson, Anneka Tomlinson, Chloe Piercy, Nazanin Doostdar, Ben Grayson, Sam Marsh,
Marianne Leger, Katie Murry, Giovanni Podda, Lisa Heany, William Watremez, Joshua Jackson,
Joanna Oladipo, Daniel Squirrel, Sander Tanni, Maria Constantinou and Ines Das Neves for their
support and assistance. This work would not have been possible without them and the support of
the biological services facility (BSF) and the University of Manchester more broadly. Additionally, I
would like to thank Rayna Samuels of Prof. Hugh Piggins‟ lab for her advice during the C-Fos
project and Dr Jaleel Miyan for his advice and support while writing my thesis.
More personally I thank my friends for their emotional support throughout the entire process,
particularly Oktawia Borecka who has been a tireless proof reader and emotional support
throughout.
I would not have gotten to this stage without the lifelong support of my parents, sister and wider
family. Thank you all for your help and support.
12
1 General Introduction
1.1 Context of Thesis
The failure of projects after many years in development has cost pharmaceutical companies
millions and moved many drug companies‟ priorities for drug discovery away from psychiatric
research (Insel and Sahakian, 2012, Nutt and Goodwin, 2011, Neill and Hendrie, 2012). This is
particularly problematic for psychiatric conditions as they are difficult to treat, leading to patients‟
needs not being fully met. For these reasons, the National Institute of Mental Health has produced
the Research Domain Criteria (RDoC) to identify transdiagnostic domains of mental health (Young
et al., 2016). Through a multilevel analysis of cells, circuitry and genes this allows research to
move away from grouping by a collection of symptoms and instead enables analysis of the
common mechanisms involved. Attention is one such domain outlined by RDoC, where the study of
attention deficits can be applied to multiple conditions such as schizophrenia (Luck and Gold, 2008,
Orzack and Kornetsky, 1966, Phillips et al., 2015), attention deficit hyperactivity disorder (ADHD)
(Bernfort et al., 2008, Epstein et al., 2003, Kooij et al., 2010), anorexia nervosa (Dalmaso et al.,
2015) and neurodegenerative disorders (Sharma et al., 2016) including Alzheimer‟s disease (Foldi
et al., 2002, Perry et al., 2000, Vasquez et al., 2011). This thesis, therefore, focuses on developing
a translatable model of inattention in rodents and the application of this to pharmacological studies.
The work then progresses to studying the functional neurobiology of attention and applies this to
study the effects of putative pro-attentive compounds.
1.2 Attention
To survive, organisms have developed many ways to detect changes in the environment. Due to
limited neural resources, this information cannot all enter conscious awareness. Attention is the
filter used to prioritise information adaptively (Petersen and Posner, 2012). How this filter works is a
fundamental issue to understand, firstly due to its role in shaping our perception of the world and
secondly due to the involvement of inattention in many neuropsychiatric conditions such as
schizophrenia, Alzheimer‟s disease and ADHD.
1.2.1 History of attention
Early research into visual attention by Broadbent (1958) proposed a dual-mechanism hypothesis in
humans with two stages; spatial and feature-based processing (Mazer, 2011). Spatial processing
extracts low-level detail such as colour, luminescence and orientation and is fairly effortless.
However, to extract high-level details such as facial features, a more focused approach is needed.
Feature-based processing allows small details to be picked out from a cluttered environment.
Feature-based attention was later developed by Treisman and Gelade (1980) to the “attention
spotlight theory”. This theory proposed a restricted zone of heightened visual processing which
could move independently of eye movement (Mazer, 2011). The spotlight could also be moved by
13
conscious (top-down) control to build an awareness of an environment. In contrast, external
warning signals can cause a bottom-up control system to widen the spotlight.
One important distinction between the models of Broadbent (1958) and Treisman and Gelade
(1980) is the point at which information is filtered. Broadbent (1958) suggested that, as perception
is limited, low-level details are essential to filter at the early stages in the process and suppression
of stimuli is achieved by not perceiving them initially (Lachter et al., 2004). However, the Treisman
and Gelade (1980) model argues that perception is not the “bottleneck” and that large amounts of
data is perceived, but does not enter conscious awareness due to higher level cognitive filters. By
cognitive filtering they allow for a change in the size of the attention spotlight, which happens in
periods of heightened alertness. A wealth of studies now show that it is, in fact, a combination of
the two models in which attention can be filtered early or late depending on the stimulus (Murphy et
al., 2016). Stimuli that are either complex or similar to distractors are said to have a high perceptual
load. In low perceptual load, moving of the spotlight is sufficient, but if a high perceptual load is
encountered then cognitive filters have to be used to alter the size of the spotlight by cognitively
filtering out the irrelevant information.
Petersen and Posner (2012) bring together these ideas into a three-part process. Firstly, an
alerting signal controls arousal of attention to a state prepared to detect an upcoming signal.
Alerting is particularly evident when a warning signal is used and increases the reaction time and
accuracy of the participant and can be thought of as a temporary increase in the limit of perceptual
load. Secondly, orienting is the process of using internal (top-down) and external (bottom-up)
information to orient attention to the required stimulus. Shifts of the spotlight can occur either with
(overt) or without (covert) corresponding eye movement. Finally, the executive stage is where the
attended information can be cognitively filtered before entering conscious awareness. Executive
networks are also the site of top-down control and conflict monitoring for the other two stages.
While the three stages are presented as separate, there is evidence that interaction between them
is essential for goal-directed behaviours (Xuan et al., 2016).
1.2.2 How attention is used
The widely used terms of selective, sustained and divided attention, as well as vigilance, refer to
the pattern of use of the attentional spotlight. Focusing the brain‟s resources to a single stimulus
can be advantageous to goal-directed behaviour (Robbins, 2002). Selective attention allows us to
ignore task-irrelevant and focus on task-relevant stimuli. An example of this is managing to ignore a
noisy surrounding and focus on and listen to an individual‟s voice in a crowd (the cocktail party
effect; Treisman and Gelade (1980)). Selectivity is achieved by a dual mechanism of enhancing the
representation of relevant stimuli and inhibiting irrelevant stimuli (Chelazzi et al., 2013, Shioiri et al.,
2016, Klimesch, 2012).
The environment also often requires prolonged periods of attention. This is accounted for by two
forms of attention adaptation (Egeland et al., 2009). Firstly, sustained attention allows a focus to be
maintained for long periods, for example when reading a book (Carter et al., 2013). The second
14
form is vigilance, which refers to the ability to remain alert toward incoming unpredictable stimuli
(Egeland et al., 2009). This state of arousal can be induced by an initial unexpected stimulus,
termed a warning signal (Petersen and Posner, 2012). An example could be, if a deer sees
movement in the bushes, it would enter a vigilant state to be prepared for further sensory input that
would warn of danger from a predator. Vigilance is also used to search for rare but important
events such as in the classic studies of Mackworth (1948) assessing the performance of World
War II radar operators.
Finally divided attention, also commonly called multi-tasking, is the ability to pay attention to
multiple stimuli at once (Finley et al., 2014). An example of this is, being able to walk and hold a
conversation. However, as attention is limited, this involves rapidly reorienting attention between
the stimuli and severely decreases performance compared to selective attention. A prominent
example of this is using a mobile phone while driving, which significantly increases reaction times
and chances of accident even if used in a hands-free mode (Horrey and Wickens, 2006).
1.2.3 Neurobiology of attention
The locus coeruleus (LC) is a key region in the alerting of attention (Howells et al., 2012, Aston-
Jones and Cohen, 2005, Aston-Jones et al., 1997). Neurones of this region usually fire tonically at
a rate of around 1-3 Hz. When a warning signal is detected, they enter a phasic state of firing which
can promote a prolonged fixed focus (sustained attention). This is possibly due to the broad
noradrenergic innervation to much of the neocortex, particularly the prefrontal (PFC) and posterior
parietal (PPC) cortices (Aston-Jones and Cohen, 2005). It is by balancing tonic and phasic states
that the locus coeruleus is thought to actively control attention by adjusting the synaptic gain across
the cortex (Howells et al., 2012). This optimisation can be modelled as an inverted U-shaped
relationship between locus coeruleus activity and behavioural performance (Aston-Jones and
Cohen, 2005). There is evidence that the error monitoring processes of the anterior cingulate
cortex can modulate LC neuronal firing mode allowing continuous adaptive optimisation of arousal
(Petersen and Posner, 2012, Aston-Jones and Cohen, 2005, Cohen et al., 2000).
Stimuli can be split into two categories, top down/internal cues (e.g. memory) and bottom
up/external cues (e.g, sudden movement). Attention is oriented differently for these two categories.
Corbetta and Shulman (2002) used functional magnetic resonance imaging (fMRI) to show that
there are two orienting systems (Petersen and Posner, 2012). The ventral system includes the
ventral frontal cortex and the temporoparietal junction and is activated by external cues. For
memory related internal cues, Corbetta and Shulman (2002) found the intraparietal sulcus, superior
parietal lobe and the frontal eye fields were activated and termed this the dorsal attention network.
There are two current theories about the anatomy of the executive control in attention (Petersen
and Posner, 2012). The first suggests that the anterior cingulate cortex monitors conflict and the
lateral frontal areas resolve conflict when found (Botvinick et al., 2001). The alternate theory argues
15
that there are two top-down networks controlling executive function (Dosenbach et al., 2007,
Dosenbach et al., 2008). Top down networks include the frontoparietal network made up of the
dorsolateral prefrontal cortex and posterior parietal cortex. This network shows activation in start-
cue and error-related activity and may initiate and adapt control on a trial-by-trial basis. The second
is the cingulo-opercular, including the dorsal anterior cingulate/medial superior frontal cortex,
anterior insula/frontal operculum, and anterior prefrontal cortex. The network seems to show
activation in stabilising and maintenance of focus.
Figure 1.1: Brain regions involved in (a) alerting, (b) orienting and (c) executive networks. (a) Projections of the locus coeruleus in the macaque. Dense projections are sent to the frontal and parietal cortices. (b) The intraparietal sulcus/superior parietal lobe (IPS/SPL) and frontal eye fields (FEF) comprise the dorsal orienting network. The ventral orienting system consists of regions in the temporoparietal junction (TPJ) and the ventral frontal cortex (VFC). (c) In yellow are regions of the frontoparietal network: dorsolateral and anterior prefrontal cortex (dlPFC and aPFC), the dorsal frontal cortex (dFC), the IPS and the intraparietal lobe (IPL). In black are the regions of the cingulo-opercular system: dorsal anterior cingulate cortex (dACC), thalamus and anterior PFC (aPFC). Figure is taken from Petersen and Posner (2012)
1.2.4 Neurotransmitters of attention
The alerting signal derives from the locus coeruleus, which is the main course for noradrenergic
innervation in the brain (Aston-Jones and Cohen, 2005). In an awake state LC neurones fire action
potentials tonically, but to promote sustained attention they switch to phasic firing. Phasic firing
releases more noradrenaline throughout the cortex than tonic firing, particularly the prefrontal and
parietal cortices. It has been proposed that noradrenaline release increases the signal to noise
ratio of the attention system allowing rapid orienting, then sustained attention (Aston-Jones and
Cohen, 2005, Sara and Bouret, 2012). There is also a relationship between task performance and
levels of tonic activity in the LC (Aston-Jones and Cohen, 2005). Low levels of tonic activity
promote inattention, increasing tonic activity improves performance in attentive tasks by promoting
task engagement to an optimal point, beyond the optimal point increases of tonic activity promote a
distractible state that reduces performance. It is also essential to have a balance between phasic
and tonic firing for optimal performance (Howells et al., 2012).
Dopamine neurotransmission has a well-established role in reward and control of motivation
(Bromberg-Martin et al., 2010). In this role, dopamine release enhances synaptic plasticity to
reinforce firing patterns that result in positive outcomes (e.g. food). However, it has more recently
been suggested that this role in synaptic plasticity can also be used in response to non-rewarding
16
and aversive stimuli (Ventura et al., 2007, Bromberg-Martin et al., 2010). This would suggest that
rather than being a reward signal, dopamine signals motivational salience. In agreement with this,
different populations of dopamine neurones have been found to respond to positive or both positive
and negative events (Brischoux et al., 2009, Bromberg-Martin et al., 2010, Matsumoto and
Hikosaka, 2009). These can be considered as value-coding dopamine neurones and salience-
coding dopamine neurones, respectively. These signals for motivational salience are involved in
orienting attention, particularly if rewards are expected (Anderson et al., 2016).
Increases of serotonin (5-Hydroxytryptamine; 5-HT) have been linked to the reduction of vigilance.
Selective serotonin reuptake inhibitors are commonly used for the treatment of depression, but
have also been shown to reduce vigilance in humans performing the Mackworth clock task. An
functional magnetic resonance imaging (fMRI) study found that these compounds act on the frontal
cortex, which is a key area for attention. The exact mechanism by which serotonin modulates
attention is not clear, but it may be through interaction with the 5-HT2 receptors (e.g., 5-HT2a),
which causes indirect changes in dopamine neurotransmission (Wingen et al., 2007).
Acetylcholine is also important in attention; this is shown by lesions of cholinergic input from the
basal forebrain to the prefrontal cortex promoting distractibility and reducing performance under
high attentional demand in rodents (Dalley et al., 2004b, Newman and McGaughy, 2008). This is
further reinforced by detection of changes in acetylcholine levels during tasks of attention (Parikh et
al., 2007). As these changes can be detected broadly and transiently, acetylcholine neurones may
also fire either tonically and phasically, but further research is needed.
Local injection of GABAA agonists, antagonists and inverse agonists into the PFC impairs accuracy
and increases omissions in rats performing the 5 choice-serial reaction time task (5C-SRTT)
(Murphy et al., 2012, Paine et al., 2011, Pehrson et al., 2013, Pezze et al., 2014). This shows that
Gamma aminobutyric acid (GABA) neurotransmission is involved in attention and that, like
dopamine and noradrenaline, it may operate as an inverted U-shaped relationship with attention.
The mechanism of GABA neurotransmission in attention is not yet fully understood but may be due
to dysregulation of the dopaminergic system, (Pezze et al., 2014).
1.2.5 Pharmacology of attention
In ADHD, inattention is a core symptom and, therefore, the treatments for this condition are largely
aimed at improving attention. Methylphenidate and amphetamine salts are first and third line
treatments in the National Institute of Clinical Excellence (NICE) guidelines, respectively (Palanivel
et al., 2009). There is a large body of evidence that these stimulants are very effective in children
with ADHD with an improvement of inattentive symptoms in around 75% of patients (Faraone and
Antshel, 2008, Leucht et al., 2012, Pietrzak et al., 2006, Schachter et al., 2001).
Amphetamine salts such as d-amphetamine (dextroamphetamine) and lisdexamphetamine are
efficacious but are used less due to adverse effects and abuse liability (Heal et al., 2009, Heal et
al., 2013). The main side effects encountered include loss of appetite, insomnia, and abdominal
17
pain (Punja et al., 2016). D-amphetamine inhibits dopamine and noradrenaline transporters, but
also increases the release of these neurotransmitters and inhibits monoamine oxidase, which
degrades them. Methylphenidate selectively inhibits the dopamine and noradrenaline reuptake
transporters to increases the length of time that these neurotransmitters are present at synapses
after release (Heal et al., 2009). Through stimulating dopamine and noradrenaline
neurotransmission, stimulants can promote resting state connectivity of attention-related networks,
such as the frontoparietal network and connections from the visual cortex to the frontoparietal and
executive networks (Mueller et al., 2014, An et al., 2013). Stimulants also increase blood flow to the
frontoparietal network when performing attention and working memory-based tasks (Tomasi et al.,
2011).
The second-line treatment for ADHD is the non-stimulant, atomoxetine, which alleviates impulsive
measures in the continuous performance, stop signal and delay discounting tasks in humans (Barry
et al., 2009, Chamberlain et al., 2006). However, for attentive measures, response rates and effect
sizes are lower than for stimulants at around 64% of ADHD patients (Heal et al., 2009, Cunill et al.,
2013). Atomoxetine is a selective inhibitor of the noradrenaline reuptake transporter. As the main
source of noradrenaline in the brain, the LC is an important site of action for atomoxetine, as well
as the previously mentioned compounds. Via action at this site, noradrenaline can promote both
spontaneous and sensory-evoked discharges of the LC neurones in the rat, thus, increasing phasic
firing and reducing tonic firing (Bari and Aston-Jones, 2013). This causes arousal of attention and
an increase in selective attention seen in many studies using this compound (Tomlinson et al.,
2014, Blondeau and Dellu-Hagedorn, 2007, Cunill et al., 2013, Heal et al., 2009). However, this is
not the entire picture, as atomoxetine administered locally to the PFC still promotes noradrenaline
and dopamine release, suggesting that it directly and indirectly alters neurotransmission in the PFC
(Bymaster et al., 2002).
The α2 adrenoceptor is abundant in the LC and the PFC and blockade of this receptor results in
increased distractibility and prevents the improvement of attention by atomoxetine (Arnsten et al.,
2007). Guanfacine and clonidine are α2 adrenoceptor agonists that can improve inattentive and
impulsive symptoms in ADHD (Bolea-Alamanac et al., 2014, Huss et al., 2016). Less commonly
used treatments include modafinil, desipramine and bupropion (Minzenberg, 2012, Palanivel et al.,
2009). These drugs are not recommended by NICE due to a poor understanding of how they
improve attention.
In schizophrenia, the majority of treatments focus on the positive symptoms with the cognitive
symptoms being poorly treated to date. However, the atypical antipsychotic clozapine is also able
to improve cognitive symptoms including inattention. The improvement in attention is due to the
increase in the ability to orient attention (Spagna et al., 2015). Clozapine has a complex
pharmacology with binding affinity to dopamine (D1, D2 and D4), serotonin (5-HT2a and 2C),
noradrenaline (α1) and cholinergic (M1 and M5) receptors (Baviera et al., 2008). It is therefore not
clear whether modulation of single or multiple transmission systems/ receptors mediate this
improvement in the orienting of attention (Spagna et al., 2015).
18
Donepezil, galantamine and rivastigmine are inhibitors of cholinesterase. As cholinesterase is an
enzyme that degrades acetylcholine, its inhibition increases acetylcholine availability (Sharma et
al., 2016). Cholinesterase inhibitors are regularly used for Alzheimer's disease to reduce symptoms
of memory loss and inattention (Lee et al., 2015). It is effective at reducing cognitive decline across
the spectrum of Alzheimer‟s disease progression, but has a limited duration of effectiveness (Lee et
al., 2015).
1.2.6 Neural Oscillations
As attention is a rapid and transient process, it is challenging to study with some imaging methods
such as fMRI/ positron emission tomography (PET) due to their low temporal resolution. As the
brain communicates using instantaneous changes in electrical potential, detection of changes in
electrical activity is the most direct way to measure changes in brain activity. Also, as electricity
propagates at the speed of light, detection of these changes offers insight into patterns and timing
of neuronal firing. The net potential difference of an area of neuronal firing is a local field potential,
which can propagate through the brain, skull and scalp. In electroencephalography (EEG) an array
of scalp-fixed electrodes can record these changes. An event-related potential is an electrical
signal detected by scalp electrodes, which corresponds to an event within a task. However, due to
the various tissues the signal has travelled through it is not possible to locate its source with high
certainty, which is called the „inverse problem‟. There are methods that attempt to overcome this
issue using computationally intense approaches (Castano-Candamil et al., 2015, Grech et al.,
2008). However, these methods often have limitations in spatial or temporal dimensions or may be
insensitive or distorted by non-stationary signal components (Castano-Candamil et al., 2015). The
latter is a particular issue as non-stationary signals are fundamental to neurophysiological
recordings (Woolrich et al., 2013). The computations also rely on estimating the number of
variables such as the number of current dipoles, but this is impossible to know, and the estimations
cannot be validated fully (Grech et al., 2008). Therefore, to study the function of a region,
electrodes need to be inserted into it to record the local field potential (LFP).
As described so far, regions involved in the process of attention are dispersed across the brain.
This raises the question of how they can be dynamically entrained. Due to the rapid timescale of
attention, anatomical changes would be impossible. However, the theory of communication through
coherence suggests that, by synchronising firing of large groups of neurones to form neural
oscillations, two distant regions can transfer information selectively (Fries, 2015). Due to this
observation a large research effort has tried to link oscillations at particular frequencies to a range
of cognitive processes. For attention, the processes of selection of task-relevant stimuli and
suppression of irrelevant stimuli are essential. For inhibition, alpha has been an area of strong
interest (~8-15 Hz) (Jensen and Mazaheri, 2010). Alpha was originally thought to be generated by
the thalamus, but more recent studies have found isolated brain slice preparations containing
cortical pyramidal neurones capable of generating alpha independently of the thalamus (Bollimunta
et al., 2008, Haegens et al., 2015). Regardless of its origin, evidence shows that the presence of
alpha inhibits the firing of neurones (Jensen and Mazaheri, 2010), possibly via the recruitment of
19
local inhibitory neurones (Lorincz et al., 2009). Through inhibitory mechanisms, well-timed alpha
oscillations in the PFC facilitate suppression and selection of stimuli (Klimesch, 2012, Zanto et al.,
2011). Alpha oscillation timing can also be recorded on electrodes over the parietal and occipital
cortices preceding a stimulus, with event-related synchronisation to the alpha band, which
desynchronises upon stimulus presentation to enable shifting of attention (Schack et al., 2005,
Broussard and Givens, 2010, Klimesch, 2012). In agreement with this, (a) more than 60% of inter-
subject variability in perceptual learning can be explained by the relative engagement of alpha
rhythm (Freyer et al., 2013) and (b) children and adults with ADHD show reduced event-related
alpha synchronisation compared to controls (Hasler et al., 2016, Klimesch, 2012). However, as
attention is a complex behaviour, it is unlikely that it is controlled by a single oscillation. Therefore,
the interaction between different frequencies needs to be analysed. Gamma oscillations are one
such consideration (30-90 Hz). This is because alpha power increases during anticipation of a
stimulus, upon stimulus presentation gamma oscillations increase in power as alpha reduces (Fries
et al., 2001, Muller et al., 2000). The increase in gamma power is predictive of task performance
(Siegel et al., 2008). Magnetoencephalography in primate subjects has proven a useful tool in
elucidating the nature of this interaction. A study by Siegel et al. (2008) showed firstly that visual
attention modulates the oscillations found between frontal eye fields, intraparietal sulcus and visual
cortex and secondly that the intraparietal sulcus is the main source of alpha pre-stimulus and
gamma post-stimulus.
1.3 Animal Modelling
The work in this thesis focuses on the modelling of low attention in rodents, and so the following
review paper contextualises the need for animal models, some approaches that have been taken
and how the area can progress in the future.
20
Low attentive and high impulsive rats: a translational animal model of ADHD and disorders of attention and impulse control
Authors: Andrew Hayward, Anneka Tomlinson and Joanna C Neill
[Published in Pharmacology and Therapetics:
HAYWARD, A., TOMLINSON, A. & NEILL, J. C. 2016. Low attentive and high impulsive rats: A
translational animal model of ADHD and disorders of attention and impulse control. Pharmacol
Ther, 158, 41-51.]
21
1.3.1 Abstract
Many human conditions such as attention deficit hyperactivity disorder (ADHD), schizophrenia and
drug abuse are characterised by deficits in attention and impulse control. Carefully validated animal
models are required to enhance our understanding of the pathophysiology of these disorders,
enabling the development of improved pharmacotherapy. Recent models have attempted to
recreate the psychopathology of these conditions using chemical lesions or genetic manipulations.
In a diverse population, where the aetiology is not fully understood and is multifactorial, these
methods are restricted in their ability to identify novel targets for drug discovery. Two tasks of visual
attention and impulsive action typically used in rodents and based on the human continuous
performance task (CPT) include, the well-established 5 choice serial reaction time task (5C-SRTT)
and the more recently validated, 5 choice continuous performance task (5C-CPT) which provides
enhanced translational value. We suggest that separating animals by behavioural performance into
high and low attention and impulsivity cohorts using established parameters in these tasks offers a
model with enhanced translational value. In this review, methods to separate animals are
compared and the results discussed to highlight advantages over more constrained models, in
addition to potential future directions for enhanced validation. Advantages include reliability,
flexibility and enhanced translation to clinical conditions, all important considerations in modelling
ADHD, schizophrenia and drug abuse, conditions with multifactorial aetiology. Based on the
existing evidence, we suggest that future studies should incorporate an element of behavioural
separation when studying the constructs of visual attention and impulsive action of relevance to
human disorders.
22
1.3.2 Introduction
The focus of this review is attention deficit hyperactivity disorder (ADHD) an illness clearly
characterised by abnormal hyperactivity in addition to deficits in attention and impulsivity, although
many of the constructs we discuss are relevant for other disorders such as drug abuse and
schizophrenia.
Psychiatric disorders remain poorly managed and present a large economic burden. Brain
disorders cost €141 billion per annum in the UK, with a total 2010 cost (in million € purchasing
power parity) of €134 (Fineberg et al., 2013). ADHD is often confounded by other conditions such
as generalised anxiety disorder, oppositional defiant disorder and learning difficulties (Larsson et
al., 2011). The estimated direct and indirect costs per child or adolescent patient are estimated to
be €9,860 to €14,483 per year (Le et al., 2014). Of this, between 8-25% are estimated to be direct
healthcare costs, showing that a large proportion of the economic cost is due to indirect costs such
as education and social services. When the condition persists into adulthood, symptoms begin to
include poor occupational performance (Kuriyan et al., 2013) and higher risk taking behaviour
(Flory et al., 2006, Groen et al., 2013) resulting in even poorer financial and societal outcomes
(Doshi et al., 2012). Factors such as these contribute to a reduced quality of life for ADHD patients,
which often persist through middle age into old age if left untreated (Brod et al., 2012).
ADHD affects 5-8% of school aged children and can persist to adulthood in 50-80% of cases
(Faraone et al., 2000, Barkley et al., 1990, Fayyad et al., 2007, Bari et al., 2008). It is recognised
as a heterogeneous disorder characterised by three main symptoms (inattention, hyperactivity and
impulsivity), which forms the basis of the current psychiatric classification system. However a
recent review describes the many other different methods of characterising ADHD, including; types
and traits, age, gender, treatment, duration and variations in neuroanatomical structure and
function (Heidbreder, 2015). ADHD is most commonly sub-classified as; 1. Predominantly
inattentive (ADHD-I), 2. Predominantly hyperactive/impulsive (ADHD-HI) 3. Combined type
presentation (ADHD-C). In a large meta-analysis investigating the validity of the subtypes and
symptom dimensions, the authors concluded that separation into subtypes is a useful, convenient
way to describe the functional and behavioural aspects of ADHD, however care must be taken
when defining the subtypes as discrete distinct forms of the disorder (Willcutt et al., 2012).
There is a large evidence-base confirming the heterogeneity between subtypes in ADHD. This
evidence suggests the existence of neurocognitive, gender, neural correlate and behavioural
differences (different phenotypes) between subtypes. The gender differences reported show that
women are more likely than men to be diagnosed with the inattentive subtype of ADHD (Biederman
et al., 2002) and men are more likely to be diagnosed with the combined subtype (Ramtekkar et al.,
2010b). Interestingly, the male to female ratio differs between children and adults. In children,
ratio estimates differ depending on the sample; in community surveys male to female ratio is higher
(as high as 10:1) than in a clinical sample (as low as 3:1). However in adults, this ratio is reduced
(1:1 to 2:1) (Williamson and Johnston, 2015a). The importance of recognising gender differences
in research is highlighted in a recent comprehensive review, which recommends adequate
23
representation of both genders in clinical samples (Williamson and Johnston, 2015a). It is vital that
both male and female subjects are included in future studies, including in animal studies to
represent the true clinical picture of ADHD and to more carefully assess the contribution of gender
to the pathophysiology of the disease.
The ADHD-200 Consortium investigating heterogeneity in ADHD have shown a number of specific
neural correlates within the subtypes of ADHD (ADHD-200-Consortium, 2012). The ADHD-I
subtype was found to have variations in specific brain regions including mainly dorsolateral
prefrontal and cerebellar regions, known to be associated with control systems. ADHD-C subtype
patients were noted to have atypical connectivity in midline default network components and the
insular cortex. These results support the hypothesis that differential connectivity disturbances
underpin each ADHD subtype (Fair et al., 2013). It has also been shown that an abnormality in the
motor circuit represents the major difference between ADHD-C and ADHD-I subtypes (Lei et al.,
2014).
Existing pharmacotherapy is not designed for the diversity of symptom presentations, and therefore
lacks efficacy and also has a large side effect burden. Critical to the development of improved
therapy is improved understanding of the aetiology and neurobiology of these disorders, achieved
through carefully validated animal models, which are currently lacking.
Stimulant medication such as methylphenidate and amphetamine salts are currently first and third
line treatment in the National Institute of Clinical Excellence (NICE) guidelines respectively
(Palanivel et al., 2009). They control attentive symptoms in approximately 75% of patients, but do
not adequately control impulsive symptoms and, at high doses and in certain patient groups, may
even exacerbate them (Winstanley, 2011, Faraone and Antshel, 2008). They are also
contraindicated for patients with a history of drug abuse due to their abuse liability. The second line
treatment is the non-stimulant, atomoxetine which alleviates impulsive measures in the continuous
performance, stop signal, and delay discounting tasks in humans (Barry et al., 2009, Chamberlain
et al., 2006), however, for attentive measures response rates and effect sizes are lower than for
stimulants (Heal et al., 2009, Cunill et al., 2013). A recent meta-analysis of atomoxetine treatment
concluded that the risk-benefit ratio for atomoxetine is even lower in an adult ADHD population as it
has only minimal effect on clinically meaningful endpoints (e.g. job performance or sociability)
(Cunill et al., 2013). Other non-stimulant treatments, guanfacine and clonidine are only
recommended in North America for ADHD (Bolea-Alamanac et al., 2014). Their omission from
NICE and other European guidelines is likely due to the need for clearer evidence of their efficacy
across patient groups, particularly in adulthood. With adult ADHD being increasingly recognised to
have a significant socioeconomic burden, it is important that we find new pharmacological
treatments with efficacy, safety and tolerability in the adult ADHD population, or enhance the use of
available medication by delivering the most appropriate treatment to each subtype of patient (Kooij
et al., 2010). Less commonly used treatments include modafinil, desipramine, and buproprion
(Minzenberg, 2012, Palanivel et al., 2009). These drugs are not recommended by NICE as their
mechanism of action is not yet fully understood. Taking all this into account, it is important to
24
conduct further research using animal models in order to define patient populations with symptom
profiles who may benefit most from the different types of medication.
1.3.2.1 Value of Animal Models and the Need for Improved Translation
Animal models offer the ability to control and manipulate experimental parameters more precisely
and rapidly than in human subjects (Markou et al., 2009, Moore, 2010). This level of manipulation
supports hypothesis driven research, with the overall aim of improving understanding of
neuropathology, and developing better treatments for clinical conditions. The high attrition rate of
drug development programmes and the withdrawal of the pharmaceutical industry from psychiatric
drug discovery has been attributed to poor translation from the animal models into the clinic (Neill
and Hendrie, 2012, Nutt and Goodwin, 2011, Insel and Sahakian, 2012). However, a more likely
explanation is that not enough time and resources have been provided to improve validity of the
animal models and appropriateness of the tests employed, a situation currently being addressed
(eg see Neill et al., 2014, Stuart et al., 2013, and Tomlinson et al., 2014 for ADHD). Progress in
this area is essential to enable much needed progression in psychiatric drug development. This
section of our review aims to summarise recent efforts to enhance translation from rodents to the
clinic in ADHD, using the separation of rodents into high and low performing groups based on
attentive and impulsive measures. The tests used for this include the standard 5- choice serial
reaction time task (5C-SRTT) and a task with enhanced translation, the 5- choice continuous
performance task (5C-CPT).
There are three factors suggested by Willner (1986) to consider when reviewing an animal model:
face, construct and predictive validity. Firstly, face validity requires the model to produce the same
phenotype as the human condition. In the case of ADHD, deficits in measures designed to translate
to human attention and impulsivity are key to producing the model. Furthermore, models designed
to align with clinical subtypes further increase this translational power. Secondly, construct validity
is required to mimic aetiology and neurobiology of the human condition in the animal. ADHD as an
example is considered highly heritable, with estimates of 80% or greater heritability (Biederman et
al., 1990, Kieling et al., 2008, Langner et al., 2013). Multiple genetic factors have been linked to
this heritability such as a dopamine D4 receptor (DRD4) seven transmembrane repeat (Woolley et
al., 2008, Faraone et al., 2001, Li et al., 2006, Gornick et al., 2007) and the dopamine transporter
type 1 (Lim et al., 2006). This suggests that several genetic factors combine to produce the
observed phenotype, or that other factors may play a role including environmental factors and
epigenetics (Maher, 2008). Due to the multifactorial and diverse nature of the aetiology of ADHD
and many other neuropsychiatric conditions, aetiological genetic modelling would be a restrictive
method preventing exploration of alternative mechanisms (Bari and Robbins, 2011, Nestler and
Hyman, 2010). One alternative is lesion modelling, but this suffers the same issue of restriction of
mechanism, and lesion or manipulation of brain regions with neurotoxins only mimics a small group
of ADHD patients who encounter toxins such as heavy metals. Ultimately these methods
investigate the effect of the manipulation rather than attempting to replicate the core symptoms of
ADHD (Jupp et al., 2013).
25
One theory of ADHD, and indeed other disorders, is that it represents the extreme of a continuum
present within the general population (Bari and Robbins, 2011). As rats tested in the 5C-SRTT and
5C-CPT show a continuum of performance, a method of modelling with stronger validity and the
potential for translation is behavioural separation based upon differences in performance, as it
avoids assumption of mechanisms (Puumala et al., 1996, Dalley et al., 2007, Tomlinson et al.,
2014). A variety of separation methods have been used to produce differing models based on the
condition under investigation or hypothesis studied. Taking into account the important differences
between the subtypes identified by behavioural separation, optimum animal models need to
characterize specific symptoms (i.e. impulsivity) or model the individual subtypes of ADHD. We
review the main methods used to date and assess their translational value in the following sections.
1.3.2.2 Measuring Attention and Impulsive Action in Rodents
In humans, the CPT is the most widely utilised task used to measure sustained attention (errors of
omission and accuracy of response), impulsive action (errors of commission) and vigilance (d
prime-d′), the ability to remain aware of changing situations in order to respond correctly (Epstein et
al., 2003, Riccio et al., 2002). All of these are readily measured in animals. The most widely used
test is the rodent 5C-SRTT (Carli et al., 1983, Robbins, 2002, Bari et al., 2008). This test has
recently been further validated by reverse-translation to humans, as the four choice serial reaction
time task (Voon et al., 2014). For full details of the methods for 5C-SRTT see Bari et al. (2008), in
brief a rodent, usually a rat, is placed into a standard five-hole operant chamber (Fig 1). During the
task a light stimulus is lit briefly in the back of a randomly selected aperture. The animal‟s ability to
detect and respond, usually via a nose poke response, to the correct aperture is recorded as a
measure of accuracy, which is considered an analogue of selective attention (Robbins, 2002). A
second measure provided by this task is premature responding during the waiting period, before
the next stimulus presentation. This is considered a measure of impulsive action and will be termed
waiting impulsivity for the purposes of this review. This task has been widely adopted as a
preclinical task of attention and waiting impulsivity (Robbins, 2002, Winstanley, 2011).
A variation of this task is the 5 choice continuous performance task (5C-CPT) first described in
mice (Young et al., 2009), and recently validated in our laboratory for rats (Barnes et al., 2012a,
Barnes et al., 2012b). It has also been reverse translated to humans, demonstrating translational
validity (McKenna et al., 2013). In the CPT, the same response to a single light stimulus is required
during most trials, but infrequent non-target trials add an additional challenge. In non-target trials,
the pre-potent signal of all five lights requires no response by the animal in order to gain a food
reward (i.e. the animal is rewarded for withholding its impulse to respond). Responding during this
trial, rather than withholding responding, is termed a false alarm, and interpreted as response
disinhibition (Fig 1). Response disinhibition closely corresponds with human errors of commission
in the CPT (Riccio et al., 2002). Response disinhibition is a form of impulsive action, but is
genetically and pharmacologically distinct from waiting impulsivity as measured by premature
responses in the 5C-SRTT and 5C-CPT (Young et al., 2011). When DRD4 expression is reduced,
response disinhibition increases without affecting the premature responses. Pharmacologically,
SD242084 (a 5-HT2C receptor antagonist) increases premature responses without affecting false
26
alarms. The addition of non-target trials also enables the assessment of vigilance (measured as
sensitivity index-SI) by robust statistical analyses using signal detection theory in the same way as
in the human CPT (discussed in detail in Young et al. (2009). The 5C-CPT in rodents offers
enhanced translation to humans because of its close resemblance to the human CPT. Due to this
enhanced translational value and lengthy training procedure, it has been suggested that the 5C-
CPT is best used at later stages of the drug development process and that the 5C-SRTT is more
suited to earlier stages, due to its higher throughput (Lustig et al., 2013).
Figure 1.2: The 5 Choice Serial Reaction Time Task, left panel (5C-SRTT) only uses trials where a reward is earned by responding in the hole in which a light stimulus appears, either during the stimulus duration, or immediately afterwards, in the remaining limited hold period. The 5 Choice Continuous performance task (5C-CPT) uses these target trials and adds infrequent non target trials where all five lights are lit and no response is required to earn a food reward, no go trials are shown in the right panel. Adapted with permission from Tomlinson et al. (2015).
1.3.3 Methods of Separation
1.3.3.1 Models of High Impulsive Action
The most commonly used method was first described by Dalley et al. (2007) and has been used in
several studies since, demonstrating its reproducibility. This method involves separating rats based
on premature responses in the 5C-SRTT, as an analogue of impulsive action. Dalley and
colleagues use adult male Lister Hooded rats in this model. In order to separate the animals, they
were trained to standard parameters of the task (as above) using a 5 seconds (s) fixed inter-trial
interval (ITI), they were then cycled through 3 weeks of daily testing with two days at a 5 s fixed ITI,
one day at a 7 s fixed ITI and then two days at a 5 s fixed ITI again (Fig 2). The use of two ITIs
demonstrates the rats‟ performance at baseline (5 s) and when challenged (7 s) (Dalley et al.,
2007). HI animals were originally shown to have twice as many premature responses as LI animals
at 7 s ITI (Dalley et al., 2007). They have also been characterised as having 50% premature
responses across the three weeks at 7 s ITI with Low Impulsive (LI) animals having less than 30%
premature responses (Caprioli et al., 2013, Caprioli et al., 2014) or as being in the top 15th
27
percentile for impulsivity (Isherwood et al., 2015). This HI model was originally used to model
aspects of drug taking behaviour, such as abstinence and relapse in addition to investigation of the
neuroanatomical basis of impulsive behaviour, more specifically, the role of prefrontal cortex and
ventral striatum circuitry in impulsive behaviour (Jupp et al., 2013). This method results in
approximately 7% of the animals being classified as high impulsive, which is then a stable trait
when retested over time (Dalley et al., 2007).
Another method of separation for impulsivity is described by Diergaarde et al. (2008) who
separated a group of 32 male Wistar rats based on premature responding by upper and lower
quartiles using five stable training sessions. This is a shorter baseline period than that used by
Dalley et al. (2007). Training sessions used a standard 5 s fixed ITI but a 1 s stimulus duration
(SD), which is longer than the standard SD (Bari et al., 2008). The primary effect of altered SD is
on accuracy and, as the limited hold is 1 s in both cases, there would be an expected minimal
effect on premature responses. This method produced 8 HI and 8 LI animals, removing 16
moderately performing animals from the comparison. This model has been used to study nicotine
dependence and non-drug related addiction by sucrose self-administration (discussed later). The
advantage of this method is that it offers a simple means of separation with a substantial difference
in performance between the subgroups. While use of cohort characteristics such as upper and
lower quartile makes comparison within a cohort robust, it also means that whole cohort
performance is an important factor when comparing between studies.
Winstanley et al. (2010) used a similar method, 41 male Long-Evans rats were trained to a stable
five-day baseline in the 5C-SRTT. They were then separated according to the median premature
responses. Using the median reduces the effect of extreme values and allows re-establishment of
a baseline between experiments. However, it also produces an unequal HI:LI ratio. The authors
state that this method was not designed to rigorously separate animals into extreme groups but to
offer a simple separation method and also has the ethical benefit of utilising all members of a
Figure 1.3: Baseline separation of HI-filled circles/LI-empty
circle animals using the Dalley et al. (2007) method. This
establishes a baseline of 5 s ITI and then challenges the
animals with 7 s long ITI. Under this challenge, HI animals can
be separated. The 5 day schedule is repeated for three
consecutive weeks for reliability. Reproduced with permission
from Dalley et al. (2007).
28
cohort (Winstanley et al., 2010). This model has been used to study obsessive compulsive disorder
and bipolar disorder.
Tomlinson et al. (2014) recently proposed a model to account for impulsive symptoms by using the
parameters of premature responses (waiting impulsivity) and probability of false alarms (pFA;
response inhibition) offering a unique modelling method using the highly translatable 5C-CPT in
female Lister Hooded rats. For this model, the parameters were premature responses above or
below 10 and probability of false alarms above or below 0.5 for high and low impulsive groups
respectively. As response inhibition is a critical factor of ADHD impulsive and combined subtype
presentation in people (Epstein et al., 2001) this model is designed for enhanced translation to the
clinical CPT (Young et al., 2009, Lustig et al., 2013, Tomlinson et al., 2014, Tomlinson et al., 2015).
1.3.3.2 Models of Low Attention
To separate animals for attentive performance, Granon et al. (2000) used the measure of accuracy
and selected animals performing above or below 75% accuracy to produce high and low baseline
performance groups respectively. Male Lister Hooded rats were trained to 5 s SD as in the
standard 5C-SRTT protocol (Bari et al., 2008). Once rats had been trained, stability of performance
was measured over a five day baseline. Animals in which performance deviated <5% from the
mean were progressed to a reduced SD (not lower than 0.25 s). Before testing, animals performed
stably between 0.5-0.25 s SD, this varied between animals so is a confounding factor for analysis,
but high attentive (HA) and low attentive (LA) groups were well balanced in SD. Ensuring stability is
important for accurate interpretation of the results, but testing the animals based on different SDs
alters the nature of the separation as the ability to respond at shorter SDs is seen as an attentive
trait which directly affects accuracy (Robbins, 2002, Young et al., 2013).
Paterson et al. (2011) used male Long Evans rats to produce a sub-optimally performing group.
This, as with the previous model, was based on less than 75% accuracy across a 5 day baseline,
but this study only looked at the low performing group. To control for lack of trials completed they
also stipulated that at least 50 trials had to be completed in each of five consecutive sessions. They
state that 25% of the whole group of rats met these criteria making it similar to a lower quartile
separation.
Grottick & Higgins (2000) produced a model to test the hypothesis that nicotine has a pro-attentive
effect that cannot be measured in normal animals using the 5C-SRTT. They chose male Lister
Hooded rats with < 80% accuracy and > 20% omissions. They are the only group to separate using
omissions as a measure of ability to sustain attention. Although it is debated whether omissions
represent sustained attention or motivation in this task (Robbins, 2002), this demonstrates the
flexibility to separate by a particular parameter of interest.
The 5C-CPT has also been used recently to model the attentive deficits in ADHD, which are known
to persist into adulthood more consistently than impulsive symptoms (Ingram et al., 1999, Wilens et
al., 2004, Kooij et al., 2010). Tomlinson et al. (2014) produced a model of the inattentive subtype
29
of ADHD using the 5C-CPT by separating female Lister Hooded rats using parameters of above or
below 90% accuracy with a SI of above or below 0.3, for high attentive and low attentive groups
respectively across 5 baseline training sessions. The SI adds an important factor to the model by
accounting for vigilance, which has been shown to be a significant factor in all ADHD symptom
subtypes (Collings, 2003). This therefore produces a model closer to the clinical situation, with the
aim of improved assessment of compounds with differential efficacy in certain ADHD impulsive
subtype patients.
1.3.3.3 Models Most Relevant to ADHD Symptomatology
The earliest example of behavioural separation using the 5C-SRTT is by Puumala et al. (1996).
Their method translates well to combined subtype of ADHD. They produced the model by
separating male Han:Wistar rats into an ADHD and a control group. To do this they used
percentage correct responses and percentage premature responses. An important feature of their
method is that they used two parameters to define the model, correctly recognising that ADHD
commonly presents with both inattention and impulsivity, particularly in combined subtype patients
(Epstein et al., 2003).
In an attempt to represent the spectrum of ADHD symptoms, Blondeau and Dellu-Hagedorn (2007)
used a cluster analysis method to isolate an array of phenotypes (Fig 3). This method produced
the efficient performing (E) animals (high percentage correct, and low percentage premature), the
moderate performing (M) group (with average percentage correct and percentage premature), the
inattentive (IA) group (low percentage correct, but average percentage premature) and finally the
inattentive and impulsive (IA-I) group (low percentage correct and high percentage premature).
These groups are similar to those found in a clinical and non-clinical population when people were
tested with a self-report questionnaire based on disease criteria (Marsh and Williams, 2004).
Tomlinson et al. (2015) also produced a model of adult ADHD combined subtype, this model used
accuracy to separate by attentiveness, SI to separate by vigilance and pFA to separate by
impulsive action. In ADHD-C type patients these three symptoms combine to produce the most
persistent and difficult to treat subgroup (Molina et al., 2009, Kooij et al., 2010). Therefore an
animal model of this subtype is particularly valuable to enhance our understanding of the
neurobiological basis of this subtype and so improve treatment strategies.
30
Figure 1.4: a dimensional analysis
method of performance separation.
This method is particularly interesting
as it produces multiple groups which
correspond to clinical ADHD
subgroups. Reproduced with
permission, from Blondeau and Dellu-
Hagedorn (2007).
31
Table 1.1: Summary of published methods for modelling impulsivity, attention and combined symptom subtypes using performance based separation techniques, as discussed above.
Abbreviations: 5 choice-serial reaction time task (5C-SRTT), 5 choice continuous performance task (5C-CPT), high/ low impulsive (HI/LI), high/ low performance (HP/LP),
attention deficit hyperactivity disorder (ADHD), ADHD inattentive subtype (ADHD-I), ADHD combined subtype (ADHD-C), inattentive (IA), IA impulsive (IA-I), moderate
performers (M), efficient performers (E), Lister Hooded (LH), premature response (PR), probability of false alarms (p[FA]), accuracy (Acc), sensitivity index (SI).
Developed by Test Model Rat sex & strain Parameter Rule Additional
Notes
Imp
uls
ivit
y
Dalley et al. (2007)
5C
-SR
TT
HI Male LH PR >2 at 7 s ITI Tested over 3
weeks
Diergaarde et al. (2008) HI/ LI Male Wistar PR Upper/ Lower
Quartiles
Winstanley et al. (2010) HI/ LI Male Long-Evans PR Above/ Below
Median
Tomlinson et al. (2014)
5C
-
CP
T
HI/ LI Female LH PR >/ < 10
Mean of 5 tests p[FA] >/ < 0.5
Att
en
tio
n
Granon et al. (2000)
5C
-SR
TT
HP/ LP Male LH Acc >/ < 75%
accuracy
When stable
over 5 sessions
Paterson et al. (2011) LP Male Long-Evans Acc
< 75%
accuracy and >
50 trials
Similar to
Granon et al.
(2000), but only
used LP group
Grottick and Higgins
(2000) LP Male LH
Acc < 75% Stable for 2
weeks. Omissions > 20%
32
Tomlinson et al. (2014)
5C
-CP
T
ADHD-I/ Control Female LH
Acc >/ < 90%
Mean of 5 tests
SI >/ < 0.3
Att
en
tio
n &
Im
pu
lsiv
ity
Puumala et al. (1996)
5C
-SR
TT
ADHD/ Control Male Han:Wistar Acc < 60%/ > 75%
PR > 40%/ < 30%
Blondeau and Dellu-
Hagedorn (2007)
IA-I/ IA/ M/ E Male Sprague-
Dawley
Acc
Cluster analysis
Control and
three subtypes
of ADHD PR
Tomlinson et al. (2015)
5C
-CP
T
ADHD-C/
Control Female LH
Acc </ > 90% Attention,
Vigilance and
Impulsive action
SI </ > 0.3
p[FA] >/ < 0.5
33
1.3.4 Neurobiology Identified Using Separation Methods
1.3.4.1 Using Impulsive Models
1.3.4.1.1 Behaviour
Dalley et al. (2007) produced the most widely used behavioural separation model using the 5C-
SRTT. It was proposed that differences in neurobiology between the animals in each group could
drive escalated cocaine self-administration. Drug naive HI animals showed reduced binding
potential for [18
F]fallypride using positron emission tomography imaging, suggesting that HI animals
had a reduced number of D2 and/or D3 receptors in the ventral striatum, showing an
endophenotype that may contribute to the behavioural differences. High impulsivity has been linked
to reduced D2/D3 availability in the striatum of ADHD patients, which adds construct validity to the
model (Ghahremani et al., 2012, Buckholtz et al., 2010). This difference was also observed
following cocaine self-administration, which was enhanced in this group and correlated with
reduced dopamine release in the nucleus accumbens core (NacC) of HI animals. Diergaarde et al.
(2008) used their model to show that high impulsive action predicted greater overall nicotine self-
administration and longer maintenance compared with animals with lower impulsive action. They
also showed that HI animals show escalated sucrose self-administration (Diergaarde et al., 2009).
This shows that high impulsive animals are more responsive to reinforcement which is not limited
to drugs of abuse. Dalley et al. (2007) showed that the HI phenotype is linked to increased cocaine
self-administration. One caveat of previous findings is that in humans it was not possible to
determine whether this high impulsivity is due to the drug taking itself or a predefined trait that led
to escalating drug taking behaviour (Perry and Carroll, 2008). As impulsivity is screened before
drug administration these findings support the latter theory. Belin et al. (2008) examined how HI
animals differed in attempts to stop compulsive drug taking using the negative stimulus of a foot
shock. They found that HI animals took more negative stimuli to stop drug taking behaviour
showing that impulsivity is also linked to compulsive behaviour.
The delay discounting task challenges animals to decide how long they are willing to wait for a
larger reward compared with a smaller and more immediate reward (Winstanley et al., 2003). This
shows how impulsive the animals are, but is distinct from waiting impulsivity already discussed.
This type of reward sensitivity is called impulsive choice, and is the ability to delay gratification for a
better long term solution when faced with an immediate payoff. When HI animals in the Dalley et al.
(2007) study were tested in the delay discounting task they showed a steeper discounting curve,
showing that they are less willing to wait for a larger delayed reward as the delay increased
(Robinson et al., 2009). However these rats showed no deficit in the stop signal reaction time task.
This measures the speed at which an action can be cancelled. This highlights the separable nature
of the different forms of impulsivity, which has previously been demonstrated in lesion studies
(Cardinal et al., 2001, Christakou et al., 2004, Eagle and Robbins, 2003). Separation by
performance using the delay discounting task has also been studied and is discussed in Jupp et al.
(2013); here we focus on impulsive action in 5C-SRTT and 5C-CPT. Robinson et al. (2009) also
34
demonstrated the ability of atomoxetine to reduce premature responses (waiting impulsivity) and
flatten the delay discounting curve (impulsive choice).
1.3.4.2 Pharmacology/ Neurobiology
High waiting impulsivity in the Dalley et al. (2007) HI model can be modulated by dopaminergic and
noradrenergic pharmacological agents such as quinpirole and sumanirole (both D2/3 receptor
agonists), which both reduce premature responses (Fernando et al., 2012). In that study,
atomoxetine (a NA reuptake inhibitor) and guanfacine (an α2 adrenoceptor agonist) dose
dependently reduced premature responses in HI animals. GBR-12909 (a DA reuptake inhibitor)
increased impulsivity in HI and LI animals, whereas the DA and NA reuptake inhibitor
methylphenidate had no effect on impulsive responses. These results show that both NA and DA
systems are involved in impulsive behaviour and shows that activation of D2/3 or α2 adrenoceptors
are involved in impulse control. The modulation of impulsivity by these pharmacological agents is
very similar to effects seen in humans with impulse control disorders and ADHD demonstrating
predictive validity of the model (Cunill et al., 2013, Wood et al., 2014). Glutamate is the most
abundant excitatory neurotransmitter in the brain and shows involvement in a number of conditions
which express high levels of impulsivity (Isherwood et al., 2015). Modulation of metabotropic
glutamate receptors is therefore a target for investigation. Recent work examined the effect of
metabotropic glutamate receptor 5 negative allosteric modulators RO4917523 and 3-[(2-methyl-
1,3-thiazol-4-yl)ethynyl]pyridine (MTEP) and the positive allosteric modulator ADX47273 on trait
impulsivity (Isherwood et al., 2015). The authors demonstrated that RO4917523 and MTEP
reduced premature responses but at the same time increased omissions and response latencies,
suggesting a non-specific motor effect rather than a true reduction in waiting impulsivity. ADX47273
reduced premature responses in the HI group, where the effect was larger when animals were
challenged with a 7 s rather than 5 s ITI. While in this group, omissions were not affected, correct
latency was significantly increased at the same doses that reduced premature responses. A
locomotor activity assessment was simultaneously conducted. The only dose of drug not to show
any effect on locomotor activity was ADX47273 at 80 mg/kg, which also had a highly significant
effect on correct latency. Due to these locomotor effects of the metabotropic glutamate receptor 5
modulators, it is challenging to separate effects on impulsivity from general effects on motor
behaviour. Future work could investigate region specific infusion and compounds with affinity for
other metabotropic receptors to analyse the role of glutamate in impulsivity without the confounding
motor effects.
Based on the initial [18
F]fallypride PET study, the NAc, as part of the ventral striatum was found to
be a key target, in particular, the D2/3 system (Dalley et al., 2007). The left NAc was found to have
reduced grey matter density in HI animals (Caprioli et al., 2014). Infusion of aripiprazole, a D2/3
receptor partial agonist and atypical antipsychotic, and nafadotride, a D2/3 receptor antagonist, into
the NAcC and nucleus accumbens shell (NAcS) separately showed a functional opposition within
the NAc (Besson et al., 2010). For HI animals, nafadotride increased premature responses when
infused into the NAcS, but reduced them when infused into the NAcC while aripiprazole had no
35
effect when directly infused into either region. This contrasting effect may explain why systemic
administration of nafadotride had no effect on premature responses. This has been further
investigated using quinpirole, a D2/3 receptor full agonist, which was found to increase premature
responses when infused into the NAcC, in HI animals only but increased locomotor activity in both
HI and LI animals (Moreno et al., 2013). In contrast, infusion of quinpirole into the NAcS promoted
an increase in locomotor activity only in HI and not in LI animals. Nafadotride did not attenuate the
increase in impulsivity caused by quinpirole infused into the NAcC, however it did attenuate the
locomotor stimulant effect of NAcS infusion in HI animals. The inhibitory neurotransmitter γ amino
butyric acid (GABA) has recently been found to play a role in impulsivity by modulation in the
NAcC. This was shown by a reduction of the enzymes needed to produce GABA, glutamate
decarboxylase GAD 65/67, as well as reduced markers of dendritic spines and microtubules in HI
compared to LI animals which was further strengthened by LI animals becoming more impulsive
when GAD65/67 gene expression was reduced by infusion of antisense GAD67 and GAD65
oligonucleotides into the NAcC (Caprioli et al., 2014). This reveals a novel mechanism of
impulsivity in rats and warrants further investigation.
Patients with damage to the orbital frontal cortex (OFC) have high scores on impulsiveness
questionnaires (Berlin et al., 2005) and so Winstanley et al. (2010) investigated whether this region
could be mediating behaviour in HI animals. OFC infusion of the dopamine D1 receptor antagonist,
SCH23390 decreased impulsive responding in HI animals, but not in controls. When infused into
the OFC, quinpirole only affected non-impulsive parameters, while a D1 receptor agonist, SKF-
81297 had no effect. This shows that D1 receptors in the OFC do not fully mediate effects on
impulsivity suggesting that brain regions outside the OFC are important, and that D2 receptors in
the OFC are not involved in impulsivity. This study again demonstrated that baseline performance
of the individual animal alters the effect of dopaminergic compounds.
Electrical signalling of clusters of neurons has also been linked to high and low impulsive behaviour
(Donnelly et al., 2014). Silicone multielectrode arrays were implanted into the NAcC or NAcS and
the infralimbic and prelimbic regions within the prefrontal cortex of HI and LI male Lister Hooded
rats and low frequency (<200 Hz) activity recorded (Donnelly et al., 2014). By correlating electrical
changes in the low frequency oscillations or local field potential with operant responses made in the
5C-SRTT, a new level of understanding was achieved. The authors discovered firstly that, across
the NAc and prefrontal cortex, gamma (50-60 Hz) and theta (7-9 Hz) frequency oscillations
increased during visual search in anticipation of stimulus presentation and again following a correct
response. Secondly they found that gamma and delta activity coupling strengthened following error
responses. Thirdly they found that these differed significantly when a premature response was
about to be made. Finally they found that in HI animals error responses were stronger, and
responses linked to correct responses were weaker. This elegant work shows that both the medial
prefrontal cortex and the NAc are important regions for attention when scanning for a target and
are linked to impulsive responses by recognition of error response. This series of experiments
shows that, not only can behavioural separation offer a platform to test and compare future
36
treatments, but enable research into new mechanisms which cause behavioural deficits. Other
models such as dopaminergic lesion would not have allowed insight into the GABAergic or
electrical activity within discrete brain areas which are becoming of increasing relevance in the
search for alternative treatment mechanisms (Hayes et al., 2014, Jupp et al., 2013, Donnelly et al.,
2014).
1.3.4.3 Using Models of Attention
Granon et al. (2000) used the parameters of above or below 75% accuracy to analyse the effect of
three dopaminergic compounds via bilateral microinfusion into the medial prefrontal cortex. They
found that, while sulpiride (a D2 receptor antagonist) had no effect on task performance, SKF-
38393 (a D1 receptor agonist) enhanced accuracy selectively in the low performing animals and
SCH 23390 (a D1 receptor antagonist) had the opposite effect to reduce accuracy selectively in the
high performing animals. Although SKF-38393 had no effect here in unseparated animals, it may
be due to the task used, as it has an effect on vigilance in 5C-CPT (Barnes et al., 2012a). The
selectivity of effect suggests that there could be differences in dopamine system engagement in
high and low performing animals, where low performing animals have reduced dopamine activity in
the prefrontal cortex making them susceptible to receptor activation and high performing animals
have increased levels of dopamine and are thus less susceptible to receptor activation, but more
susceptible to antagonism.
Work by Paterson et al. (2011) expanded this by also analysing effects of d-amphetamine,
methylphenidate and atomoxetine in poor performing animals. They used similar methods of
separation to Granon et al. (2000), but only studied poor performing animals. They also compared
the effect of these clinically efficacious compounds to tolcapone (a Catechol O Methyl Transferase
(COMT) inhibitor). They found that a systemically administered acute dose of d-amphetamine
increased premature responding in the poor performing rats. Atomoxetine reduced the number of
premature responses. Methylphenidate reduced percentage omissions at low doses (0.1 and 0.5
mg/kg). One issue with this study, however, is the use of only the poor performing animals and not
both poor and high performing animals which would allow a more robust comparison. In ADHD
(discussed later) an oppositional effect is seen where ADHD-like animals show improved
performance and good performers show reduced performance in response to methylphenidate.
Using low performing animals, Grottick and Higgins (2000) found that sub-chronic dosing of
nicotine and SIB 1765F, an α4β2 nicotinic acetylcholine receptor agonist, improved accuracy and
speed of response, whereas AR-R 17779, an α7 nicotinic receptor agonist, had no effect. Nicotine
and SIB 1765F both increased premature responding during sub-chronic treatment, while nicotine
also reduced omissions. This shows that nicotine and SIB 1765F improve selective attention,
speed of response processing but increase waiting impulsivity, suggesting that the effects of
nicotine are mediated via the α4β2 nicotinic acetylcholine receptor and not the α7 receptor. The
effect of nicotine on omissions may be the result of an effect on motor activity (Grottick et al., 2003,
Young et al., 2013). It has since been found in unseparated mice and in mice with a scopolamine
37
(muscarinic acetylcholine receptor agonist) induced vigilance deficit that nicotine has a primary
effect on vigilance (Young et al., 2013). This would make it an interesting pharmacological agent to
test using animals separated in the 5C-CPT due to the ability to robustly calculate vigilance in that
task.
1.3.4.4 Using Models of ADHD
Lower right frontal cortex DOPAC/DA ratio has been observed in animals with lower accuracy,
whereas in the left frontal cortex serotonin levels were higher in animals with low accuracy
(Puumala and Sirvio, 1998). Furthermore, serotonin utilization in the right frontal cortex was higher
in highly impulsive animals. This suggests that pharmacological agents developed to improve poor
performance should increase dopaminergic neurotransmission and reduce serotonergic
neurotransmission in the frontal cortex. Methylphenidate (MPH) increases DA and NA by blocking
the dopamine transporter (DAT) and noradrenaline transporter (NAT) thus reducing reuptake of DA
and NA leading to increased activation of D1 and α2 receptors respectively. This has a positive
effect on choice accuracy in low attentive animals (Puumala et al., 1996, Blondeau and Dellu-
Hagedorn, 2007). Doses of 100 and 1000 µg/kg MPH were used in the 5C-SRTT, the lower dose
also reduced premature responses in low performing animals. However it has also been found that
MPH can increase premature responding at high doses of 1 mg/kg (Blondeau and Dellu-Hagedorn,
2007). The differential effect of methylphenidate in high and low attentive animals is interesting as it
can be understood using the inverted U shaped hypothesis of optimal arousal, where alertness can
only be increased to a point, beyond which further stimulation of the system promotes negative
effects such as distractibility, impulsiveness and anxiety (Wood et al., 2014). A neurobiological
theory of ADHD has been suggested based on this, where prefrontal cortex catecholamine levels
follow this inverted U shaped relationship (Arnsten, 2009, de Jongh et al., 2008). This also shows
that separation of the animals is essential to determine doses of compounds which may selectively
improve the performance of low and high performing animals. Future work should aim to establish
the mechanisms behind the difference in sensitivity to monoaminergic compounds.
Atomoxetine is a second line treatment for ADHD and acts in a similar way to methylphenidate, but
is selective for NAT, with minimal affinity for DAT, thus activating α2 adrenoceptors in the prefrontal
cortex and minimising striatal psychostimulant effects (Bymaster et al., 2002, Heal et al., 2009). In
animals separated by dimensional analysis into efficient (E), moderate (M), Inattentive (IA) and
Inattentive and Impulsive (IA-I) groups, atomoxetine had no effect on accuracy, but reduced
premature responses in all groups except E, at doses of 0.1, 0.5 and 1 mg/kg. This suggests that
atomoxetine requires a certain threshold of premature responding to have a beneficial effect.
1.3.4.5 Using Models developed in 5C-CPT
A selective effect of MPH has also been shown in the 5C-CPT in female Lister Hooded rats.
Accuracy increased in the LA animals (representing the ADHD-I subtype) at 2 mg/kg, while SI
increased only at 0.5 and 1 mg/kg (Tomlinson et al., 2014). Firstly, this suggests a pharmacological
38
separation between selective attention and vigilance, secondly, the lack of effect on vigilance at 2.0
mg/kg may again be explained by the inverted U shaped hypothesis of optimal arousal (Wood et
al., 2014). Furthermore in LI animals, MPH increased premature responses at the highest dose of 2
mg/kg. In animals with high levels of premature responding and pFA (HI), MPH reduced premature
responses at 0.5 and 1 mg/kg. As no effect on pFA was observed at any dose of MPH, this can be
seen as an effect on waiting impulsivity and not response disinhibition.
In the same study, atomoxetine increased accuracy at 2 mg/kg and SI at 1 and 2 mg/kg in LA
animals, but reduced accuracy in HA animals at 1 mg/kg (Tomlinson et al., 2014). This adds to
(Blondeau and Dellu-Hagedorn, 2007) showing no effect up to 1 mg/kg for accuracy in their model
using the 5C-SRTT and for the first time uses a robust calculation of vigilance which is increased
by atomoxetine. Atomoxetine also reduced premature responding in HI animals at all doses and
reduced pFA in HI (1 and 2 mg/kg) and LI (2 mg/kg) animals. This differential effect of atomoxetine
further emphasises the need to separate animals based on their performance in order to clearly
identify effects of pharmacological agents.
Tomlinson et al. (2015) produced a low attentive, low vigilance and high response disinhibition
model of ADHD-C using female Lister Hooded rats in the 5C-CPT. A-412997, a selective DRD4
receptor agonist improved vigilance and reduced pFA at 0.3 and 1 µmol/kg. No effect was seen on
any parameter in the HA animals, suggesting that targeting the DRD4 could be an effective
treatment for the combined subtype of ADHD. This is a particularly important finding as a genetic
mutation of the 7 transmembrane repeat of DRD4 is linked to adult ADHD, (Franke et al., 2012),
and several D4 receptor antagonists have improved behavioural hyperactivity in an ADHD animal
model (Zhang et al., 2002), highlighting the importance of this receptor system in ADHD
pathophysiology. The COMT inhibitor, tolcapone has previously been shown to be ineffective in the
5C-SRTT (Paterson et al., 2011). The previous method of separation was based only on accuracy
and used the 5C-SRTT, where Tomlinson et al. (2015) used tolcapone in the ADHD-C model, in
the 5C-CPT. In that study, accuracy and SI improved in ADHD-C animals (at 10 and 15 mg/kg), but
decreased in HA animals at the highest dose of 15 mg/kg. There are several possible explanations
for this discrepancy; firstly the 5C-CPT compared to the 5C-SRTT is a more difficult task, whereby
the extra difficulty may have challenged the animals enough to see an effect (Lustig et al., 2013).
Secondly, the Tomlinson et al. (2015) method used a particularly rigorous model of ADHD subtype
ADHD-C, this involved the two extra parameters of SI and pFA and a higher level of accuracy,
resulting in a different model. This highlights the potential flexibility of separation by performance to
produce two very different models depending on the parameters used for separation. It is important
that the most appropriate method of separation is used depending on the experimental question
being investigated. This subject is addressed below.
39
Table 1.2: Summary of pharmacological effects in models based on behavioural separation.
Abbreviations: Correct Latency (CL), noradrenaline (NA), INA transporter (NAT), Premature Responses (PR), dopamine transporter (DAT), catechol-O-methyl transferase (COMT) premature responses (PR), Nucleus accumbens (Na), Na core (NaC), Na shell (NaS), metabotropic glutamate receptor 5 (mGluR5), prelimbic cortex (PrL).
Model Drug Mechanism Administration
(Route) Measure Effect (Dose) Notes Reference
Imp
uls
ivit
y
Dalley et al.
(2007)
HI animals
Quinpirole
D2 Agonist Systemic (s.c.)
PR
↓ (0.01, 0.03, 0.1
mg/kg)
(Fernando et
al., 2012)
Sumanirole ↓ (0.1,0.3,1 mg/kg)
Also Increased
CL at 0.3 and 1
mg/kg
Atomoxetine NAT
Inhibition
Systemic (i.p.)
↓ (1,3 mg/kg) Also Increased
CL at 3 mg/kg
Guanfacine
α2 NA
Receptor
Agonist
↓ (0.3,1 mg/kg)
GBR 12909 DAT
Inhibition ↑ (2.5, 5 mg/kg)
Methylphenidate DAT/NAT
Inhibition - -
40
Aripiprazole D2/3 Partial
Agonist
NaC - -
(Besson et al.,
2010)
NaS - -
Nafadotride D2/3
Antagonist
NaC PR ↓ (0.1 µg)
NaS PR ↑ (0.1 µg)
Quinpirole D2/3 Full
Agonist NaC PR ↑ (0.3, 1 µg)
(Moreno et al.,
2013)
RO4917523
mGluR5
Negative
Allosteric
Modulator
Systemic (p.o.) PR ↓ (0.1,0.3 mg/kg)
Increases in
omissions, correct
latency and
locomotor activity
(Isherwood et
al., 2015) MTEP Systemic (i.p.) PR -
Increases in
omissions, CL
and locomotor
activity, reduction
in accuracy
ADX47273 mGluR5
Positive
Allosteric
Systemic (p.o.) PR ↓ (40, 60, 80, 100
mg/kg) Increases CL
41
Modulator
Tomlinson et al.
(2014) HI
animals
Methylphenidate DAT/NAT
Inhibition
Systemic (i.p.)
PR ↓ (0.5, 1 mg/kg)
(Tomlinson et
al., 2014)
Atomoxetine NAT
Inhibition
PR ↓ (0.5, 1, 2 mg/kg)
pFA ↓ (1, 2 mg/kg)
SI ↑ (1, 2 mg/kg)
Att
en
tio
n
Granon et al.
(2000)
SCH 23390 D1
Antagonist
Bilateral PrL
infusion
- -
(Granon et al.,
2000) SKF 38393 D1 Agonist
Accuracy ↑ (0.06 µg/side)
Omissions ↓ (0.06 µg/side)
Sulpiride D2
Antagonist - -
Paterson et al.
(2011) D-Amphetamine
Catecholami
ne release Systemic (i.p.)
PR
↑ (0.56, 1 mg/kg) (Paterson et
al., 2011) Perservativ
e
Responses
42
Methylphenidate DAT/NAT
Inhibition Omissions ↓ (0.1, 0.5 mg/kg)
Atomoxetine NAT
Inhibition PR ↓ (0.5,1,2 mg/kg)
Tolcapone COMT
Inhibition Omissions ↑ (3,30 mg/kg)
Grottick and
Higgins (2000)
Nicotine Nicotinic
Agonist
Subchronic
Systemic (s.c.)
for 20 days
Accuracy ↑ (0.2 mg/kg/day)
(Grottick and
Higgins, 2000)
Correct
Latency ↓ (0.2 mg/kg/day)
PR ↑ (0.2 mg/kg/day)
SIB1765F Nicotinic
α4β2 agonist
Accuracy ↑ (5 mg/kg/day)
Correct
Latency ↓ (5 mg/kg/day)
PR ↑ (5 mg/kg/day)
AR-R 17779 Nicotinic α7
Agonist - -
43
Tomlinson et al.
(2014) HA
animals
Methylphenidate DAT/NAT
Inhibition
Systemic (i.p.)
Accuracy ↑ (2 mg/kg)
(Tomlinson et
al., 2014)
SI ↑ (0.5, 1 mg/kg)
Atomoxetine NAT
Inhibition Accuracy ↑ (2 mg/kg)
Att
en
tio
n &
Im
pu
lsiv
ity
Puumala et al.
(1996) Methylphenidate
DAT/NAT
Inhibition
Systemic (s.c.)
Accuracy ↑ (0.1, 1 mg/kg) (Puumala et
al., 1996)
Blondeau and
Dellu-Hagedorn
(2007)
Methylphenidate DAT/NAT
Inhibition PR ↑ (1 mg/kg)
Largest effect in
IA-I group (Blondeau and
Dellu-
Hagedorn,
2007) Atomoxetine NAT
Inhibition PR ↑ (0.1, 0.5, 1 mg/kg) In all but E group
Tomlinson et al.
(2015)
A-412997 D4 Agonist
Systemic (i.p.)
SI ↑ (0.3, 1 µmol/kg)
(Tomlinson et
al., 2015)
pFA ↓ (0.3, 1 µmol/kg)
Tolcapone COMT
Inhibition
Accuracy ↑ (10, 15 mg/kg)
pFA ↓ (15 mg/kg)
SI ↑ (15 mg/kg)
44
1.3.5 Selection of the most appropriate method of separation
When deciding which model is best, the first question is what is being modelled. As shown
throughout this review modelling by performance is flexible allowing many measures to be used
individually or collectively. This decision also dictates which task to use. For example for a model
concerning deficits in vigilance, as seen in ADHD or schizophrenia, then the 5C-CPT is the most
viable choice. Also models involving impulsive action should strongly consider the important
difference between response disinhibition and waiting impulsivity (discussed above) as these have
been shown to be separate constructs, which can be manipulated independently (discussed in
Young et al. (2011)). However, for models of selective or sustained attention, the 5C-SRTT
measures of accuracy and omissions are well established parameters. The reduced length of
training time in the 5C-SRTT compared to the 5C-CPT can also be a determining factor in choice of
test.
Separation method differences are clear from the studies outlined in this review. To decide
between these factors there are a series of questions to ask. What level of deficit is required?
Winstanley et al. (2010) produced a model with a milder impairment using a simple median split. In
contrast upper and lower quartile separation used by Diergaarde et al. (2008) removes a middle
performing group to produce a model with a larger deficit. A larger deficit can also be produced by
increasing the challenge in the task. Dalley et al. (2007) provide an excellent example of
introducing an additional challenge in the form of longer ITIs to further separate the animals,
producing a larger deficit as shown by their model producing the highest 15th percentile of the
group (Isherwood et al., 2015). When choosing a more rigorous method it is important to consider
the principles of replacement reduction and refinement. A larger starting cohort will be required to
maintain statistical power in the group selected as a model. This contradicts the principle of
reduction of number of animals used, as recommended by the National Council of Reduction,
Refinement and Replacement (3Rs); so the two must be carefully balanced.
Methods of comparison vary between studies; a key factor to consider is within study referencing
compared to between study referencing. A within study reference is a more common approach.
This involves using measures such as mean, median, quartiles or percentiles to use the group‟s
performance to determine which animals are high versus low performers within that group. A main
advantage of this method includes consistent group numbers in high or low groups, making
comparisons more robust. Also it provides the opportunity for translation of this modelling method
between strain, gender or species which may have different baseline levels of performance. If the
groups differ in baseline performance then this too is a confounding factor. To account for this,
whole group baseline performance is important to consider when comparing between studies using
this method, and the separation levels should be clearly stated.
In between study referencing a line is produced from the data (e.g. <75%) and is then used in that
study and for future studies. This method provides more robust comparison between experiments.
However, how these reference lines are determined is often not clearly explained, questioning the
45
validity of this approach. Finally, a very interesting unique method is the dimensional analysis used
by (Blondeau and Dellu-Hagedorn, 2007). This method allows an unbiased approach to selecting
groups with different profiles using multiple measurements. As they have demonstrated, this allows
a range of profiles to be produced within a population. The use of additional measures such as
those in the 5C-CPT would allow new types of models to emerge from this method.
1.3.6 Conclusion
Animal models have several advantages including accurate control of variables and manipulation
of subjects. Recent models have attempted to recreate the symptomatology of disorders
characterised by attention and impulsivity deficits, using chemical lesions or genetic manipulations.
In a diverse population, where the aetiology is unknown, these methods restrict the ability to
identify novel targets for drug discovery. Separating animals based on performance in 5C-SRTT
and 5C-CPT offers a method of modelling without the limitation in assuming a mechanism of
action. Behavioural separation offers the further advantage of enhanced flexibility to investigate
many conditions which are characterised by low levels of attention or high impulsivity such as
ADHD, drug abuse, obsessive compulsive disorder, Parkinson‟s disease and schizophrenia.
Through continued use of these models, mechanisms underlying impulsivity have been identified
such as fronto-striatal interaction and the involvement of GABAergic signalling within the NAc. The
translational validity of these methods has been clearly demonstrated, particularly in relation to the
HI model of Dalley et al. (2007). It is hoped that many of the other models will be used to the same
level in order to determine reliability and improve translation. The series of models recently
produced using the 5C-CPT strive to improve translation of this work into a clinical setting by
mimicking more symptoms and offering specialised models for patient subtypes (Tomlinson et al.
2014; 2015). This aims to move towards a more individualised treatment method where
compounds can be pre-clinically tested for a variety of symptom subtypes. Some progress has
already been made in this area for the standard ADHD treatments, MPH and atomoxetine.
Separation of differential drug effects in animals separated by performance corresponds with the
inverted U shaped hypothesis of optimal arousal. Indeed this hypothesis highlights why behavioural
separation is so important. Pharmacological treatments clearly act differently in animal subjects
with poor attention or impulsivity and so need to be studied in these clinical populations. Finally,
one of the major considerations in this area is determination of the neurobiological basis of these
differences of performance in preclinical models and clinical population.
46
1.4 General Aims of Thesis
i) To further develop a method of modelling attention deficits in rats.
There are a number of strategies when choosing how to divide animals by their 5
choice task performance. Hayward et al. (2016) present the argument that dividing
above and below a within-study reference (i.e. median, or quartiles) offers higher
statistical and translational validity than a set reference point (i.e. 75% accuracy).
Therefore, we aimed to use this reference to refine the method used by Tomlinson
et al. (2014), (2015). An additional refinement was to use the more translational
measure of vigilance d′. To validate these design changes rats were tested in the
5C-CPT using methylphenidate and atomoxetine. The hypothesis of this experiment
was that if this method is valid, then the results of this would be comparable to
Tomlinson et al. (2014) as well as other studies with the same compounds, but with
increased translational validity.
ii) To use the high and low attentive model to study the pharmacology of
attention
It has previously been established that monoaminergic modulators promote
attention in low attentive animals performing the 5C-CPT. However, nicotine and
caffeine are widely used psychoactive stimulants that work via different
mechanisms. They have both been shown to improve sustained attention in the 5C-
SRTT, so we aimed to assess if nicotine and caffeine could also promote attention
or vigilance in low attentive animals in the 5C-CPT. We hypothesised that
improvements in sustained attention would be greater in low attentive animals as
has been seen with monoaminergic compounds.
The nicotinic acetylcholine α7 receptor has been implicated as a target for the
remediation of the cognitive symptoms of schizophrenia. Several compounds have
been developed for this purpose and seem to show improvement in a number of
cognitive parameters, including attention. Encenicline is a partial agonist of the α7
receptor and was in phase II clinical trials. We aimed to establish if pro-cognitive
effects of encenicline included improvement of inattention in low attentive animals.
iii) To examine the involvement of the frontoparietal network in performance of
the 5C-CPT and effects of attention promoting agents
There is a broad agreement of the vital involvement of the frontoparietal network in
humans and non-human primates. However, this requirement has not been shown
specifically for the 5C-CPT. The experimental hypothesis was that performance of
the task would promote activity in both regions and that this activity would be
dependent on task performance. If the frontoparietal network is important for
attention, then compounds that promote attention will promote changes in the
activity of neurones in these regions. To assess this, we compared frontoparietal
network activity in anaesthetized rats in response to injection with an attention
promoting dose of methylphenidate versus vehicle injection. We hypothesised that
47
we would see changes in activity in both PFC and PPC when treated with MPH
compared to vehicle treatment and, further, that the largest effect would be in the
attention-related alpha band.
48
2 General Methods
For Chapters 3, 4 and 5 the 5C-CPT was used to assess the accuracy, vigilance and impulsive
action of female Lister Hooded rats. The 5C-SRTT is a popular task with a large literature
demonstrating its usefulness in assessing attention and impulsivity (Robbins, 2002, Bari et al.,
2008, Dalley et al., 2007) ; however, without a „no go‟ stimulus it is unable to assess vigilance in
a robust manner (Young et al., 2009). The task, therefore, cannot translate its findings towards
the human CPT that it is based on. For this reason, the task is seen as less translatable for the
assessment of attention compared with the 5C-CPT, which does include „no go‟ trials, thereby
allowing calculation of d′ as an estimation of vigilance as used clinically (Lustig et al., 2013,
Hayward et al., 2016). It is for this key advantage of translation that the 5C-CPT is used in this
thesis.
2.1.1 Animals
All animals used in the studies presented in this thesis are adult female Hooded Lister rats
(Charles River, UK; weighing 210 ± 20 g at the start of experiments). Animals were housed in
groups of up to five for the duration of the studies. The environment was maintained at 21 ± 2
oC, 55 ± 5 % humidity. The diet was standard rat chow (Special Diet Services, UK) controlled to
maintain 90% free feeding weight throughout training and testing (10-12 g/rat/day). Home cages
were individually ventilated cages with two levels (GR1800 Double-Decker Cage, Techniplast,
UK) and testing was completed under a standard 12 hour light: dark cycle (lights on at 7:00 am).
Female rats were used as the 5C-CPT has been carefully validated in female rats in our
laboratory (Tomlinson et al., 2014, Tomlinson et al., 2015, Barnes et al., 2012a, Barnes et al.,
2012b). Some claim that to use females the stages of the estrous cycle need to be carefully
monitored and accounted for, however Prendergast et al. (2014) found that across an array of
different methods that female mice were less variable than males even without accounting for
the oestrus cycle. Additionally, they are poorly represented in preclinical research.
2.1.2 Animal Usage
For 5C-CPT experiments a total of 80 animals started training (Table 2.1). Animals used in the
electrophysiology experiment were ordered for those experiments only. Training for 5C-CPT
took approximately 8 months in total; at the time of the encenicline study, the animals were
approximately 12 months old, for nicotine and caffeine they were 14 months old and during the
final c-fos experiment they were approximately 17 months old.
49
Figure 2.1: Animals used for 5C-CPT experiments and the order of use. Between each experiment
complete health assessments were completed by a veterinarian and animals were tested to ensure
performance was maintained. Animals were excluded from tests for poor health or inconsistent
performance.
2.1.3 Apparatus
Training and testing 5C-CPT was conducted in eight standard nine hole chambers (Figure
2.2A; Campden Instruments Ltd, UK) with holes 2, 4, 6, and 8 occluded. Chambers were
controlled and data collected by ABet II Standard software (Lafayette Instrument Company, IL,
USA). Animal actions were detected by infrared photobeams across each apperture and an
infrared beam which was broken by movement of the magazine door to detect food collection
and initiate each trial. Animals were rewarded with 45 mg rodent reward pellets (Test Diet, MO,
USA).
Figure 2.2: 5C-CPT Apparatus for training and pharmacological testing. (A) A rodent in the
chamber being presented with a go trial. (B) Schematic representation of go and no go trials in the
chamber.
Go
TriaNo Go
Trial
A B
50
2.1.4 5C-CPT Task Structure
The task structure is shown in Figure 2.1. The animal initiates a trial by nosepoke into the
magazine tray. A variable inter trial interval (ITI) of 7-12 s follows in which response in any one
of the five is counted as a premature response and a 5s time out follows. If no response is
made the software randomly presents a stimulus in one of the five holes (go trial; 70%
weighting) or to present the light stimulus in all 5 holes (no go trial; 30% weighting). The
stimulus will stay on for 1s in the standard version of the task. The limited hold is a 5s timer from
stimulus presentation during which responses can be made. In go trials, response in the same
hole as the light was presented is rewarded, and a response in any other hole or no response
during the limited hold is punished with a time out. During time out the house light is illuminated
for 5s. In no go trials a response in any hole during the limited hold is punished and no
response is rewarded. After the reward is dispensed or time out concludes, the magazine tray is
lit and a response is needed to initiate the next trial. The task lasts either 120 trials or 30
minutes, whichever comes first.
Figure 2.3: 5C-CPT Task Structure. Ovals represent timing points, rectangles represent task measures
and triangles represent outputs to the subject.
51
2.1.5 5C-CPT Training
Animals were trained in a similar way to previously published (Barnes et al., 2012a, Barnes et
al., 2012b, Tomlinson et al., 2014, Tomlinson et al., 2015). They started with only go trials at a
stimulus duration and limited hold of 60 s, which were progressively shortened to 1 s (Table
2.1). The median duration from beginning of training to final criteria was 164 days in a
representative cohort (Table 2.1). Once the animals had reached criterion, performance was
maintained by once weekly training until and between testing.
52
Table 2.1: Training stages for 5C-CPT. Animals are progressively introduced to the different elements of
the task. Firstly nose poking response to a stimulus (stage 1-3) then no go trials (stage 4-5), then a
variable stimulus duration (stage 6-7) then final task criteria with short stimulus duration and a longer
variable inter-trial interval (stage 8).
Training Stage Stimulus Duration
(s)
Limited Hold
(s)
ITI (s) Criteria Trial Types
1 60 60 2 30 correct Go Only
2 45 45 2 30 correct
3 30 30 2 30 correct
4 10 12 5 <40% FA
>70%
Accuracy
<25%
Omissions
>50 Trials
Go and No
Go
64% Go Trials
5 8 10 5 <40% FA
>70%
Accuracy
<25%
Omissions
>50 Trials
p[HR]>p[FA]
6 5 5 5-7 <40% FA
>70%
Accuracy
<25%
Omissions
>50 Trials
Go and No
Go
70 % Go
Trials
53
7 2 5 5-7 <40% FA
>70%
Accuracy
<25%
Omissions
>50 Trials
8 1 5 7-12 <40% FA
>70%
Accuracy
<25%
Omissions
>50 Trials
0 5 0 1 0 0 1 5 0
0
2
4
6
8
T ra in in g L e n g th (D a y s )
Tra
inin
g S
tag
e
Figure 2.4: Representative progression of training. The median number of days from the start of
training to reaching criteria for each stage for n=40 animals (cohort used in Nicotine/Caffeine, EVP-
6124 and C-Fos Studies).
2.1.1 Signal Detection Theory
There are multiple versions of the clinical CPT available, but they are all unified by the principle
of detecting the difference between a target stimulus and a non-target stimulus (Young et al.,
2009, Riccio et al., 2002). In response to these two events, there can be four possible
54
responses. A target stimulus can be correctly responded to (hit) or not responded to correctly/
not responded to (miss) and non-target stimuli can either be not responded to (correct rejection)
or erroneously responded to (false alarm). Signal detection theory is used to interpret these
different responses and estimate the traits of sensitivity and bias to the stimuli (Marston, 1996,
Young et al., 2009). Sensitivity is how well the subject can determine the difference between
target and non-target stimuli. To have high sensitivity the number of hits needs to be high, as
well as the number of false alarms, need to be low. There are two ways to estimate the
sensitivity. If the hits and false alarms are normally distributed, then d′ can be used (calculation
in Table 2.2). If not then the sensitivity index (SI) must be used:
( ) ( )
When the 5C-CPT was first proposed Young et al. (2009) suggest SI was most suitable to avoid
assumptions of normality. However, in later papers, they began to use d′ as the data was in fact
normally distributed and this is a more translatable measure which is commonly used in the
clinical versions of CPT (Young et al., 2013). To estimate bias responsivity index is used. This
variable is necessary for establishing if the animals have a tendency to respond to all stimuli
(liberal response strategy) or withhold from all stimuli (conservative response strategy) which
the factors d′ and SI do not account for (Young et al., 2009, Young et al., 2013).
d′ is calculated as the z score of pHR minus the z score of pFA, as shown in Table 2.2. The z
scores are calculated as the individual‟s performance minus the average performance of the
cohort (across all treatments) divided by the standard deviation of the cohort.
Table 2.2: Summary of measures in the 5C-CPT, how they are calculated and what they mean.
Correct, Incorrect, Missed, Premature Response, False Alarm and Correct Rejection are as shown
in Figure 2.3.
Measure Calculation Meaning
Go Trials Accuracy
Selective/
Sustained
Attention
Omissions
Sustained
attention (and/or
motivation and/or
sedation)
Hit Rate (p[HR])
Used in signal
detection theory
55
calculations
Premature
Responses (PR)
Number of premature responses Waiting Impulsivity
(Impulsive Action)
Correct Latency Average time from stimulus to correct
response
Processing speed/
Motor Effects
Incorrect Latency Average time from stimulus to
incorrect response
Processing speed/
Motor Effects
Magazine Latency
(Go)
Average time from correct response
to reward collection
Motivation/ Motor
Effects
No Go Trials Probability of
False Alarms
(p[FA])
Response
Disinhibition
(Impulsive Action)
False Alarm
Latency
Average time from stimulus to false
alarm response
Processing speed/
Motor Effects
Magazine Latency
(No Go)
Average time from correct rejection to
reward collection
Motivation/ Motor
Effects
Signal
Detection
Theory
d′ ( ) ( ) Vigilance
Responsivity
Index (RI)
( )
Response Bias/
Strategy
2.1.2 Variable Stimulus Duration
In Chapters 4 and 5 a variable stimulus duration was used along with extended session length
and number of trials. This was done to provide a much greater challenge to the animals when
tested with nicotinic compounds. This has been done for a number of different assessments of
different assessments of nicotinic compound effect . The same stimulus durations (0.75, 1.25
and 2s), number of trials (250) and length of session (60mins) were previously used for testing
nicotinic acetylcholine receptor agonists in mice performing the 5C-CPT (Young et al., 2013).
The increase in attentional load promotes omissions in α7 nicotinic acetylcholine receptor
knock-out mice more than controls suggesting a link between high attentive load and the
nicotinic receptor system. We aimed to examine this further.
56
2.1.3 Division into subgroups
Animals were divided into high and low performing groups based on performance when treated
with the vehicle data of each study. The best performing ~25% of animals for d′ and accuracy
formed the high-attentive (HA) group and the worst performing ~25% of animals formed the low-
attentive (LA) group. This was designed based on an upper/lower quartile split, which
represents a careful balance between having significantly different groups and the number of
animals used (see Hayward et al. (2016)). To achieve this, animals were ordered by d′ and the
highest/ lowest ~25% animals were chosen, providing their accuracy was also above/below the
cohort average. This method is based on Blondeau and Dellu-Hagedorn (2007) dimensional
analysis grouping to produce combinations of symptoms based on the topology of the whole
cohort. The number of subjects used for HA/ LA comparison was based on examples from the
literature where group sizes are typically 8-12 (Dalley et al., 2007, Fernando et al., 2012,
Blondeau and Dellu-Hagedorn, 2007). For testing which involved a variable stimulus duration
the most challenging stimulus duration of 0.75 s was used to group the animals as it represents
the highest attentive load therefore causing the animals to separate most clearly.
57
3 Pharmacological Validation of a Refined
Low Attentive Model using
Methylphenidate and Atomoxetine
Authors: Andrew Hayward, Lisa Adamson, Joanna C. Neill
[Unpublished data]
3.1 Contributions
Andrew Hayward and Joanna C. Neill planned the study. Training of the animals was completed
by Andrew Hayward and Lisa Adamson. The study and data analysis was conducted by Andrew
Hayward. The manuscript was prepared by Andrew Hayward and Joanna C. Neill.
58
3.2 Context
As Chapter 1 has outlined, grouping animals by performance in 5 choice tasks offers many
advantages over other methods of modelling, particularly in low attention as the causes of
clinical inattention are still poorly understood. Low attentive models in the 5C-CPT exist in the
literature, here we repeat and refine one of the models to establish a model with the aim of
higher translation and a clearer deficit.
3.3 Abstract
Inattention is a symptom in a number of conditions including attention deficit hyperactivity
disorder, schizophrenia, bipolar disorder and Alzheimer‟s disease. Grouping animals based on
performance in an attentive task (5 choice tasks) enables a low attentive model where the
mechanism of action is not assumed. We present an improved method of grouping the animals
to produce a larger difference between groups and thus a greater understanding of the
mechanism of action of pharmacological agents that enhance attention. Adult female Lister
Hooded rats were trained to perform the 5 Choice – Continuous Performance Task (5C-CPT)
and tested with methylphenidate 0.5, 1 and 2 mg/kg or vehicle over four weeks. The order of
the treatments was randomised and counterbalanced. The 10 animals with the highest vigilance
(d′) and attention (accuracy) following vehicle treatment were classified as high attentive (HA)
and the 10 with the lowest performance were low attentive (LA). In HA animals,
methylphenidate reduced accuracy and d′ as well as increasing omissions, however, in LA
animals accuracy, and d′ were increased and omissions reduced. The same test was performed
for atomoxetine (0.5, 1 and 2 mg/kg) which also reduced d′ in HA animals, and incresed both
accuracy and d′ in LA animals. Atomoxetine also reduced the number of premature responses
in the LA animals. These results were largely in agreement with the findings of previous studies
of methylphenidate in LA animals separated according to these measures. However, we
additionally observed a reduction in omitted trials in LA animals following methylphenidate, likely
due to our removal of average performing animals resulting in a clearer separation of the two
groups. These findings suggest that future studies may benefit from taking a similar approach
when studying the differences in HA and LA animals, therefore this approach is taken
throughout the work presented in this thesis
59
3.4 Introduction
Attention can be defined as the ability to filter out irrelevant stimuli in order to optimise limited
neurochemical resources (Luck and Gold, 2008). Inattention is a core symptom of attention
deficit/ hyperactivity disorder (ADHD) which affects 5-8% of school-aged children and persists
into adulthood in roughly 50% of cases (Faraone et al., 2000, Barkley et al., 1990, Fayyad et al.,
2007, Bari et al., 2008). ADHD is most commonly separated into three sub-types; 1.
Predominantly inattentive (ADHD-I), 2. Predominantly hyperactive/impulsive (ADHD-HI) 3.
Combined type presentation (ADHD-C) (American Psychiatric Association., 2013). Subtypes in
children are often unstable, and inattentive symptoms increase with age, and impulsive
symptoms reduce, highlighting the importance of understanding inattention in adults (Larsson et
al., 2011). Inattention is also an important symptom in highly disabling conditions such as
schizophrenia, bipolar disorder and Alzheimer‟s disease (Luck and Gold, 2008, Foldi et al.,
2002). Therefore, increased understanding of the neurobiology of attention has an important
and broad application.
Methylphenidate is the first line treatment in the National Institute of Clinical Excellence (NICE)
guidelines (Palanivel et al., 2009). Methylphenidate (Ritalin™) acts via inhibition of
noradrenaline (NA) and dopamine (DA) transporters reducing their removal and increasing
levels at the synapse (Heal et al., 2009, Heal et al., 2008). It improves inattention in around 70-
80% of ADHD patients but due to increased dopamine in the mesolimbic system has addictive
potential. However, at clinical doses, it has been shown to reduce impulsivity in clinical studies
(Heal et al., 2009, Winstanley, 2011). Due to fear of misuse or diversion of the medication, it is
contraindicated in those with a history of drug addiction. The second line treatment for ADHD is
the non-stimulant, atomoxetine which is a selective noradrenaline transporter inhibitor which
causes an increase the length of time noradrenaline remains at the synapse. It also indirectly
increases dopamine, but only in the prefrontal cortex eliminating abuse liability (Bymaster et al.,
2002, Heal et al., 2009). Atomoxetine also alleviates impulsive measures in the continuous
performance, stop signal, and delay discounting tasks in humans (Barry et al., 2009,
Chamberlain et al., 2006). However, for attentive measures response rates and effect sizes are
lower than for stimulants (Heal et al., 2009, Cunill et al., 2013). A meta-analysis of atomoxetine
treatment concluded that the risk-benefit ratio for atomoxetine is even lower in an adult ADHD
population than in children as it has only a minimal effect on clinically meaningful endpoints
(e.g. job performance or sociability) (Cunill et al., 2013). With adult ADHD being increasingly
recognised to have a significant socioeconomic burden, it is important that we find new
pharmacological treatments with improved efficacy, safety and tolerability in the adult ADHD
population, or enhance the use of available medication by delivering the most appropriate
treatment to each subtype of patient (Kooij et al., 2010).
To enhance our understandings of the neurobiology underpinning attention, animal models of
low attention are required. Many models use pharmacological, physical or genetic
60
manipulations to impair attention in animals (reviewed by Bari and Robbins (2011)). However, in
conditions which are highly diverse and the causes not fully understood, these methods are
restrictive to the possible mechanisms studied (Jupp et al., 2013, Hayward et al., 2016).
Grouping animals based on performance in a behavioural task enables a holistic approach with
minimal assumptions of the mechanism (for review see Hayward et al. (2016)).
Tomlinson et al. (2014) recently demonstrated the flexibility of this method by producing low
attentive and high impulsive groups using the 5 choice continuous performance task (5C-CPT).
The 5C-CPT is designed to be highly translatable from the widely used clinical continuous
performance task (Young et al., 2009, Riccio et al., 2002, Lustig et al., 2013). This task has both
stimuli that require a response (go trials) and similar stimuli that require withholding of a
response (no-go trials) allowing assessment of attention, impulsivity and vigilance (Young et al.,
2009).Tomlinson et al. (2014) then proposed a third model of ADHD combined subtype-like
animals, showing that by careful choice of thresholds, different phenotypes can be produced
(Tomlinson et al., 2015). However, central to these studies is the very careful choice of
variables, for example in the ADHD-C type model the criteria were: accuracy <90%, sensitivity
index <0.3 and probability of false alarms >0.5, but no justification was given for the choice of
these thresholds. The same is true of a number of proposed behavioural separation models
(Hayward et al., 2016, Grottick and Higgins, 2000, Puumala et al., 1996, Paterson et al., 2011).
However, more flexibility is possible if animals are chosen based on mean, median, quartiles or
other percentiles of performance (Dalley et al., 2007, Diergaarde et al., 2008, Winstanley et al.,
2010). This can give consistent numbers in each group, which makes statistical analysis more
robust and allows easier translation of results between labs, strains and species. Improving the
robustness and reproducibility of animal models is essential to produce results that can be
translated to humans and so enable the development of enhanced therapeutic strategies.
Our objective was to determine if statistically significant differences in attention and vigilance
altered the effect of attention promoting compounds. We aimed to validate an alternative and
more flexible method to determine whether it can produce a model with predictive validity. To
judge validity, we compare to a previous method of separation by performance in 5C-CPT used
by Tomlinson et al. (2014) and the response to first and second-line treatments for ADHD
methylphenidate and atomoxetine. This method would allow the testing of novel compounds in
future experiments to advance understanding of attention.
3.5 Methods
3.5.1 Animals
38 female Hooded Lister rats (Charles River, UK; weighing 210 ± 20 g at the start of training)
were housed in groups of four in individually ventilated cages with two levels (Techniplast, UK)
with a standard 12-hour light-dark cycle (lights on 7:00 am). The environment was maintained at
21 ± 2 °C, 55 ± 5 % humidity. Diet was standard rat chow (Special Diet Services, UK) controlled
61
to maintain 90% free feeding weight throughout training and testing (usually 10 g/rat/day).
Water was available ad libitum for the duration of the study. All experiments were conducted
between 07:00-19:00 in the light phase and were conducted in accordance with the UK Animals
(Scientific Procedures) 1986 Act and University ethical guidelines.
3.5.2 Apparatus
Tests were conducted in eight standard nine hole chambers (Campden Instruments Ltd, UK)
with holes 2, 4, 6, and 8 occluded. Chambers were controlled and data collected by ABet II
Touch software (Lafayette Instrument Company, IL, USA). Animals were rewarded with 45 mg
rodent reward pellets (Test Diet, MO, USA).
3.5.3 5C-CPT Training Procedure
Animals were trained as described previously in Barnes et al. (2012a), Barnes et al. (2012b)
and Appendix. In brief, animals were trained for approximately 24 weeks to stability on final
training parameters of 1 s stimulus duration (SD), 2 s limited hold (LH), 5 s variable inter-trial
interval (vITI). This was done in a series of stages, gradually shortening the SD and LH, each
stage was achieved by meeting criteria of >70% accuracy, <25% omissions and >65% correct
rejections. Once meeting final criteria, animals were trained once per week to maintain
performance.
3.5.4 Testing Procedure
Testing 5C-CPT parameters were the same as the training procedure, with two alterations: a
reduced 1 s SD; and increased 7-12 s variable ITI. Methylphenidate and then atomoxetine were
tested in the same 38 animals following a two-week washout, a full health check and a 5C-CPT
training session to ensure performance was not different to before the study. A within subjects
latin square design was used to randomise the order of drugs. Each test day was separated by
a seven day washout period. Atomoxetine hydrochloride (Sigma, UK) and methylphenidate
hydrochloride (Sigma, UK) solutions were dissolved in saline (0.9% NaCl) as a dose of free
base to produce concentrations of 0.5, 1 and 2 mg/ml for each drug. Compounds were
administered via a 1 ml/kg intraperitoneal injection 30 minutes before testing.
3.5.5 Division into Subgroups
Animals were divided into high and low performing groups based on performance when treated
with vehicle (0.9% saline). The best performing 10 animals for d′ and accuracy formed the high-
attentive (HA) group, and the worst performing 10 animals formed the low-attentive (LA) group.
This is an upper/lower quartile split, which represents a careful balance between having
significantly different groups and the number of animals used (see Introduction). To achieve
this, animals were ordered by d′, and the highest/ lowest 10 animals were chosen, providing
their accuracy was also above/below average. In the methylphenidate group, six animals were
62
excluded due to >80% omissions or low trial numbers during vehicle treatment and for
atomoxetine four were excluded for the same reason.
3.5.6 Statistical Analysis
The method for calculation of all parameters is described in detail in Tomlinson et al. (2014) and
Young et al. (2013). The signal detection theory measure d′ was used in this case only after
confirming that probability hit rate (pHR) and the probability of false alarms (pFA) were normally
distributed, as described in Young et al. (2013). Here, d′ is calculated as described in Young et
al. (2013) as the z-score of pHR minus the z-score of pFA.
( ) ( )
( )
Using d′ is a further translational step towards making the rodent 5C-CPT closer to the human
CPT, compared to using sensitivity index (Young et al., 2013). Measures were calculated
individually for each rat at each SD for accuracy, omissions, pHR, pFA, d′ and responsivity
index (RI) where z scores were based on whole cohort data. After relevant measures were
calculated, data were analysed using InVivo Stat (Version 3.4), the most appropriate statistical
package for in vivo behavioural experiments (Clark et al., 2012). The analysis used a two-way
repeated measures mixed model ANOVA with GROUP as the treatment factor and
TREATMENT as the repeated factor for each measurement followed by LSD planned
comparisons for each treatment compared to vehicle. Normality and Sphericity of the data were
judged using the normality plot and the residual vs. predicted plots. Arcsine, square root or
Log10 transformations were used where appropriate. Data are presented as F values, and p
values for TREATMENT*GROUP interactions and effects are considered statistically significant
if the p-value from Fisher protected least squared difference (LSD) planned comparisons were <
0.05.
63
3.6 Results
3.7 Methylphenidate
3.7.1 Group Performance
It is important to establish overall group performance before separating the animals into groups,
particularly when within group reference points are used (Hayward et al., 2016), using
parameters such as upper/lower quartiles (Table 3.1).
Table 3.1: Group vehicle treated baseline before HA/LA grouping for methylphenidate
Measure Mean Standard
error
Median Lower
quartile
Upper
quartile
Accuracy (%) 93.9 1.06 96.2 91.3 100
Omission (%) 41.4 2.50 40.1 29.2 54.2
Premature
(Number)
13.7 2.15 7 5 23
p[FA] 0.360 0.0339 0.400 0.171 0.48
p[HR] 0.550 0.0227 0.583 0.458 0.644
d' 0.0787 0.136 -0.0183 -0.509 0.306
Responsivity Index -0.106 0.0541 -0.0846 -0.374 0.0875
Processed Trials
(Number)
101 3.42 107 85 120
Correct Latency (s) 0.885 0.0172 0.898 0.799 0.947
Incorrect Latency
(s)
0.797 0.105 0.997 0 1.38
Magazine Latency
(Go) (s)
1.81 0.525 1.16 1.05 1.48
False Alarm
Latency (s)
0.793 0.0263 0.8 0.704 0.901
64
Magazine Latency
(No Go) (s)
14.4 2.94 8.63 1.62 18.8
3.7.2 High/ Low Comparison
For Methylphenidate the ANOVA showed a number of significant group* treatment interactions,
which were followed by LSD planned comparisons to vehicle (saline) control. Firstly, accuracy
(Figure 3.1A, F1,18=13.4,p=0.0018) was significantly reduced by 0.5 mg/kg in HA animals
(p<0.05) and in LA animals increased at 0.5 and 1 mg/kg (p<0.05). Omissions (Figure 3.1B,
F3,54=4.59, p=0.0062) showed a significant increase at 2 mg/kg in HA animals, but reduced by 1
(p<0.05) and 2 mg/kg in LA animals (p<0.01). Hit Rate (Figure 3.2C, F3,54=5.79,p=0.0017) was
reduced by 2 mg/kg in HA animals (p<0.01), but increased by 0.5 (p<0.05), 1 (p<0.01) and 2
mg/kg (p<0.01) in LA animals. d′ (Figure 3.2A, F3,54=2.82, p=0.0477) was increased by 0.5
(p<0.05) and 1 mg/kg (p<0.01) in LA animals, but in HA animals d′ was reduced by 2 mg/kg
(p<0.01).
In the group level effects, HA animals showed significantly higher accuracy (Figure 3.1A,
F1,18=13.4,p=0.0018) and d′ (Figure 3.2A, F1,18=5.9, p=0.0258) compared with LA animals. LA
animals showed significantly higher premature responses (Figure 3.1C, F1,18=6.98, p=0.0166),
incorrect latency (Figure 3.3B, F1,18=4.5, p=0.0481) and magazine latency for no-go trials
(Figure 3.3D, F1,18=4.75, p=0.0429). Significant reduction of correct latency with higher doses
was seen irrespective of group (Figure 3.3A, F3,54, p=0.0335) which in planned comparisons
was significant for 1 mg/kg dose (p<0.01).
65
A c c u r a c y
7 0
8 0
9 0
1 0 0
0 0 .5 1 2 0 0 .5 1 2
H A L A
Pe
rc
en
tag
e
#
# # #
**
AO m is s io n s
2 0
3 0
4 0
5 0
6 0
Pe
rc
en
tag
e
0 0 .5 1 2 0 0 .5 1 2
H A L A
#
***
B
P re m a tu r e R e s p o n s e s
0
1 0
2 0
3 0
4 0
Nu
mb
er
0 0 .5 1 2 0 0 .5 1 2
H A L A
+
+
C P r o c e s s e d T r ia ls
8 0
8 5
9 0
9 5
1 0 0
1 0 5
1 1 0
1 1 5
1 2 0
1 2 5
Nu
mb
er
0 0 .5 1 2 0 0 .5 1 2
H A L A
D
Figure 3.1: Methylphenidate increased accuracy at 0.5 and 1 mg/kg in LA animals. (A) 0.5 mg/kg
increased accuracy in LA animals, but reduced it in HA animals, 1 mg/kg increased accuracy in LA
animals. (B) Methylphenidate increased omissions in HA animals at 2 mg/kg, but reduced omissions in LA
animals at 1 and 2 mg/kg. (C) Significant differences were observed between HA and LA animals, but
were not significantly altered by methylphenidate. (D) No significant changes were seen on number of
trials processed. Data are shown as mean ± SEM of n=10 HA, and n=10 LA animals each tested at all
doses. * and ** denote significance levels p< 0.05, 0.01 respectively for LSD planned comparisons of
vehicle treated LA animals to the three treatment levels following significance in GROUP*TREATMENT
interaction in a repeated measures two-way ANOVA and #, ## and ### denote the same, but to the HA
vehicle.
66
d '
-0 .7 5
-0 .5 0
-0 .2 5
0 .0 0
0 .2 5
0 .5 0
0 .7 5
1 .0 0
1 .2 5
Sc
ore
0 0 .5 1 2 0 0 .5 1 2
H A L A
# #
# #
A
* * *
R e s p o n s iv ity In d e x
-0 .4
-0 .2
0 .0
0 .2
0 .4
Sc
ore
0 0 .5 1 2 0 0 .5 1 2
H A L A
B
p [H R ]
0 .3
0 .4
0 .5
0 .6
0 .7
Pe
rc
en
tag
e
0 0 .5 1 2 0 0 .5 1 2
H A L A
# ##
*** **
C p [F A ]
0 .2
0 .3
0 .4
0 .5
0 .6
Pro
ba
bil
ity
0 0 .5 1 2 0 0 .5 1 2
H A L A
D
Figure 3.2: Vigilance is increased by 0.5 and 1 mg/kg in LA animals. (A) In HA animal‟s d′ is reduced
by 2 mg/kg methylphenidate and In LA animals it is increased at 0.5 and 1 mg/kg. (B) No significant
differences in responsivity index. (C) In HA animals p[HR] is reduced by 2 mg/kg methylphenidate, and in
LA animals 0.5, 1 and 2 mg/kg increased p[HR]. (D) No significant differences were observed in p[FA].
Data are shown as mean ± SEM of n=10 HA, and n=10 LA animals each tested at all doses. * and **
denote significance levels p< 0.05, 0.01 respectively for LSD planned comparisons of vehicle treated LA
animals to the three treatment levels following significance in GROUP*TREATMENT interaction in a
repeated measures two-way ANOVA and #, ## and ### denote the same, but to the HA vehicle.
67
C o rr e c t L a te n c y
0 .7 5
0 .8 0
0 .8 5
0 .9 0
0 .9 5
1 .0 0
1 .0 5
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
A
+
+
In c o r re c t L a te n c y
0 .5
0 .6
0 .7
0 .8
0 .9
1 .0
1 .1
1 .2
1 .3
1 .4
1 .5
1 .6
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
+ +
BM a g a z in e L a te n c y (G o )
1 .0 0
1 .0 5
1 .1 0
1 .1 5
1 .2 0
1 .2 5
1 .3 0
1 .3 5
1 .4 0
1 .4 5
1 .5 0
1 .5 5
1 .6 0
1 .6 5
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
C
F a ls e A la rm L a te n c y
0 .7
0 .8
0 .9
1 .0
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
C M a g a z in e L a te n c y (N o G o )
0
5
1 0
1 5
2 0
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
+
+
D
Figure 3.3: No significant differences were seen on the latency measures due to methylphenidate.
(A) No significant differences in correct latency due to methylphenidate. (B) HA and LA were significantly
different to one another for incorrect latency, but this was not dependent on treatment. (C) No significant
differences in magazine latency for go trials. (D) For False alarm latency, no significant differences were
seen. (E) For Magazine latency in no-go trials, a significant difference between HA and LA animals was
seen, but it was not dependent on treatment. + denotes a significant difference between groups
irrespective of treatment. Data are shown as mean ± SEM of n=10 HA and n=10 LA animals each tested at
all doses.
3.8 Atomoxetine
3.8.1 Group Performance
Table 3.2 is presented below to show the vehicle treated baseline performance of the whole
cohort before behavioural grouping.
Table 3.2: Group vehicle treated baseline before HA/LA grouping for atomoxetine.
Measure Mean Standard
error
Median Lower
quartile
Upper
quartile
Accuracy (%) 87.9 3.02 91.6 79.1 99.1
Omission (%) 46.4 3.68 44.4 35.4 59.3
Premature 11.7 2.40 8.5 3 18
68
(Number)
p[FA] 0.334 0.0349 0.296 0.2 0.452
p[HR] 0.486 0.0425 0.547 0.316 0.631
d' 0.0389 0.263 0.0681 -0.688 0.922
Responsivity
Index
-0.192 0.0662 -0.222 -0.391 -0.037
Processed
Trials (Number)
111 2.59 115 103 120
Correct Latency
(s)
0.848 0.0272 0.872 0.779 0.915
Incorrect
Latency (s)
0.756 0.138 0.718 0.051 1.30
Magazine
Latency (Go) (s)
1.72 0.277 1.23 1.08 1.70
False Alarm
Latency (s)
0.753 0.0321 0.756 0.636 0.854
Magazine
Latency (No
Go) (s)
20.9 4.69 14.6 5.30 30.2
3.8.2 High/ Low Comparison
For atomoxetine the ANOVA showed a number of significant group*treatment interactions, LSD
planned comparisons revealed significant differences for each dose compared to saline treated
control. Accuracy (Figure 3.4A, F3,54=3.61,p=0.0189) was increased by 0.5 (p<0.01) and 1
mg/kg (p<0.05) in LA animals. Premature responses (Figure 3.4C, F3,54=3.56,p=0.0201) were
reduced by 0.5 and 2 mg/kg in LA animals (both p<0.05). d′ (Figure 3.5A, F3,54=3.56,p=0.0201)
was reduced by 2 mg/kg in HA animals (p<0.01) and increased by 0.5 and 1 mg/kg (both
p<0.05) in LA animals.
In group level effects, HA animals showed significantly higher accuracy (Figure 3.4A,
F1,18=19.96,p=0.0003), hit rate (Figure 3.5C, F1,18=9.4,p<0.0067), and d′ (Figure 3.5A,
F1,18=11.32,p=0.0035). However, LA animals showed significantly higher omissions (Figure
69
3.4B, F1,18=5.58,p=0.0296) and premature responses (Figure 3.4C, F1,18=4.94,p=0.0394).
Treatment effects irrespective of group were seen for magazine latency in go trials (Figure 3.6C,
F3,54=4.38,p=0.0079) which was reduced by 0.5 mg/kg of atomoxetine (p<0.01); false alarm
latency (Figure 3.6C, F3,54=2.88, p=0.0441) but was not significant in planned comparisons and
magazine latency in no go trials (Figure 3.6D, F3,54=2.89,p<0.0437) which was significantly
reduced by 0.5 (p<0.05) and 2 mg/kg (p<0.01) of atomoxetine.
A c c u r a c y
7 0
8 0
9 0
1 0 0
0 0 .5 1 2 0 0 .5 1 2
H A L A
Pe
rc
en
tag
e
# # #
***
A O m is s io n s
3 0
3 5
4 0
4 5
5 0
5 5
6 0
6 5
Pe
rc
en
tag
e0 0 .5 1 2 0 0 .5 1 2
H A L A
*
#
B
P re m a tu r e R e s p o n s e s
2 .5
5 .0
7 .5
1 0 .0
1 2 .5
1 5 .0
1 7 .5
2 0 .0
2 2 .5
Nu
mb
er
0 0 .5 1 2 0 0 .5 1 2
H A L A
*
*
# # #
C P r o c e s s e d T r ia ls
9 0
1 0 0
1 1 0
1 2 0
1 3 0
Nu
mb
er
0 0 .5 1 2 0 0 .5 1 2
H A L A
D
Figure 3.4: In LA animals Accuracy is increased and premature responses reduced by atomoxetine.
(A) Accuracy was increased by 0.5 and 1 mg/kg of atomoxetine in LA animals. (B) HA animals had
significantly fewer omissions, but this was not dependent on treatment. (C) In LA animals 0.5 and 2 mg/kg
of atomoxetine reduced the number of premature responses. (D) No significant differences were seen for
the number of processed trials. Data are shown as mean ± SEM of n=10 HA and n=10 LA animals each
tested at all doses. * and ** denote significance levels p< 0.05, 0.01 respectively for LSD planned
comparisons of vehicle treated LA animals to the three treatment levels following significance in
GROUP*TREATMENT interaction in a repeated measures two-way ANOVA and #, ## and ### denote the
same, but to the HA vehicle group. + denotes a significant difference between groups irrespective of
treatment.
70
d '
-1 .5
-1 .0
-0 .5
0 .0
0 .5
1 .0
1 .5
Sc
ore
0 0 .5 1 2 0 0 .5 1 2
H A L A
# #
*
# # #
*
A R e s p o n s iv ity In d e x
-0 .5
-0 .4
-0 .4
-0 .3
-0 .3
-0 .2
-0 .2
-0 .1
-0 .0 5
0
0 .0 5
Sc
ore
0 0 .5 1 2 0 0 .5 1 2
H A L A
B
p [H R ]
0 .2
0 .3
0 .4
0 .5
0 .6
0 .7
Pe
rc
en
tag
e
0 0 .5 1 2 0 0 .5 1 2
H A L A
C p [F A ]
0 .2
0 .3
0 .4
0 .5
Pro
ba
bil
ity
0 0 .5 1 2 0 0 .5 1 2
H A L A
D
Figure 3.5: Atomoxetine reduced d′ in HA animals, but increased it in LA animals. (A) In HA animals
2 mg/kg of atomoxetine reduced d′ significantly, in LA animal‟s 0.5 and 1 mg/kg increased d′. (B) No
significant effect was seen on responsivity index. (C) No significant effect of atomoxetine was seen on hit
rate. (D) Atomoxetine has no significant effect on the probability of false alarm. Data are shown as mean ±
SEM of n=10 HA and n=10 LA animals each tested at all doses. * and ** denote significance levels p<
0.05, 0.01 respectively for LSD planned comparisons of vehicle treated LA animals to the three treatment
levels following significance in GROUP*TREATMENT interaction in a repeated measures two-way ANOVA
and #, ## and ### denote the same, but to the HA vehicle group. + denotes a significant difference
between groups irrespective of treatment.
71
C o rr e c t L a te n c y
0 .8
0 .9
1 .0
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
A In c o r re c t L a te n c y
0 .0 0
0 .2 5
0 .5 0
0 .7 5
1 .0 0
1 .2 5
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
B M a g a z in e L a te n c y (G o )
0 .7 5
1 .0 0
1 .2 5
1 .5 0
1 .7 5
2 .0 0
2 .2 5
2 .5 0
2 .7 5
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
C
F a ls e A la rm L a te n c y
0 .4
0 .5
0 .6
0 .7
0 .8
0 .9
1 .0
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
C M a g a z in e L a te n c y (N o G o )
0
1 0
2 0
3 0
4 0
Tim
e (
ms
)
0 0 .5 1 2 0 0 .5 1 2
H A L A
D
Figure 3.6: Atomoxetine had no significant effects on the latency measures in HA or LA animals.
3.9 Discussion
This study has shown that methylphenidate increases attention (accuracy) and vigilance (d′)
and reduces omissions (low attention/ motivation) in LA animals. The opposite effect was seen
in HA animals which had reduced performance on accuracy and d′. Atomoxetine also increased
accuracy and d′ in LA animals, but without a significant effect on omissions. However, it also
reduced premature responses in the same group. In HA animals, atomoxetine only reduced d′.
Therefore, in LA animals, both compounds showed an effect to improve selective attention and
vigilance. Methylphenidate also improved the ability to sustain attention, and atomoxetine
reduced waiting impulsivity.
The aim of this study was to assess how a novel grouping method altered the study of well-
characterised attention promoting compounds methylphenidate and atomoxetine. Tomlinson et
al. (2014) showed that in LA animals accuracy was significantly increased by 2 mg/kg
methylphenidate only and vigilance was significantly increased by 0.5 and 1 mg/kg doses, they
did not see a significant difference in HA animal performance. In LA animals our study agrees
with this entirely for the vigilance measure d′ (Figure 3.2A), but in the present study accuracy
was not affected at 2 mg/kg but was significantly increased at lower doses of 0.5 and 1 mg/kg
(Figure 3.1A). In our study vehicle accuracy was higher in HA animals and lower in LA animals
(HA: 98.1, LA: 77.64) when compared to Tomlinson et al. (2014) (HA: 96.5, LA 84.2), this may
72
be explained by our use of only the top/ bottom 10 animals and removing the animals in
between. This method produces a stronger separation between animals. The significant
reduction of accuracy by 0.5 mg/kg of methylphenidate may be due to the high baseline
accuracy and suggests that very high performing animals are impaired only by low doses of
methylphenidate. Tomlinson et al. (2014) also did not report any significant changes in
omissions due to treatment. Here we see a significant increase in HA animals given 2 mg/kg
methylphenidate, but a dose-dependent reduction in LA animals. Due to the stricter criteria used
in this study, there is a larger difference between the animals than in Tomlinson et al. (2014)
allowing this subtle, but opposing effect to be revealed. The theory of optimal arousal posits
that too low levels of forebrain monoamines result in an inability to maintain focus and that
increasing the level of monoamines improves the ability to sustain attention, but too much
causes too much reorienting of attention leading to distractibility (Wood et al., 2014). This is
often called an inverted U-shaped relationship or the Yerkes-Dodson effect (Wood et al., 2014).
So, if HA and LA animals had different levels of forebrain monoamines, this would explain why
the same compound could cause opposing effects. In LA animals it is improving towards the
optimum, where in HA it is moving them towards distractibility, causing more omitted trials or
less accurate performance. It has not yet been tested if 5C-CPT behavioural grouping can
produce groups with different baseline monoamine signalling. However, animals grouped by the
waiting impulsivity measure of premature responses have been shown to have significantly
reduced D2/3 receptors in the nucleus accumbens (Dalley et al., 2007). Similar studies to
analyse the receptor abundance and neurotransmitter levels in animals grouped in the 5C-CPT
would be highly informative. This could also include analysis of the neural circuitry involved in
response to the task as has been done for animals grouped for waiting impulsivity in the 5C-
SRTT by recording the extracellular neural potentials (Donnelly et al., 2014, Donnelly et al.,
2015).
In summary, we have shown that, with a different method of grouping animals based on
behavioural performance, we can recreate the major findings a previous paper. This method of
modelling has also shown novel differences not seen in the previous study exemplifying the
flexibility of this method of modelling. Future studies should focus on expanding upon this
method to allow novel compounds to be studied as well as increasing our understanding of what
can cause differences in baseline performance and how that mediates differences in response
to pharmacological treatment.
73
4 Nicotine and Caffeine in the 5 Choice
continuous performance task
Andrew Hayward, Lisa Adamson and Joanna C. Neill
[Unpublished data]
4.1 Contributions
Andrew Hayward and Joanna C. Neill planned the study. Training of the animals was completed
by Andrew Hayward and Lisa Adamson. The study was conducted by Andrew Hayward. The
manuscript was prepared by Andrew Hayward and Joanna C. Neill.
74
4.2 Context
Chapter 3 established the validity of the model to produce a deficit which is sensitive to
compounds used to treat inattention clinically and that the results are comparable to a previous
model, but with a larger deficit and higher translational potential. We, therefore, extend this to
study two compounds that have not been tested in a behaviourally grouped model of low
attention, nicotine and caffeine.
4.3 Abstract
Inattention is a debilitating symptom in many psychiatric conditions. Grouping animals by
performance has emerged as a powerful method to analyse the effect of compounds on low or
high attentive individuals with strong predictive validity. Nicotine and caffeine are the most
widely used pro-cognitive compounds. Nicotine improves sustained attention in humans and
rodents, with a larger effect in low performing individuals. The effect of caffeine on attention has
been poorly studied in rodents compared to nicotine, but also seems to improve sustained
attention. The effect of acute doses of nicotine (0.01, 0.02 mg/kg) and caffeine (2.5, 5 mg/kg)
were assessed in adult female Lister hooded rats using a variable stimulus duration (vSD; 0.75,
1.25 and 2 s) version of the 5 choice continuous performance task (5C-CPT). This showed that
0.02 mg/kg nicotine reduces omissions (sustained attention), increases premature responses
(impulsivity) and increases the probability of false alarms (pFA; impulsivity). Caffeine (2.5
mg/kg) increased pFA and reduces latency to false alarm (impulsivity). We also grouped the
adult female Lister Hooded rats by performance in measures of selective attention and vigilance
to produce low attentive (LA) and high attentive (HA) groups. This grouping showed the
increase of pFA due to 0.01 mg/kg nicotine and 2.5 mg/kg caffeine is more distinct in HA than
LA animals. These changes show that nicotine and caffeine promote impulsivity in female Lister
hooded rats. Future studies should aim to use selective agonists of the nicotinic acetylcholine
receptors to understand their involvement in this effect.
75
4.4 Introduction
Nicotine is an active ingredient of tobacco and widely used in the form of cigarette smoking.
Nicotine acts at nicotinic acetylcholine receptors (nACh) to produce pro-cognitive effects such
as enhanced sustained attention and vigilance. These effects have been demonstrated in non-
smoking adults using the continuous performance task (CPT) (Rezvani and Levin, 2001,
Heishman et al., 2010, Levin et al., 1998). Pro-cognitive effects have also been shown in
rodents using the 5C-CPT (Young et al., 2013) and the 5 choice serial reaction time task (5C-
SRTT) (Grottick and Higgins, 2000). In healthy non-smokers, the attention-enhancing effects of
nicotine are most pronounced in low performing individuals (Wignall and de Wit, 2011, Poltavski
and Petros, 2006). Attention improvement has also been demonstrated using the 5C-SRTT in
male Lister Hooded rats where nicotine produces a greater improvement in low-performing
animals (Grottick and Higgins, 2000).
Caffeine is the active ingredient in coffee and energy drinks, making it one of the most widely
used psychoactive drugs in the world (Fulgoni et al., 2015). It is regularly used to promote
wakefulness, and this is achieved by non-selective antagonist action at adenosine A1 and A2A
receptors primarily (Ioannidis et al., 2014). Adenosine A1 receptors are present pre and
postsynaptically across many neurone types in the putamen, mediodorsal thalamus and
throughout the neocortex (Bauer et al., 2003). There are two subtypes of A2A receptors, the
typical, found in the striatum, and the atypical, located in the hippocampus and throughout the
cortex including the prefrontal cortex (Fredholm et al., 2003, Riccioni et al., 2010). Clinical
reports of its behavioural effect are mixed; with initial reports showing an improvement in ADHD
symptoms of inattention and low vigilance (Schnackenberg, 1973, Ioannidis et al., 2014), but
more recent reports have both supported and refuted this claim (reviewed in Ioannidis et al.
(2014)). In rodents, caffeine has been shown to reduce the number of omissions made in the
5C-SRTT at 2.5 mg/kg and severely impaired performance at 20 mg/kg (Bizarro et al., 2004).
However, another study using Long-Evans and CD rats found no significant changes of
omissions in either strain, but did report a significant reduction of correct latency at 3 mg/kg
dose (Higgins et al., 2007). In the same task, age-related reduction in performance is reversed
by caffeine including a reduction in omissions (Grottick and Higgins, 2002).
The aims of this study were to firstly assess if nicotine and caffeine cause alterations of
performance in the 5C-CPT when a variable stimulus duration is used in female Lister Hooded
rats. Secondly, we aimed to evaluate the effect of the two compounds in a population grouped
based on low attention and vigilance. We hypothesised that both compounds would improve
sustained attention shown by reduction of omissions and that for nicotine the effect would be
larger in low attentive animals as seen in clinical populations.
76
4.5 Experimental Procedures
4.6 Animals
37 female Lister Hooded rats (Charles River, UK; weighing 210 ± 20 g at the start of training)
were housed in groups of four in individually ventilated cages with two levels (GR1800 Double-
Decker Cage, Techniplast, UK) under a standard 12 hour light: dark cycle (lights on 7:00 am).
We used female rats as the 5C-CPT has been carefully validated in female rats in our
laboratory (Tomlinson et al., 2014, Tomlinson et al., 2015, Barnes et al., 2012a, Barnes et al.,
2012b). Also, male rats grow rapidly, which precludes social housing over the lengthy period it
takes to train and test rats in the 5C-CPT. Individual housing is stressful for this social species
and best avoided (Holson et al., 1991, Brown and Grunberg, 1995). The environment was
maintained at 21 ± 2 oC, 55 ± 5 % humidity. The diet was standard rat chow (Special Diet
Services, UK) controlled to maintain 90% free feeding weight throughout training and testing
(typically 10 g/rat/day). Water was available ad libitum for the duration of the study, except in
the operant chambers. All experiments were conducted between 09:00-17:00 in the light phase.
All experiments were conducted in accordance with the UK Animals (Scientific Procedures)
1986 Act and local University ethical guidelines.
4.6.1 Apparatus
Tests were conducted in eight standard nine-hole chambers (Campden Instruments Ltd, UK)
with holes 2, 4, 6 and 8 occluded. Chambers and data collection were controlled by ABet II
Touch software (Lafayette Instrument Company, IL, USA). Animals‟ behaviour was reinforced
with 45 mg rodent reward pellets (Test Diet, MO, USA).
4.6.2 5C-CPT Training Procedure
Animals were trained as described in previously (Barnes et al. (2012a), Tomlinson et al. (2014)
and in Appendix). In brief, animals were trained for approximately 24 weeks to set criteria on
final training parameters of 2 s stimulus duration (SD), 2 s limited hold, 5 s inter-trial interval
(ITI). Training was conducted in a series of stages with successively longer SD and limited hold
where each stage was reached by meeting criteria of >70% accuracy, <25% omissions and
>65% correct rejections (see Barnes et al. (2012a), Barnes et al. (2012b)). Once rats reached
these final criteria, they were trained once per week to maintain performance and avoid over-
training.
4.6.3 Testing Procedure
Testing parameters were as per training, with a variable 0.75, 1.25 and 2 s SD and increased 7-
12 s variable ITI. As previous studies have found separable effects of nicotine in genetically
modified α7 receptor mutations at different SDs, a variable SD was used here (Young et al.,
77
2013, Young et al., 2007). To gather sufficient data for three SDs, the session was extended to
1 hour or 250 trials, whichever came first.
(-) Nicotine hydrogen tartrate salt (Sigma, UK) was prepared as a dose of free base to
concentrations of 0.1 and 0.2 mg/kg in saline solution (0.9% NaCl) and administered
subcutaneously in a volume of 1 ml/kg one hour before testing. Anhydrous caffeine (Sigma, UK)
was prepared in saline solution (0.9% NaCl) to concentrations of 2.5 and 5 mg/kg in a volume of
1 ml/kg 30 minutes before testing via the intraperitoneal route. The four treatments and a control
of saline solution were administered to each animal over 5 weeks. Each test day was separated
by a seven day washout period (no drug administration, or training in 5C-CPT). Treatment order
was randomised and counterbalanced using a within-subjects Latin Square design. For both
compounds doses were selected from previous effects in rats performing the 5C-SRTT (Bizarro
et al., 2004).
4.6.4 Division into Subgroups
Animals were divided into high and low performing groups based on performance when treated
with vehicle (0.9% saline). The best performing ten animals for d′ and accuracy formed the high-
attentive (HA) group, and the worst performing ten animals formed the low-attentive (LA) group.
This is an upper/lower quartile split, which represents a careful balance between having
significantly different groups and the number of animals used (see Hayward et al. (2016)). To
achieve this, animals were ordered by d′, and the highest/ lowest ten animals were chosen,
providing their accuracy was also above/below average. As vigilance is a measure of overall
performance across go and no-go trials, accuracy was also used to give more weight to the
inattentive phenotype. Two animals were excluded for making >80% omissions.
4.6.5 Statistical Analysis
The method for calculation of all parameters is described in detail in Tomlinson et al. (2014),
Young et al. (2013) and Appendix. The signal detection theory measure d′ was used in this case
only after confirming that probability hit rate (pHR) and the probability of false alarms (pFA) were
normally distributed, as described in Young et al. (2013). Here, d′ is calculated in the same way
as described in Young et al. (2013) as the z score of pHR minus the z-score of pFA.
( ) ( )
78
( )
Measures were calculated individually for each rat at each SD for accuracy, omissions, pHR,
pFA, d′ and responsivity index (RI) where z scores were based on full cohort data. After
important measures had been calculated, data were analysed using InVivo Stat (Version 3.4),
as the most appropriate statistical package for in vivo behavioural experiments (Clark et al.,
2012). Overall performance (Table 4.1) was compared using a one-way Analysis of Variance
(ANOVA) comparing treatment as a factor and using animal ID as a blocking factor to account
for the within-subjects design. Between-group comparisons used least square difference (LSD)
tests adjusted for multiple comparisons by a Benjamini-Hochberg‟s procedure to reduce the
false discovery rate. Measures compared by this method include accuracy and d′ for three SDs.
Grouped analysis used a two-way repeated measures mixed model ANOVA with GROUP as
the treatment factor and TREATMENT as the repeated factor for each measurement followed
by LSD planned comparisons for each treatment compared to vehicle. Normality and Sphericity
of the data were judged using the normality plot and the residual vs. predicted plots. Square
root, Arcsine, Log10 or RANK transformations were used when appropriate. Data are presented
as F values, and p values for TREATMENT*GROUP interactions and effects are considered
statistically significant if the p-value from Fisher protected least square difference planned
comparisons were < 0.05.
4.7 Results
4.7.1 Whole Group Baseline
In the whole cohort (Table 4.1), an ANOVA with LSD planned comparisons found that nicotine
significantly reduced omissions (F4=3.18, p=0.0156) at 0.2 mg/kg (p<0.05), but significantly
increased the number of premature responses (F4=3.29, p=0.0131) at 0.2 mg/kg (p<0.001),
increased responsivity index in response to 0.75 s SD (F4=2.48, p=0.0467) at 0.1 (p<0.05) and
0.2 mg/kg (p<0.01). In response to 1.25 s SD, responsivity index (F4=3.71, p=0.0067) was also
increased by 0.2 mg/kg (p<0.01). Nicotine also increased the probability of false alarms during
0.75 s SD (F4=2.48, p=0.0467) at 0.1 (p<0.01) and 0.2 mg/kg (p<0.05), 1.25 s SD probability of
false alarms (F4=3.71, p=0.0067) was also increased by 0.2 mg/kg (p<0.01) and false alarm
latency (F4=3.62, p=0.0077) was significantly reduced by 0.1 mg/kg (p<0.01). Nicotine also
reduced the number of trials completed in the first 20 minutes (F4=14.54, p=0.0001) at a dose
of 0.2 mg/kg (p<0.001). Caffeine had two significant effects on the whole cohort, it increased
probability of false alarms (F4=2.48, p=0.0467) during 0.75 s SD trials at 2.5 mg/kg dose
(p<0.05) and it reduced the time to make a false alarm response (F4=3.62, p=0.0077) for the
same dose (p<0.01).
79
Table 4.1: Whole cohort (n=35) 5C-CPT performance following vehicle compared to nicotine (0.1
and 0.2 mg/kg) and caffeine (2.5 and 5 mg/kg). Performance following the four treatments is compared
to performance after saline vehicle treatment. Significant Fisher protected LSD planned comparisons are
represented in bold, and asterisks denote the significance level. * p<0.05, ** p<0.01 and *** p<.001.
Vehicle Nicotine 0.1
mg/kg
Nicotine 0.2
mg/kg
Caffeine 2.5
mg/kg
Caffeine 5
mg/kg
Accuracy (%)
0.75 89.4 ± 1.7 84.1 ± 2.56 87.7 ± 1.94 90.6 ± 1.62 92.4 ± 1.2
1.25 94.7 ±
1.25 92.2 ± 1.5 92.8 ± 1.65 95.2 ± 0.821 95 ± 1.34
2 95.2 ±
1.26 93.1 ± 1.99 93.7 ± 1.36 96.6 ± 1.02 96 ± 1.08
Omissions
(%)
0.75 47.9 ±
2.76 44.4 ± 3.08 44.2 ± 3.31 46.8 ± 3.09
47.4 ±
2.91
1.25 40.3 ±
3.14 39.5 ± 2.99 34.3 ± 3.07
* 40.5 ± 2.84 40 ± 2.9
2 37.4 ±
3.14 34.4 ± 2.8 34.3 ± 3.25 37.7 ± 2.86
38.6 ±
2.78
Premature (Number) 9.66 ±
1.68 14.3 ± 2.68 19.7 ± 2.94
*** 12.3 ± 1.66
12.5 ±
1.81
d'
0.75 0.139 ±
0.149
-0.295 ±
0.224
-0.174 ±
0.221
-0.0985 ±
0.174
0.0738 ±
0.181
1.25 0.0185 ±
0.225
-0.262 ±
0.216
-0.177 ±
0.23
-0.132 ±
0.193
0.0951 ±
0.2
2 -0.0043 ±
0.204
-0.09 ±
0.238
-0.104 ±
0.211
-0.148 ±
0.215
0.0385 ±
0.179
Responsivity
Index
0.75 -0.387 ±
0.0487
-0.263 ±
0.0532*
-0.258 ±
0.0601**
-0.295 ±
0.0562
-0.326 ±
0.0483
1.25 -0.219 ±
0.0563
-0.158 ±
0.0582
-0.0617 ±
0.0627**
-0.184 ±
0.0534
-0.216 ±
0.045
80
2 -0.147 ±
0.0552
-0.134 ±
0.0588
-0.0969 ±
0.0557
-0.133 ±
0.0548
-0.198 ±
0.0464
pFA
0.75 0.187 ±
0.0282
0.281 ±
0.0315**
0.276 ±
0.0371*
0.249 ±
0.0311*
0.222 ±
0.0279
1.25 0.251 ±
0.0335
0.301 ±
0.0348
0.337 ±
0.0391**
0.279 ±
0.0317
0.242 ±
0.024
2 0.264 ±
0.0284
0.29 ±
0.0364
0.296 ±
0.0302
0.289 ±
0.0334
0.244 ±
0.0234
Correct Latency (s) 0.816 ±
0.0221
0.803 ±
0.0269
0.801 ±
0.0253
0.782 ±
0.023
0.796 ±
0.021
Incorrect Latency (s) 1.05 ±
0.0523
0.95 ±
0.0482
0.989 ±
0.0563
0.877 ±
0.0606
0.998 ±
0.0556
False Alarm Latency
(s)
0.763 ±
0.0265
0.689 ±
0.0322**
0.738 ±
0.0287
0.677 ±
0.0294**
0.719 ±
0.024
Magazine Latency
(Go) (s)
1.07 ±
0.0307
1.08 ±
0.0247 1.25 ± 0.115
1.06 ±
0.0321
1.27 ±
0.205
Magazine Latency
(No Go) (s)
10.8 ±
1.57 10.8 ± 1.55 8.81 ± 1.51 12.4 ± 1.8 9.6 ± 1.55
Trials
Completed
(Number)
0-20 67.3 ±
1.33 65.2 ± 2.08 53.1 ± 2.71
*** 67.6 ± 1.13
67.6 ±
1.14
21-40 67.8 ±
1.14 69.4 ± 1.07 68.3 ± 1.24 67.3 ± 1.63
66.5 ±
1.76
41-60 66.5 ±
1.39 68.9 ± 0.734 68.3 ± 1.13 68.4 ± 1.11 64.7 ± 2.1
4.7.2 High/Low Comparison
LSD comparisons adjusted for multiple comparisons showed a significant difference between
LA and HA animals for accuracy at 0.75 s SD (p<0.05) and for d′ at 0.75 (p<0.001), 1.25 and 2 s
SD (p<0.05; Figure 4.1).
81
V e h ic le A c c u ra c y
0.7
5
1.2
5
2.0
0
7 0
8 0
9 0
1 0 0
1 1 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e (
%)
A
*
V e h ic le d '
0.7
5
1.2
5
2.0
0
-1 .5
-1 .0
-0 .5
0 .0
0 .5
1 .0
1 .5
S tim u lu s D u ra tio n
Sc
ore
B
L A
H A
* * * *
*
Figure 4.1: Comparison of accuracy and d′ after vehicle treatment in HA and LA groups when
grouped by these parameters. (A) Accuracy significantly differs between HA and LA animals only at the
most challenging SD of 0.75 s. (B) d′ significantly differed at all SDs, but most significantly at the most
challenging SD of 0.75 s. LSD comparisons were used and adjusted by the Benjamini-Hochberg method
for multiple comparisons. Significance levels are represented as * for p<0.05 and *** for p<0.001 after
adjustment.
A significant difference due to treatment and group interaction was also seen for omissions at 2
s SD (F4,72=2.64, p=0.0407; Figure 4.2B/D). However, LSD planned comparisons showed no
significant difference between any treatment and control, although a reduction in omissions,
when treated with nicotine 0.2 mg/kg, approached significance (p=0.053).
A c c u ra c y H A
0.7
5
1.2
5
2.0
0
7 0
8 0
9 0
1 0 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e
A O m is s io n H A
0.7
5
1.2
5
2.0
0
0
1 0
2 0
3 0
4 0
5 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e
B
V e h ic le
N ic o tin e 0 .1 m g /k g
N ic o tin e 0 .2 m g /k g
C a ffe in e 2 .5 m g /k g
C a ffe in e 5 m g /k g
A c c u ra c y L A
0.7
5
1.2
5
2.0
0
7 0
8 0
9 0
1 0 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e
C O m is s io n L A
0.7
5
1.2
5
2.0
0
0
2 0
4 0
6 0
8 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e
D P re m a tu r e
HA
LA
0
5
1 0
1 5
2 0
Nu
mb
er
E
82
Figure 4.2: Performance measures of the go trials in 5C-CPT when treated with saline, nicotine (0.1
and 0.2 mg/kg) and caffeine (2.5 and 5 mg/kg). A significant group treatment interaction was seen for
omissions (B/D), but planned comparisons were not significant.
For pFA, a significant group*treatment interaction showed an increase at 0.75 s SD (F4,72=3.32,
p=0.0149; Figure 3B), planned comparisons showed this to be significant in HA animals when
treated with 0.2 mg/kg nicotine (p<0.01) and 2.5 mg/kg caffeine (p<0.01). Responsivity index
also showed a significant group treatment*interaction for the 2 s stimulus duration (F4,72=2.79,
p=0.0328; Figure 3D)) and planned comparisons showed a significant reduction when HA
animals were treated with 5 mg/kg caffeine (p<0.05).
83
Figure 4.3: Signal Detection Theory related measures of the 5C-CPT. (B) A significant increase in pFA was seen in HA animals following treatment with nicotine 0.2 mg/kg and caffeine
2.5 mg/kg. (D) 5 mg/kg of caffeine significantly reduced responsivity index in HA animals representing a more conservative response strategy. Significance levels are denoted as ** for p<0.01
and * for p<0.05.
p H R H A
0.7
5
1.2
5
2.0
0
0 .0
0 .2
0 .4
0 .6
0 .8
S tim u lu s D u ra tio n
Pro
ba
bil
ity
A p F A H A
0.7
5
1.2
5
2.0
0
0 .0
0 .1
0 .2
0 .3
0 .4
S tim u lu s D u ra tio n
Pro
ba
bil
ity
B
* *
* *
d ' H A
0.7
5
1.2
5
2.0
0
-0 .5
0 .0
0 .5
1 .0
1 .5
S tim u lu s D u ra tio n
Sc
ore
C R I H A
0.7
5
1.2
5
2.0
0
-0 .6
-0 .4
-0 .2
0 .0
0 .2
V e h ic le
N ic o tin e 0 .1 m g /k g
N ic o tin e 0 .2 m g /k g
C a ffe in e 2 .5 m g /k g
S tim u lu s D u ra tio n
Sc
ore
C a ffe in e 5 m g /k g
D
*
p H R L A
0.7
5
1.2
5
2.0
0
0 .0
0 .2
0 .4
0 .6
S tim u lu s D u ra tio n
Pro
ba
bil
ity
E p F A L A
0.7
5
1.2
5
2.0
0
0 .0
0 .1
0 .2
0 .3
S tim u lu s D u ra tio n
Pro
ba
bil
ity
Fd ' L A
0.7
5
1.2
5
2.0
0
-1 .0
-0 .5
0 .0
0 .5
S tim u lu s D u ra tio n
Sc
ore
G R I L A
0.7
5
1.2
5
2.0
0
-0 .8
-0 .6
-0 .4
-0 .2
0 .0
S tim u lu s D u ra tio n
Sc
ore
H
84
A significant group*treatment interaction was seen for magazine latency in go trials (F4,72=2.52,
p=0.0485; Figure 4D) LSD planned comparisons showed a significant increase due to 0.2
mg/kg of nicotine in LA animals (p<0.001).
C o rr e c t L a te n c y
HA
LA
0 .6
0 .7
0 .8
0 .9
1 .0
Tim
e (
s)
A In c o r re c t L a te n c y
HA
LA
0 .0
0 .5
1 .0
1 .5
Tim
e (
s)
B F a ls e A la rm L a te n c y
HA
LA
0 .0
0 .2
0 .4
0 .6
0 .8
1 .0
Tim
e (
s)
C
M a g a z in e L a te n c y (G o )
HA
LA
0 .0
0 .5
1 .0
1 .5
2 .0
Tim
e (
s)
D
* * *
M a g a z in e L a te n c y (N o G o )
HA
LA
0
5
1 0
1 5
2 0
N ic o tin e 0 .1 m g /k g
N ic o tin e 0 .2 m g /k g
C a ffe in e 2 .5 m g /k g
Tim
e (
s)
C a ffe in e 5 m g /k g
E
V e h ic le
Figure 4.4: Latency measures of the 5C-CPT following treatment with vehicle, nicotine (0.1 and 0.2
mg/kg) or caffeine (2.5 and 5 mg/kg). (D) A significant increase in magazine latency in go trials was seen
following 0.2 mg/kg of nicotine.
T r ia ls C o m p le te d H A
1-2
0
21-4
0
41-6
0
0
2 0
4 0
6 0
8 0
T im e (m in s )
Nu
mb
er
A T r ia ls C o m p le te d L A
1-2
0
21-4
0
41-6
0
0
2 0
4 0
6 0
8 0
T im e (m in s )
Nu
mb
er
B
V e h ic le
N ic o tn in e 0 .1 m g /k g
N ic o tn in e 0 .2 m g /k g
C a ffe in e 2 .5 m g /k g
C a ffe in e 5 m g /k g
Figure 4.5: Number of trials completed per 20 minutes of the 5C-CPT. There was no significant
difference in the number of trials completed that differed between the HA and LA animals. A significant
treatment level effect can be seen in both (A) and (B) or 0.2 mg/kg nicotine to reduce the number of trials
completed (also see Table 1).
85
4.8 Discussion
This study has shown that acute doses of nicotine can cause a reduction in omissions showing
improved sustained attention and an increase in premature responses and pFA showing an
increase in impulsive action due to the treatment. A significant change was also seen in
responsivity index, which is used to assess response strategy, where the increase in this value
represents a more liberal response strategy after treatment (Young et al., 2009, Marston, 1996).
This evidence suggests a phenotype of high impulsivity, which is further supported by the
observed reduction in latency to make a false alarm response.
The grouped analysis showed that the HA animals were more susceptible to the increase in
pFA than the LA animals. While both groups had a similar pFA at 0.75 s SD, they may have
used different strategies to achieve this. The strategy can be deduced from the ratio of correct
rejections (1-pFA) in no-go trials to omissions in go trials as both represent a lack of response.
The HA group had much lower omissions than correct rejections. However, LA animals had
higher omissions showing they were not responding to go or no-go trials. Therefore in the LA
animals, the pFA may represent an inability to sustain attention and therefore respond to go or
no-go trials.
Bizarro et al. (2004) showed that the same doses of nicotine reduce omissions and increase
premature responses in the 5C-SRTT in male rats. In the 5C-CPT we found similar effects in
female rats. However, we did not see the same increase in accuracy at 0.2 mg/kg or a
significant reduction of correct latency at 0.1 mg/kg. However there are two factors which may
explain the difference, firstly, a higher baseline accuracy in this study compared to Bizarro et al.
(2004) would make increase less likely, but reduction more likely due to a ceiling effect.
Secondly, this may represent a difference in response between male and female rats. Female
rats are more susceptible to the rewarding effects of nicotine and thus compulsively administer
nicotine faster and at lower doses than males (Donny et al., 2000). A study by Taylor and
Maloney (2010) showed a synergistic effect of estradiol and nicotine in overectomised female
rats to improve performance in a maze based version of the 5C-SRTT. The effect of estradiol
suggests that nicotine needs to be studied in both males and females to compare gender-
related differences directly. Nicotine has been tested in the 5C-CPT by Young et al. (2013)
using four-day sub-chronic treatment with doses of 0.001-0.1 mg/kg in male mice. They found
significant improvement d′ at the 0.01 mg/kg dose using 0.75, 1.25 and 2 s SDs, which
significantly reduced omissions. They also analysed the performance on the first day of dosing
to ascertain acute effects of nicotine and found none. However, their acute data was not
presented in full to enable a detailed comparison, but in the data set non-significant differences
were observed for the 0.1 mg/kg dose in this study, and it is possible that their lower n number
of 15 did not have the statistical power to detect differences in the more variable acute data set,
which we can observe with 35 animals. Also, species differences cannot be ruled out
particularly in respect to the strategy of response, shown by a recent study which found rats to
86
be more dependent on a temporal strategy than mice (Cope et al., 2016). It has previously been
shown that acute doses of nicotine can promote task disruptive effects, whereas these are not
seen if subjects have previously received chronic or subchronic doses of nicotine (Hahn and
Stolerman, 2002). Differences between chronic and acute mechanisms of action have been
suggested to be via a number of different mechanisms. Firstly acute administration causes an
increase in plasma adrenocorticotropic hormone and cortisone; however, repeated injections do
not cause this response (Matta et al., 2007). Secondly, chronic dosing of nicotine facilitates
long-term potentiation and active up and down regulation of receptors (Fujii et al., 1999,
Mansvelder and McGehee, 2000). Future assessments could investigate this further by trying
subchronic or chronic doses to overcome disruptive effects or by analysing the effect of more
selective compounds to assess the role of nicotinic receptors with fewer off-target effects.
Caffeine, in the whole cohort, increased impulsive action (pFA) when under high attentive load
(i.e. low SD) and also reduced the false alarm latency. The increase in pFA was only observed
at the lowest dose (2.5 mg/kg) and at the most challenging SD (0.75 s) showing that under high
attentive load caffeine can promote impulsive behaviour. In the grouped analysis impulsive
action was also only significantly increased in HA animals, which could be explained by
differences in baseline omissions between LA and HA animals (as discussed above for
nicotine). Caffeine has been shown to improve 5C-SRTT accuracy in aged rats, which show a
reduction of accuracy (Grottick and Higgins, 2002). In healthy adult rats, no improvement in
accuracy has been observed, only a reduction in omissions at the doses tested here (Bizarro et
al., 2004). An increase in premature responses was also seen by Bizarro et al. (2004) at 5
mg/kg, but only when a fixed ITI was used, no difference was observed when a variable ITI was
used. Here we used a variable ITI and in agreement with their study found no significant effect
on premature responses. This measure is often used as an indicator of impulsive action, more
specifically waiting impulsivity. pFA is also a measure of impulsive action (response
disinhibition) it is, therefore, possible that the increase seen in pFA is related to the increase in
premature responses seen by Bizarro et al. (2004) with a fixed ITI. The increase in pFA would
support the hypothesis that, in healthy animals, caffeine can increase impulsive action.
However, premature responses can also represent timing strategies which would explain why
the effect is not seen with a variable ITI both in the Bizarro et al. (2004)‟s study and the work
presented here (Cope et al., 2016, Bizarro et al., 2004). Therefore the increase in pFA caused
by caffeine is independent of changes in premature responses showing an increase in response
disinhibition, but not waiting impulsivity. The mechanism by which caffeine can cause changes
in impulsivity is not known. However, the colocalisation of adenosine A2A receptors and
dopamine D2 receptors in the striatum offers one potential mechanism (Fink et al., 1992). The
selective removal of A2A receptors from the nucleus accumbens core prevents caffeine-induced
arousal (Lazarus et al., 2011). However, further research is needed to understand its
relationship with impulsivity.
87
In conclusion, acute doses of both nicotine and caffeine increased impulsive behaviour as
measured in the 5C-CPT in animals which have not been grouped based on behaviour. In
animals which have been grouped by behaviour nicotine increased impulsivity in HA animals.
Nicotine also reduced omissions, thus increasing sustained attention as has been previously
reported. These findings support previous work, but also highlight the power of the 5C-CPT to
assess impulsive action in a robust way and overcome issues of timing strategy. Future work
should compare acute to chronic or subchronic dosing for both compounds to see how this
differs their effect. Additionally, using the same method with more selective nicotinic
acetylcholine agonists would be important to establish the role of different receptor subtypes in
nicotine's action.
88
5 Partial agonism at the α7 nicotinic
acetylcholine receptor improves attention,
impulsive action and vigilance in low
attentive rats
Authors: Andrew Hayward, Lisa Adamson, Joanna C Neill
[Published in the Journal of European Neuropsychopharmacology:
HAYWARD, A., ADAMSON, L. & NEILL, J. C. 2017. Partial agonism at the alpha7 nicotinic
acetylcholine receptor improves attention, impulsive action and vigilance in low attentive rats.
Eur Neuropsychopharmacol, 27, 325-335.]
5.1 Contributions
Andrew Hayward and Joanna C. Neill planned the study. Training of the animals was completed
by Andrew Hayward and Lisa Adamson. The study was conducted by Andrew Hayward. The
manuscript was prepared by Andrew Hayward and Joanna C. Neill and approved by Lisa
Adamson.
89
5.2 Context
With nicotine showing a limited effect in the task and no specific effect in low attentive animals
one theory is that non-specific effects of nicotine promote a stress response. Testing more
specific compounds is one way around this and knock-outs of the α7 nicotinic acetylcholine
receptor result in an inattentive phenotype making it a receptor of interest (Young et al., 2007).
5.3 Abstract
Inattention is a disabling symptom in conditions such as schizophrenia and attention
deficit/hyperactivity disorder. Nicotine can improve attention and vigilance but is unsuitable for
clinical use due to abuse liability. Genetic knockout of the α7 nicotinic acetylcholine receptor
(nAChR) induces attention deficits, therefore, selective agonism may improve attention, without
the abuse liability associated with nicotine. The α7 nAChR partial agonist encenicline (formerly
EVP-6124) enhances memory in rodents and humans. Here we investigate, for the first time,
the efficacy of encenicline to improve attention and vigilance in animals behaviourally grouped
for low attentive traits in the 5 choice-continuous performance task (5C-CPT). Female Lister
Hooded rats were trained to perform the 5C-CPT with a variable stimulus duration (SD).
Animals were then grouped based on performance into upper and lower quartiles of d′
(vigilance) and accuracy (selective attention), producing high-attentive (HA) and low-attentive
(LA) groups. LA animals showed an increase in selective attention and vigilance at 0.3 mg/kg
encenicline, a reduction in impulsive action (probability of false alarms) and an increase in
vigilance following 1 mg/kg at 0.75 s SD. At 1 mg/kg, HA animals had reduced selective
attention at 0.75 s SD and reduced vigilance at 0.75 and 1.25 s SD. Improvement of attention,
vigilance and impulsive action in LA animals demonstrates that encenicline has pro-attentive
properties dependent on baseline levels of performance. Our work suggests that α7 nAChR
partial agonism may improve attention particularly in conditions with low attention.
90
5.4 Introduction
Attention can be broadly defined as the ability to filter out irrelevant stimuli in order to optimise
limited neurochemical resources (Luck and Gold, 2008). Inattention is an early symptom in
highly disabling conditions such as schizophrenia, bipolar disorder, Alzheimer‟s disease and
attention deficit/ hyperactivity disorder (ADHD) (Luck and Gold, 2008, Foldi et al., 2002).
Nicotine acts at nicotinic acetylcholine receptors (nACh) to produce pro-cognitive effects, in
particular, enhanced selective attention. This effect has been demonstrated in non-smoking
adults using the continuous performance task (CPT) (Rezvani and Levin, 2001), in rodents
using the 5 choice continuous performance task (5C-CPT) (Young et al., 2013) and in the 5
choice serial reaction time task (5C-SRTT) (Grottick and Higgins, 2000). In healthy non-
smokers, nicotine improves attention most in low performing individuals (Wignall and de Wit,
2011, Poltavski and Petros, 2006). In animals, this has been demonstrated using the 5C-SRTT
where nicotine produces a greater improvement in low-performing animals (Grottick and
Higgins, 2000).
Smoking in schizophrenia patients may be a form of self-medication (Segarra et al., 2011,
Parikh et al., 2016, Young and Geyer, 2013). This idea is supported by the higher prevalence of
smoking in schizophrenia patients (86-90 %) compared to healthy controls (15-25 %) and
improvement of cognitive domains, including vigilance and attention, in schizophrenia patients
and healthy subjects following nicotine treatment in the CPT (Rezvani and Levin, 2001,
Heishman et al., 2010, Segarra et al., 2011). However, nicotine has a high abuse liability and
many unpleasant side effects including dizziness, headache, agitation, anxiety and loss of
appetite (White and Levin, 1999, Watkins et al., 2000). This makes it unsuitable as a treatment,
particularly for schizophrenia and ADHD patients who often also exhibit high levels of
impulsivity, which further increases the risk of drug addiction.
The α4β2 and α7 receptors are the most abundant nicotinic acetylcholine receptor (nAChR)
subtypes in the CNS. The α7 nAChR is a pentameric ligand-gated ion channel located both pre
and post synaptically throughout the cortex enabling fast synaptic neurotransmission (Vizi and
Lendvai, 1999). Mice with a deletion of the α7 nAChR show cognitive deficits including reduced
sustained attention, supporting a role for α7 nAChR mechanisms in attention (Young et al.,
2007, Hoyle et al., 2006). It has also been shown that attention deficits in α7 receptor knockout
mice can be reversed by β2 containing receptor agonists (Kolisnyk et al., 2015). This suggests
that activation of β2 nAChRs can bypass attentional deficits due to α7nAChR deficiency.
However, the β2-containing heterodimeric (largely α4β2) nAChRs are predominantly implicated
in nicotine dependence via activation of the mesolimbic dopamine system (Picciotto et al., 1998,
Bloem et al., 2014, Brioni et al., 1996, Pons et al., 2008). Therefore α7 receptor activation
represents a more favourable target.
91
We have shown previously that in sub-chronic phencyclidine (PCP) treated rats (a well-
validated model for schizophrenia, McLean et al., 2011) and in healthy control animals (McLean
et al., 2016, McLean et al., 2011) the α7 full agonist PNU-282987 improves performance in
novel object recognition, a visual recognition memory task. In contrast, the same compound has
been shown not to improve attention or vigilance in the 5C-CPT in both normal mice and those
with a scopolamine-induced deficit (Young et al., 2013). This lack of efficacy is in agreement
with other studies using α7 full agonists in tasks of attention; the data suggest therefore that
visual recognition memory, not attention is improved by full agonists of the α7 receptor (Young
and Geyer, 2013).
However, the α7 partial agonist ABT-126 produced a significant increase in attention in a phase
II clinical study when assessed using the MATRICS Consensus Cognitive Battery (Haig et al.,
2016), and this is supported by the effects of tropisetron and R-3487, which are also α7 receptor
partial agonists. Both compounds are also antagonists at the 5-HT3 receptor, but antagonism at
the 5-HT3 receptor does not improve attention (Muir et al., 1995). Tropisetron improved
sustained attention in schizophrenia patients assessed with the Cambridge Neuropsychological
Test Automated Battery (CANTAB) (Shiina et al., 2010) and R3487 improved sustained
attention in a visual signal detection task in rodents; these results await clinical confirmation
(Rezvani et al., 2009). Partial agonists at the α7 receptor have not been studied in animals
using a specific task of attention such as the 5C-CPT, nor have their effects been investigated
in low attentive animals.
Encenicline is a potent and selective α7 nAChR partial agonist and 5-HT3A receptor antagonist
(Prickaerts et al., 2012). However, at low nanomolar concentrations in the brain, it is thought
that 5-HT3 receptor binding is minimal, but could reduce side-effects such as nausea via
peripheral action (Prickaerts et al., 2012). In rats, a dose of 0.3 mg/kg of encenicline improves
visual recognition memory in the novel object recognition task with a 24-hour inter-trial interval
in normal rats and reverses a scopolamine-induced deficit in the same task (Prickaerts et al.,
2012).
The 5C-CPT is a test of sustained/selective attention, impulsive action and vigilance, developed
from the clinically used CPT. It is also based on the widely-used rodent 5C-SRTT. However, it
provides enhanced validity and translation to the clinic compared with the 5C-SRTT by including
trials where no response is required. Comparing the two, the 5C-CPT is the more translational,
whereas the 5C-SRTT offers higher throughput (Lustig et al., 2013). In both of these tasks,
several studies have separated animals into groups based on variation in their task
performance, which are stable across multiple test sessions (Blondeau and Dellu-Hagedorn,
2007, Dalley et al., 2007, Grottick and Higgins, 2000, Tomlinson et al., 2014); reviewed in
Hayward et al. (2016)). We have demonstrated that animals respond differentially to
pharmacological treatments, in particular, methylphenidate and atomoxetine, drugs currently
used to treat ADHD in patients, based on this natural variation of performance when the test
92
parameters of the 5C-CPT are manipulated (Tomlinson et al., 2014, Tomlinson et al., 2015). It is
our view that this paradigm offers an unbiased method to study attentional deficits of relevance
to several disorders without assuming a mechanism, as is the case with genetic or chemical
manipulations (Hayward et al., 2016).
Effects of a partial agonist at the α7 nAChR on attention and vigilance have yet to be studied
using the 5C-CPT and, as nicotine has a greater effect to improve attention in low-performing
groups, our aim was to compare the effects of encenicline in high and low attentive rats. We
hypothesised that encenicline would improve attention in the 5C-CPT, particularly in low-
attentive animals.
5.5 Experimental Procedures
5.5.1 Animals
37 female Lister Hooded rats (Charles River, UK; weighing 210 ± 20 g at the start of training)
were housed in groups of four or five in individually ventilated cages with two levels (GR1800
Double-Decker Cage, Techniplast, UK) under a standard 12 hour light: dark cycle (lights on
7:00 am). We used female rats as the 5C-CPT has been carefully validated in female rats in our
laboratory (Tomlinson et al., 2014, Tomlinson et al., 2015, Barnes et al., 2012a, Barnes et al.,
2012b). In addition, male rats grow rapidly, which precludes social housing over the lengthy
period of time it takes to train and test rats in the 5C-CPT. Individual housing is stressful for this
social species and best avoided (Holson et al., 1991, Brown and Grunberg, 1995). The
environment was maintained at 21 ± 2 oC, 55 ± 5 % humidity. The diet was standard rat chow
(Special Diet Services, UK) controlled to maintain 90% free feeding weight throughout training
and testing (typically 10 g/rat/day). Water was available ad libitum for the duration of the study,
except in the operant chambers. All experiments were conducted between 09:00-17:00 in the
light phase. All experiments were conducted in accordance with the UK Animals (Scientific
Procedures) 1986 Act and local University ethical guidelines.
5.5.2 Apparatus
Tests were conducted in eight standard nine-hole chambers (Campden Instruments Ltd, UK)
with holes 2, 4, 6, and 8 occluded. Chambers and data collection were controlled by ABet II
Touch software (Lafayette Instrument Company, IL, USA). Animals‟ behaviour was reinforced
with 45 mg rodent reward pellets (Test Diet, MO, USA).
5.5.3 5C-CPT Training Procedure
Animals were trained as described previously (Barnes et al. 2012a; 2012b; Tomlinson et al.
2014). In brief, animals respond to one of five holes when they detect a light stimulus. A nose
poke either while the stimulus is on (stimulus duration (SD)) or within a set period from it first
93
coming on (limited hold) causes release of a food reward. Rare no-go trials are also
interspersed (30% of trials), in these all 5 lights illuminate and no response is required to get a
food reward. Incorrect responses, missed go trials and responded to no-go trials are punished
with a 5 s time out with house light on. When food is dispensed or at the end of a time out the
food tray lights up and entry into the tray starts a timer before the next trial (inter-trial interval
(ITI)). Response during this period is a premature response and also results in a time out.
Animals were trained for approximately 24 weeks to set criteria on final training parameters of 2
s SD, 2 s limited hold, 5 s ITI. This was conducted in a series of stages with successively longer
SD and limited hold where each stage was reached by meeting criteria of >70% accuracy,
<25% omissions and >65% correct rejections (see Barnes et al. (2012a), Barnes et al. (2012b).
Once rats reached these final criteria, they were trained once per week in order to maintain
performance and avoid over-training.
5.5.4 Testing Procedure
Testing parameters were as per final training criteria, with a variable 0.75, 1.25 and 2 s SD and
increased 7-12 s variable ITI. As a previous study found separable genetic effects of α7
receptor mutations at different SDs, a variable SD was used here (Young et al., 2013, Young et
al., 2007). To gather sufficient data for three SDs, the session was extended to 1 hour or 250
trials, whichever came first.
Encenicline ((R)-7-chloro-N-(quinuclidin-3-yl)benzo[b]thiophene-2-carboxamide; kindly donated
by Autifony Therapeutics, Verona, Italy), was prepared at concentrations of 0.1, 0.3 and 1
mg/kg in distilled water and administered by oral gavage 30 min before testing in a volume of 2
ml/kg. Each test day was separated by a seven day washout period (no drug administration, or
training in 5C-CPT). Treatments were randomly assigned based on a within-subjects Latin
square design. Doses were selected from published studies showing efficacy in novel object
recognition in rats (Prickaerts et al., 2012, van Goethem et al., 2015).
5.5.5 Division into Subgroups
Animals were divided into high and low performing groups based on performance when treated
with vehicle (0.9% saline). The best performing 10 animals for d′ and accuracy formed the high-
attentive (HA) group and the worst performing 10 animals formed the low-attentive (LA) group.
This is an upper/lower quartile split, which represents a careful balance between having
significantly different groups and the number of animals used (see Hayward et al. (2016)). To
achieve this, animals were ordered by d′ and the highest/ lowest 10 animals were chosen,
providing their accuracy was also above/below average. As vigilance is a measure of overall
performance across go and no-go trials, accuracy was also used to separate the group into an
attention phenotype. Two animals were excluded for making >80% omissions.
5.5.6 Statistical Analysis
94
The method for calculation of all parameters is described in detail in Tomlinson et al. (2014) and
Young et al. (2013). The signal detection theory measure d′ was used in this case only after
confirming that probability hit rate (pHR) and the probability of false alarms (pFA) were normally
distributed, as described in Young et al. (2013). Here, d′ is calculated in the same way as
described in Young et al. (2013) as the z-score of pHR minus the z-score of pFA.
( ) ( )
( )
Using d′ is a further translational step towards making the rodent 5C-CPT closer to the human
CPT, compared to using sensitivity index (Young et al., 2013). Measures were calculated
individually for each rat at each SD for accuracy, omissions, pHR, pFA, d′ and responsivity
index (RI) where z scores were based on whole cohort data. After relevant measures were
calculated, data were analysed using InVivo Stat (Version 3.4), as the most appropriate
statistical package for in vivo behavioural experiments (Clark et al., 2012). Overall performance
(Table 2) was compared using a one-way Analysis of Variance (ANOVA) comparing treatment
as a factor and using animal ID as a blocking factor to account for the within-subjects design.
Between-group comparisons used least square difference (LSD) tests adjusted for multiple
comparisons by a Benjamini-Hochberg‟s procedure. Measures compared by this method
include accuracy, pFA and d′ for three SDs. Grouped analysis used a two-way repeated
measures mixed model ANOVA with GROUP as the treatment factor and TREATMENT as the
repeated factor for each measurement followed by LSD planned comparisons for each
treatment compared to vehicle. Normality and Sphericity of the data were judged using the
normality plot and the residual vs. predicted plots. Arcsine, square root or Log10 transformations
were used when appropriate. Data are presented as F values and p values for
TREATMENT*GROUP interactions and effects are considered statistically significant if the p-
value from Fisher protected least square difference planned comparisons were < 0.05.
5.6 Results
5.6.1 Group Performance
95
It is important to establish overall group performance before separating the animals into the
groups, particularly when within group reference points are used (Hayward et al., 2016), using
parameters such as upper/lower quartiles (Table 5.1). Comparison for all (n=37) animals
showed no significant difference between treatment levels following a one-way ANOVA with
animal ID as a blocking factor.
Table 5.1: When considered as a single cohort no treatment effect of encenicline was observed.
Measure Variant Vehicle 0.1 mg/kg 0.3 mg/kg 1.0 mg/kg
Accuracy (%) 0.75 s 93±1.5 90±1.6 93±1.5 90±1.5
1.25 s 95±1.2 97±0.62 97±0.57 96±0.95
2 s 96±0.90 97±0.91 98±0.53 97±0.94
Omissions (%) 0.75 s 44±2.8 47±2.6 43±2.6 48±2.6
1.25 s 38±2.9 40±3.0 37±2.8 41±2.8
2 s 36±3.1 38±2.9 33±3.2 36±2.5
pFA 0.75 s 0.23±0.028 0.20 ±0.024 0.21±0.027 0.19±0.024
1.25 s 0.24±0.026 0.22±0.019 0.26±0.027 0.22±0.027
2 s 0.23±0.020 0.21±0.02 0.23±0.023 0.24±0.023
d′ 0.75 s 0.029±0.22 -0.11±0.23 0.059±0.23 -0.053±0.21
1.25 s 0.065±0.22 -0.092±0.18 -
0.033±0.21
-0.077±0.25
2 s -0.086±0.22 0.14±0.23 0.094±0.20 -0.32±0.20
RI 0.75 s 0.32±0.055 0.21±0.056 0.26±0.052 0.18±0.053
1.25 s 0.35±0.048 0.26±0.050 0.38±0.045 0.29±0.043
2 s 0.36±0.046 0.33±0.047 0.39±0.044 0.33±0.048
Trials (Number) 0-20 mins 69±0.98 68±1.1 70±0.98 66±2.0
21-40 mins 68±1.1 69±1.5 69±1.1 67±1.1
96
41-60 mins 56±2.0 54±2.0 56±2.2 56±2.0
Premature
(Number)
- 11±1.4 7.4±1.1 11±2.1 11±1.7
Correct Latency
(s)
- 0.82±0.022 0.85±0.023 0.82±0.023 0.83±0.022
Incorrect Latency
(s)
- 0.95±0.069 1.2±0.054 1.0±0.054 1.0±0.065
Magazine
Latency (Go) (s)
- 1.1±0.024 1.1±0.038 1.0±0.029 1.0±0.028
False Alarm
Latency (s)
- 0.74 ±0.031 0.75±0.031 0.73±0.032 0.71±0.030
Magazine
Latency (No Go)
(s)
- 10±1.5 9.8±1.4 11±1.5 13.±1.8
5.6.2 High/ Low Comparison
Comparison of the main measures (Accuracy, pFA and d′; Figure 5.1) using adjusted LSD tests
found that HA and LA groups differed significantly in accuracy and d′ at 0.75 (both p<0.01) and
1.25 s (both p<0.05) SD. The data show most robust separation at the most challenging
(shortest) SD (0.75 s).
97
V e h ic le A c c u ra c y
0.7
5
1.2
5
2.0
0
7 0
8 0
9 0
1 0 0
1 1 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e (
%)
** *
V e h ic le d '
0.7
5
1.2
5
2.0
0
-1 .5
-1 .0
-0 .5
0 .0
0 .5
1 .0
S tim u lu s D u ra tio n
Sc
ore
** *
L A
H A
A B
Figure 5.1: Group differences between high and low attentive groups (HA/LA) at increasing SDs.
During saline treatment, the difference between HA/LA is most significant at the most challenging (i.e.,
shortest) stimulus duration (0.75 s), with LA showing significantly lower attention (Accuracy; A) and d′
(vigilance; B). Data are shown as mean ± SEM. n=10 HA and n=10 LA animals. *, **, denote significance
levels p< 0.05, 0.01 of Benjamini-Hochberg‟s procedure adjusted LSD planned comparisons comparing
HA with LA groups at each SD.
A c c u ra c y H A
0.7
5
1.2
5
2.0
0
7 0
8 0
9 0
1 0 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e *
O m is s io n H A
0.7
5
1.2
5
2.0
0
0
2 0
4 0
6 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e
**
d ' H A
0.7
5
1.2
5
2.0
0
-1 .5
-1 .0
-0 .5
0 .0
0 .5
1 .0
1 .5
V e h ic le
E V P 0 .1 m g /k g
E V P 0 .3 m g /k g
E V P 1 m g /k g
S tim u lu s D u ra tio n
Sc
ore
** *
A c c u ra c y L A
0.7
5
1.2
5
2.0
0
7 0
8 0
9 0
1 0 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e
*
O m is s io n L A
0.7
5
1.2
5
2.0
0
0
2 0
4 0
6 0
S tim u lu s D u ra tio n
Pe
rc
en
tag
e
*
d ' L A
0.7
5
1.2
5
2.0
0
-1 .5
-1 .0
-0 .5
0 .0
0 .5
1 .0
1 .5
S tim u lu s D u ra tio n
Sc
ore
**
A B C
D E F
Figure 5.2: Encenicline improved selective attention and vigilance in low attentive (LA) animals
only. In HA animals, 1 mg/kg encenicline significantly reduced performance in accuracy (A), omissions (B)
and d′ (D) at 0.75 s SD, d′ was also reduced by the same dose at 1.25 s SD. In LA animals, accuracy
(selective attention; D) was significantly increased at 0.3 mg/kg for 0.75, d′ (F) is significantly increased at
0.3 and 1 mg/kg of encenicline at 0.75 s SD and omissions (E) are increased at 0.1 mg/kg at 0.75 s SD.
Data are shown as mean ± SEM of n=10 HA and n=10 LA animals each tested at all doses. * and **
denote significance levels p< 0.05, 0.01 respectively for LSD planned comparisons of vehicle treated
98
animals to the three treatment levels following significance in GROUP*TREATMENT interaction in a
repeated measures two way ANOVA.
In HA animals encenicline impaired performance in a dose-dependent manner as demonstrated
by significantly reduced accuracy at 0.75 SD (Figure 5.2A; F3,54=3.91,p=0.013) an effect that
was significant at 1 mg/kg (p<0.05) of encenicline compared with the vehicle-treated group,
Figure 2. A significant increase in omissions was also observed at 0.75 s SD (Figure 5.2B; F3,
54=2.99, p=0.039) again significant at 1 mg/kg of encenicline (p<0.01). These two factors led to
a reduction of the signal detection theory parameter d′ (d prime) at all SDs: 0.75 (Figure 5.2C;
F3, 54=9.34, p<0.0001), 1.25 s (F3, 54=3.24, p=0.029) 2 s (F3,54=3.11,p=0.034), planned
comparisons showed significantly reduced d‟ for 0.75 and 1.25 s SDs at the 1 mg/kg dose
(p<0.01 and p<0.05 respectively). As d′ is a robust measure of vigilance, this suggests that the
HA animals became less vigilant as well as less attentive when treated acutely with 1 mg/kg of
encenicline. These effects, combined with increased omissions and a lack of significant
changes in any of the latency measures and total trials completed suggests that partial agonism
of 7 nicotinic receptors induces a reduction in the ability to sustain attention throughout the
task in HA animals.
In contrast, encenicline enhanced performance in LA animals as shown by a significant
increase in accuracy at 0.75 s SD (Figure 5.2D; F3,54=3.1, p=0.034), an effect that was
significant at 0.3 mg/kg of encenicline (p<0.05). This change corresponds to an increase in
selective attention (Robbins, 2002). LA animals also showed increased omissions at 0.75 s SD
(Figure 5.2E; F3, 54=2.99, p=0.039). This effect was significant following treatment with the
lowest dose of 0.1 mg/kg of encenicline (p<0.05), this dose did not produce any other significant
effects on attentive measures in either HA or LA animals. Similar to accuracy, d′ showed the
opposite effect in LA compared with HA animals and was increased (i.e., vigilance was
enhanced) at 0.75 s SD (Figure 5.2F; F3, 54=9.34, p<0.0001) with a significant effect produced
at 0.3 and 1 mg/kg of encenicline (p<0.05).
99
p F A H A
0.7
5
1.2
5
2.0
0
0 .0
0 .1
0 .2
0 .3
E V P 0 .3 m g /k g
E V P 1 m g /k g
V e h ic le
E V P 0 .1 m g /k g
S tim u lu s D u ra tio n
Pro
ba
bil
ity
p F A L A
0.7
5
1.2
5
2.0
0
0 .0
0 .1
0 .2
0 .3
S tim u lu s D u ra tio n
Pro
ba
bil
ity
*
P re m a tu r e
HA
LA
0 .0
2 .5
5 .0
7 .5
1 0 .0
1 2 .5
1 5 .0
1 7 .5
Nu
mb
er
A
B C
Figure 5.3: Encenecline reduces impulsivity in LA animals at 0.75 s SD. In HA animals pFA (A)
trended to significance for 0.75 s SD and 0.1 mg/kg (p=0.081). The probability of false alarms (pFA; B)
significantly reduced in low attentive animals (LA) at 1 mg/kg encenicline at 0.75 SD. No significant effects
were seen on premature responses. Data are shown as mean ± SEM of n=10 HA and n=10 LA animals
each tested at all doses. *, denotes significance levels p< 0.05 for LSD planned comparisons of vehicle
treated animals to the three treatment levels following significance in GROUP*TREATMENT interaction in
a repeated measures two-way ANOVA.
The impulsivity measure, pFA, was significantly reduced in LA animals at 0.75 s SD (Figure
5.3B; F3, 54=3.1, p=0.034) following treatment with 1 mg/kg encenicline (p<0.05) compared to
the saline control group. As false alarms represent an impulsive behaviour, this shows reduced
response disinhibition, i.e., enhanced inhibitory control (for impulsive action).
100
T r ia ls C o m p le te d H A
1-2
0
21-4
0
41-6
0
0
2 5
5 0
7 5
T im e (m in s )
Nu
mb
er
T r ia ls C o m p le te d L A
1-2
0
21-4
0
41-6
0
0
2 5
5 0
7 5
T im e (m in s )
Nu
mb
er
V e h ic le
E V P 0 .1 m g /k g
E V P 0 .3 m g /k g
E V P 1 m g /k g
A B
C o rr e c t L a te n c y
HA
LA
0 .6
0 .7
0 .8
0 .9
1 .0
Tim
e (
s)
In c o r re c t L a te n c y
HA
LA
0 .0 0
0 .2 5
0 .5 0
0 .7 5
1 .0 0
1 .2 5
Tim
e (
s)
F a ls e A la rm L a te n c y
HA
LA
0 .0
0 .1
0 .2
0 .3
0 .4
0 .5
0 .6
0 .7
0 .8
0 .9
Tim
e (
s)
M a g a z in e L a te n c y (G o )
HA
LA
0 .0 0
0 .2 5
0 .5 0
0 .7 5
1 .0 0
1 .2 5
Tim
e (
s)
M a g a z in e L a te n c y (N o G o )
HA
LA
0
5
1 0
1 5
2 0
Tim
e (
s)
V e h ic le
E V P 0 .1 m g /k g
E V P 0 .3 m g /k g
E V P 1 m g /k g
C D E
F G
Figure 5.4: Latency measures of the 5C-CPT showed no significant TREATMENT*GROUP
interaction. Data are shown as mean ± SEM of n=10 HA and n=10 LA animals each tested at all doses.
No significant effect was seen on the number of trials completed or on any of the latency
measures for GROUP*TREATMENT interaction (Figure 5.4). This is most likely due to high
within-subject variability in these measures as when comparisons between subjects were made
(LSD) they were significant for „no go‟ magazine latency and incorrect latency. However, the
design of this study uses Fisher protection to reduce false positives, so these are not
considered valid differences.
5.7 Discussion
To study the effect of encenicline in high and low attentive animals we firstly established that,
after grouping, the animals were significantly different in the measures for attention and
vigilance. Treatment with encenicline showed opposite effects between HA and LA groups for
accuracy, the correlate of selective attention, where LA animals had enhanced selective
attention and HA animals had reduced selective attention. A similar effect was observed for
vigilance via the measure d′. For both of these measures 0.3 mg/kg was the optimal dose, i.e.,
was most effective at the most challenging SD of 0.75 s. An effect was also observed on pFA,
101
but not on premature responses (both measures of impulsive action), indeed these have
previously been shown to be separable constructs pharmacologically and genetically (Young et
al., 2011). Therefore, our present study shows that LA animals are more able to withhold from
responding to a prepotent stimulus but are not significantly different in their ability to wait for the
stimulus when treated with encenicline. When assessing false alarms, it is important to evaluate
whether reduced responses in „no go‟ trials are also seen in „go‟ trials, reflecting reduced
motivation rather than enhanced response inhibition. Changes in omissions and pFA occurred
at different concentrations of the drug suggesting that these effects are independent of one
another. The changes we observed in omissions were apparent only for the challenging 0.75 s
SD and were observed at 1 mg/kg in HA animals and 0.1 mg/kg in LA animals. Changes in
omissions can reflect sustained attention, motivation, sedative or motor effects; in order to
ascertain which of these are affected other measures are required (Robbins, 2002). Sedative
and motor effects should be detected as changes across the latencies; motivational changes
would alter the magazine latencies. Incorrect latency and no go magazine latency were
significantly different in planned comparisons, but not via the GROUP*TREATMENT interaction
(Figure 5.4). This is most likely due to high within-subject variability, which is only accounted for
in the repeated measures ANOVA. This high variability would require larger group sizes in order
to determine whether an effect is robust. However, sedative or motor effects would also be
expected to alter correct latency and motivational effects would be expected to alter „go‟
magazine latency. As neither of these was observed, the changes in motivation may represent
lapses in the ability to sustain attention; however, this requires further testing. The difference in
dose required to elicit the effect in HA and LA animals may represent differences in
neurobiology between the groups, in agreement with opposing effects seen on other measures.
In summary, encenicline improves selective attention and vigilance in LA animals at 0.3 mg/kg
and reduces impulsivity at 1 mg/kg.
As we had extended the number of trials to 250 rather than the standard 120 trials, satiety/
reduced motivation are important factors for consideration. Looking at Table 1 and Figure 4 A
and B the number of trials is lower at the final time point. However this was not different
between treatment groups suggesting that it did not contribute significantly to the drug effect.
This conclusion is further supported by our observation of no significant effect between
treatments on magazine latency, a measure of satiety. Using a variable SD enabled us to
sample a span of difficulty levels, however, in the 2 second SD, both groups showed >90%
accuracy suggesting that a ceiling effect may play a role at this SD preventing increases from
being detected.
The 5-HT3 receptor is a homologue of the α7 receptor and so α7 agonists often show
appreciable affinity and act as antagonists at this receptor subtype (Prickaerts et al., 2012).
However, the selective 5-HT3 receptor antagonist ondansetron does not improve attention in
rats in the 5C-SRTT, suggesting that attention promoting effects seen here are not due to an
interaction with the 5-HT3 receptor (Muir et al., 1995).
102
Increased accuracy at 0.75 s stimulus duration in LA animals is interesting as it seems to follow
an inverted U-shaped dose response trend which has been reported for other nicotinic
compounds (Kolisnyk et al., 2015, Olincy and Freedman, 2012, Picciotto, 2003). A recent
microdialysis study by Huang et al. (2014) found that the effect of encenicline on
neurotransmitter efflux in the medial prefrontal cortex differs depending on dose. In that study,
0.1 mg/kg of encenicline increased dopamine, acetylcholine and glutamate release in the
prefrontal cortex, whereas only acetylcholine was increased at the higher dose of 0.3 mg/kg.
This may be due to receptor populations existing both pre- and post-synaptically, giving the α7
receptor a modulatory role which differs depending on the concentration of acetylcholine at the
synapse (Vizi and Lendvai, 1999). An inverted U-shaped trend (Yerkes-Dodson effect) is also a
property of the arousal of attention (Wood et al., 2014) which has been linked to prefrontal
monoamine levels (Arnsten, 2009). The theory of optimal arousal and neurotransmitter
differences may contribute to the outcomes of this study and would make it interesting to see a
combined microdialysis or fast scan voltammetry and attentive task study using encenicline,
particularly comparing high/low performing animals.
A previous study found no effect of the α7 full agonist, PNU-282987 in the 5C-CPT in either
untreated mice or in those with a scopolamine-induced attention/ vigilance deficit, in agreement
with our study when we considered the entire cohort together (Young et al., 2013). However,
once we separated animals into HA and LA groups, the α7 receptor partial agonist encenicline
clearly improved attention, vigilance and response inhibition in LA animals. There are two key
differences between the work of Young et al. (2013) and the present study; firstly, the model
used in our study is a behavioural separation, whereas Young et al. (2013) used a
pharmacologically-mediated deficit. Scopolamine has a number of non-specific effects, which
may also affect performance, such as increased locomotor activity, pupil dilation and mnemonic
changes (Klinkenberg and Blokland, 2010). The second difference is the mechanism of action
of PNU-282987, which is a full agonist, whereas encenicline is a partial agonist, allowing it to
act as an agonist or antagonist, depending on endogenous acetylcholine levels (Prickaerts et
al., 2012, Hajos et al., 2005). Furthermore, as outlined earlier, other α7 nAChR partial agonists
have been shown to improve attention (Rezvani et al., 2009, Haig et al., 2016, Shiina et al.,
2010), but have not been tested in specific tasks of attention/ vigilance and not in animals
separated according to performance.
In a small study of schizophrenia patients, encenicline added onto regular medication had no
effect on the attentive measures in the CogState schizophrenia battery compared to placebo
(Preskorn et al., 2014). The small sample size in this study where the placebo group composed
of 4 patients was a serious limitation and the authors recommend caution for these measures.
Indeed, a larger study, using the same doses found significant effects of encenicline across a
broad range of cognitive tests, including attention, using the CogState battery (Keefe et al.,
2015). In a small pilot study in schizophrenia patients both mismatch negativity (a negative
potential following an odd stimulus in a sequence) and P300 (positive inflection after a standard
103
stimulus presentation) which can be correlated to the severity of cognitive deficits (Ford et al.,
1999, Baldeweg et al., 2004) were shown to be normalised by encenicline (Preskorn et al.,
2014). However, a recent phase III trial did not reach significance for primary cognitive
endpoints in schizophrenia patients compared to placebo (Pharmaceuticals, 2016). Also, in the
Alzheimer‟s disease Phase III trial, rare, but serious, gastrointestinal side effects have put
testing on hold (Pharmaceuticals, 2015). These findings suggest that encenicline itself is not
likely to be used clinically in either of these conditions, perhaps due to specific properties of the
molecule, rather than the target itself.
Our data and other studies suggest that the α7 receptor is a valuable target for improving
attention in disorders where this is impaired, and supports efforts to find a clinically-viable
molecule. Development is still on-going for new α7 partial agonists which may prove more
successful (Bristow et al., 2016). An alternative route of investigation is the positive allosteric
modulators of the α7 receptor but these have yet to be tested in attention tasks. It is important
to note that no drug has yet received a license specifically for improving cognition in
schizophrenia; this is clearly a particularly challenging unmet clinical need. In our view, it is most
likely that pharmacological combined with psychological and other interventions, such as
cognitive remediation and exercise therapy, providing a holistic approach to treatment is most
likely to produce positive effects for this symptom domain and improve quality of life for patients.
In summary, this study demonstrates, for the first time, that separation by performance in the
5C-CPT is feasible using a variable SD. The difference in performance was greatest when the
animals were most challenged; in this case with a 0.75 s SD. Encenicline showed contrasting
effects on attention and vigilance in LA and HA animals. In LA animals, encenicline improves
selective attention, vigilance and response inhibition (impulsive action), but in HA animals
selective attention and vigilance are reduced. These findings show for the first time, that an α7
nAChR partial agonist improves attention when tested in a specific task of attention designed for
high translation (5C-CPT) and that this depends on baseline levels of attention. Our data
suggests that partial agonism at the α7 nAChR has the potential to enhance attention in patients
stratified according to low attentional performance.
104
6 Frontoparietal neural activation correlates
with vigilance in the 5C-CPT
Authors: Andrew Hayward, Ines Das Neves, Joanna C. Neill , John
Gigg
[Unplublished data]
6.1 Contributions
Andrew Hayward, John Gigg and Joanna C. Neill planned the study. Training of the animals
was completed by Andrew Hayward. The behavioural aspects of the study were conducted by
Andrew Hayward; histological aspects were completed by Ines Das Neves and Andrew
Hayward. The manuscript was prepared by Andrew Hayward and John Gigg.
105
6.2 Context
The previous chapters have focused on pharmacology in the 5C-CPT. Here we aim to establish
if the frontoparietal network is involved in 5C-CPT task performance as an initial step towards
understanding the pharmacological response to these compounds.
6.3 Abstract
The medial prefrontal, cingulate and posterior parietal cortices form the frontoparietal network,
which is essential for orienting attention. Vigilance is the ability to adaptively allocate attention
and respond appropriately to unpredictable stimuli. This process is impaired in clinical
conditions such as attention deficit hyperactivity disorder, schizophrenia and bipolar disorder.
Vigilance is assessed clinically using the continuous performance task (CPT). This has been
translated to rodents as the 5 choice continuous performance task (5C-CPT). As the ability to
translate between rodent and human research is key to drug discovery, a reverse translation of
the 5C-CPT has been produced. In both the reverse translated and widely used Conner‟s CPT
the frontoparietal network is engaged in the human brain. To date, this network has not been
probed in rodents using 5C-CPT. We aimed, therefore, to assess whether task engagement in
the 5C-CPT correlates with activation of the frontoparietal network as shown previously shown
for humans. Female Lister Hooded rats (n=30) were trained to criterion in the 5C-CPT. We then
assessed whether their performance measures in the 5C-CPT (e.g., vigilance; d‟) correlated
with neuronal activity in prelimbic/infralimbic and parietal cortices, as measured by expression of
the immediate early gene c-Fos. C-Fos staining showed a significant correlation with the
vigilance parameter d′. The probability of false alarms (pFA) showed a negative correlation,
which approached significance and was more pronounced in the prefrontal than the posterior
parietal cortex. No correlation was seen for hit rate (pHR) or number of premature responses.
This study identifies the importance of the rat medial prefrontal (mPFC) and posterior parietal
cortex (PPC) for task engagement during the 5C-CPT. In particular, cellular activity correlates
positively with vigilance. This suggests that as task engagement/performance increases this
requires a further matched recruitment of mPFC and PPC circuits. This cellular metric is in
agreement with fMRI studies in clinical versions of the task. Therefore, this study further
supports a vital function of these regions in sustained attention as well as the utility of the 5C-
CPT as a highly translatable task.
106
6.4 Introduction
Vigilance is the ability to remain alert to changing situations to optimally either elicit or withhold
a response to a given stimulus, also referred to as “psychological readiness”. Its importance
was highlighted for the first time when studying the performance of radar operators in World
War II who had to monitor radar outputs over extended periods for rare but important events
(Mackworth, 1948). Vigilance is strongly related to attention and is often impaired in psychiatric
conditions such as attention deficit/ hyperactivity disorder (ADHD) (Huang-Pollock et al., 2012,
Riccio et al., 2002) and schizophrenia (Nuechterlein et al., 2015, Luck and Gold, 2008). The
continuous performance task is the most widely used paradigm to measure vigilance in human
subjects (d prime; d′) alongside selective attention (accuracy of response) (Riccio et al., 2002,
Nuechterlein et al., 2008).
Increased activity in the frontoparietal network, as assessed by functional magnetic resonance
imaging (fMRI), has been observed across a diverse array of cognitive functions, including
memory, arithmetic and language, suggesting that it is fundamental for engagement in many
cognitive tasks (Duncan, 2010, 2013). This network comprises the dorsolateral prefrontal,
anterior cingulate and posterior parietal cortices (Petersen and Posner, 2012, Duncan, 2010). A
clinical study by Fedorenko et al. (2013) compared fMRI activity patterns across a similarly
diverse array of tasks, where each had an easy and a difficult version. This showed that more
challenging tasks caused corresponding elevated activity in the frontoparietal network,
presumably due to the higher level of engagement required. Convergent research shows that
the main function of the frontoparietal network is to decide where to orient attention by
combining prefrontal „top-down‟ goal-directed information with a saliency map of the
environment produced by the posterior parietal cortex (Ptak, 2012, Marshall et al., 2015,
Petersen and Posner, 2012, Buschman and Miller, 2007).
In rodents, the frontoparietal network appears to function in a similar way to that in primates,
with increased neuronal firing in the posterior parietal cortex (PPC) when an animal attends to a
stimulus (Broussard, 2012). This is shown by PPC neurones in a sustained attention task firing
in association with successful responses to stimuli compared to misses and further reinforced
by higher firing in response to task-relevant stimuli compared to visually similar distractors
(Broussard et al., 2006, Broussard, 2012). The 5 choice continuous performance task (5C-CPT)
is a translational test used in rodents to assess attention, vigilance and impulsivity (Young et al.,
2009). The task involves either responding to a single visual stimulus („go‟) or withholding a
response when multiple visual stimuli co-occur („no-go‟). The 5C-CPT is the only task to
measure attention, impulsive action and vigilance in rodents simultaneously. As it is based on
the clinical CPT, it has high translational validity and relevance (Lustig et al., 2013, McKenna et
al., 2013, Young et al., 2009). To highlight the latter, McKenna et al. (2013) produced a reverse-
translated version of the 5C-CPT for human use. With this version of the task, they used fMRI to
analyse the functional correlates of vigilance in humans. They found differential activation of the
107
frontoparietal network by the target and non-target trials, demonstrating that the posterior
parietal cortex is activated in both trial types and the dorsolateral prefrontal cortex is activated
during non-target trials (McKenna et al., 2013). This is in agreement with an fMRI study which
used the Conner‟s CPT and also showed activation in both prefrontal and posterior parietal
cortex, thus, frontoparietal engagement in the task (Ogg et al., 2008). As the frontoparietal
network has not been studied previously in rodents for the 5C-CPT, this result is ripe for
translation, which would allow comparison of rodent and human functional circuitry as well as an
assessment of the translational validity of the task.
This study aimed to add further translational validity to the 5C-CPT by assessing the role of the
frontoparietal network in the 5C-CPT. For this two key hypotheses were tested: (1) 5C-CPT
performance will engage neuronal activity across the frontoparietal network. (2) the level of
neuronal activity will correlate positively with successful performance of the task. To detect
frontoparietal engagement, we measured immunohistochemical staining for the immediate early
gene c-fos, which is widely used to indicate neuronal activation or plasticity (Burnham et al.,
2010, Sagar et al., 1988).
6.5 Methods
6.5.1 Animals
Female Lister Hooded rats (Charles River, UK; n=30, weighing 215 ± 20 g at the start of
training) were housed in groups of four or five in individually ventilated cages with two levels
(GR1800 Double-Decker Cage, Techniplast, UK) under a standard 12 hour light: dark cycle
(lights on 7:00 am). The environment was maintained at 21 ± 2 °C, 55 ± 5 % humidity. Animals
were food restricted to 90% free-feeding body weight (usually 10g/rat/day) for training and
testing. All experiments were conducted in accordance with the UK Animals (Scientific
Procedures) 1986 Act and local University ethical guidelines and permissions.
6.5.2 5C-CPT Apparatus, Training and Testing
Standard nine-hole operant conditioning chambers (Campden Instruments Ltd, UK) were
controlled by ABet II Standard software (Lafayette Instrument Company, IL, USA) and used as
previously described (Hayward et al., 2017). Final task parameters (2s stimulus duration, 2s
limited hold, 5s inter-trial interval) were reached over a training period of approximately 24
weeks (Barnes et al., 2012a). Animals were rewarded with 45 mg rodent pellets (Test Diet, MO,
USA). After meeting final criteria of 75% accuracy, <40% false alarms and >20% omissions,
animals were trained once per week to maintain performance and avoid over-training. Animals
were then randomised and allocated in a random but counterbalanced (by task performance)
way to two groups. One group was tested with a standard 5C-CPT procedure (5s limited hold,
1s stimulus duration and 10s variable inter-trial interval; n=23) and the other housed individually
108
in their home cage for the same period as a control (n=7). The order in which animals were
tested was randomised.
6.5.3 Tissue Extraction
One hour after testing or control situation, animals were deeply anaesthetised with urethane
(30%, w/v in 0.9% saline, 2 g/kg) and perfused with phosphate buffered saline (1x) followed by
paraformaldehyde (4% w/v) and brains were extracted. Brains were stored in paraformaldehyde
for 48 hours, then sucrose solution (30% w/v) for 24 hours, rapidly frozen in isopentane (Fisher
Scientific, UK) and stored at -80 oC until sectioned. Free-floating sections (40μm) were prepared
using a freezing sledge microtome (Leica, UK) and stored in a cryoprotectant solution (30%
Glycerol, 30% Glycol Ethylene, 40% PBS 1x) at -20oC until staining. The brains were
randomised by animal id before slicing.
6.5.4 Immunohistochemistry
Sections were selected for staining based on the rat brain atlas (Paxinos, 2006) to include
medial prefrontal cortex (mPFC; a combination of cingulate cortex 1 (Cg1) and prelimbic cortex
(PrL)) and posterior parietal cortex (PPC; medial parietal associative cortex; MPtA) (Figure 6.1).
These regions were chosen based on anatomical and functional evidence that they represent
homologues of the human posterior parietal cortex (Corwin and Reep, 1998, Broussard, 2012)
and prefrontal cortex (Uylings et al., 2003). Tissue was randomised by rat identification number
into four batches for staining.
To stain the tissue for c-Fos activity, sections were first washed in 0.1M PBS with 0.1% Triton
(PBS-T) and then immersed in 0.3% hydrogen peroxide (Riedel-de Haën, Germany) diluted in
100% methanol for 15 minutes, to block endogenous peroxidase activity. This was followed by
3x five-minute washes in PBS-T. Sections were then incubated for an hour in a blocking
solution of 10% normal goat serum (Vector Laboratories, UK) and 1% bovine serum albumin
(BSA; Sigma-Aldrich, UK) in PBS-T. Primary incubation with the c-Fos antibody (sc-52
polyclonal rabbit anti c-Fos, Santa Cruz Biotechnology, USA; 1:1000) was carried out overnight
at 4oC. This dilution factor was selected following a series of pilot experiments. After 3x five-
minute washes in PBS-T, sections were incubated with the secondary antibody (Biotinylated
anti-rabbit Ig6 (H+L), Vector Laboratories, UK; 1:200) for 1h at 4°C. Both the primary and
secondary antibodies were used in solution with 1% Goat Serum and 1% BSA in PBS-T.
Following 3x ten-minute washes in PBS-T, an ABC kit solution (Vectastain ABC Kit, Elite PK-
600, standard; Vector Laboratories, UK) was applied for 30 minutes. After 3x ten-minute
washes, a diaminobenzidine solution (DAB substrate kit for peroxidases, sk-4100; Vector
Laboratories, UK) was applied until the tissue had darkened (1.5-2min) and brain sections were
then washed in distilled water. All steps in the staining process were carried out under light
agitation. Once three washes had been completed, the stained sections were mounted onto
polylysine slides (Fisher Scientific, UK) and sequentially dehydrated in 70%, 90% and 100%
109
ethanol (5 minutes per concentration). The clearing agent, Histo-clear II (National Diagnostics,
UK) was applied for 5 minutes, and the slides were finally coverslipped with DPX (Sigma-
Aldrich, UK) as the mounting medium.
Figure 6.1: Regions analysed for C-Fos quantification. Representative brain sections to delineate
regions (grey) analysed for c-Fos-positive cells in mPFC (PrL and Cg1) and posterior parietal cortex
(PPC). (B) A representative example of c-Fos-positive staining from the prefrontal cortex. Key: Cg1,
cingulate; PrL, prelimbic cortex; PPC, posterior parietal cortex. The level of sections labelled relative to
Bregma. Adapted from Paxinos (2006).
110
6.5.1 Stain Optimisation
To maximise stain quality the concentration of primary antibody, length of incubation and
distribution of sections in staining wells was optimised. In the first optimisation experiment, the
procedure outlined in the immunohistochemistry section was carried out using primary antibody
concentrations of 1:50, 1:300, and 1:500. For this, rat brain sections that did not contain the
areas of interest for the study were selected from 4 animals. For each animal, 3 sections of
brain tissue were stained at each of the selected antibody concentrations. A comparison of 24
or 48 hour incubation was conducted for each concentration. As staining for the two conditions
was similar, 24 hours was used to improve speed and tissue integrity. Frontal and parietal
sections (from the same level as those to be used in the main experiment) were selected from 2
of the four animals. The sections from one animal were separated into two wells; 6 frontal
sections in one and 3 parietal sections in the other. The second animal‟s sections were
combined in the same well. The tissue was incubated in primary antibody (1:300) for 24 hours.
The combined method resulted in uneven staining, so frontal and parietal tissues were
separated in further staining.
The second optimisation experiment was based on the results of the first. Here, brain tissue
was stained at the following concentrations: 0 (negative control), 1:300, 1:500 and 1:1000.
Primary incubation was carried out overnight. All sections used in this experiment were from
between the frontal and parietal regions. The 1:1000 primary antibody concentration provided
the clearest distinction between nuclei and background and so was used in the main
experiment. The negative control showed no specific staining.
6.5.2 Quantification
Staining and counting were performed blind to animal grouping and task performance to prevent
experimenter bias. Images were acquired via an Olympus BX51 microscope and counting was
conducted online to allow adjustment of illumination/focus and improve identification (Image-Pro
Plus v6.3, Media Cybernetics Inc.). Regions were identified bilaterally and selected at 4x
magnification. Counting was completed on a median of sixty 120x120μm squares randomly
positioned bilaterally within each region of interest and spread across three sections for each
region for each rat (Figure 6.1). Cells were counted and the median for each region (mPFC and
PPC) calculated.
6.5.3 Statistical Analysis
The method for calculation of all parameters in relation to behavioural performance was as
described in detail in Tomlinson et al. (2014) and Young et al. (2013). The signal detection
theory measure d′ was used in this case only after confirming that probability of hit rate (pHR)
and false alarms (pFA) (see below) was normally distributed (Young et al. (2013). d′ was
calculated in the same way as by Young et al. (2013) as the z score of pHR minus the z-score
of pFA.
111
( ) ( )
As d′ is used as the measure of vigilance in clinical CPT, the use here is a further step to
improve translatability of preclinical research compared to using the sensitivity index (Young et
al., 2013).
Statistical tests were conducted using InVivoStat v3.4 (Clark et al., 2012). Graphs were
produced in GraphPad Prism for Windows 7 (GraphPad Software, USA). A repeated measures
Analysis of Covariance (ANCOVA) was used to compare the tested (n=23) to non-tested (n=7)
animals, with testing as a between-subject factor and region (mPFC vs. PPC) as the within-
subject factor. The four staining runs were used as a blocking factor to account for any variation
due to differential staining and count order was used as a covariate due to a trend in count
order and the number of cells (Figure 6.2). The analysis used Fisher protection as this only
reports the results of planned comparisons if model effects were significant. Thus, it reduces the
number of comparisons and, therefore, the chance of false positives. For all comparisons
normality of data was judged using Q-Q plots and sphericity of data by residual vs. predicted
plots.
0 1 0 2 0 3 0
0
2 0
4 0
6 0
8 0
C o v a r ia te
C o u n t O rd e r
C-F
os
po
sit
ive
ce
lls
(m
ed
ian
)
m P F C
P P C
Figure 6.2: Correlation between count order and the number of identified c-Fos-positive neurones.
A practice effect was clear where the number of identified cells increased with further assessments of
sections. Due to randomisation of count order, this did not bias the data; also count order was used as a
covariate to account for this trend statistically.
To analyse correlation of c-Fos-positive staining with task performance, a multivariate analysis
of covariance (MANCOVA) was used. In this, region (mPFC vs PPC) was compared to the
continuous variable d′, p[FA], p[HR] or premature responses; the blocking factor of staining
112
batch and the covariate of count order were also used as in the previous comparison. For
reporting MANCOVA results, the r2 value was used to show how much variance is explained by
this model. The r2 was adjusted for the number of variables used. The significance of any factor
in the model is reported as F value with degrees of freedom and p-value.
6.6 Results
Significant model effects were seen for region (Figure 6.3; F1,26=29.95,p<0.0001) as well as the
interaction between testing and region (F1,26=6.8, p=0.015). Planned comparisons were
significantly different between frontal and parietal in tested (p<0.001) and non-tested animals
(p<0.01). However, whether rats were from tested or home cage groups had no effect on the
number of c-Fos-positive cells (Figure 6.3; F1,26=0.01, p=0.93).
m P F C P P C
0
2 0
4 0
6 0
8 0
1 0 0
C-F
os
po
sit
ive
ce
lls
(m
ed
ian
)
T ested
N o n -T e s ted******
Figure 6.3: C-Fos-positive cell count was higher in PPC compared to PFC. Tested animals (n=23)
and non-tested group (n=7). A significant difference was seen between prefrontal cortex (mPFC) and
posterior parietal cortex (PPC). C-Fos-positive cell count was similar between tested and non-tested
groups. Boxes extend from the lower to the upper quartile of each data set; the midline corresponds to the
median and whiskers indicate the data range (min to max). *** denotes the model level effect of p<0.001
in the factor REGION.
For d′ the MANCOVA model fit the data well (r2=0.741) and showed a significant positive
correlation between the number of c-Fos-positive cells and the vigilance measure d′ (Figure
6.4C; F1=12.57, p=0.0011), where cell counts increased with d′. There was no significant
difference between regions (F1=2.42, p=0.1285) or any interaction between region and d′
(F1=0.319, p=0.94).
No significant influence of pHR (r2= 0.6632, F1=0.37, p=0.55) or premature response was seen
(r2=0.6733, F1=2.5, p=0.1222). For pFA, the model fits well (r
2=0.6911), and a negative
correlation of less staining with higher pFA was observed that closely approached significance
(F1= 3.77, p=0.0599). While the interaction between region and pFA was not significant, it can
be seen that this trend was more pronounced in the mPFC than the PPC (Figure 6.4).
113
0 .0 0 .2 0 .4 0 .6 0 .8
0
2 0
4 0
6 0
8 0
1 0 0
p H R
p H R
C-F
os
po
sit
ive
ce
lls
(m
ed
ian
)
m P F C
P P C
r2= 0 .6 6 3 2 p = 0 .5 4 6 3
A
0 .0 0 .2 0 .4 0 .6 0 .8
0
2 0
4 0
6 0
8 0
1 0 0
p F A
p F A
C-F
os
po
sit
ive
ce
lls
(m
ed
ian
)
r2= 0 .6 9 1 1 p = 0 .0 5 9 9
B
-2 -1 0 1 2
0
2 0
4 0
6 0
8 0
1 0 0
d '
d '
C-F
os
po
sit
ive
ce
lls
(m
ed
ian
)
r2= 0 .7 4 1 p < 0 .0 1
C
0 5 1 0 1 5 2 0 2 5
0
2 0
4 0
6 0
8 0
1 0 0
P re m a tu r e R e s p o n s e s
P re m a tu re R e s p o n s e sC
-Fo
s p
os
itiv
e c
ell
s (
me
dia
n)
r2= 0 .6 7 3 3 p = 0 .1 2 2 2
D
Figure 6.4: Vigilance (d′) correlates positively with the number of c-Fos-positive cells in mPFC and
PPC (C). (A) There was no significant correlation in tested rats between pHR and number of c-Fos-positive
cells. (B) A negative correlation trended towards significance for pFA, and the correlation was stronger in
the PFC than the PPC. (C) A significant positive correlation was seen comparing d′ to the number of c-
Fos-positive cells, which shows that animals with higher d′ also had higher abundance of c-Fos-positive
cells. (D) No significant correlation was seen when comparing the number of premature responses to c-
Fos-positive cell count. R2 values were adjusted to account for the number of comparisons and denote
how well the data fit the ANCOVA model, p values denote the significance of the model effect for each
behavioural measure.
6.7 Discussion
This study has shown a positive correlation between vigilance (d′) and c-Fos expression in
mPFC and PPC. Thus, activation of the frontoparietal network (mPFC and PPC) of the rat
during 5C-CPT correlates with performance in the task. This shows, as hypothesised that neural
activity is linked to and correlated with performance in the 5C-CPT. However; the study also
revealed no difference in c-Fos-positive cell count between tested and untested animals, which
suggests the system is not involved in task performance. Based on previous studies
investigating neurotransmitter release (Broussard, 2012, Dalley et al., 2004a, Robbins, 2002),
fMRI BOLD signal (Duncan, 2013, Petersen and Posner, 2012) and the electrophysiological
measures of single unit firing and local field potentials (Broussard and Givens, 2010, Donnelly
114
et al., 2014, Donnelly et al., 2015) in these regions during performance on tasks of attention,
we would expect to see a difference between tested and untested animals. Therefore, we can
conclude one or more of the following: (1) the animals within their home cage utilised the same
regions for related or unrelated processes (e.g. attending to stimuli around the home cage); (2)
The method used did not have high enough spatial or temporal resolution to detect changes
and other methods with higher temporal or spatial resolution should be considered in the future
(3) variation was too high and so larger group sizes are needed (4) inter-individual differences
were too great to allow between subject comparison, and a method should be considered which
allows within-subject comparisons of changes in activation at multiple time points. For example,
as used in fMRI studies such as that of McKenna et al. (2013), the signal is normalised to a
baseline from within the same subject. It is not possible with post-mortem techniques such as c-
Fos immunohistochemical staining to assess multiple time points (i.e., attentive and non-
attentive) or (5) that rats do not utilise the frontoparietal network to perform the 5C-CPT.
The final suggestion above is unlikely to be true. A number of studies have shown that lesion of
either prelimbic or cingulate cortex of the rodent severely reduces sustained attention in the 5
choice serial reaction time task (effectively the 5C-CPT without „no-go‟ trials) (Bari and Robbins,
2011, Muir et al., 1996). This is supported by PPC recordings that show that more neurones
fire during a sustained attention task when a response is made to a stimulus compared to
omitted trials and that less firing is seen in response to a distracting stimulus compared to task-
relevant stimuli (Broussard et al., 2006, Broussard, 2012). This demonstrates the involvement of
the rodent frontoparietal network in tasks similar to the 5C-CPT, both using different methods.
This is further shown by the correlation of c-Fos with vigilance in our study and may suggest
that the control was unsuitable or too variable (note the high variance in the home cage group:
Figure 6.3).
In the tested group, c-Fos staining in both the mPFC and PPC was strongly and specifically
correlated with performance in the 5C-CPT via the measure d′, which represents vigilance. As
vigilance requires both the allocation of sustained attention and the ability to withhold
responding, it is the most challenging construct tested by the 5C-CPT. Therefore, we suggest
that this increase in the number of c-Fos-positive cells represents increased utilisation of the
frontoparietal network due to higher task engagement. This is consistent with the clinical study
by Fedorenko et al. (2013) where task difficulty was positively correlated with larger responses
from the multiple demand system, which includes the frontoparietal network. The prefrontal
cortex is thought to use event-related inhibitory alpha oscillations to reduce the response to
salient, but task-irrelevant stimuli (Klimesch, 2012, Jensen and Bonnefond, 2013). So the near
significant (p=0.0599) negative correlation between c-Fos count and pFA may be of interest to
pursue with methods of higher temporal resolution, as hypofrontality or decrease in mPFC
activity is linked to increased impulsive/hyperactive symptoms in ADHD patients (Huang-Pollock
et al., 2012, Riccio et al., 2002).
115
In conclusion, we have shown for the first time that the level of mPFC and PPC cellular activity
correlates positively with vigilance in the 5C-CPT. This is strongly supported by previous
experiments in humans and non-human primates using indirect measures of neuronal activity
such as fMRI. However, spatially and temporally specific methods are needed to ascertain how
the frontoparietal network is involved in task performance in rodents, where a large portion of
novel ideas are tested. The 5C-CPT offers a valuable translational methodology, but more
research is required to determine how changes in task performance are represented in the
brain. Advancement in this area would greatly improve the translation of research from rodents
towards clinical application.
116
7 Oscillations in cognition: an attention-
promoting dose of methylphenidate
augments alpha rhythm in the dorsal
attention network
Authors: Andrew Hayward, Joanna C. Neill, John Gigg
[Unpublished data]
7.1 Contributions
Andrew Hayward and Joanna C. Neill and John Gigg planned the study. The experimental
procedures were conducted by Andrew Hayward and John Gigg. Analysis was completed by
Andrew Hayward. The manuscript was prepared by Andrew Hayward and John Gigg.
117
7.2 Context
The previous chapter established that the frontoparietal cortex is more engaged in animals
which perform better in the task so we wanted to study if methylphenidate increased activity in
these regions as a possible endophenotype of attention modulating compounds.
7.3 Abstract
The frontoparietal network orients attention in both humans and rodents. In humans,
methylphenidate is the first line treatment for attention deficit hyperactivity disorder (ADHD) due
to its ability to improve attention. It increases functional magnetic resonance imaging blood
oxygen level dependent signal (fMRI BOLD) in the frontoparietal network, both during visual
attention tasks and in resting state in ADHD patients. The increase in activity across the
frontoparietal network correlates with improvement in ADHD symptoms.
Electroencephalography in patients has shown that methylphenidate increases alpha
oscillations in frontal and parietal regions as well as increasing beta in frontal regions, which
correlates with improved performance in attentive tasks. This study aimed to assess if
methylphenidate can increase activity in the frontoparietal network in rats, and whether this
activity change is seen in particular within the alpha and beta frequency bands.
We recorded the intracranial electrical activity of the PFC and PPC in five urethane
anaesthetised female Lister Hooded rats after saline (vehicle) and then methylphenidate
(1mg/kg) injection. Subsequent analysis found that around 30 minutes after methylphenidate
injection a significant increase of alpha band oscillations arose in the PFC, alongside this in
PFC the slower theta and delta bands reduced in power and the high-frequency gamma band
also reduced in power. In the PPC high gamma band oscillations significantly increased at the
same time point.
Alpha oscillations are related to selection and suppression of visual stimuli in attention and so
the presence of these in the PFC is in agreement with our hypothesis. As previous experiments
used scalp EEG, the lack of gamma oscillations in these is not surprising as they do not travel
far from the source. Finally, beta oscillations were seen in EEG, but not here, as beta
oscillations are linked to cognitive focus, they may only be seen during active task engagement.
In conclusion, we have shown for the first time that a pro-attentive dose of MPH has a
significant effect on alpha, delta and gamma oscillations within the frontoparietal network of the
rat.
118
7.4 Introduction
Attention is an essential process for adaptively allocating limited neural resources across a
broad range of incoming stimuli. It is an essential component of goal-directed behaviours,
including verbal and spatial working memory, response inhibition and decision-making.
Techniques such as functional magnetic resonance imaging (fMRI) and electroencephalography
(EEG) have been used to study brain regions involved in attentional processing. Indeed, fMRI
data across a diverse array of tasks has identified a set of commonly recruited brain regions.
These regions form the multiple demand system and are involved in orienting to and processing
stimuli (Duncan, 2010, Duncan, 2013). The key cortical regions of this system for orienting of
attention are the prefrontal cortex (anterior cingulate cortex and dorsolateral prefrontal cortex)
and the posterior parietal cortex (intraparietal sulcus), collectively termed the frontoparietal
network (Duncan, 2010, Petersen and Posner, 2012, Duncan, 2013). Altered activity has been
identified in this network in conditions with inattention such as attention deficit hyperactivity
disorder (ADHD) (Lin et al., 2015) and schizophrenia (Poppe et al., 2016, Tu et al., 2013). The
importance of the network is also shown by its preservation through evolution, as it has been
shown in humans (Duncan, 2013, Petersen and Posner, 2012, Scolari et al., 2015), non-human
primates (Duncan, 2010), rats (Broussard et al., 2006, Broussard, 2012, Broussard and Givens,
2010) and other rodents (Sellers et al., 2016).
Methylphenidate (MPH) is the current first line treatment for ADHD in the national institute of
clinical excellence (NICE) guidelines due to its ability to improve attention (Palanivel et al.,
2009). In rodents the ability of MPH to improve attention has been demonstrated clearly in
animals with low performance in the 5 choice serial reaction time task (5C-SRTT) and the 5C-
CPT (Tomlinson et al., 2014, Puumala et al., 1996, Blondeau and Dellu-Hagedorn, 2007,
Caballero-Puntiverio et al., 2017). MPH acts at dopamine and noradrenaline transporters to
inhibit dopamine and noradrenaline reuptake at presynaptic terminals in the prefrontal cortex
and nucleus accumbens (Heal et al., 2009). This action promotes a large efflux of dopamine
and noradrenaline, which peaks around 30 minutes after i.p. injection. MPH-evoked
neurotransmitter changes were also linked to increased fMRI BOLD signal within the
frontoparietal network during working memory and visual attention tasks compared to saline
treatment in healthy adults (Tomasi et al., 2011). Also, MPH can normalise resting state fronto-
parieto-cerebellar network dysfunction in children with ADHD and the normalisation of this
network correlates strongly with a reduction of ADHD symptoms (An et al., 2013). Studies using
electroencephalography (EEG) to examine neural activity as the synchronised electrical
oscillations of groups of neurones have shown that MPH increases alpha oscillations at central
and parietal electrodes as well as increasing beta at frontal electrodes, which correlates with
improved performance in attentive tasks (Loo et al., 2004). EEG offers a very high temporal
resolution (but poor spatial resolution) as well as insight into the various activation states
promoted by different frequencies. However, it cannot easily reveal information about the
operation of circuits within the brain, which requires a targeted and invasive approach.
119
In this study, we recorded the local field potential (LFP) from PFC and PPC using multi-
electrode array depth recordings in urethane-anaesthetized rats. We assessed the effects of
MPH on these regions to establish whether they are the source regions for previously-described
MPH-induced alterations in EEG oscillations. Based on the fMRI data we expected to see
increased activity in both PFC and PPC when treated with MPH compared to vehicle treatment
and from the EEG studies we expected the main effect of MPH to be in the alpha and beta
frequency ranges.
7.5 Methods
7.5.1 Animals
5 female Lister Hooded rats (275g ± 30g on the day of surgery) were group housed in
individually ventilated cages with two levels (GR1800 Double-Decker Cage, Techniplast, UK)
under a standard 12 hour light: dark cycle (lights on 7:00 am). The environment was maintained
at 21 ± 2 °C, 55 ± 5 % humidity. Animals were anaesthetised between 9:00-12:00 during their
light phase. All experiments were conducted in accordance with the UK Animals (Scientific
Procedures) 1986 Act and local University ethical guidelines.
7.5.2 Surgery
Initial anaesthesia was induced by urethane (30%, w/v in 0.9% saline, 1.3 g/kg) and up to two
top-up doses of urethane (0.1 ml) were administrated at approximately 30-min intervals until
areflexia was achieved. Body temperature was maintained at 37 °C using a homoeothermic
heating pad. The rat was head-fixed in a stereotaxic frame and a 2-mm diameter craniotomy
was drilled above mPFC (Bregma + 2.5-3.0, ML 2.5, DV 3.0) and PPC (Bregma - 4.5, ML 2.0,
DV 1.0). The dura was carefully incised and a 2 x 16 multi-electrode array (NeuronexusTech,
USA) was inserted at each site. The electrode had two 10mm shanks, each with 16 iridium
contacts (recording area 177μm2) spaced linearly (100μm apart) along each shank with 500μm
between shanks. The frontal electrode was inserted at a 30° angle from vertical, in the coronal
plane. The posterior electrode was inserted vertically in the sagittal plane.
7.5.3 Recording
Electrodes were coated using the lipophilic fluorescent compound Di-I before insertion, and
placement was confirmed post hoc (DiCarlo et al., 1996). Electrode arrays were attached to an
electrode board and preamplifier (Plexon, USA) with a fixed gain of 20x. An AC amplifier was
adjusted to give a total gain of 2000x (Recorder64, Plexon, USA). Data were processed online
using a Windows computer running (Recorder64 (Plexon, USA). LFPs were digitised at 1 kHz
and bandpass filtered (1-500Hz). Once stable signals were achieved each animal received a
saline injection (1 ml/kg, i.p.) and the LFP was recorded for 30 min. Animals then received an
injection of theo-methylphenidate hydrochloride (Tocris Biosciences, UK) made up in saline
120
(0.9% NaCl) to a dose of 1 mg/ml and an injection volume of 1 ml/kg (i.p.). LFP was then
recorded for a further 60 minutes.
7.5.4 Histology
After recording animals were perfused with phosphate buffered saline (1x) then
paraformaldehyde (4% w/v) and brains were extracted. Brains were stored in paraformaldehyde
for 48 hours, then sucrose solution (30% w/v) for 24 hours. Free-floating sections were sliced at
40μm using a freezing sledge microtome (Leica, UK) and imaged Olympus BX51 microscope
using a Texas red filter. Sections were later Nissl stained to confirm anterior-posterior position.
7.5.5 Analysis
The analysis was completed using custom written MATLAB scripts. Signals were down-sampled
and filtered using the function decimate to 250 Hz and the power spectral density (PSD)
calculated using the Chronux toolbox function for power spectral density (PSD; mtspecgramc)
using a sampling window of 2 seconds, a shift of 1.8 s, a time-bandwidth product of 4 and 2.5
multi-tapers. Average PSD across the 30 minute saline treatment was calculated and for each
time point in the methylphenidate analysis, the PSD was divided by the average saline PSD to
normalise each animal to its saline treated baseline (Figure 7.2). The relative PSD was then
averaged within the band's delta (1-4 Hz) theta (4-8 Hz) alpha (8-15 Hz) beta (15-30 Hz) low
gamma (30-45 Hz) and high gamma (55-90 Hz). Relative PSD of the methylphenidate recording
for each band was then averaged into three 20 min bins. Time points and regions were
compared using a two-way ANOVA with animal ID as a blocking factor to produce a pseudo-
repeated design (InVivo Stat, v3.4; Clark et al. (2012)). Least Square Difference planned
comparisons between the first time point and the other two for each region were adjusted with
the Benjamini-Hochberg procedure to reduce the false discovery rate (Benjamini and Hochberg,
1995).
7.6 Results
Initial processing with the Chronux function mtspecgramc showed that the saline control was
stable across the 30 min recording, but with some fluctuation, particularly in the high gamma
range (Figure 7.1). Therefore, the average across the 30 minutes is highly representative of the
state of the regions after saline application. This supports the normalisation of methylphenidate
data to the saline control in order to show drug effects.
121
Figure 7.1: representative power spectral densitiy for saline recording. This shows a stable recording during saline application allowing this data to be used to normalise the methylphenidate data to. Colour bar represents power spectral density in microvolts per hert.
The PSDs were normalised to saline control and showed a clear deviation from this baseline
approximately 30min after MPH injection (Figure 7.2). This change was seen to start in all
animals 20-40 minutes into the MPH recording. From this point onwards a clear difference was
seen, which represents the neural action of methylphenidate. For statistical analysis, the signal
was decomposed into the relevant frequency bands (delta, theta, alpha, beta, low gamma and
high gamma) and also into three 20-minute time bins to account for the observed time-
dependent changes.
122
In the mPFC, the delta band significantly reduced over time (F2=7.18, p=0.0045; Figure 7.3A)
adjusted LSD comparisons agreed for the mPFC at 20-40 and 40-60 minutes (p<0.05 and
p<0.01, respectively) when compared to 0-20 minutes. The theta band (Figure 7.3B) showed
no significant model level effects, but adjusted LSD comparisons for the mPFC were
significantly different for 20-40 and 40-60 minutes (p<0.05 for both) compared to 1-20 minutes.
In the alpha band a significant increase over time was seen (F2=9.64, p=0.0012; Figure 7.3C)
and adjusted LSD comparisons showed a significant difference for mPFC between 1-20 and 40-
60 minutes (p<0.01). In the beta (Figure 7.3D) and low gamma (Figure 7.3E) bands, no
significant differences were seen. However, in high gamma a significant interaction between
time and region was seen (F2=7.06, p=0.0048; Figure 7.3F) and LSD comparisons showed this
to be a significant reduction at 40-60 minutes compared to 1-20 minutes for mPFC (p<0.05), but
an increase in the same comparison for PPC (p<0.05).
Figure 7.2: Methylphenidate takes 20-40 minutes to cause changes in the neural oscillations. A representative spectrogram from a single rat shows the onset of LFP changes in the mPFC and PPC approximately 30 minutes after MPH injection (injection at 0min). The main effect in this example is a reduction of power in low frequency delta and theta and increased power in alpha within the mPFC. In the PPC power is increased for theta and high gamma bands. Colours represent the power relative to saline.
123
0-2
0
20-4
0
40-6
0
0 .0
0 .5
1 .0
1 .5
D e lta (1 -4 H z )
T im e (m in s )
Po
we
r
Re
lati
ve
to
S
ali
ne
* ***
m P F C
P P C
0-2
0
20-4
0
40-6
0
0 .0
0 .5
1 .0
1 .5
2 .0
T h e ta (4 -8 H z )
T im e (m in s )
Po
we
r
Re
lati
ve
to
S
ali
ne
* *
0-2
0
20-4
0
40-6
0
0
1
2
3
A lp h a (8 -1 2 H z )
T im e (m in s )
Po
we
r
Re
lati
ve
to
S
ali
ne **
0-2
0
20-4
0
40-6
0
0 .0
0 .5
1 .0
1 .5
2 .0
B e ta (1 2 -3 0 H z )
T im e (m in s )
Po
we
r
Re
lati
ve
to
S
ali
ne
0-2
0
20-4
0
40-6
0
0 .8
1 .0
1 .2
1 .4
1 .6
L o w G a m m a (3 0 - 4 5 H z )
T im e (m in s )
Po
we
r
Re
lati
ve
to
S
ali
ne
0-2
0
20-4
0
40-6
0
0 .0
0 .5
1 .0
1 .5
2 .0
H ig h G a m m a (5 5 -9 0 H z )
T im e (m in s )
Po
we
r
Re
lati
ve
to
S
ali
ne
*
*
A B
C D
E F
Figure 7.3: Methylphenidate significantly increases power in the alpha band, particularly in the
mPFC, while reducing power at lower frequencies. Delta, Theta and high Gamma bands were
significantly reduced in mPFC 40-60 minutes after dosing, while alpha band oscillations were significantly
increased. In the PPC methylphenidate promotes increased high gamma band activity. * p<0.05 ***
p<0.01.
7.7 Discussion
In this study, the LFP oscillations in the mPFC and PPC were studied following administration of
methylphenidate, the current first-line treatment for ADHD. The main effect of MPH dosing was
a dramatic change in neural oscillations that started around 30min after injection. The latency of
this change is consistent with that previously reported for peak dopamine and noradrenaline
release in the striatum and mPFC following i.p. MPH injection (Heal et al., 2009, Berridge et al.,
124
2006). The timing of the change supports the hypothesis that the effects seen after that point
were due to MPH-evoked catecholamine signalling. In the mPFC reduced low-frequency delta
and theta oscillations were seen alongside increasing alpha band activity, which more than
doubled in the period 40-60min post-injection compared to the vehicle reference. Some
changes in beta oscillations were seen, but were not consistent across animals and so did not
reach significance. Finally, for mPFC, the high gamma band reduced in the final 20min period
post-injection compared to the first. However, in PPC a significant increase was seen in the high
gamma band (Figure 7.3F).
Our observed changes in LFP power following MPH dosing are likely to be relevant to the
clinical effectiveness of this drug. Delta oscillations are thought to represent basic homeostatic
and motivational processes (Knyazev, 2012) and so reduction of this is likely the result of power
moving into other frequency bands. Theta at rest is thought to represent under arousal or
drowsiness (Barry et al., 2003, Lazzaro et al., 1999, Sergeant, 2000, Tye et al., 2014) –
effectively disengagement from a task - and its presence over frontal and central regions is
strongly correlated to ADHD symptoms in children and adults with the condition (Barry et al.,
2003, Shi et al., 2012, Tye et al., 2014). Therefore, the reduction seen here in the PFC may be
related to the stimulant effect of methylphenidate to promote arousal and reduce theta power.
Activity in the alpha band is thought to be inhibitory in nature (Jensen and Mazaheri, 2010),
allowing it to fit a filtering role for suppression and selection of stimuli in attention (Klimesch,
2012). Thus, the alpha power increase seen here may represent the mechanism by which MPH
facilitates the filtering of sensory input to support attention. In rodent studies, increases in alpha
were seen during anticipation of stimuli and around distraction signals, supporting a role in
inhibition of response in rodents, as in humans (Broussard and Givens, 2010). In relation to
MPH, this could relate to increased sustained attention, which has been shown behaviourally
(Tomlinson et al., 2014, Heal et al., 2013). Furthermore, alpha oscillations are thought to
negatively modulate excitatory gamma band activity (Jensen and Mazaheri, 2010). In the
context of the current study, this may explain why a reduction of high gamma in the mPFC was
seen alongside an increase in alpha, whereas, in the PPC, the reverse was observed. Recent
evidence using magnetoencephalography in humans performing an attentive task further
supports the theory that high gamma (70-120 Hz) in the mPFC modulates the PPC which, in
turn, modulates the visual cortices at the same frequency band; this results in an increase in
synaptic gain, thereby enabling allocation of attention (Marshall et al., 2015). It is our hypothesis
that MPH promotes sustained and selective attention through control of alpha and gamma
within the frontoparietal network. It is through this mechanism that we propose well-documented
increases in attention and vigilance may be mediated (Tomlinson et al., 2014, Puumala et al.,
1996, Blondeau and Dellu-Hagedorn, 2007, Caballero-Puntiverio et al., 2017).
Moving forward with this research there are two key directions: Firstly, establishing how MPH
alters ongoing electrical oscillations within the active brain, which could be performed with
125
simultaneous recording of electrical activity with 5C-CPT performance as has recently been
done in the study of impulsivity in the 5C-SRTT (Donnelly et al., 2014, Donnelly et al., 2015).
This would allow insight into whether MPH alters the timing of oscillation patterns, such as
event-related synchronisation and desynchronisation in the alpha band (Klimesch, 2012).
Secondly, other compounds could be studied in the same way to add a much deeper insight
into how these changes are caused. For example, atomoxetine is the third line treatment for
ADHD but has a more selective pharmacology by only acting at noradrenaline reuptake
transporters. Response to this drug could help dissociate the noradrenergic versus
dopaminergic effects of MPH.
In conclusion, we have shown for the first time that a pro-attentive dose of MPH has a
significant effect on EEG oscillations within the frontoparietal network of the rat. In particular,
MPH reduced delta, theta and high gamma as well as increased alpha in the mPFC and
increased high gamma in the PPC. These effects suggest that MPH could facilitate the
„normalisation‟ of the EEG in low-performing subjects to improve sustained attention and reduce
distractibility.
126
8 General Discussion
In the first part of this thesis modelling of attention deficits by grouping rats based on
performance in the 5C-CPT was reviewed. This method was then refined and validated
(Chapter 3) to facilitate this type of modelling. For this, three compounds were assessed for the
first time in this model (Chapters 4 and 5), the involvement of the frontoparietal network in the
5C-CPT was verified (Chapter 6) and the effect of methylphenidate on this network was
established (Chapter 7). This work was conducted to fulfil three general aims:
iv) To further develop a method of modelling high and low attention in rats.
v) To use the high and low attentive model to study the pharmacology of
attention
vi) To examine the involvement of the frontoparietal network in performance of
the 5C-CPT and the effects of attention-promoting agents
The work here has established a more translatable model that is effective for studying
compounds with attention enhancing potential.
8.1 Aim 1: To further develop a method of modelling
attention deficits in rats.
In Chapter 3 the method of characterizing animals by performance as a model of low attention
was refined from, and compared to, the work of Tomlinson et al. (2014). There were two
changes made to the experimental design of that work: firstly, rather than grouping animals as
above or below a series of set values (e.g., 70% accuracy), we chose to use the upper and
lower quartiles of the group and exclude animals with performance close to the median. The
exclusion of the moderate performers resulted in a larger seperation between groups; secondly,
we used the measure d′ from signal detection theory to establish vigilance in the animals. This
is important as d′ is the measure used clinically in the CPT, where sensitivity index (SI) was
used previously in pre-clinical studies. This is due to SI being robust to non-parametric data,
however, animals produce normally distributed data sets so d′ is more applicable (Young et al.,
2013). Therefore, the work outlined in Chapter 3 establishes a refined model that has a clearer
deficit of attention and higher translational potential than the previous model. The work in
Chapters 4 and 5 extends this to test compounds that have not previously been tested in 5C-
CPT using animals grouped by performance.
The study in chapter 3 confirmed that methylphenidate increases selective attention and
vigilance in LA animals, but there were a number of differences to Tomlinson et al. (2014) on
other measures. For example, a reduction of omissions was seen with the refined method,
which represents an increase in sustained attention. This was not observed in the study by
Tomlinson et al. (2014), but is reproduced in other studies using the 5C-SRTT in low-performing
127
animals (Paterson et al., 2011). Improvements in sustained attention are also seen in ADHD
patients treated with methylphenidate and in low attentive individuals without a diagnosis of
ADHD (Agay et al., 2014). We can, therefore, propose that the robust deficit produced by the
refined model presented here enhances the translational potential of findings in this model. For
atomoxetine, the same improvement in LA animals for accuracy and d′ was shown in Chapter 3
and by Tomlinson et al. (2014). The reduction of premature responses was not seen in
Tomlinson et al. (2014). During vehicle treatment, the number of premature responses was
significantly different between HA and LA animals, which shows that LA animals also presented
with high waiting impulsivity. Therefore, the reduction in premature responses is in agreement
with the results of atomoxetine in high impulsive animals using a different separation method
(Tomlinson et al., 2014).
Due to the similarities in the effects of both methylphenidate and atomoxetine to those reported
in the literature, the conclusion derived is that this method is capable of studying
pharmacological treatments in a translational manner and represents a refinement in modelling
low attention. It was used, therefore, to address the second aim of the Thesis.
8.2 Aim 2: To use the high and low attentive model to study
the pharmacology of attention
Due to the wide range of conditions with inattention as a symptom, it is essential to continue the
development of novel modulators of attention and to study how well-known compounds interact
in the LA model validated in Chapter 3. Research has primarily focussed on monoaminergic
compounds as attention promoting agents due to the clinical success of amphetamine,
methylphenidate and atomoxetine in ADHD. However, the ability to produce a robust deficit
without chemical or genetic manipulation allows an unbiased assessment of under-researched
mechanisms. In Chapter 4, therefore, we studied nicotine and caffeine, two established
psychoactive stimulants that facilitate sustained attention (Rezvani and Levin, 2001, Heishman
et al., 2010, Levin et al., 1998).
Acute doses of both nicotine and caffeine increased 5C-CPT impulsive behaviour as measured
by pFA in animals that were not grouped based on their performance. In animals which were
grouped by performance, nicotine increased impulsivity only in HA animals. In addition, nicotine
also reduced omissions, thereby, increasing sustained attention, as previously reported.
Therefore, these findings not only support previous work but also highlight the power of the 5C-
CPT to use both pFA and premature responses as robust measures of impulsivity. This is
important as waiting impulsivity (premature responses) and response disinhibition (pFA) are
separate constructs, which can be manipulated independently (Young et al., 2011). Also, recent
research suggests that premature responses are linked more to response timing strategy than
impulsivity (Cope et al., 2016). It is suggested that nicotine in particular should be researched
further with chronic dosing regimens as a high level of variability in the present nicotine study
128
may have been due to the acute effects of nicotine on stress markers such as corticosterone,
which are not present in chronic dosing (Matta et al., 2007).
Nicotine has shown some potential to modulate attention in the 5C-CPT and genetic knock-out
of the α7 nicotinic acetylcholine receptor induces attention deficits. Encenicline is a partial
agonist of this receptor with cognition improving properties (Prickaerts et al., 2012), but has not
yet been assessed for its effect on attention in LA subjects. In Chapter 5 a detailed analysis
using the model developed in Chapter 3 revealed that encenicline has a contrasting effect in HA
and LA animals. In LA animals it improves performance across measures of accuracy, response
disinhibition and vigilance, but in HA animals it impaired performance on the same measures.
This suggests that encenicline may be an effective compound for the reduction of inattentive
symptoms in states such as ADHD, schizophrenia or Alzheimer‟s disease and that baseline
inattention should be a factor when deciding which treatment to use. As there are few
compounds available to alleviate LA in schizophrenia, this could be an important compound for
pharmacotherapy. However, encenicline has not shown efficacy in stage II clinical trials for
schizophrenia and in another trial elderly Alzheimer‟s patients developed severe gastrointestinal
side effects. As the α7 nicotinic acetylcholine receptor is an area of increasing research for
schizophrenia, it is hoped we will see compounds that will overcome the issues faced by
encenicline (Young and Geyer, 2013). One promising compound has been recently
characterised demonstrating a continued interest in the receptor (Bristow et al., 2016).
8.3 Aim 3: To examine the involvement of the frontoparietal
network in performance of the 5C-CPT and effects of
attention promoting agents
The frontoparietal network is involved in the process of orienting attention both in human and
non-human primates. However, its involvement in the performance of the 5C-CPT in rodents
has not been investigated. In Chapter 6 this was probed by ex vivo immunohistochemistry for a
marker of neural activity, the immediate early gene C-Fos. The present results showed a
correlation between C-Fos expression in the frontoparietal network and vigilance in the 5C-CPT,
which suggests involvement of the medial prefrontal cortex and posterior parietal cortex in 5C-
CPT performance. This is not clear though as there is no difference between untested controls
and animals performing 5C-CPT. In the context of previous studies showing the involvement of
both regions in humans and non-human primates using indirect measures of neuronal activity
such as fMRI (Duncan, 2010, Duncan, 2013, Petersen and Posner, 2012), it seems possible
that flaws in the untested control may explain this, but this requires further analysis.
In Chapter 7 the response of the frontoparietal network to an attention-promoting dose of
methylphenidate was assessed in anaesthetised animals. This study showed for the first time
that methylphenidate simultaneously promotes alpha and gamma EEG rhythms whilst reducing
both delta and theta oscillations in the frontoparietal network. Alpha oscillations have been
129
associated with the suppression and selection of stimuli to promote attention. Increases in theta
have been associated with a loss of attentional focus (introspection). Therefore, the effects seen
in this study may be a mechanism by which methylphenidate alters the firing patterns of
neurones in the frontoparietal network to improve attention.
8.4 Limitations and Future Research
The majority of preclinical research using 5-choice tasks has used male rodents. Childhood
ADHD is diagnosed more frequently in males (2.6-5:1) than females, however, in adulthood, this
ratio reduces (1-2:1) (Bachmann et al., 2017, Williamson and Johnston, 2015b). This may be
due to the symptom profile of ADHD in males and females being different. Females tend to
show higher levels of inattentive symptoms, whereas, males tend to show a higher prevalence
of hyperactive and impulsive symptoms (Williamson and Johnston, 2015b, Ramtekkar et al.,
2010a). As the latter reduce more with ageing than inattentive symptoms, females are more
likely to persist into adulthood with ADHD.
Based on this, there is a female population with high inattention in adulthood, which is poorly
represented in the preclinical research. Researchers have avoided using female rodents mainly
due to fear of increased variation due to the stages of the oestrous cycle. This can be easily
measured and, thus, accounted for in study design and statistics. However, to maintain study
power, this may require larger groups, which adds cost and time to a study. In contrast, recent
evidence has shown that female rats and mice are not more variable across a vaiety of
behavioural tasks than males even if the oestrous cycle is not accounted for (Prendergast et al.,
2014, Becker et al., 2016). Therefore, the representation of females needs to be increased in
neuroscience research to improve the translatability of the findings.
Validation of the new method compared to that of Tomlinson et al. (2014) is important in order
to establish similarities and differences in the methods. However, we still do not know the
causes of the differences between HA and LA animals. The differences in response to
pharmacological challenge suggest there may be differences in neurobiology between the two
populations. As discussed in the introduction, the Dalley et al. (2007) high impulsivity model has
been shown to differ in pharmacological response (Fernando et al., 2012), but it has also shown
differences in D2/3 receptor binding as a mechanism by which the difference is produced
(Dalley et al., 2007). Future work with the LA animal model presented in Chapter 3 could assess
whether there are differences in the expression of, or binding to, proteins important for attention.
Clinical ADHD heritability studies have highlighted some potential targets for this investigation
such as DAT and dopamine D4 receptor (Klein et al., 2017). Analysing differences in HA and LA
animals of receptors such as DAT and DRD4 would help to understand (a) the neurobiology
that underlies performance variation in rodent populations and (b) whether these mechanisms in
rodents are similar to those in human cases, therefore reinforcing the construct validity of the
model.
130
Another question highlighted in the model of Dalley et al. (2007) is whether differences between
animals are a state or a trait (Crabbe and Cunningham, 2007). That is, whether differences are
inherent to the animals (trait) or induced by environment and/or training (state). Once
biomarkers such as DAT and dopamine receptor D4 are discovered these could be traced
through genetics and epigenetics to establish if these markers are predictive of poor
performance. However, future work could also attempt to further validate the method by
establishing stability of individuals and groups and then addressing any instability with improved
training methods. One such improvement would be a subject-driven training method, as
suggested by Martin et al. (2015) for the 5C-SRTT. With this method, task difficulty is increased
or reduced based on the performance of the individual on a trial-by-trial basis. This would
include the reduction of stimulus duration and the gradual introduction of „no go‟ trials. This has
three key benefits: 1) The adaptation after incorrect responses reduces the likelihood of
disengagement; 2) based on its use in 5C-SRTT, it would be expected to reduce training time;
3) the standardisation of training will make it more reliable and results more reproducible; and 4)
training automation allows analysis of differential training speeds with much more detail and
confidence. The final point is of particular interest, as it would allow links to be researched
between learning speed and final performance.
Another important consideration when using 5 choice tasks are ceiling and floor effects. These
may occur if it becomes much more difficult or impossible to pass a certain performance point.
For example, if an animal is performing the task to 98% accuracy, then increasing performance
beyond that level is unlikely, purely due to the restraints of the task. Because of this, even were
the animal to improve in performance the task would be unable to detect it. This factor is
particularly important when looking at very high or low performing individuals. However, this
potential issue may be overcome in a number of ways: 1) Careful control of training, for
example, the subject-driven methods discussed above, as over long periods of training animals
will trend towards higher performance over time 2) Increasing the task difficulty, for example, by
reducing stimulus duration, increasing inter-trial intervals or using audible or visual distractors.
Both of these methods would shift performance away from the artificial limits of detection by the
task making assessments more reliable.
This work has mostly focused on the HA and LA animals in each cohort to assess any
differential response to pharmacological challenge. To answer the question of how attention
interacts with performance, other methods of analysis could also be used. For example,
regression analysis would allow the assessment of relationships between baseline performance
and changes from the baseline. Regression analysis would also have higher power due to use
of the full cohort rather than just 25% of the cohort for each group (HA/LA). However, it is
important to remember that the goal of this research is to be translational towards conditions of
low attention such as ADHD and schizophrenia. These conditions are extreme manifestations of
low attention, which are modelled well by the LA group. Therefore, as changes in the LA group
are of the most clinical relevance these need to be studied in isolation.
131
Chapter 6 highlighted the involvement of the frontoparietal network in the 5C-CPT. However,
the method used does not allow comparison of an animal to itself and can offer no insight into
the evolution of neuronal activity changes during task performance. This could be improved in
future work by using a different methodology. The recording of electrical activity in awake
animals performing the 5C-SRTT has recently been achieved (Donnelly et al., 2014, 2015). As
neurones transmit information as electrical potentials, measuring electrical changes is a more
direct way to analyse activity than C-Fos expression. Also, the fact the animals are awake and
performing the task allows within-subject comparison of attentive and non-attentive states. This
would be a powerful way to analyse the involvement of the frontoparietal network in the
performance of the 5C-CPT.
In Chapter 7 we showed that changes in neural electrical oscillatory activity are evoked by
methylphenidate in the frontoparietal network. However, we have not yet established any
transfer of information between these regions. This is something that will be addressed in future
work. Information transfer entropy is an analysis method designed to model the complex
dependencies of a system in a directional manner (Schreiber, 2000). As neural oscillations are
complex non-linear systems, the application of information transfer entropy is ideal to establish
if causal interactions exist between different frequencies of oscillations (Besserve et al., 2010).
By application of this method of analysis to the existing data presented in Chapter 7, the
hypothesis that methylphenidate alters the transfer of information between nodes of the
frontoparietal network can be tested. We would hypothesise that the regions would synchronise
as they do in non human primates (Buschman and Miller, 2007). However, as the directionality
of the synchronisation is dependent on if the stimuli are „top-down‟ or „bottom-up‟. Therefore, it
would be very interesting to establish how methylphenidate alters the directionality of
frontoparietal synchronisation.
The work presented in this thesis could be extended in different directions. Firstly, future studies
could compare the effect of methylphenidate to other compounds that alter attention. This could
include the selective noradrenaline transporter inhibitor atomoxetine to a dopaminergic
transporter inhibitor. In this experiment, the mechanism of generation of alpha and gamma
oscillations can be further understood in relation to attention-promoting compounds. Secondly,
the experiment could be repeated, but with animals already grouped based on performance in
the 5C-CPT to assess whether the electrical oscillations of the regions differ in the two
populations and if methylphenidate normalises any differences found as it does clinically (An et
al., 2013, Loo et al., 2004). Finally, to assess the involvement of the frontoparietal cortex in
attention, it will be essential to record its activity during performance of the 5C-CPT. A method
that could be used for this type of assessment has recently been published for the 5C-SRTT
(Donnelly et al., 2014). This would allow the analysis of changes in oscillatory state as attention
is allocated, offering tremendous insight into the involvement of the frontoparietal network.
Future studies could then assess if and how methylphenidate and other attention modulating
132
compounds alter neural oscillations in real time and if the response is different in high attentive
and low attentive animals.
8.5 Conclusion
The work presented in this thesis has refined a method for modelling low attention in rodents,
which is of relevance to many clinical conditions such as ADHD, schizophrenia and Alzheimer‟s
disease. The refinements outlined have brought it closer to the clinical version of the task by
calculating the same parameter (d′) to estimate vigilance and by using upper and lower quartiles
to produce a clearer deficit. The utility of the model has been shown by examination of the α7
nicotinic acetylcholine receptor partial agonist encenicline; results show for the first time that this
compound improves both attention and vigilance whilst reducing impulsive action in LA animals.
This work has highlighted partial agonists of the α7 receptor as potential attention-promoting
agents. This is important as there is a large body of research into this receptor for the alleviation
of cognitive symptoms in schizophrenia (Young and Geyer, 2013). Additionally, the present
study highlights important differences between the full agonist PNU-282987, which has no effect
on attention, and the partial agonist encenicline, which improves attention. It is hoped this can
help to guide the assessment of novel compounds for their ability to improve inattention during
the early stages of drug development.
The present studies on activity of the frontoparietal network have shown that it correlates
positively with 5C-CPT performance and that methylphenidate promotes a dramatic change in
the frequencies of neural oscillations across this network. The reduction of theta and increase in
alpha is consistent with the effect of methylphenidate in human EEG and so offers insight into
the localisation/mechanism of action of the effect of methylphenidate. This finding has
highlighted the importance of studying neural oscillations in relation to attention-promoting
compounds and has scope to be extended by analysis into how these regions communicate.
Finally, the application of the methods used here to study animals as they perform the 5C-CPT
would be a powerful tool to study the effects of attention-promoting agents and how they alter
communication across underlying brain regions.
133
9 References
ADHD-200-CONSORTIUM 2012. he ADHD-200 Consortium: a model to advance the translational potential of neuroimaging in clinical neuro- science. Frontiers in Systems Neuroscience, 6, doi: 10.3389/fnsys.2012.00062.
AGAY, N., YECHIAM, E., CARMEL, Z. & LEVKOVITZ, Y. 2014. Methylphenidate enhances cognitive performance in adults with poor baseline capacities regardless of attention-deficit/hyperactivity disorder diagnosis. J Clin Psychopharmacol, 34, 261-5.
AMERICAN PSYCHIATRIC ASSOCIATION. 2013. Diagnostic and statistical manual of mental disorders : DSM-5, Washington, D.C., American Psychiatric Association.
AN, L., CAO, X. H., CAO, Q. J., SUN, L., YANG, L., ZOU, Q. H., KATYA, R., ZANG, Y. F. & WANG, Y. F. 2013. Methylphenidate normalizes resting-state brain dysfunction in boys with attention deficit hyperactivity disorder. Neuropsychopharmacology, 38, 1287-95.
ANDERSON, B. A., KUWABARA, H., WONG, D. F., GEAN, E. G., RAHMIM, A., BRASIC, J. R., GEORGE, N., FROLOV, B., COURTNEY, S. M. & YANTIS, S. 2016. The Role of Dopamine in Value-Based Attentional Orienting. Curr Biol, 26, 550-5.
ARNSTEN, A. F. 2009. Toward a new understanding of attention-deficit hyperactivity disorder pathophysiology: an important role for prefrontal cortex dysfunction. CNS Drugs, 23 Suppl 1, 33-41.
ARNSTEN, A. F., SCAHILL, L. & FINDLING, R. L. 2007. alpha2-Adrenergic receptor agonists for the treatment of attention-deficit/hyperactivity disorder: emerging concepts from new data. J Child Adolesc Psychopharmacol, 17, 393-406.
ASTON-JONES, G. & COHEN, J. D. 2005. An integrative theory of locus coeruleus-norepinephrine function: adaptive gain and optimal performance. Annual Review of Neuroscience, Vol 35, 28, 403-50.
ASTON-JONES, G., RAJKOWSKI, J. & KUBIAK, P. 1997. Conditioned responses of monkey locus coeruleus neurons anticipate acquisition of discriminative behavior in a vigilance task. Neuroscience, 80, 697-715.
BACHMANN, C. J., WIJLAARS, L. P., KALVERDIJK, L. J., BURCU, M., GLAESKE, G., SCHUILING-VENINGA, C. C., HOFFMANN, F., AAGAARD, L. & ZITO, J. M. 2017. Trends in ADHD medication use in children and adolescents in five western countries, 2005-2012. Eur Neuropsychopharmacol.
BALDEWEG, T., KLUGMAN, A., GRUZELIER, J. & HIRSCH, S. R. 2004. Mismatch negativity potentials and cognitive impairment in schizophrenia. Schizophr Res, 69, 203-17.
BARI, A. & ASTON-JONES, G. 2013. Atomoxetine modulates spontaneous and sensory-evoked discharge of locus coeruleus noradrenergic neurons. Neuropharmacology, 64, 53-64.
BARI, A., DALLEY, J. W. & ROBBINS, T. W. 2008. The application of the 5-choice serial reaction time task for the assessment of visual attentional processes and impulse control in rats. Nat Protoc, 3, 759-767.
BARI, A. & ROBBINS, T. W. 2011. Animal models of ADHD. Curr Top Behav Neurosci, 7, 149-85. BARKLEY, R. A., DUPAUL, G. J. & MCMURRAY, M. B. 1990. Comprehensive Evaluation of
Attention-Deficit Disorder with and without Hyperactivity as Defined by Research Criteria. Journal of Consulting and Clinical Psychology, 58, 775-789.
BARNES, S. A., YOUNG, J. W. & NEILL, J. C. 2012a. D-1 receptor activation improves vigilance in rats as measured by the 5-choice continuous performance test. Psychopharmacology, 220, 129-141.
BARNES, S. A., YOUNG, J. W. & NEILL, J. C. 2012b. Rats tested after a washout period from sub-chronic PCP administration exhibited impaired performance in the 5-Choice
134
Continuous Performance Test (5C-CPT) when the attentional load was increased. Neuropharmacology, 62, 1432-1441.
BARRY, R. J., CLARKE, A. R., HAJOS, M., MCCARTHY, R., SELIKOWITZ, M. & BRUGGEMANN, J. M. 2009. Acute atomoxetine effects on the EEG of children with attention-deficit/hyperactivity disorder. Neuropharmacology, 57, 702-7.
BARRY, R. J., CLARKE, A. R. & JOHNSTONE, S. J. 2003. A review of electrophysiology in attention-deficit/hyperactivity disorder: I. Qualitative and quantitative electroencephalography. Clin Neurophysiol, 114, 171-83.
BAUER, A., HOLSCHBACH, M. H., MEYER, P. T., BOY, C., HERZOG, H., OLSSON, R. A., COENEN, H. H. & ZILLES, K. 2003. In vivo imaging of adenosine A1 receptors in the human brain with [18F]CPFPX and positron emission tomography. Neuroimage, 19, 1760-9.
BAVIERA, M., INVERNIZZI, R. W. & CARLI, M. 2008. Haloperidol and clozapine have dissociable effects in a model of attentional performance deficits induced by blockade of NMDA receptors in the mPFC. Psychopharmacology (Berl), 196, 269-80.
BECKER, J. B., PRENDERGAST, B. J. & LIANG, J. W. 2016. Female rats are not more variable than male rats: a meta-analysis of neuroscience studies. Biol Sex Differ, 7, 34.
BELIN, D., MAR, A. C., DALLEY, J. W., ROBBINS, T. W. & EVERITT, B. J. 2008. High impulsivity predicts the switch to compulsive cocaine-taking. Science, 320, 1352-5.
BENJAMINI, Y. & HOCHBERG, Y. 1995. Controlling the False Discovery Rate - a Practical and Powerful Approach to Multiple Testing. Journal of the Royal Statistical Society Series B-Methodological, 57, 289-300.
BERLIN, H. A., ROLLS, E. T. & IVERSEN, S. D. 2005. Borderline personality disorder, impulsivity, and the orbitofrontal cortex. Am J Psychiatry, 162, 2360-73.
BERNFORT, L., NORDFELDT, S. & PERSSON, J. 2008. ADHD from a socio-economic perspective. Acta Paediatr, 97, 239-45.
BERRIDGE, C. W., DEVILBISS, D. M., ANDRZEJEWSKI, M. E., ARNSTEN, A. F., KELLEY, A. E., SCHMEICHEL, B., HAMILTON, C. & SPENCER, R. C. 2006. Methylphenidate preferentially increases catecholamine neurotransmission within the prefrontal cortex at low doses that enhance cognitive function. Biol Psychiatry, 60, 1111-20.
BESSERVE, M., SCHOLKOPF, B., LOGOTHETIS, N. K. & PANZERI, S. 2010. Causal relationships between frequency bands of extracellular signals in visual cortex revealed by an information theoretic analysis. J Comput Neurosci, 29, 547-66.
BESSON, M., BELIN, D., MCNAMARA, R., THEOBALD, D. E., CASTEL, A., BECKETT, V. L., CRITTENDEN, B. M., NEWMAN, A. H., EVERITT, B. J., ROBBINS, T. W. & DALLEY, J. W. 2010. Dissociable control of impulsivity in rats by dopamine d2/3 receptors in the core and shell subregions of the nucleus accumbens. Neuropsychopharmacology, 35, 560-9.
BIEDERMAN, FARAONE, S. V., KEENAN, K., KNEE, D. & TSUANG, M. T. 1990. Family-genetic and psychosocial risk factors in DSM-III attention deficit disorder. J Am Acad Child Adolesc Psychiatry, 29, 526-33.
BIEDERMAN, J., MICK, E., FARAONE, S. V., BRAATEN, E., DOYLE, A. & SPENCER, T. 2002. Influence of gender on attention deficit hyperactivity disorder in children referred to a psychiatric clinic American Journal of Psychiatry, 159, 36-42.
BIZARRO, L., PATEL, S., MURTAGH, C. & STOLERMAN, I. P. 2004. Differential effects of psychomotor stimulants on attentional performance in rats: nicotine, amphetamine, caffeine and methylphenidate. Behavioural Pharmacology, 15, 195-206.
BLOEM, B., POORTHUIS, R. B. & MANSVELDER, H. D. 2014. Cholinergic modulation of the medial prefrontal cortex: the role of nicotinic receptors in attention and regulation of neuronal activity. Front Neural Circuits, 8, 17.
BLONDEAU, C. & DELLU-HAGEDORN, F. 2007. Dimensional analysis of ADHD subtypes in rats. Biol Psychiatry, 61, 1340-50.
BOLEA-ALAMANAC, B., NUTT, D. J., ADAMOU, M., ASHERSON, P., BAZIRE, S., COGHILL, D., HEAL, D., MULLER, U., NASH, J., SANTOSH, P., SAYAL, K., SONUGA-BARKE, E., YOUNG,
135
S. J. & BRITISH ASSOCIATION FOR, P. 2014. Evidence-based guidelines for the pharmacological management of attention deficit hyperactivity disorder: update on recommendations from the British Association for Psychopharmacology. J Psychopharmacol, 28, 179-203.
BOLLIMUNTA, A., CHEN, Y., SCHROEDER, C. E. & DING, M. 2008. Neuronal mechanisms of cortical alpha oscillations in awake-behaving macaques. J Neurosci, 28, 9976-88.
BOTVINICK, M. M., BRAVER, T. S., BARCH, D. M., CARTER, C. S. & COHEN, J. D. 2001. Conflict monitoring and cognitive control. Psychological Review, 108, 624-652.
BRIONI, J. D., KIM, D. J. & O'NEILL, A. B. 1996. Nicotine cue: lack of effect of the alpha 7 nicotinic receptor antagonist methyllycaconitine. Eur J Pharmacol, 301, 1-5.
BRISCHOUX, F., CHAKRABORTY, S., BRIERLEY, D. I. & UNGLESS, M. A. 2009. Phasic excitation of dopamine neurons in ventral VTA by noxious stimuli. Proc Natl Acad Sci U S A, 106, 4894-9.
BRISTOW, L. J., EASTON, A. E., LI, Y. W., SIVARAO, D. V., LIDGE, R., JONES, K. M., POST-MUNSON, D., DALY, C., LODGE, N. J., GALLAGHER, L., MOLSKI, T., PIESCHL, R., CHEN, P., HENDRICSON, A., WESTPHAL, R., COOK, J., IWUAGWU, C., MORGAN, D., BENITEX, Y., KING, D., MACOR, J. E., ZACZEK, R. & OLSON, R. 2016. The Novel, Nicotinic Alpha7 Receptor Partial Agonist, BMS-933043, Improves Cognition and Sensory Processing in Preclinical Models of Schizophrenia. PLoS One, 11, e0159996.
BROADBENT, D. 1958. Perception and Communication, London, Pergamon Press. BROD, M., SCHMITT, E., GOODWIN, M., HODGKINS, P. & NIEBLER, G. 2012. ADHD burden of
illness in older adults: a life course perspective. Qual Life Res, 21, 795-9. BROMBERG-MARTIN, E. S., MATSUMOTO, M. & HIKOSAKA, O. 2010. Dopamine in motivational
control: rewarding, aversive, and alerting. Neuron, 68, 815-34. BROUSSARD, J., SARTER, M. & GIVENS, B. 2006. Neuronal correlates of signal detection in the
posterior parietal cortex of rats performing a sustained attention task. Neuroscience, 143, 407-17.
BROUSSARD, J. I. 2012. Posterior parietal cortex dynamically ranks topographic signals via cholinergic influence. Front Integr Neurosci, 6, 32.
BROUSSARD, J. I. & GIVENS, B. 2010. Low frequency oscillations in rat posterior parietal cortex are differentially activated by cues and distractors. Neurobiol Learn Mem, 94, 191-8.
BROWN, K. J. & GRUNBERG, N. E. 1995. Effects of housing on male and female rats: crowding stresses male but calm females. Physiol Behav, 58, 1085-9.
BUCKHOLTZ, J. W., TREADWAY, M. T., COWAN, R. L., WOODWARD, N. D., LI, R., ANSARI, M. S., BALDWIN, R. M., SCHWARTZMAN, A. N., SHELBY, E. S., SMITH, C. E., KESSLER, R. M. & ZALD, D. H. 2010. Dopaminergic network differences in human impulsivity. Science, 329, 532.
BURNHAM, K. E., BANNERMAN, D. M., DAWSON, L. A., SOUTHAM, E., SHARP, T. & BAXTER, M. G. 2010. Fos expression in the brains of rats performing an attentional set-shifting task. Neuroscience, 171, 485-95.
BUSCHMAN, T. J. & MILLER, E. K. 2007. Top-down versus bottom-up control of attention in the prefrontal and posterior parietal cortices. Science, 315, 1860-2.
BYMASTER, F. P., KATNER, J. S., NELSON, D. L., HEMRICK-LUECKE, S. K., THRELKELD, P. G., HEILIGENSTEIN, J. H., MORIN, S. M., GEHLERT, D. R. & PERRY, K. W. 2002. Atomoxetine increases extracellular levels of norepinephrine and dopamine in prefrontal cortex of rat: a potential mechanism for efficacy in attention deficit/hyperactivity disorder. Neuropsychopharmacology, 27, 699-711.
CABALLERO-PUNTIVERIO, M., FITZPATRICK, C. M., WOLDBYE, D. P. & ANDREASEN, J. T. 2017. Effects of amphetamine and methylphenidate on attentional performance and impulsivity in the mouse 5-Choice Serial Reaction Time Task. J Psychopharmacol, 269881116684339.
136
CAPRIOLI, D., HONG, Y. T., SAWIAK, S. J., FERRARI, V., WILLIAMSON, D. J., JUPP, B., ADRIAN CARPENTER, T., AIGBIRHIO, F. I., EVERITT, B. J., ROBBINS, T. W., FRYER, T. D. & DALLEY, J. W. 2013. Baseline-dependent effects of cocaine pre-exposure on impulsivity and D2/3 receptor availability in the rat striatum: possible relevance to the attention-deficit hyperactivity syndrome. Neuropsychopharmacology, 38, 1460-71.
CAPRIOLI, D., SAWIAK, S. J., MERLO, E., THEOBALD, D. E., SPOELDER, M., JUPP, B., VOON, V., CARPENTER, T. A., EVERITT, B. J., ROBBINS, T. W. & DALLEY, J. W. 2014. Gamma aminobutyric acidergic and neuronal structural markers in the nucleus accumbens core underlie trait-like impulsive behavior. Biol Psychiatry, 75, 115-23.
CARDINAL, R. N., PENNICOTT, D. R., SUGATHAPALA, C. L., ROBBINS, T. W. & EVERITT, B. J. 2001. Impulsive choice induced in rats by lesions of the nucleus accumbens core. Science, 292, 2499-2501.
CARLI, M., ROBBINS, T. W., EVENDEN, J. L. & EVERITT, B. J. 1983. Effects of Lesions to Ascending Noradrenergic Neurons on Performance of a 5-Choice Serial Reaction Task in Rats - Implications for Theories of Dorsal Noradrenergic Bundle Function Based on Selective Attention and Arousal. Behavioural Brain Research, 9, 361-380.
CARTER, L., RUSSELL, P. N. & HELTON, W. S. 2013. Target predictability, sustained attention, and response inhibition. Brain and Cognition, 82, 35-42.
CASTANO-CANDAMIL, S., HOHNE, J., MARTINEZ-VARGAS, J. D., AN, X. W., CASTELLANOS-DOMINGUEZ, G. & HAUFE, S. 2015. Solving the EEG inverse problem based on space-time-frequency structured sparsity constraints. Neuroimage, 118, 598-612.
CHAMBERLAIN, S. R., MULLER, U., BLACKWELL, A. D., CLARK, L., ROBBINS, T. W. & SAHAKIAN, B. J. 2006. Neurochemical modulation of response inhibition and probabilistic learning in humans. Science, 311, 861-3.
CHELAZZI, L., PERLATO, A., SANTANDREA, E. & DELLA LIBERA, C. 2013. Rewards teach visual selective attention. Vision Research, 85, 58-72.
CHRISTAKOU, A., ROBBINS, T. W. & EVERITT, B. J. 2004. Prefrontal cortical-ventral striatal interactions involved in affective modulation of attentional performance: implications for corticostriatal circuit function. J Neurosci, 24, 773-80.
CLARK, R. A., SHOAIB, M., HEWITT, K. N., STANFORD, S. C. & BATE, S. T. 2012. A comparison of InVivoStat with other statistical software packages for analysis of data generated from animal experiments. Journal of Psychopharmacology, 26, 1136-42.
COHEN, J. D., BOTVINICK, M. & CARTER, C. S. 2000. Anterior cingulate and prefrontal cortex: who's in control? Nat Neurosci, 3, 421-3.
COLLINGS, R. D. 2003. Differences between ADHD inattentive and combined types on the CPT. Journal of Psychopathology and Behavioral Assessment, 25, 177-189.
COPE, Z. A., HALBERSTADT, A. L., VAN ENKHUIZEN, J., FLYNN, A. D., BREIER, M., SWERDLOW, N. R., GEYER, M. A. & YOUNG, J. W. 2016. Premature responses in the five-choice serial reaction time task reflect rodents' temporal strategies: evidence from no-light and pharmacological challenges. Psychopharmacology (Berl), 233, 3513-25.
CORBETTA, M. & SHULMAN, G. L. 2002. Control of goal-directed and stimulus-driven attention in the brain. Nature Reviews Neuroscience, 3, 201-215.
CORWIN, J. V. & REEP, R. L. 1998. Rodent posterior parietal cortex as a component of a cortical network mediating directed spatial attention. Psychobiology, 26, 87-102.
CRABBE, J. C. & CUNNINGHAM, C. L. 2007. Trait or state? Science, 317, 1033-5; author reply 1033-5.
CUNILL, R., CASTELLS, X., TOBIAS, A. & CAPELLA, D. 2013. Atomoxetine for attention deficit hyperactivity disorder in the adulthood: a meta-analysis and meta-regression. Pharmacoepidemiol Drug Saf, 22, 961-9.
DALLEY, J. W., CARDINAL, R. N. & ROBBINS, T. W. 2004a. Prefrontal executive and cognitive functions in rodents: neural and neurochemical substrates. Neurosci Biobehav Rev, 28, 771-84.
137
DALLEY, J. W., FRYER, T. D., BRICHARD, L., ROBINSON, E. S., THEOBALD, D. E., LAANE, K., PENA, Y., MURPHY, E. R., SHAH, Y., PROBST, K., ABAKUMOVA, I., AIGBIRHIO, F. I., RICHARDS, H. K., HONG, Y., BARON, J. C., EVERITT, B. J. & ROBBINS, T. W. 2007. Nucleus accumbens D2/3 receptors predict trait impulsivity and cocaine reinforcement. Science, 315, 1267-70.
DALLEY, J. W., THEOBALD, D. E., BOUGER, P., CHUDASAMA, Y., CARDINAL, R. N. & ROBBINS, T. W. 2004b. Cortical cholinergic function and deficits in visual attentional performance in rats following 192 IgG-saporin-induced lesions of the medial prefrontal cortex. Cereb Cortex, 14, 922-32.
DALMASO, M., CASTELLI, L., FRANCHETTI, L., CARLI, L., TODISCO, P., PALOMBA, D. & GALFANO, G. 2015. Altered orienting of attention in anorexia nervosa. Psychiatry Res, 229, 318-25.
DE JONGH, R., BOLT, I., SCHERMER, M. & OLIVIER, B. 2008. Botox for the brain: enhancement of cognition, mood and pro-social behavior and blunting of unwanted memories. Neurosci Biobehav Rev, 32, 760-76.
DICARLO, J. J., LANE, J. W., HSIAO, S. S. & JOHNSON, K. O. 1996. Marking microelectrode penetrations with fluorescent dyes. J Neurosci Methods, 64, 75-81.
DIERGAARDE, L., PATTIJ, T., NAWIJN, L., SCHOFFELMEER, A. N. & DE VRIES, T. J. 2009. Trait impulsivity predicts escalation of sucrose seeking and hypersensitivity to sucrose-associated stimuli. Behav Neurosci, 123, 794-803.
DIERGAARDE, L., PATTIJ, T., POORTVLIET, I., HOGENBOOM, F., DE VRIES, W., SCHOFFELMEER, A. N. & DE VRIES, T. J. 2008. Impulsive choice and impulsive action predict vulnerability to distinct stages of nicotine seeking in rats. Biol Psychiatry, 63, 301-8.
DONNELLY, N. A., HOLTZMAN, T., RICH, P. D., NEVADO-HOLGADO, A. J., FERNANDO, A. B., VAN DIJCK, G., HOLZHAMMER, T., PAUL, O., RUTHER, P., PAULSEN, O., ROBBINS, T. W. & DALLEY, J. W. 2014. Oscillatory activity in the medial prefrontal cortex and nucleus accumbens correlates with impulsivity and reward outcome. PLoS One, 9, e111300.
DONNELLY, N. A., PAULSEN, O., ROBBINS, T. W. & DALLEY, J. W. 2015. Ramping single unit activity in the medial prefrontal cortex and ventral striatum reflects the onset of waiting but not imminent impulsive actions. Eur J Neurosci, 41, 1524-37.
DONNY, E. C., CAGGIULA, A. R., ROWELL, P. P., GHARIB, M. A., MALDOVAN, V., BOOTH, S., MIELKE, M. M., HOFFMAN, A. & MCCALLUM, S. 2000. Nicotine self-administration in rats: estrous cycle effects, sex differences and nicotinic receptor binding. Psychopharmacology (Berl), 151, 392-405.
DOSENBACH, N. U. F., FAIR, D. A., COHEN, A. L., SCHLAGGAR, B. L. & PETERSEN, S. E. 2008. A dual-networks architecture of top-down control. Trends in Cognitive Sciences, 12, 99-105.
DOSENBACH, N. U. F., FAIR, D. A., MIEZIN, F. M., COHEN, A. L., WENGER, K. K., DOSENBACH, R. A. T., FOX, M. D., SNYDER, A. Z., VINCENT, J. L., RAICHLE, M. E., SCHLAGGAR, B. L. & PETERSEN, S. E. 2007. Distinct brain networks for adaptive and stable task control in humans. Proceedings of the National Academy of Sciences of the United States of America, 104, 11073-11078.
DOSHI, J. A., HODGKINS, P., KAHLE, J., SIKIRICA, V., CANGELOSI, M. J., SETYAWAN, J., ERDER, M. H. & NEUMANN, P. J. 2012. Economic impact of childhood and adult attention-deficit/hyperactivity disorder in the United States. J Am Acad Child Adolesc Psychiatry, 51, 990-1002 e2.
DUNCAN, J. 2010. The multiple-demand (MD) system of the primate brain: mental programs for intelligent behaviour. Trends Cogn Sci, 14, 172-9.
DUNCAN, J. 2013. The structure of cognition: attentional episodes in mind and brain. Neuron, 80, 35-50.
138
EAGLE, D. M. & ROBBINS, T. W. 2003. Inhibitory control in rats performing a stop-signal reaction-time task: Effects of lesions of the medial striatum and d-amphetamine. Behavioral Neuroscience, 117, 1302-1317.
EGELAND, J., JOHANSEN, S. N. & UELAND, T. 2009. Differentiating between ADHD sub-types on CCPT measures of sustained attention and vigilance. Scandinavian Journal of Psychology, 50, 347-354.
EPSTEIN, J. N., ERKANLI, A., CONNERS, C. K., KLARIC, J., COSTELLO, J. E. & ANGOLD, A. 2003. Relations between Continuous Performance Test performance measures and ADHD behaviors. J Abnorm Child Psychol, 31, 543-54.
EPSTEIN, J. N., JOHNSON, D. E., VARIA, I. M. & CONNERS, C. K. 2001. Neuropsychological assessment of response inhibition in adults with ADHD. J Clin Exp Neuropsychol, 23, 362-71.
FAIR, D. A., NIGG, J. T., BATHULA, D., MILLS, K. L. & AL., E. 2013. Distinct neural signatures detected for ADHD subtypes after controlling for micro-movements in resting state functional connectivity MRI data. Frontiers in Systems Neuroscience, 6, 1-31.
FARAONE, S. V. & ANTSHEL, K. M. 2008. Diagnosing and treating attention-deficit/hyperactivity, disorder in adults. World Psychiatry, 7, 131-136.
FARAONE, S. V., BIEDERMAN, J., FEIGHNER, J. A. & MONUTEAUX, M. C. 2000. Assessing symptoms of attention deficit hyperactivity disorder in children and adults: Which is more valid? Journal of Consulting and Clinical Psychology, 68, 830-842.
FARAONE, S. V., DOYLE, A. E., MICK, E. & BIEDERMAN, J. 2001. Meta-analysis of the association between the 7-repeat allele of the dopamine D-4 receptor gene and attention deficit hyperactivity disorder. American Journal of Psychiatry, 158, 1052-1057.
FAYYAD, J., DE GRAAF, R., KESSLER, R., ALONSO, J., ANGERMEYER, M., DEMYTTENAERE, K., DE GIROLAMO, G., HARO, J. M., KARAM, E. G., LARA, C., LEPINE, J. P., ORMEL, J., POSADA-VILLA, J., ZASLAVSKY, A. M. & JIN, R. 2007. Cross-national prevalence and correlates of adult attention-deficit hyperactivity disorder. Br J Psychiatry, 190, 402-9.
FEDORENKO, E., DUNCAN, J. & KANWISHER, N. 2013. Broad domain generality in focal regions of frontal and parietal cortex. Proc Natl Acad Sci U S A, 110, 16616-21.
FERNANDO, A. B., ECONOMIDOU, D., THEOBALD, D. E., ZOU, M. F., NEWMAN, A. H., SPOELDER, M., CAPRIOLI, D., MORENO, M., HIPOLITO, L., ASPINALL, A. T., ROBBINS, T. W. & DALLEY, J. W. 2012. Modulation of high impulsivity and attentional performance in rats by selective direct and indirect dopaminergic and noradrenergic receptor agonists. Psychopharmacology (Berl), 219, 341-52.
FINEBERG, N. A., HADDAD, P. M., CARPENTER, L., GANNON, B., SHARPE, R., YOUNG, A. H., JOYCE, E., ROWE, J., WELLSTED, D., NUTT, D. J. & SAHAKIAN, B. J. 2013. The size, burden and cost of disorders of the brain in the UK. J Psychopharmacol, 27, 761-70.
FINK, J. S., WEAVER, D. R., RIVKEES, S. A., PETERFREUND, R. A., POLLACK, A. E., ADLER, E. M. & REPPERT, S. M. 1992. Molecular cloning of the rat A2 adenosine receptor: selective co-expression with D2 dopamine receptors in rat striatum. Brain Res Mol Brain Res, 14, 186-95.
FINLEY, J. R., BENJAMIN, A. S. & MCCARLEY, J. S. 2014. Metacognition of multitasking: How well do we predict the costs of divided attention? J Exp Psychol Appl, 20, 158-65.
FLORY, K., MOLINA, B. S., PELHAM, W. E., JR., GNAGY, E. & SMITH, B. 2006. Childhood ADHD predicts risky sexual behavior in young adulthood. J Clin Child Adolesc Psychol, 35, 571-7.
FOLDI, N. S., LOBOSCO, J. J. & SCHAEFER, L. A. 2002. The effect of attentional dysfunction in Alzheimer's disease: theoretical and practical implications. Semin Speech Lang, 23, 139-50.
FORD, J. M., MATHALON, D. H., MARSH, L., FAUSTMAN, W. O., HARRIS, D., HOFF, A. L., BEAL, M. & PFEFFERBAUM, A. 1999. P300 amplitude is related to clinical state in severely and moderately ill patients with schizophrenia. Biol Psychiatry, 46, 94-101.
139
FRANKE, B., FARAONE, S. V., ASHERSON, P., BUITELAAR, J., BAU, C. H., RAMOS-QUIROGA, J. A., MICK, E., GREVET, E. H., JOHANSSON, S., HAAVIK, J., LESCH, K. P., CORMAND, B., REIF, A. & INTERNATIONAL MULTICENTRE PERSISTENT, A. C. 2012. The genetics of attention deficit/hyperactivity disorder in adults, a review. Mol Psychiatry, 17, 960-87.
FREDHOLM, B. B., CUNHA, R. A. & SVENNINGSSON, P. 2003. Pharmacology of adenosine A2A receptors and therapeutic applications. Curr Top Med Chem, 3, 413-26.
FREYER, F., BECKER, R., DINSE, H. R. & RITTER, P. 2013. State-dependent perceptual learning. J Neurosci, 33, 2900-7.
FRIES, P. 2015. Rhythms for Cognition: Communication through Coherence. Neuron, 88, 220-35.
FRIES, P., REYNOLDS, J. H., RORIE, A. E. & DESIMONE, R. 2001. Modulation of oscillatory neuronal synchronization by selective visual attention. Science, 291, 1560-3.
FUJII, S., JI, Z., MORITA, N. & SUMIKAWA, K. 1999. Acute and chronic nicotine exposure differentially facilitate the induction of LTP. Brain Res, 846, 137-43.
FULGONI, V. L., 3RD, KEAST, D. R. & LIEBERMAN, H. R. 2015. Trends in intake and sources of caffeine in the diets of US adults: 2001-2010. Am J Clin Nutr, 101, 1081-7.
GHAHREMANI, D. G., LEE, B., ROBERTSON, C. L., TABIBNIA, G., MORGAN, A. T., DE SHETLER, N., BROWN, A. K., MONTEROSSO, J. R., ARON, A. R., MANDELKERN, M. A., POLDRACK, R. A. & LONDON, E. D. 2012. Striatal dopamine D(2)/D(3) receptors mediate response inhibition and related activity in frontostriatal neural circuitry in humans. J Neurosci, 32, 7316-24.
GORNICK, M. C., ADDINGTON, A., SHAW, P., BOBB, A. J., SHARP, W., GREENSTEIN, D., AREPALLI, S., CASTELLANOS, F. X. & RAPOPORT, J. L. 2007. Association of the dopamine receptor D4 (DRD4) gene 7-repeat allele with children with attention-deficit/hyperactivity disorder (ADHD): An update. American Journal of Medical Genetics Part B-Neuropsychiatric Genetics, 144B, 379-382.
GRANON, S., PASSETTI, F., THOMAS, K. L., DALLEY, J. W., EVERITT, B. J. & ROBBINS, T. W. 2000. Enhanced and impaired attentional performance after infusion of D1 dopaminergic receptor agents into rat prefrontal cortex. J Neurosci, 20, 1208-15.
GRECH, R., CASSAR, T., MUSCAT, J., CAMILLERI, K. P., FABRI, S. G., ZERVAKIS, M., XANTHOPOULOS, P., SAKKALIS, V. & VANRUMSTE, B. 2008. Review on solving the inverse problem in EEG source analysis. J Neuroeng Rehabil, 5, 25.
GROEN, Y., GAASTRA, G. F., LEWIS-EVANS, B. & TUCHA, O. 2013. Risky behavior in gambling tasks in individuals with ADHD - a systematic literature review. Plos One, 8, e74909.
GROTTICK, A. J., HAMAN, M., WYLER, R. & HIGGINS, G. A. 2003. Reversal of a vigilance decrement in the aged rat by subtype-selective nicotinic ligands. Neuropsychopharmacology, 28, 880-7.
GROTTICK, A. J. & HIGGINS, G. A. 2000. Effect of subtype selective nicotinic compounds on attention as assessed by the five-choice serial reaction time task. Behav Brain Res, 117, 197-208.
GROTTICK, A. J. & HIGGINS, G. A. 2002. Assessing a vigilance decrement in aged rats: effects of pre-feeding, task manipulation, and psychostimulants. Psychopharmacology (Berl), 164, 33-41.
HAEGENS, S., BARCZAK, A., MUSACCHIA, G., LIPTON, M. L., MEHTA, A. D., LAKATOS, P. & SCHROEDER, C. E. 2015. Laminar Profile and Physiology of the alpha Rhythm in Primary Visual, Auditory, and Somatosensory Regions of Neocortex. J Neurosci, 35, 14341-52.
HAHN, B. & STOLERMAN, I. P. 2002. Nicotine-induced attentional enhancement in rats: effects of chronic exposure to nicotine. Neuropsychopharmacology, 27, 712-22.
HAIG, G. M., BAIN, E. E., ROBIESON, W. Z., BAKER, J. D. & OTHMAN, A. A. 2016. A Randomized Trial to Assess the Efficacy and Safety of ABT-126, a Selective alpha7 Nicotinic Acetylcholine Receptor Agonist, in the Treatment of Cognitive Impairment in Schizophrenia. Am J Psychiatry, appiajp201515010093.
140
HAJOS, M., HURST, R. S., HOFFMANN, W. E., KRAUSE, M., WALL, T. M., HIGDON, N. R. & GROPPI, V. E. 2005. The selective alpha7 nicotinic acetylcholine receptor agonist PNU-282987 [N-[(3R)-1-Azabicyclo[2.2.2]oct-3-yl]-4-chlorobenzamide hydrochloride] enhances GABAergic synaptic activity in brain slices and restores auditory gating deficits in anesthetized rats. J Pharmacol Exp Ther, 312, 1213-22.
HASLER, R., PERROUD, N., MEZIANE, H. B., HERRMANN, F., PRADA, P., GIANNAKOPOULOS, P. & DEIBER, M. P. 2016. Attention-related EEG markers in adult ADHD. Neuropsychologia, 87, 120-33.
HAYES, D. J., JUPP, B., SAWIAK, S. J., MERLO, E., CAPRIOLI, D. & DALLEY, J. W. 2014. Brain gamma-aminobutyric acid: a neglected role in impulsivity. European Journal of Neuroscience.
HAYWARD, A., ADAMSON, L. & NEILL, J. C. 2017. Partial agonism at the alpha7 nicotinic acetylcholine receptor improves attention, impulsive action and vigilance in low attentive rats. Eur Neuropsychopharmacol, 27, 325-335.
HAYWARD, A., TOMLINSON, A. & NEILL, J. C. 2016. Low attentive and high impulsive rats: A translational animal model of ADHD and disorders of attention and impulse control. Pharmacol Ther, 158, 41-51.
HEAL, D. J., CHEETHAM, S. C. & SMITH, S. L. 2009. The neuropharmacology of ADHD drugs in vivo: insights on efficacy and safety. Neuropharmacology, 57, 608-18.
HEAL, D. J., SMITH, S. L., GOSDEN, J. & NUTT, D. J. 2013. Amphetamine, past and present - a pharmacological and clinical perspective. Journal of Psychopharmacology, 27, 479-496.
HEAL, D. J., SMITH, S. L., KULKARNI, R. S. & ROWLEY, H. L. 2008. New perspectives from microdialysis studies in freely-moving, spontaneously hypertensive rats on the pharmacology of drugs for the treatment of ADHD. Pharmacol Biochem Behav, 90, 184-97.
HEIDBREDER, R. 2015. ADHD symptomatology is best conceptualized as a spectrum: a dimensional versus unitary approach to diagnosis. ADHD Attention Deficit Hyperactivity Disorder, DOI 10.1007/s 12402-015-0171-4.
HEISHMAN, S. J., KLEYKAMP, B. A. & SINGLETON, E. G. 2010. Meta-analysis of the acute effects of nicotine and smoking on human performance. Psychopharmacology (Berl), 210, 453-69.
HIGGINS, G. A., GRZELAK, M. E., POND, A. J., COHEN-WILLIAMS, M. E., HODGSON, R. A. & VARTY, G. B. 2007. The effect of caffeine to increase reaction time in the rat during a test of attention is mediated through antagonism of adenosine A2A receptors. Behav Brain Res, 185, 32-42.
HOLSON, R. R., SCALLET, A. C., ALI, S. F. & TURNER, B. B. 1991. "Isolation stress" revisited: isolation-rearing effects depend on animal care methods. Physiol Behav, 49, 1107-18.
HORREY, W. J. & WICKENS, C. D. 2006. Examining the impact of cell phone conversations on driving using meta-analytic techniques. Hum Factors, 48, 196-205.
HOWELLS, F. M., STEIN, D. J. & RUSSELL, V. A. 2012. Synergistic tonic and phasic activity of the locus coeruleus norepinephrine (LC-NE) arousal system is required for optimal attentional performance. Metab Brain Dis, 27, 267-74.
HOYLE, E., GENN, R. F., FERNANDES, C. & STOLERMAN, I. P. 2006. Impaired performance of alpha7 nicotinic receptor knockout mice in the five-choice serial reaction time task. Psychopharmacology (Berl), 189, 211-23.
HUANG-POLLOCK, C. L., KARALUNAS, S. L., TAM, H. & MOORE, A. N. 2012. Evaluating vigilance deficits in ADHD: a meta-analysis of CPT performance. J Abnorm Psychol, 121, 360-71.
HUANG, M., FELIX, A. R., FLOOD, D. G., BHUVANESWARAN, C., HILT, D., KOENIG, G. & MELTZER, H. Y. 2014. The novel alpha7 nicotinic acetylcholine receptor agonist EVP-6124 enhances dopamine, acetylcholine, and glutamate efflux in rat cortex and nucleus accumbens. Psychopharmacology (Berl).
141
HUSS, M., CHEN, W. & LUDOLPH, A. G. 2016. Guanfacine Extended Release: A New Pharmacological Treatment Option in Europe. Clin Drug Investig, 36, 1-25.
INGRAM, S., HECHTMAN, L. & MORGENSTERN, G. 1999. Outcome issues in ADHD: Adolescent and adult long-term outcome. Mental Retardation and Developmental Disabilities Research Reviews, 5, 243-250.
INSEL, T. R. & SAHAKIAN, B. J. 2012. Drug research: a plan for mental illness. Nature, 483, 269. IOANNIDIS, K., CHAMBERLAIN, S. R. & MULLER, U. 2014. Ostracising caffeine from the
pharmacological arsenal for attention-deficit hyperactivity disorder--was this a correct decision? A literature review. Journal of Psychopharmacology, 28, 830-6.
ISHERWOOD, S. N., PEKCEC, A., NICHOLSON, J. R., ROBBINS, T. W. & DALLEY, J. W. 2015. Dissociable effects of mGluR5 allosteric modulation on distinct forms of impulsivity in rats: interaction with NMDA receptor antagonism. Psychopharmacology (Berl).
JENSEN, O. & BONNEFOND, M. 2013. Prefrontal alpha- and beta-band oscillations are involved in rule selection. Trends in Cognitive Sciences, 17, 10-12.
JENSEN, O. & MAZAHERI, A. 2010. Shaping functional architecture by oscillatory alpha activity: gating by inhibition. Front Hum Neurosci, 4, 186.
JUPP, B., CAPRIOLI, D. & DALLEY, J. W. 2013. Highly impulsive rats: modelling an endophenotype to determine the neurobiological, genetic and environmental mechanisms of addiction. Dis Model Mech, 6, 302-11.
KEEFE, R. S., MELTZER, H. A., DGETLUCK, N., GAWRYL, M., KOENIG, G., MOEBIUS, H. J., LOMBARDO, I. & HILT, D. C. 2015. Randomized, Double-Blind, Placebo-Controlled Study of Encenicline, an alpha7 Nicotinic Acetylcholine Receptor Agonist, as a Treatment for Cognitive Impairment in Schizophrenia. Neuropsychopharmacology, 40, 3053-60.
KIELING, C., GONCALVES, R. R. F., TANNOCK, R. & CASTELLANOS, F. X. 2008. Neurobiology of attention deficit hyperactivity disorder. Child and Adolescent Psychiatric Clinics of North America, 17, 285-+.
KLEIN, M., ONNINK, M., VAN DONKELAAR, M., WOLFERS, T., HARICH, B., SHI, Y., DAMMERS, J., ARIAS-VA SQUEZ, A., HOOGMAN, M. & FRANKE, B. 2017. Brain imaging genetics in ADHD and beyond - mapping pathways from gene to disorder at different levels of complexity. Neurosci Biobehav Rev.
KLIMESCH, W. 2012. alpha-band oscillations, attention, and controlled access to stored information. Trends Cogn Sci, 16, 606-17.
KLINKENBERG, I. & BLOKLAND, A. 2010. The validity of scopolamine as a pharmacological model for cognitive impairment: a review of animal behavioral studies. Neurosci Biobehav Rev, 34, 1307-50.
KNYAZEV, G. G. 2012. EEG delta oscillations as a correlate of basic homeostatic and motivational processes. Neurosci Biobehav Rev, 36, 677-95.
KOLISNYK, B., AL-ONAIZI, M. A., PRADO, V. F. & PRADO, M. A. 2015. alpha7 nicotinic ACh receptor-deficient mice exhibit sustained attention impairments that are reversed by beta2 nicotinic ACh receptor activation. Br J Pharmacol, 172, 4919-31.
KOOIJ, S. J., BEJEROT, S., BLACKWELL, A., CACI, H., CASAS-BRUGUE, M., CARPENTIER, P. J., EDVINSSON, D., FAYYAD, J., FOEKEN, K., FITZGERALD, M., GAILLAC, V., GINSBERG, Y., HENRY, C., KRAUSE, J., LENSING, M. B., MANOR, I., NIEDERHOFER, H., NUNES-FILIPE, C., OHLMEIER, M. D., OSWALD, P., PALLANTI, S., PEHLIVANIDIS, A., RAMOS-QUIROGA, J. A., RASTAM, M., RYFFEL-RAWAK, D., STES, S. & ASHERSON, P. 2010. European consensus statement on diagnosis and treatment of adult ADHD: The European Network Adult ADHD. BMC Psychiatry, 10, 67.
KURIYAN, A. B., PELHAM, W. E., JR., MOLINA, B. S., WASCHBUSCH, D. A., GNAGY, E. M., SIBLEY, M. H., BABINSKI, D. E., WALTHER, C., CHEONG, J., YU, J. & KENT, K. M. 2013. Young adult educational and vocational outcomes of children diagnosed with ADHD. J Abnorm Child Psychol, 41, 27-41.
142
LACHTER, J., FORSTER, K. I. & RUTHRUFF, E. 2004. Forty-five years after Broadbent (1958): still no identification without attention. Psychol Rev, 111, 880-913.
LANGNER, I., GARBE, E., BANASCHEWSKI, T. & MIKOLAJCZYK, R. T. 2013. Twin and Sibling Studies Using Health Insurance Data: The Example of Attention Deficit/Hyperactivity Disorder (ADHD). Plos One, 8.
LARSSON, H., DILSHAD, R., LICHTENSTEIN, P. & BARKER, E. D. 2011. Developmental trajectories of DSM-IV symptoms of attention-deficit/hyperactivity disorder: genetic effects, family risk and associated psychopathology. J Child Psychol Psychiatry, 52, 954-63.
LAZARUS, M., SHEN, H. Y., CHERASSE, Y., QU, W. M., HUANG, Z. L., BASS, C. E., WINSKY-SOMMERER, R., SEMBA, K., FREDHOLM, B. B., BOISON, D., HAYAISHI, O., URADE, Y. & CHEN, J. F. 2011. Arousal effect of caffeine depends on adenosine A2A receptors in the shell of the nucleus accumbens. J Neurosci, 31, 10067-75.
LAZZARO, I., GORDON, E., LI, W., LIM, C. L., PLAHN, M., WHITMONT, S., CLARKE, S., BARRY, R. J., DOSEN, A. & MEARES, R. 1999. Simultaneous EEG and EDA measures in adolescent attention deficit hyperactivity disorder. Int J Psychophysiol, 34, 123-34.
LE, H. H., HODGKINS, P., POSTMA, M. J., KAHLE, J., SIKIRICA, V., SETYAWAN, J., ERDER, M. H. & DOSHI, J. A. 2014. Economic impact of childhood/adolescent ADHD in a European setting: the Netherlands as a reference case. Eur Child Adolesc Psychiatry, 23, 587-98.
LEE, J. H., JEONG, S. K., KIM, B. C., PARK, K. W. & DASH, A. 2015. Donepezil across the spectrum of Alzheimer's disease: dose optimization and clinical relevance. Acta Neurol Scand, 131, 259-67.
LEI, D., MA, J., DU, X. & SHEN 2014. Microstructural Abnormalities in the Combined and Inattentive Subtypes of Attention Deficit Hyperactivity Disorder: a Diffusion Tensor Imaging Study. Scientific Reports, 4.
LEUCHT, S., HIERL, S., KISSLING, W., DOLD, M. & DAVIS, J. M. 2012. Putting the efficacy of psychiatric and general medicine medication into perspective: review of meta-analyses. Br J Psychiatry, 200, 97-106.
LI, D. W., SHAM, P. C., OWEN, M. J. & HE, L. 2006. Meta-analysis shows significant association between dopamine system genes and attention deficit hyperactivity disorder (ADHD). Human Molecular Genetics, 15, 2276-2284.
LIM, M. H., KIM, H. W., PAIK, K. C., CHO, S. C., YOON, D. Y. & LEE, H. J. 2006. Association of the DAT1 polymorphism with attention deficit hyperactivity disorder (ADHD): A family-based approach. American Journal of Medical Genetics Part B-Neuropsychiatric Genetics, 141B, 309-311.
LIN, H. Y., TSENG, W. Y. I., LAI, M. C., MATSUO, K. & GAU, S. S. F. 2015. Altered Resting-State Frontoparietal Control Network in Children with Attention-Deficit/Hyperactivity Disorder. Journal of the International Neuropsychological Society, 21, 271-284.
LOO, S. K., HOPFER, C., TEALE, P. D. & REITE, M. L. 2004. EEG correlates of methylphenidate response in ADHD: association with cognitive and behavioral measures. J Clin Neurophysiol, 21, 457-64.
LORINCZ, M. L., KEKESI, K. A., JUHASZ, G., CRUNELLI, V. & HUGHES, S. W. 2009. Temporal framing of thalamic relay-mode firing by phasic inhibition during the alpha rhythm. Neuron, 63, 683-96.
LUCK, S. J. & GOLD, J. M. 2008. The construct of attention in schizophrenia. Biol Psychiatry, 64, 34-9.
LUSTIG, C., KOZAK, R., SARTER, M., YOUNG, J. W. & ROBBINS, T. W. 2013. CNTRICS final animal model task selection: control of attention. Neurosci Biobehav Rev, 37, 2099-110.
MACKWORTH, N. H. 1948. The Breakdown of Vigilance during Prolonged Visual Search. Quarterly Journal of Experimental Psychology, 1, 6-21.
MAHER, B. 2008. Personal genomes: The case of the missing heritability. Nature, 456, 18-21. MANSVELDER, H. D. & MCGEHEE, D. S. 2000. Long-term potentiation of excitatory inputs to
brain reward areas by nicotine. Neuron, 27, 349-57.
143
MARKOU, A., CHIAMULERA, C., GEYER, M. A., TRICKLEBANK, M. & STECKLER, T. 2009. Removing obstacles in neuroscience drug discovery: the future path for animal models. Neuropsychopharmacology, 34, 74-89.
MARSH, P. J. & WILLIAMS, L. M. 2004. An investigation of individual typologies of attention-deficit hyperactivity disorder using cluster analysis of DSM-IV criteria. Personality and Individual Differences, 36, 1187-1195.
MARSHALL, T. R., BERGMANN, T. O. & JENSEN, O. 2015. Frontoparietal Structural Connectivity Mediates the Top-Down Control of Neuronal Synchronization Associated with Selective Attention. PLoS Biol, 13, e1002272.
MARSTON, H. M. 1996. Analysis of cognitive function in animals, the value of SDT. Brain Res Cogn Brain Res, 3, 269-77.
MARTIN, T. J., GRIGG, A., KIM, S. A., RIRIE, D. G. & EISENACH, J. C. 2015. Assessment of attention threshold in rats by titration of visual cue duration during the five choice serial reaction time task. J Neurosci Methods, 241, 37-43.
MATSUMOTO, M. & HIKOSAKA, O. 2009. Two types of dopamine neuron distinctly convey positive and negative motivational signals. Nature, 459, 837-41.
MATTA, S. G., BALFOUR, D. J., BENOWITZ, N. L., BOYD, R. T., BUCCAFUSCO, J. J., CAGGIULA, A. R., CRAIG, C. R., COLLINS, A. C., DAMAJ, M. I., DONNY, E. C., GARDINER, P. S., GRADY, S. R., HEBERLEIN, U., LEONARD, S. S., LEVIN, E. D., LUKAS, R. J., MARKOU, A., MARKS, M. J., MCCALLUM, S. E., PARAMESWARAN, N., PERKINS, K. A., PICCIOTTO, M. R., QUIK, M., ROSE, J. E., ROTHENFLUH, A., SCHAFER, W. R., STOLERMAN, I. P., TYNDALE, R. F., WEHNER, J. M. & ZIRGER, J. M. 2007. Guidelines on nicotine dose selection for in vivo research. Psychopharmacology (Berl), 190, 269-319.
MAZER, J. A. 2011. Spatial Attention, Feature-Based Attention, and Saccades: Three Sides of One Coin? Biological Psychiatry, 69, 1147-1152.
MCKENNA, B. S., YOUNG, J. W., DAWES, S. E., ASGAARD, G. L. & EYLER, L. T. 2013. Bridging the bench to bedside gap: validation of a reverse-translated rodent continuous performance test using functional magnetic resonance imaging. Psychiatry Research-Neuroimaging, 212, 183-191.
MCLEAN, S. L., GRAYSON, B., IDRIS, N. F., LESAGE, A. S., PEMBERTON, D. J., MACKIE, C. & NEILL, J. C. 2011. Activation of alpha7 nicotinic receptors improves phencyclidine-induced deficits in cognitive tasks in rats: implications for therapy of cognitive dysfunction in schizophrenia. Eur Neuropsychopharmacol, 21, 333-43.
MCLEAN, S. L., GRAYSON, B., MARSH, S., ZARROUG, S. H., HARTE, M. K. & NEILL, J. C. 2016. Nicotinic alpha7 and alpha4beta2 agonists enhance the formation and retrieval of recognition memory: Potential mechanisms for cognitive performance enhancement in neurological and psychiatric disorders. Behav Brain Res, 302, 73-80.
MINZENBERG, M. J. 2012. Pharmacotherapy for attention-deficit/hyperactivity disorder: from cells to circuits. Neurotherapeutics, 9, 610-21.
MOLINA, B. S., HINSHAW, S. P., SWANSON, J. M., ARNOLD, L. E., VITIELLO, B., JENSEN, P. S., EPSTEIN, J. N., HOZA, B., HECHTMAN, L., ABIKOFF, H. B., ELLIOTT, G. R., GREENHILL, L. L., NEWCORN, J. H., WELLS, K. C., WIGAL, T., GIBBONS, R. D., HUR, K., HOUCK, P. R. & GROUP, M. T. A. C. 2009. The MTA at 8 years: prospective follow-up of children treated for combined-type ADHD in a multisite study. J Am Acad Child Adolesc Psychiatry, 48, 484-500.
MOORE, H. 2010. The role of rodent models in the discovery of new treatments for schizophrenia: updating our strategy. Schizophr Bull, 36, 1066-72.
MORENO, M., ECONOMIDOU, D., MAR, A. C., LOPEZ-GRANERO, C., CAPRIOLI, D., THEOBALD, D. E., FERNANDO, A., NEWMAN, A. H., ROBBINS, T. W. & DALLEY, J. W. 2013. Divergent effects of D(2)/(3) receptor activation in the nucleus accumbens core and shell on impulsivity and locomotor activity in high and low impulsive rats. Psychopharmacology (Berl), 228, 19-30.
144
MUELLER, S., COSTA, A., KEESER, D., POGARELL, O., BERMAN, A., COATES, U., REISER, M. F., RIEDEL, M., MOLLER, H. J., ETTINGER, U. & MEINDL, T. 2014. The effects of methylphenidate on whole brain intrinsic functional connectivity. Human Brain Mapping.
MUIR, J. L., EVERITT, B. J. & ROBBINS, T. W. 1995. Reversal of visual attentional dysfunction following lesions of the cholinergic basal forebrain by physostigmine and nicotine but not by the 5-HT3 receptor antagonist, ondansetron. Psychopharmacology (Berl), 118, 82-92.
MUIR, J. L., EVERITT, B. J. & ROBBINS, T. W. 1996. The cerebral cortex of the rat and visual attentional function: dissociable effects of mediofrontal, cingulate, anterior dorsolateral, and parietal cortex lesions on a five-choice serial reaction time task. Cerebral Cortex, 6, 470-81.
MULLER, M. M., GRUBER, T. & KEIL, A. 2000. Modulation of induced gamma band activity in the human EEG by attention and visual information processing. Int J Psychophysiol, 38, 283-99.
MURPHY, E. R., FERNANDO, A. B., URCELAY, G. P., ROBINSON, E. S., MAR, A. C., THEOBALD, D. E., DALLEY, J. W. & ROBBINS, T. W. 2012. Impulsive behaviour induced by both NMDA receptor antagonism and GABAA receptor activation in rat ventromedial prefrontal cortex. Psychopharmacology (Berl), 219, 401-10.
MURPHY, G., GROEGER, J. A. & GREENE, C. M. 2016. Twenty years of load theory-Where are we now, and where should we go next? Psychon Bull Rev, 23, 1316-1340.
NEILL, J. C., HARTE, M. K., HADDAD, P. M., LYDALL, E. S. & DWYER, D. M. 2014. Acute and chronic effects of NMDA receptor antagonists in rodents, relevance to negative symptoms of schizophrenia: a translational link to humans. Eur Neuropsychopharmacol, 24, 822-35.
NEILL, J. C. & HENDRIE, C. A. 2012. Bringing Order to Disorder. Public Service Review: UK Science and Technology
NESTLER, E. J. & HYMAN, S. E. 2010. Animal models of neuropsychiatric disorders. Nat Neurosci, 13, 1161-9.
NEWMAN, L. A. & MCGAUGHY, J. 2008. Cholinergic deafferentation of prefrontal cortex increases sensitivity to cross-modal distractors during a sustained attention task. J Neurosci, 28, 2642-50.
NUECHTERLEIN, K. H., GREEN, M. F., CALKINS, M. E., GREENWOOD, T. A., GUR, R. E., GUR, R. C., LAZZERONI, L. C., LIGHT, G. A., RADANT, A. D., SEIDMAN, L. J., SIEVER, L. J., SILVERMAN, J. M., SPROCK, J., STONE, W. S., SUGAR, C. A., SWERDLOW, N. R., TSUANG, D. W., TSUANG, M. T., TURETSKY, B. I. & BRAFF, D. L. 2015. Attention/vigilance in schizophrenia: performance results from a large multi-site study of the Consortium on the Genetics of Schizophrenia (COGS). Schizophr Res, 163, 38-46.
NUECHTERLEIN, K. H., GREEN, M. F., KERN, R. S., BAADE, L. E., BARCH, D. M., COHEN, J. D., ESSOCK, S., FENTON, W. S., FRESE, F. J., 3RD, GOLD, J. M., GOLDBERG, T., HEATON, R. K., KEEFE, R. S., KRAEMER, H., MESHOLAM-GATELY, R., SEIDMAN, L. J., STOVER, E., WEINBERGER, D. R., YOUNG, A. S., ZALCMAN, S. & MARDER, S. R. 2008. The MATRICS Consensus Cognitive Battery, part 1: test selection, reliability, and validity. Am J Psychiatry, 165, 203-13.
NUTT, D. & GOODWIN, G. 2011. ECNP Summit on the future of CNS drug research in Europe 2011: report prepared for ECNP by David Nutt and Guy Goodwin. Eur Neuropsychopharmacol, 21, 495-9.
OGG, R. J., ZOU, P., ALLEN, D. N., HUTCHINS, S. B., DUTKIEWICZ, R. M. & MULHERN, R. K. 2008. Neural correlates of a clinical continuous performance test. Magnetic Resonance Imaging, 26, 504-512.
OLINCY, A. & FREEDMAN, R. 2012. Nicotinic mechanisms in the treatment of psychotic disorders: a focus on the alpha7 nicotinic receptor. Handb Exp Pharmacol, 211-32.
145
ORZACK, M. H. & KORNETSKY, C. 1966. Attention dysfunction in chronic schizophrenia. Arch Gen Psychiatry, 14, 323-6.
PAINE, T. A., SLIPP, L. E. & CARLEZON, W. A., JR. 2011. Schizophrenia-like attentional deficits following blockade of prefrontal cortex GABAA receptors. Neuropsychopharmacology, 36, 1703-13.
PALANIVEL, V., PATEL, N., ANJAY, M. A., GOODWIN, A. & PRADHAN, S. 2009. Audit of Attention Deficit Hyperactivity Disorder (Adhd) Clinics in the "Pre-Publication Era'' of National Institute of Clinical Excellence (Nice) Guidelines. Acta Paediatrica, 98, 204-205.
PARIKH, V., KOZAK, R., MARTINEZ, V. & SARTER, M. 2007. Prefrontal acetylcholine release controls cue detection on multiple timescales. Neuron, 56, 141-54.
PARIKH, V., KUTLU, M. G. & GOULD, T. J. 2016. nAChR dysfunction as a common substrate for schizophrenia and comorbid nicotine addiction: Current trends and perspectives. Schizophr Res, 171, 1-15.
PATERSON, N. E., RICCIARDI, J., WETZLER, C. & HANANIA, T. 2011. Sub-optimal performance in the 5-choice serial reaction time task in rats was sensitive to methylphenidate, atomoxetine and d-amphetamine, but unaffected by the COMT inhibitor tolcapone. Neurosci Res, 69, 41-50.
PAXINOS, G., WATSON, C. 2006. The Rat Brain in Stereotaxic Coordinates. PEHRSON, A. L., BONDI, C. O., TOTAH, N. K. & MOGHADDAM, B. 2013. The influence of NMDA
and GABA(A) receptors and glutamic acid decarboxylase (GAD) activity on attention. Psychopharmacology (Berl), 225, 31-9.
PERRY, J. L. & CARROLL, M. E. 2008. The role of impulsive behavior in drug abuse. Psychopharmacology (Berl), 200, 1-26.
PERRY, R. J., WATSON, P. & HODGES, J. R. 2000. The nature and staging of attention dysfunction in early (minimal and mild) Alzheimer's disease: relationship to episodic and semantic memory impairment. Neuropsychologia, 38, 252-71.
PETERSEN, S. E. & POSNER, M. I. 2012. The Attention System of the Human Brain: 20 Years After. Annual Review of Neuroscience, Vol 35, 35, 73-89.
PEZZE, M., MCGARRITY, S., MASON, R., FONE, K. C. & BAST, T. 2014. Too little and too much: hypoactivation and disinhibition of medial prefrontal cortex cause attentional deficits. J Neurosci, 34, 7931-46.
PHARMACEUTICALS, F. 2015. FORUM Pharmaceuticals Inc. Announces Removal of Partial Clinical Hold on Phase 3 Encenicline Program for Cognitive Impairment in Schizophrenia http://www.forumpharma.com/forum-pharmaceuticals-inc-announces-removal-of-partial-clinical-hold-on-phase-3-encenicline-program-for-cognitive-impairment-in-schizophrenia/.
PHARMACEUTICALS, F. 2016. FORUM Pharmaceuticals Inc. Provides Update on Encenicline Phase 3 Clinical Trial Program in Cognitive Impairment in Schizophrenia In: HEINS, S. (ed.). http://www.forumpharma.com/forum-pharmaceuticals-inc-provides-update-encenicline-phase-3-clinical-trial-program-cognitive-impairment-schizophrenia/.
PHILLIPS, R. C., SALO, T. & CARTER, C. S. 2015. Distinct neural correlates for attention lapses in patients with schizophrenia and healthy participants. Front Hum Neurosci, 9, 502.
PICCIOTTO, M. R. 2003. Nicotine as a modulator of behavior: beyond the inverted U. Trends Pharmacol Sci, 24, 493-9.
PICCIOTTO, M. R., ZOLI, M., RIMONDINI, R., LENA, C., MARUBIO, L. M., PICH, E. M., FUXE, K. & CHANGEUX, J. P. 1998. Acetylcholine receptors containing the beta2 subunit are involved in the reinforcing properties of nicotine. Nature, 391, 173-7.
PIETRZAK, R. H., MOLLICA, C. M., MARUFF, P. & SNYDER, P. J. 2006. Cognitive effects of immediate-release methylphenidate in children with attention-deficit/hyperactivity disorder. Neurosci Biobehav Rev, 30, 1225-45.
POLTAVSKI, D. V. & PETROS, T. 2006. Effects of transdermal nicotine on attention in adult non-smokers with and without attentional deficits. Physiol Behav, 87, 614-24.
146
PONS, S., FATTORE, L., COSSU, G., TOLU, S., PORCU, E., MCINTOSH, J. M., CHANGEUX, J. P., MASKOS, U. & FRATTA, W. 2008. Crucial role of alpha4 and alpha6 nicotinic acetylcholine receptor subunits from ventral tegmental area in systemic nicotine self-administration. J Neurosci, 28, 12318-27.
POPPE, A. B., BARCH, D. M., CARTER, C. S., GOLD, J. M., RAGLAND, J. D., SILVERSTEIN, S. M. & MACDONALD, A. W. 2016. Reduced Frontoparietal Activity in Schizophrenia Is Linked to a Specific Deficit in Goal Maintenance: A Multisite Functional Imaging Study. Schizophrenia Bulletin, 42, 1149-1157.
PRENDERGAST, B. J., ONISHI, K. G. & ZUCKER, I. 2014. Female mice liberated for inclusion in neuroscience and biomedical research. Neurosci Biobehav Rev, 40, 1-5.
PRESKORN, S. H., GAWRYL, M., DGETLUCK, N., PALFREYMAN, M., BAUER, L. O. & HILT, D. C. 2014. Normalizing effects of EVP-6124, an alpha-7 nicotinic partial agonist, on event-related potentials and cognition: a proof of concept, randomized trial in patients with schizophrenia. J Psychiatr Pract, 20, 12-24.
PRICKAERTS, J., VAN GOETHEM, N. P., CHESWORTH, R., SHAPIRO, G., BOESS, F. G., METHFESSEL, C., RENEERKENS, O. A., FLOOD, D. G., HILT, D., GAWRYL, M., BERTRAND, S., BERTRAND, D. & KONIG, G. 2012. EVP-6124, a novel and selective alpha7 nicotinic acetylcholine receptor partial agonist, improves memory performance by potentiating the acetylcholine response of alpha7 nicotinic acetylcholine receptors. Neuropharmacology, 62, 1099-110.
PTAK, R. 2012. The frontoparietal attention network of the human brain: action, saliency, and a priority map of the environment. Neuroscientist, 18, 502-15.
PUNJA, S., SHAMSEER, L., HARTLING, L., URICHUK, L., VANDERMEER, B., NIKLES, J. & VOHRA, S. 2016. Amphetamines for attention deficit hyperactivity disorder (ADHD) in children and adolescents. Cochrane Database Syst Rev, 2, CD009996.
PUUMALA, T., RUOTSALAINEN, S., JAKALA, P., KOIVISTO, E., RIEKKINEN, P., JR. & SIRVIO, J. 1996. Behavioral and pharmacological studies on the validation of a new animal model for attention deficit hyperactivity disorder. Neurobiol Learn Mem, 66, 198-211.
PUUMALA, T. & SIRVIO, J. 1998. Changes in activities of dopamine and serotonin systems in the frontal cortex underlie poor choice accuracy and impulsivity of rats in an attention task. Neuroscience, 83, 489-99.
RAMTEKKAR, U. P., REIERSEN, A. M., TODOROV, A. A. & TODD, R. D. 2010a. Sex and age differences in attention-deficit/hyperactivity disorder symptoms and diagnoses: implications for DSM-V and ICD-11. J Am Acad Child Adolesc Psychiatry, 49, 217-28 e1-3.
RAMTEKKAR, U. P., REIERSEN, A. M., TOROROV, A. A. & TODD, R. D. 2010b. Sex and age differences in attention-deficit/hyperactivity disorder symptoms and diagnoses: Implications for DSM-V and ICD-11. Journal of the American Academy of Child and Adolescent Psychiatry, 49, 217-227.
REZVANI, A. H., KHOLDEBARIN, E., BRUCATO, F. H., CALLAHAN, P. M., LOWE, D. A. & LEVIN, E. D. 2009. Effect of R3487/MEM3454, a novel nicotinic alpha7 receptor partial agonist and 5-HT3 antagonist on sustained attention in rats. Prog Neuropsychopharmacol Biol Psychiatry, 33, 269-75.
REZVANI, A. H. & LEVIN, E. D. 2001. Cognitive effects of nicotine. Biol Psychiatry, 49, 258-67. RICCIO, C. A., REYNOLDS, C. R., LOWE, P. & MOORE, J. J. 2002. The continuous performance
test: a window on the neural substrates for attention? Arch Clin Neuropsychol, 17, 235-72.
RICCIONI, T., LEONARDI, F. & BORSINI, F. 2010. Adenosine A(2A) Receptor Binding Profile of Two Antagonists, ST1535 and KW6002: Consideration on the Presence of Atypical Adenosine A(2A) Binding Sites. Front Psychiatry, 1, 22.
ROBBINS, T. W. 2002. The 5-choice serial reaction time task: behavioural pharmacology and functional neurochemistry. Psychopharmacology, 163, 362-380.
147
ROBINSON, E. S., EAGLE, D. M., ECONOMIDOU, D., THEOBALD, D. E., MAR, A. C., MURPHY, E. R., ROBBINS, T. W. & DALLEY, J. W. 2009. Behavioural characterisation of high impulsivity on the 5-choice serial reaction time task: specific deficits in 'waiting' versus 'stopping'. Behav Brain Res, 196, 310-6.
SAGAR, S. M., SHARP, F. R. & CURRAN, T. 1988. Expression of c-fos protein in brain: metabolic mapping at the cellular level. Science, 240, 1328-31.
SARA, S. J. & BOURET, S. 2012. Orienting and Reorienting: The Locus Coeruleus Mediates Cognition through Arousal. Neuron, 76, 130-141.
SCHACHTER, H. M., PHAM, B., KING, J., LANGFORD, S. & MOHER, D. 2001. How efficacious and safe is short-acting methylphenidate for the treatment of attention-deficit disorder in children and adolescents? A meta-analysis. CMAJ, 165, 1475-88.
SCHACK, B., KLIMESCH, W. & SAUSENG, P. 2005. Phase synchronization between theta and upper alpha oscillations in a working memory task. Int J Psychophysiol, 57, 105-14.
SCHNACKENBERG, R. C. 1973. Caffeine as a substitute for Schedule II stimulants in hyperkinetic children. Am J Psychiatry, 130, 796-8.
SCHREIBER, T. 2000. Measuring information transfer. Phys Rev Lett, 85, 461-4. SCOLARI, M., SEIDL-RATHKOPF, K. N. & KASTNER, S. 2015. Functions of the human
frontoparietal attention network: Evidence from neuroimaging. Curr Opin Behav Sci, 1, 32-39.
SEGARRA, R., ZABALA, A., EGUILUZ, J. I., OJEDA, N., ELIZAGARATE, E., SANCHEZ, P., BALLESTEROS, J. & GUTIERREZ, M. 2011. Cognitive performance and smoking in first-episode psychosis: the self-medication hypothesis. Eur Arch Psychiatry Clin Neurosci, 261, 241-50.
SELLERS, K. K., YU, C., ZHOU, Z. C., STITT, I., LI, Y., RADTKE-SCHULLER, S., ALAGAPAN, S. & FROHLICH, F. 2016. Oscillatory Dynamics in the Frontoparietal Attention Network during Sustained Attention in the Ferret. Cell Rep, 16, 2864-74.
SERGEANT, J. 2000. The cognitive-energetic model: an empirical approach to attention-deficit hyperactivity disorder. Neurosci Biobehav Rev, 24, 7-12.
SHARMA, K., DAVIS, T. & COULTHARD, E. 2016. Enhancing attention in neurodegenerative diseases: current therapies and future directions. Transl Neurosci, 7, 98-109.
SHI, T., LI, X., SONG, J., ZHAO, N., SUN, C., XIA, W., WU, L. & TOMODA, A. 2012. EEG characteristics and visual cognitive function of children with attention deficit hyperactivity disorder (ADHD). Brain Dev, 34, 806-11.
SHIINA, A., SHIRAYAMA, Y., NIITSU, T., HASHIMOTO, T., YOSHIDA, T., HASEGAWA, T., HARAGUCHI, T., KANAHARA, N., SHIRAISHI, T., FUJISAKI, M., FUKAMI, G., NAKAZATO, M., IYO, M. & HASHIMOTO, K. 2010. A randomised, double-blind, placebo-controlled trial of tropisetron in patients with schizophrenia. Ann Gen Psychiatry, 9, 27.
SHIOIRI, S., HONJYO, H., KASHIWASE, Y., MATSUMIYA, K. & KURIKI, I. 2016. Visual attention spreads broadly but selects information locally. Sci Rep, 6, 35513.
SIEGEL, M., DONNER, T. H., OOSTENVELD, R., FRIES, P. & ENGEL, A. K. 2008. Neuronal synchronization along the dorsal visual pathway reflects the focus of spatial attention. Neuron, 60, 709-19.
SPAGNA, A., DONG, Y., MACKIE, M. A., LI, M., HARVEY, P. D., TIAN, Y., WANG, K. & FAN, J. 2015. Clozapine improves the orienting of attention in schizophrenia. Schizophr Res, 168, 285-91.
STUART, S. A., BUTLER, P., MUNAFO, M. R., NUTT, D. J. & ROBINSON, E. S. 2013. A translational rodent assay of affective biases in depression and antidepressant therapy. Neuropsychopharmacology, 38, 1625-35.
TAYLOR, G. T. & MALONEY, S. 2010. Gender differences and the role of estrogen in cognitive enhancements with nicotine in rats. Pharmacol Biochem Behav, 95, 139-45.
TOMASI, D., VOLKOW, N. D., WANG, G. J., WANG, R., TELANG, F., CAPARELLI, E. C., WONG, C., JAYNE, M. & FOWLER, J. S. 2011. Methylphenidate enhances brain activation and
148
deactivation responses to visual attention and working memory tasks in healthy controls. Neuroimage, 54, 3101-10.
TOMLINSON, A., GRAYSON, B., MARSH, S., HARTE, M. K., BARNES, S. A., MARSHALL, K. M. & NEILL, J. C. 2014. Pay attention to impulsivity: Modelling low attentive and high impulsive subtypes of adult ADHD in the 5-choice continuous performance task (5C-CPT) in female rats. Eur Neuropsychopharmacol.
TOMLINSON, A., GRAYSON, B., MARSH, S., HAYWARD, A., MARSHALL, K. M. & NEILL, J. C. 2015. Putative therapeutic targets for symptom subtypes of adult ADHD: D4 receptor agonism and COMT inhibition improve attention and response inhibition in a novel translational animal model. Eur Neuropsychopharmacol, 25, 454-67.
TREISMAN, A. M. & GELADE, G. 1980. A feature-integration theory of attention. Cogn Psychol, 12, 97-136.
TU, P. C., LEE, Y. C., CHEN, Y. S., LI, C. T. & SU, T. P. 2013. Schizophrenia and the brain's control network: aberrant within- and between-network connectivity of the frontoparietal network in schizophrenia. Schizophr Res, 147, 339-47.
TYE, C., RIJSDIJK, F. & MCLOUGHLIN, G. 2014. Genetic overlap between ADHD symptoms and EEG theta power. Brain Cogn, 87, 168-72.
UYLINGS, H. B. M., GROENEWEGEN, H. J. & KOLB, B. 2003. Do rats have a prefrontal cortex? Behavioural Brain Research, 146, 3-17.
VAN GOETHEM, N. P., PRICKAERTS, J., WELTY, D., FLOOD, D. G. & KOENIG, G. 2015. Continuous infusion of the alpha7 nicotinic acetylcholine receptor agonist EVP-6124 produces no signs of tolerance at memory-enhancing doses in rats: a pharmacokinetic and behavioral study. Behav Pharmacol, 26, 403-6.
VASQUEZ, B. P., BUCK, B. H., BLACK, S. E., LEIBOVITCH, F. S., LOBAUGH, N. J., CALDWELL, C. B. & BEHRMANN, M. 2011. Visual attention deficits in Alzheimer's disease: relationship to HMPAO SPECT cortical hypoperfusion. Neuropsychologia, 49, 1741-50.
VENTURA, R., MORRONE, C. & PUGLISI-ALLEGRA, S. 2007. Prefrontal/accumbal catecholamine system determines motivational salience attribution to both reward- and aversion-related stimuli. Proc Natl Acad Sci U S A, 104, 5181-6.
VIZI, E. S. & LENDVAI, B. 1999. Modulatory role of presynaptic nicotinic receptors in synaptic and non-synaptic chemical communication in the central nervous system. Brain Res Brain Res Rev, 30, 219-35.
VOON, V., IRVINE, M. A., DERBYSHIRE, K., WORBE, Y., LANGE, I., ABBOTT, S., MOREIN-ZAMIR, S., DUDLEY, R., CAPRIOLI, D., HARRISON, N. A., WOOD, J., DALLEY, J. W., BULLMORE, E. T., GRANT, J. E. & ROBBINS, T. W. 2014. Measuring "waiting" impulsivity in substance addictions and binge eating disorder in a novel analogue of rodent serial reaction time task. Biol Psychiatry, 75, 148-55.
WATKINS, S. S., KOOB, G. F. & MARKOU, A. 2000. Neural mechanisms underlying nicotine addiction: acute positive reinforcement and withdrawal. Nicotine Tob Res, 2, 19-37.
WHITE, H. K. & LEVIN, E. D. 1999. Four-week nicotine skin patch treatment effects on cognitive performance in Alzheimer's disease. Psychopharmacology (Berl), 143, 158-65.
WIGNALL, N. D. & DE WIT, H. 2011. Effects of nicotine on attention and inhibitory control in healthy nonsmokers. Exp Clin Psychopharmacol, 19, 183-91.
WILENS, T. E., FARAONE, S. V. & BIEDERMAN, J. 2004. Attention-deficit/hyperactivity disorder in adults. JAMA, 292, 619-23.
WILLCUTT, E. G., NIGG, J. T., PENNINGTON, B. F., SOLANTO, M. V. & AL, E. 2012. Validity of DSM-IV attention–deficit/hyperactivity disorder symptom dimensions and subtypes. Journal of Abnormal Psychology, 121, 991-1010.
WILLIAMSON, D. & JOHNSTON, C. 2015a. Gender differences in adults with attention-deficit/hyperactivity disorder. Clinical Psychology Review, 40, 15-27.
WILLIAMSON, D. & JOHNSTON, C. 2015b. Gender differences in adults with attention-deficit/hyperactivity disorder: A narrative review. Clin Psychol Rev, 40, 15-27.
149
WILLNER, P. 1986. Validation criteria for animal models of human mental disorders: learned helplessness as a paradigm case. Prog Neuropsychopharmacol Biol Psychiatry, 10, 677-90.
WINGEN, M., KUYPERS, K. P. & RAMAEKERS, J. G. 2007. The role of 5-HT1a and 5-HT2a receptors in attention and motor control: a mechanistic study in healthy volunteers. Psychopharmacology (Berl), 190, 391-400.
WINSTANLEY, C. A. 2011. The utility of rat models of impulsivity in developing pharmacotherapies for impulse control disorders. British Journal of Pharmacology, 164, 1301-1321.
WINSTANLEY, C. A., DALLEY, J. W., THEOBALD, D. E. & ROBBINS, T. W. 2003. Global 5-HT depletion attenuates the ability of amphetamine to decrease impulsive choice on a delay-discounting task in rats. Psychopharmacology (Berl), 170, 320-31.
WINSTANLEY, C. A., ZEEB, F. D., BEDARD, A., FU, K., LAI, B., STEELE, C. & WONG, A. C. 2010. Dopaminergic modulation of the orbitofrontal cortex affects attention, motivation and impulsive responding in rats performing the five-choice serial reaction time task. Behav Brain Res, 210, 263-72.
WOOD, S., SAGE, J. R., SHUMAN, T. & ANAGNOSTARAS, S. G. 2014. Psychostimulants and cognition: a continuum of behavioral and cognitive activation. Pharmacol Rev, 66, 193-221.
WOOLLEY, M. L., WATERS, K. A., REAVILL, C., BULL, S., LACROIX, L. P., MARTYN, A. J., HUTCHESON, D. M., VALERIO, E., BATE, S., JONES, D. N. & DAWSON, L. A. 2008. Selective dopamine D4 receptor agonist (A-412997) improves cognitive performance and stimulates motor activity without influencing reward-related behaviour in rat. Behavioural Pharmacology, 19, 765-76.
WOOLRICH, M. W., BAKER, A., LUCKHOO, H., MOHSENI, H., BARNES, G., BROOKES, M. & REZEK, I. 2013. Dynamic state allocation for MEG source reconstruction. Neuroimage, 77, 77-92.
XUAN, B., MACKIE, M. A., SPAGNA, A., WU, T., TIAN, Y., HOF, P. R. & FAN, J. 2016. The activation of interactive attentional networks. Neuroimage, 129, 308-19.
YOUNG, J. W., CRAWFORD, N., KELLY, J. S., KERR, L. E., MARSTON, H. M., SPRATT, C., FINLAYSON, K. & SHARKEY, J. 2007. Impaired attention is central to the cognitive deficits observed in alpha 7 deficient mice. Eur Neuropsychopharmacol, 17, 145-55.
YOUNG, J. W. & GEYER, M. A. 2013. Evaluating the role of the alpha-7 nicotinic acetylcholine receptor in the pathophysiology and treatment of schizophrenia. Biochemical Pharmacology, 86, 1122-32.
YOUNG, J. W., LIGHT, G. A., MARSTON, H. M., SHARP, R. & GEYER, M. A. 2009. The 5-Choice Continuous Performance Test: Evidence for a Translational Test of Vigilance for Mice. Plos One, 4.
YOUNG, J. W., MEVES, J. M. & GEYER, M. A. 2013. Nicotinic agonist-induced improvement of vigilance in mice in the 5-choice continuous performance test. Behav Brain Res, 240, 119-33.
YOUNG, J. W., POWELL, S. B., SCOTT, C. N., ZHOU, X. & GEYER, M. A. 2011. The effect of reduced dopamine D4 receptor expression in the 5-choice continuous performance task: Separating response inhibition from premature responding. Behavioural Brain Research, 222, 183-192.
YOUNG, J. W., WINSTANLEY, C. A., BRADY, A. M. & HALL, F. S. 2016. Research Domain Criteria versus DSM V: How does this debate affect attempts to model corticostriatal dysfunction in animals? Neurosci Biobehav Rev.
ZANTO, T. P., RUBENS, M. T., THANGAVEL, A. & GAZZALEY, A. 2011. Causal role of the prefrontal cortex in top-down modulation of visual processing and working memory. Nat Neurosci, 14, 656-61.
150
ZHANG, K., DAVIDS, E., TARAZI, F. I. & BALDESSARINI, R. J. 2002. Effects of dopamine D4 receptor-selective antagonists on motor hyperactivity in rats with neonatal 6-hydroxydopamine lesions. Psychopharmacology (Berl), 161, 100-6.