Emotional modulation of control dilemmas: The role of positive affect, reward, and dopamine in cognitive stability and flexibility

Emotional modulation of control dilemmas: The role of positive affect,reward, and dopamine in cognitive stability and flexibility

Thomas Goschke n, Annette BolteDepartment of Psychology, Technische Universität Dresden, 01062 Dresden, Germany

a r t i c l e i n f o

Article history:Received 1 November 2013Received in revised form23 June 2014Accepted 16 July 2014

Keywords:Cognitive controlControl dilemmasEmotionPositive affectDopamineCognitive flexibilityPrefrontal cortexBasal ganglia

a b s t r a c t

Goal-directed action in changing environments requires a dynamic balance between complementarycontrol modes, which serve antagonistic adaptive functions (e.g., to shield goals from competingresponses and distracting information vs. to flexibly switch between goals and behavioral dispositions inresponse to significant changes). Too rigid goal shielding promotes stability but incurs a cost in terms ofperseveration and reduced flexibility, whereas too weak goal shielding promotes flexibility but incurs acost in terms of increased distractibility. While research on cognitive control has long been conductedrelatively independently from the study of emotion and motivation, it is becoming increasingly clear thatpositive affect and reward play a central role in modulating cognitive control. In particular, evidencefrom the past decade suggests that positive affect not only influences the contents of cognitive processes,but also modulates the balance between complementary modes of cognitive control. In this article wereview studies from the past decade that examined effects of induced positive affect on the balancebetween cognitive stability and flexibility with a focus on set switching and working memorymaintenance and updating. Moreover, we review recent evidence indicating that task-irrelevant positiveaffect and performance-contingent rewards exert different and sometimes opposite effects on cognitivecontrol modes, suggesting dissociations between emotional and motivational effects of positive affect.Finally, we critically review evidence for the popular hypothesis that effects of positive affect may bemediated by dopaminergic modulations of neural processing in prefrontal and striatal brain circuits, andwe refine this “dopamine hypothesis of positive affect” by specifying distinct mechanisms by whichdopamine may mediate effects of positive affect and reward on cognitive control. We conclude with adiscussion of limitations of current research, point to central unresolved questions and outlineperspective for future research on affective and motivational modulations of cognitive control modes.

& 2014 Published by Elsevier Ltd.

1. Introduction

The term cognitive control denotes a heterogeneous set ofmechanisms that underlie the human ability to configure behavioraldispositions according to superordinate goals or task instructions, tomaintain goals in the face of distraction, and to suppress prepotent,but unwanted habitual or impulsive responses (Banich, 2009;Goschke, 2013; Miller & Cohen, 2001). Although in the past twodecades substantial progress has been made in elucidating the factorstructure (e.g., Friedman et al., 2008; Miyake et al., 2000), computa-tional mechanisms (e.g., O'Reilly, Herd, & Pauli, 2010), and neuralbasis (e.g., Mars, Sallet, Rushworth, & Yeung, 2011) of cognitivecontrol, it is still insufficiently understood how cognitive controlprocesses are modulated by emotional and motivational factors.While research on cognitive control has long been conducted

relatively independently from the study of emotions, it becomesincreasingly clear that brain systems involved in cognitive controlsuch as the prefrontal cortex (PFC) are strongly interconnected withbrain systems involved in the processing of emotion and motivation(Banich et al., 2009; Chiew & Braver, 2011; Mars et al., 2011; Pessoa,2009; Ray & Zald, 2012). However, in contrast to a voluminous bodyof research on effects of emotions and moods on perception,attention, and creative problem-solving (for reviews see Bolte &Goschke, 2010; Fredrickson, 2013; Friedman & Förster, 2010; Isen,2007), research on how emotions specifically modulate cognitivecontrol processes has only recently gained momentum (e.g., Banichet al., 2009; Bolte & Goschke, 2010; Chiew & Braver, 2011; Dreisbach& Fischer, 2012; Mitchell & Phillips, 2007; Mueller, 2011). Impor-tantly, findings obtained during the past decade have revealed thatemotions not only influence the contents of cognitive controlprocesses (e.g., which goals are maintained in working memory),but also modulate the mode of cognitive control (e.g., how stronglygoals are shielded from distraction or how flexibly cognitive sets areupdated).

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/neuropsychologia

Neuropsychologia

http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.0150028-3932/& 2014 Published by Elsevier Ltd.

n Corresponding author. Tel.: þ49 351 4633 4695.E-mail address: [email protected] (T. Goschke).

Please cite this article as: Goschke, T., & Bolte, A. Emotional modulation of control dilemmas: The role of positive affect, reward, anddopamine in cognitive stability and flexibility. Neuropsychologia (2014), http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.015i

Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎

www.sciencedirect.com/science/journal/00283932

www.elsevier.com/locate/neuropsychologia

http://dx.doi.org/10.1016/j.neuropsychologia.2014.07.015



mailto:[email protected]





1.1. Scope of the present review

In this review we focus specifically on effects of inducedpositive affect on complementary cognitive control modes and,in particular, the balance between stable maintenance and flexibleshifting of goals and task-sets. Our guiding hypothesis is thatemotions are associated with different settings of “meta-controlparameters” that regulate the balance between complementarycontrol modes and thereby promote either flexible switching orstable maintenance of goals and cognitive sets. While our primaryfocus is on positive affect, we also provide a selective review ofrecent evidence indicating that task-irrelevant positive affect andperformance-contingent rewards exert markedly different andsometimes opposite effects on cognitive control modes. Thisdiscussion complements recent reviews on the role of positiveaffect and reward in other domains of cognitive control such asconflict monitoring (Chiew & Braver, 2011; Dreisbach & Fischer,2012). Our second main aim is to discuss possible computationaland neural mechanisms that may mediate emotional modulationsof cognitive control modes. In particular, we critically reviewevidence for the popular hypothesis that effects of positive affecton cognitive control are mediated by dopaminergic modulations ofneural processing in frontal-striatal brain circuits (e.g., Ashby, Isen,& Turken, 1999; Ashby, Valentin, & Turken, 2003).

Note that it is neither our aim to provide a comprehensiveoverview of emotional modulations of cognitive processes ingeneral nor to review effects of emotions on all aspects of cognitivecontrol. Rather, we restrict the scope of this review to studies thathave examined effects of positive affect (or reward) on task-setswitching and working memory maintenance and updating, andwe focus particularly on studies that are informative with respectto the question how positive affect and reward modulate com-plementary control modes (for reviews of the role of positive affectin other cognitive domains such as perception, attention, orproblem-solving see Bolte & Goschke, 2010; Fredrickson, 2013;Friedman & Förster, 2010; Isen, 2007).

1.2. Conceptual and methodological issues

The term emotion has been notoriously difficult to define(Hamann, 2012; LeDoux, 2012) and it has even been asked howmeaningful a categorical distinction between cognition and emo-tion (or “cognitive” and “affective” brain areas) is (Pessoa, 2008). Inthis article we use a pragmatic working definition, according towhich emotions can be conceived as psycho-physiologicalresponse patterns which involve several components, including a(conscious or unconscious) evaluation of the significance of anevent in the light of the organism's needs, motives, and goals;physiological responses of the autonomous nervous system asindicated by different indicators of increased arousal; the recruit-ment of brain circuits involved in the processing of reward, threat,or punishment; the generation of motivational tendencies that setparticular categories of action into readiness (e.g., approach vs.avoidance); specific facial and postural expressions; and often (butnot necessarily) a qualitative subjective experience (the feelingcomponent) (Bolte & Goschke, 2010). The relation between emo-tion and motivation will be discussed in more detail in Section 3.3on dissociations between positive affect and reward (see alsoChiew & Braver, 2011).

In most studies reviewed in this article positive affect wasinduced by presenting positive emotional stimuli (e.g., pictures,movie clips) either on a trial-by-trial basis or before a block oftrials, whereas a smaller number of studies examined effects ofmore enduring moods. Although phasic emotional responses and

tonic moods likely differ with respect to underlying neuralsystems, the degree to which they capture focal attention, andwhether they motivate emotion regulation strategies, to ourknowledge no studies have systematically investigated howshort-lived emotional responses to affective stimuli and tonicmoods differ in their effects on cognitive control. As there is notsufficient evidence for a systematic comparison of effects of tonicand phasic emotions, we have organized our review along thecontrol functions under investigation (set shifting; working mem-ory maintenance and updating) rather than according to emotioninduction methods.

2. An integrative theoretical framework: control dilemmasand complementary control modes

2.1. Control dilemmas

Our discussion of emotional modulations of cognitive control isguided by a theoretical framework that distinguishes differentglobal control modes, which serve complementary adaptive func-tions in goal-directed action. While the evolution of cognitivecontrol capacities dramatically increased the flexibility of humanaction, as is evident in our ability to select actions based onanticipated future goals, to rapidly reconfigure behavioral disposi-tions according to changing intentions and instructions, and tomaintain goals in the face of competing habitual or impulsiveresponses, this increase in cognitive and behavioral flexibility alsogave rise to new kinds of conflicts. We conceive of such conflicts ascontrol dilemmas to express the idea that goal-directed action in achanging environment is a multiple constraint satisfaction pro-blem that confronts agents with fundamental trade-offs betweenantagonistic adaptive requirements (Goschke, 2000, 2003, 2013;Goschke & Dreisbach, 2008; Gruber & Goschke, 2004; Kuhl &Goschke, 1994) (for related ideas see Cohen, McClure, & Yu, 2007;Cools, 2008). In this review we focus specifically on what we termthe shielding–shifting dilemma. On the one hand, goal-directedaction requires that goals (e.g., finishing a review paper) aremaintained and shielded from distracting stimuli (e.g., music froma neighborhood party) or competing response tendencies (to godancing rather than continue writing) (Gollwitzer & Bayer, 1999;Hofmann, Schmeichel, & Baddeley, 2012; Kuhl, 1985). On the otherhand, however, agents must be able to update cognitive sets,disengage from a currently active goal, and flexibly reconfigureresponse dispositions to adapt to significant changes in theenvironment or internal state (for instance, when noticing unex-pected noise in the basement while writing at night).

A core assumption of control dilemma theory is that differentcontrol modes are associated with complementary benefits andcosts. While strong goal-shielding supports behavioral persistenceand cognitive stability, it may incur a cost in terms of perseverativebehavior and impaired adaptation to changing contexts or taskdemands. Conversely, while weak goal shielding facilitates flexibleset switching, it increases distractibility and the risk of unstablebehavior that is driven by every minor change in the environment.Evidence for complementary benefits and costs of goal shieldingstems from studies of conflict-induced adjustments of cognitivecontrol. These studies have shown that response conflicts ininterference tasks (e.g., Stroop or flanker tasks) trigger theenhanced recruitment of cognitive control, as indicated by thefinding that interference from distracting information or compet-ing responses is reduced on trials immediately following aresponse conflict (e.g., Fischer, Dreisbach, & Goschke, 2008;Gratton, Coles, & Donchin, 1992; Kerns et al., 2004; Stürmer,Leuthold, Soetens, Schröter, & Sommer, 2002) (for a review see

T. Goschke, A. Bolte / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎2





Egner, 2007)1 and that response conflicts induce increased goalshielding even within the conflict trial itself (Scherbaum, Fischer,Dshemuchadse, & Goschke, 2011). According to an influentialtheoretical account (Botvinick, Braver, Barch, Carter, & Cohen,2001), such conflict-adaptation effects reflect the fact that con-flicts, which are assumed to be monitored by the anteriorcingulate cortex (ACC), serve as signals for the enhanced recruit-ment of control by the dorsolateral PFC, which is assumed tomediate increased goal shielding and top-down biasing of task-relevant processing pathways (for review see Mansouri, Tanaka, &Buckley, 2009). However, while a conflict-triggered enhancementof goal shielding is beneficial when the task and response rulesremain constant, control dilemma theory predicts that it shouldincur a cost when one must flexibly switch to a different task.Consistent with this prediction, it has been shown that conflict-induced goal shielding increases task-switch costs (Brown,Reynolds, & Braver, 2007; Goschke, 2000; Meiran, Hsieh, &Dimov, 2010) and impairs background monitoring for incidentalcues signaling the requirement to switch from an ongoing task to adifferent task (Goschke & Dreisbach, 2008).

2.2. Meta-control parameters

Within our control dilemma framework, we assume that thebalance between complementary control modes depends on asmall set of meta-control parameters, which regulate global fea-tures of the mode of information-processing (Doya, 2008;Goschke, 2013). With respect to the shielding–shifting dilemma,a critical control parameter is the updating threshold that deter-mines the stability of working memory representations. A highupdating threshold renders working memory resistant againstdistraction and strengthens the top-down biasing of perceptualprocessing towards task-relevant information (Desimone &Duncan, 1995; Gazzaley & Nobre, 2012; Miller & Cohen, 2001;Sakai, 2008). However, a high updating threshold may also preventpotentially significant information outside the current goal-directed focus of attention to gain access to working memory(Goschke & Dreisbach, 2008; Gruber, Diekhof, Kirchenbauer, &Goschke, 2010; Gruber et al., 2009). Conversely, while a lowupdating threshold facilitates flexible set shifting, it may alsoincrease distractibility and crosstalk. Note that the updatingthreshold does not directly determine which goal or task-set isencoded into working memory, but it modulates in a content-unspecific manner how easy novel information gains access toworking memory, thereby regulating the balance between main-tenance and updating.

On a computational level, the updating threshold can beconceived of in terms of the stability of attractor states in neuralnetworks. In such networks, goals or task-sets are represented asactivation patterns that can be sustained in the absence ofperceptual input by means of strong recurrent interconnectionsbetween active processing units and constitute attractors in thenetwork's state space (Durstewitz, Seamans, & Sejnowski, 2000b;O'Reilly, 2006; Rolls, 2010; Scherbaum, Dshemuchadse, Ruge, &Goschke, 2012). The dynamics of such networks is often illustratedby conceiving of the network's state space in analogy to a potentiallandscape, where the stability of an attractor state depends onthe depth of the attractor basin (Rolls, 2010; Scherbaum,Dshemuchadse, & Kalis, 2008). Thus, deep attractor basins renderself-sustaining activation patterns resistant against interference,

but also make switching between states more difficult. Conversely,shallow attractor basins facilitate shifting between different states,but also render the system vulnerable to interference and distrac-tion (Durstewitz & Seamans, 2008; Gruber & Goschke, 2004;O'Reilly, 2006). The tradeoff between stability and flexibility hasnicely been illustrated in a recent neural network modeling study(Herd et al., 2014), which showed that increasing the stability ofgoal representations in a working memory layer reduced inter-ference in a simulated Stroop task, but at the same time increasedtask switch costs. As will be discussed in Section 4, on aneurobiological level the updating threshold has been related toinfluences of the neuromodulator dopamine on neural processingin the PFC and basal ganglia thought to mediate working memorymaintenance and set shifting, respectively.

2.3. Affective modulation of complementary control modes

Control dilemma theory raises the fundamental question of howthe balance between complementary control modes is regulatedand which variables determine the settings of meta-control para-meters. While meta-control parameters depend on multiple factors(including experience-based learning, statistical features of anenvironment, personality dispositions), a central assumption ofthe control dilemma framework is that emotions, and positiveaffect in particular, play an important role in modulating theupdating threshold and the stability–flexibility balance. Outsidethe domain of cognitive control, support for this assumption stemsfrom studies suggesting that positive affect is associated with amore flexible, exploratory mode of cognitive processing (Fiedler,Martin, & Clore, 2001; Fredrickson, 2013; Isen, 2007; Isen, Dalgleish,& Power, 1999; Kuhl, 2000). For instance, mild increases of positiveaffect have been associated with increased cognitive fluency(Phillips, Bull, Adams, & Fraser, 2002), less functional fixedness inproblem-solving (Gasper, 2003; Isen, Daubman, & Nowicki, 1987),an expanded scope of attention (Fredrickson & Branigan, 2005;Friedman & Förster, 2010; Rowe, Hirsh, & Anderson, 2007), activa-tion of remote associates in semantic memory (Bolte, Goschke, &Kuhl, 2003; Isen, Johnson, Mertz, & Robinson, 1985; Rowe et al.,2007), an improved ability to overcome dominant responses(Baumann & Kuhl, 2005; Kuhl & Kazen, 1999), and a focus on globalrather than local perceptual features (Gasper, 2004; Gasper & Clore,2002) (for reviews see Bolte & Goschke, 2010; Fredrickson, 2013;Isen, 2007) (but see also Harmon-Jones, Gable, & Price, 2013;Huntsinger, 2013).

However, while these findings suggest that positive affectenhances cognitive flexibility in perception, attention, or creativeproblem-solving tasks, early studies on effects of positive mood onexecutive control functions indicated that positive mood mayactually impair performance in tasks requiring effortful control.For instance, it had been reported that an induced tonic positivemood impaired task-switching (Phillips et al., 2002) and planning(Oaksford, Morris, Grainger, & Williams, 1996), and reduced work-ing memory capacity especially in tasks requiring central executiveand articulatory control process (Spies, Hesse, & Hummitzsch,1996). Such findings were attributed to the fact that positive moodinduces task-irrelevant thoughts and draws attention and workingmemory resources away from task-relevant processing (Ellis et al.,1997; Seibert & Ellis, 1991). However, while the resource competi-tion hypothesis accounts for the reported adverse effects ofpositive mood on cognitive control, it must be noted that thetasks that were used in these studies to assess cognitive controlwere not process-pure, but involved multiple component pro-cesses, which may have been affected differentially or even inopposite ways by positive affect.

1 Alternative interpretations attribute sequential congruence effects to repeti-tion priming (Mayr, Awh, & Laurey, 2003) or feature binding (Hommel, Proctor, &Vu, 2004). Various attempts to disentangle these effects from conflict-inducedcontrol adjustments suggest that both kinds of processes contribute to sequentialmodulations of congruence effects (for a review see Egner, 2007).

T. Goschke, A. Bolte / Neuropsychologia ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3





2.4. Predictions of the control dilemma framework

According to the control dilemma framework positive affectneither generally improves performance by increasing cognitiveflexibility, nor does it generally impair performance due to resourcecompetition. Rather, we assume that positive affect is associatedwith both benefits and costs depending on the processing demandsof a task. More specifically, we assume that positive affect hasopposite effects on complementary control functions: whereasincreased cognitive flexibility presumably associated with positiveaffect should facilitate set switching and working memory updating,it should incur costs in terms of increased distractibility andimpaired maintenance of task-relevant information. As our reviewwill show, an increasing number of recent studies have used tasksthat allow investigating whether positive affect exerts differenteffects on complementary components of cognitive control.

A second assumption of the control dilemma framework holdsthat effects of positive affect on cognitive control depend on theadaptive function of affective signals during goal-directed action.In line with other proposals (Berridge & Kringelbach, 2013; Chiew& Braver, 2011; Dreisbach & Fischer, 2012) we assume that positiveaffect elicited by task-irrelevant emotional stimuli has markedlydifferent effects than positive affect elicited by cues signaling theprospect of gaining a performance-contingent reward. Althoughpositive affective stimuli have a rewarding quality, and althoughrewards as well as the anticipation of a reward elicit positive affect(Knutson & Greer, 2008), the two kinds of affective responses verylikely serve different adaptive functions in the regulation ofcognitive control. Whereas task-unrelated positive affect has beensuggested to serve as a safety signal that promotes cognitiveflexibility and exploration (Carver, 2003; Fiedler et al., 2001;Fredrickson, 2013; Goschke, 1996), cues signaling the prospect togain a performance-contingent reward serve as motivationalsignals, which promote the recruitment of effortful cognitivecontrol to increase the chances of obtaining the reward (Braver,2012; Chiew & Braver, 2011).

It should be noted that this “reward-as-motivation” hypothesisshares some aspects, but also differs in important respects fromthe motivational dimensional model of affect (Harmon-Jones, Gable,& Price, 2012). According to this model, positive affect low inapproach motivation intensity (e.g., joy) signals that goal pursuitruns smoothly or even better than expected and there is no needfor effortful control, thus promoting an explorative mode ofcontrol and a broadened scope of attention. In contrast, positiveaffect high in approach motivation intensity (as elicited, forinstance, by pictures of tasty food) is assumed to have the oppositeeffect and constrict rather than expand the scope of attention.There are two differences between the motivational dimensionalmodel and the reward-as-motivation hypothesis. First, whereasthe reward-as-motivation hypothesis predicts that reward incen-tives enhance the motivation to recruit effortful control in order toimprove performance in the potentially rewarded task, the moti-vational dimensional model assumes that positive affective stimulihigh in approach motivation intensity induce a general narrowingof the scope of attention even if the motivational content of theaffective stimuli is unrelated to the task (for instance, whenpictures of tasty food are assumed to narrow attention in anunrelated perceptual task). Secondly, whereas the motivationaldimensional model predicts that high approach motivation inten-sity generally narrows the scope of attention, the reward-as-motivation hypothesis predicts that reward incentives induce acontrol mode that is suited to improve task performance. Whilethis may show up in a narrowed scope of attention and increasedgoal shielding when a task requires filtering out distractinginformation, it may also show up in increased flexibility when atask requires intentional switching between tasks.

3. Review of studies on effects of positive affect and reward onthe shielding–shifting dilemma

During the past decade, two lines of research have examinedhow positive affect modulates the balance between stable main-tenance and flexible switching of goals and task-sets. One set ofstudies has used task-set switching paradigms in which partici-pants have to switch between different tasks or cognitive sets,while the other used working memory tasks requiring bothflexible updating and stable maintenance of task-relevant infor-mation. Although set shifting and working memory updating mayappear closely linked, latent-variable analyses of task batteriesassessing executive control functions indicate that the two con-structs of “shifting” and “updating” constitute partly separablecomponents of cognitive control (Friedman et al., 2008; Miyake etal., 2000). In the following review, we focus on the question(1) how positive affect modulates processes mediating robustmaintenance vs. flexible switching, and (2) how effects of task-irrelevant positive affect differ from those of performance-contingent reward (Table 1).

3.1. Effects of positive affect on task-set switching and dual taskcoordination

Paradigms in which participants have to switch between differ-ent tasks or cognitive sets are particularly well suited to investigatethe shielding–shifting dilemma as they maximize the trade-offbetween the antagonistic requirements to shield the current task-set from competing task-sets vs. to rapidly deactivate executed (no-longer-relevant) task-sets to prevent proactive interference(Goschke, 2000; Mayr & Keele, 2000; Meiran, 2010). The termtask-set denotes a particular configuration of mental processes thatis established as a result of receiving a task instruction, whichspecifies which stimuli or stimulus features are task-relevant andhow they are mapped to responses (Monsell, 2003). Switchingbetween different tasks is usually associated with a performancecost (and/or a repetition benefit) that shows up in increased RTsand error rates for task switches compared to task repetitions(Allport, Styles, & Hsieh, 1994; Meiran, Chorev, & Sapir, 2000;Rogers & Monsell, 1995) (for reviews see De Baene, Kuhn, & Brass,2012; Kiesel et al., 2010; Koch, Gade, Schuch, & Philipp, 2010; Ruge,Jamadar, Zimmermann, & Karayanidis, 2013; Vandierendonck,Liefooghe, & Verbruggen, 2010). Importantly, switch costs are nota process-pure indicator of the time required for the reconfigurationof task-sets but depend on multiple factors, including advancepreparation for a new task (Gruber, Karch, Schlueter, Falkai, &Goschke, 2006; Ruge et al., 2013), proactive interference fromrecently executed task-sets (Allport et al., 1994; Altmann & Gray,2008), crosstalk from currently irrelevant task-sets (Yeung,Nystrom, Aronson, & Cohen, 2006), and persisting inhibition ofpreviously suppressed but now to-be-executed task-sets (Goschke,2000; Mayr & Keele, 2000). Moreover, the processes involved inswitching differ depending on whether one switches betweendifferent sets of stimulus–response rules, perceptual features,stimulus dimensions or categories (Ravizza & Carter, 2008).

Although the ability to rapidly switch between task-sets is acentral precondition for flexible intentional action, to date fewstudies have investigated effects of emotions and moods on taskswitching performance. An early study investigated effects ofinduced mood states on a Stroop color-word interference task, inwhich participants alternated from trial to trial between readingcolor words and naming the display color of the words (Phillipset al., 2002). Participants in a positive mood showed significantlygreater switch costs than control participants in a neutral mood.While this result appears to contradict the assumption that positiveaffect increases cognitive flexibility, it must be considered that the






Table 1Overview of reviewed studies (2004–2014) on effects of positive affect and reward incentives on set switching and working memory.

Study Task Affect manipulation Main findings

Task-irrelevant positive affect: Set switching and dual taskingDreisbach and Goschke(2004)

Setswitching

Blocks with positive vs. neutral affective pictures beforeeach trial

Positive affect reduced switch cost in perseveration blocks andincreased switch costs in learnt irrelevance/distractibility blocks

Liu and Wang (2014) Setswitching

Blocks with positive vs. neutral affective pictures high orlow in approach motivation intensity before each trial

Positive affect low in motivational intensity reduced switch cost inperseveration blocks and increased switch costs in learntirrelevance/distractibility blocks; positive affect high in motivationalintensity had opposite effects

Frober and Dreisbach(2012; Experiment 3)

Cued taskswitchingwith valid(75%) andinvalid(25%) taskcues

Different groups with neutral or positive affective pictures(low vs. high in arousal) on each trial

The cue-validity effect in the first task-switching block was reducedfor task repetitions in the group with low-arousal positive picturescompared to the group with high-arousal positive pictures; the cue-validity effect in the neutral affect group lay in between

Zwosta et al. (2013) Dual-taskparadigma

Positive vs. negative film clips before a task block Increased between-task crosstalk (reduced task shielding) underpositive compared to negative mood

Task-irrelevant positive affect: Working memory maintenance and updatingGray (2001) Spatial and

verbal2-back

Approach, withdrawal, and neutral video clips Positive relative to neutral mood impaired spatial 2-back andimproved verbal 2-back task

Martin and Kerns (2011) Runningmemoryspan

Positive vs. neutral video clip Positive relative to neutral mood caused a slight impairment ofperformance

Yang et al. (2013) Operationspan

Unexpected gift or candy Positive relative to neutral mood caused a slight improvement ofperformance

Dreisbach (2006) AX-CPT Groups with positive vs. negative vs. neutral affectivepictures before each trial

Positive (compared to neutral or negative) affect reduced errors onAY trials and increased errors and RTs (in Experiment 2) on BX trials

Van Wouwe et al. (2011) AX-CPT Positive vs. neutral film clip before experimental run Positive compared to neutral affect reduced errors on AY trials; noeffect on BX trials

Frober and Dreisbach(2014)

AX-CPT Different groups with positive or neutral affective picturesbefore each trial

Positive relative to neutral affect reduced errors and RTs on AYtrials; no effect on BX trials

Chiew and Braver(2014)

AX-CPT Neutral vs. positive video clip before each task block; inaddition, on each trial a positive or neutral affectivepictures was randomly presented

Error rates in the positive relative to the neutral affect block weregenerally reduced, but slightly increased on AY trials; positive andneutral picture trials within the positive affect block did not differ inerror rates; positive relative to neutral picture trials were associatedwith generally faster RTs but slightly increased RTs on AY trials

Performance-contingent reward incentives: Set switchingMüller, Dreisbach,Goschke, Hensch,Lesch, and Brocke(2007)

Setswitching

Monetary reward cues Positive correlation between reported effort after reward cue andswitch cost in perseveration condition; negative correlation(p¼ .051) with switch cost in the distractibility condition

Kleinsorge andRinkenauer (2012)

Cued taskswitching

Monetary reward cue accompanying the task cue in onethird of the trials

Reward cues reduced switch costs

Savine et al. (2010;Experiment 1)

Cued taskswitching

Monetary reward cue accompanying the task cue on 50% oftrials

Reward cues reduced switch cost, especially with long cue-target-interval

Performance-contingent reward incentives: Set switchingSavine et al. (2010;Experiment 2)

Sternbergitemrecognitiontask

Cues signaling no, low, or high concentration liquid reward Faster RT on incentive compared to non-incentive trials; no effect onerror rates; faster RT on no-incentive trials in reward blockcompared to baseline (no reward) block

Shen and Chun (2011;Experiments 3a and3b)

Cued taskswitching

Monetary reward cues Reduced switch costs after cues signaling a higher reward than onthe previous trial

Jiang and Xu (2014) Cued taskswitching

Monetary reward cues Reward on trial N-1 increased backward inhibition (N-2 repetitioncost) relative to non-rewarded trial N-1; reward on trial N-2eliminated backward inhibition

Performance-contingent reward incentives: Working memory maintenance and updatingBeck et al. (2010) Delayed

responsetask

Monetary or liquid reward cues; no-incentive (20%), lowreward (40%), high reward (40%) trials

Faster RT for incentive trials compared to no-incentive trials; fasterRTs in high compared to low incentive trials; no effects on error rate

Locke and Braver (2008) AX-CPT Block with opportunity to gain monetary reward vs.baseline block

Reward incentives generally decreased RTs, but increased errorsselectively on AY trials


AX-CPT Monetary reward cue or neutral cue at the beginning ofeach trial

Reward incentives reduced RTs and error rates on all trial types, butincreased error rates selectively on AY trials


AX-CPT Baseline vs. reward block with monetary reward cues on50% of trials

Reward incentives decreased RTs and error rates on all trial types,but increased error rates selectively on AY trials

Frober and Dreisbach(2014)

AX-CPT Baseline vs. reward blocks with performance-contingent ornoncontingent monetary reward cues on 50% of trials

In blocks with performance-contingent reward incentives, RTs weregenerally faster but error rates were selectively increased on AY(and to a lesser degree on AX) trials compared to blocks withnoncontingent reward

Note: Only behavioral effects of positive affect or reward incentives are summarized. Only experimental conditions relevant for these effects are mentioned; some studiesmay contain additional conditions or results.






particular task-switching paradigm did not allow disentanglingcomponent processes such as perseveration of recently executedtask-sets, inhibition of prepotent responses, or the ability to switchattention back to a recently suppressed stimulus dimensions, whichmay have been differentially affected by a positive mood.

A first attempt to directly test the more specific hypothesis thatpositive affect reduces perseveration of a recently executed task-set but at the same time increases susceptibility to interferencefrom novel task-irrelevant distracters was made by Dreisbach andGoschke (2004). Participants had to respond to target stimuli in apre-specified color (e.g. red) while ignoring distracters in adifferent color (e.g. green). In a perseveration condition, after 40trials participants had to switch attention to targets in a novelcolor (blue) and ignore distracters in the previous target color(red). In this condition, perseveration of the previous task-set(presumably reflecting strong task-shielding) should increaseswitch costs, whereas weak goal shielding and a bias towardsnovel stimuli should facilitate switching. Conversely, in a learntirrelevance or distractibility condition participants had to switch totargets in the previously irrelevant color (green), while ignoringdistracters in a new color (blue). In this condition, weak goalshielding and/or a novelty bias should increase interference fromnovel distracters and thus lead to increased switch costs. Phasicemotional responses were induced by briefly presenting positiveaffective or neutral pictures at the start of each trial. The mainresult was that positive affect almost completely eliminated theswitch cost in the perseveration condition, whereas it reliablyincreased the switch cost in the distractibility condition. Thispattern indicates that phasic increases in positive affect reducedperseveration at the cost of increased distractibility.2 Of note,negative affective pictures matched in their arousal potential tothe positive pictures had no reliable effects on switch costs(if anything, on a descriptive level negative pictures had the reverseeffect than positive pictures), ruling out that effects of positive affectwere merely due to arousal. In conclusion, these findings indicatethat positive affect did neither generally improve nor impair set-switching, but rather shifted the control mode towards increasedflexibility and possibly a bias toward novel stimuli.3

This pattern of findings has recently been replicated andextended in a study (Liu & Wang, 2014), which used the sameset-switching task and affect induction method as Dreisbach andGoschke (2004), but in addition compared the effects of positiveaffective pictures low vs. high in approach motivation intensity.This was done to test the prediction of the motivational dimen-sional model of affect (Harmon-Jones et al., 2012) that positiveemotional stimuli low in approach motivation intensity broaden

the scope of visual attention, whereas positive emotional stimulihigh in approach motivation intensity constrict the scope ofattention (presumably increasing goal stability and reducing dis-tractibility) (Harmon-Jones et al., 2012). Liu and Wang fullyreplicated Dreisbach and Goschke's finding that positive affectivepictures low in approach motivation intensity reduced switchcosts in the perseveration condition, but increased switch costsin the distractibility condition. In marked contrast, positive affec-tive pictures with high approach motivation intensity had oppositeeffects, that is, they increased perseveration and reduced distract-ibility. These findings indicate that effects of positive affect on theflexibility–stability-balance in set switching are moderated in asimilar way by motivational intensity as has been demonstratedfor emotional modulations of the breadth of attention (Harmon-Jones et al., 2013).

Further support for the assumption that positive affect low inmotivational intensity attenuates goal shielding and increasesdistractibility stems from a recent dual tasking study (Zwosta,Hommel, Goschke, & Fischer, 2013). Like task switching, dualtasking puts particular demands on the adaptive regulation ofcognitive control, as efficient performance depends on a dynamicbalance between shielding the prioritized task against crosstalkfrom the secondary task while flexibly re-allocating attention tothe secondary task. Zwosta et al. (2013) investigated dual taskperformance in participants in whom either a positive or negativemood had been induced by film clips. Participants were asked tocategorize consecutively two digits presented above (S1) andbelow a fixation point (S2) as smaller vs. larger than five. As ameasure of how efficient participants shielded the primary task T1against crosstalk from the secondary task T2, the authors exam-ined the magnitude of performance decrements in T1 onresponse-incompatible trials, on which the two digits requireddifferent categorizations (cf. Hommel, 1998; Logan & Schulkind,2000). The results showed that between-task crosstalk (as indi-cated by the response compatibility effect in T1) was significantlylarger in the positive compared to the negative mood group, andthis effect could not be accounted for by differences in arousal.This finding indicates that positive relative to negative affect wasassociated with reduced task shielding and increased crosstalk.However, as the experiment did not include a neutral controlcondition, it remains open whether the results reflect reduced taskshielding under positive affect, increased task shielding undernegative affect, or a mixture of both.

3.2. Affective modulation of working memory maintenance andupdating

A second line of research on emotional modulations of theshielding–shifting balance has investigated effects of positiveaffect on working memory maintenance and updating. Like task-switching, the adaptive regulation of working memory requiresbalancing a tradeoff between the antagonistic requirements toshield task-relevant information from distraction vs. to updateworking memory when significant stimuli afford a shift of thecurrent goal or task-set.

A few studies have examined effects of induced (tonic) positivemood on working memory performance and yielded mixedresults. An early study by Gray (2001) showed that performancein a spatial 2-back task, in which participants had to monitorwhether the current stimulus in a stream stimuli was identical tothe stimulus presented two trials before, was enhanced by anegative mood and impaired by a positive mood, whereas theopposite pattern was observed for a verbal 2-back task. The authorspeculated that approach-related emotions enhance performancein verbal tasks thought to depend on the left frontal lobe, whereaswithdrawal-related emotions enhance performance in spatial

2 The reader will have noticed that the switch cost in the learnt irrelevance/distractibility condition may alternatively reflect persisting inhibition of theprevious distracter, suggesting that increased switch cost under positive affectmay have been due to enhanced distracter inhibition. However, as discussed byDreisbach and Goschke in their original paper, according to this account underpositive effect enhanced distractor inhibition before the switch should have beenoffset by reduced interference from distracters after the switch. Moreover,increased distracter suppression under positive affect should have reduced inter-ference on response-incompatible trials before the switch, which was not the case;by contrast, positive affect reliably increased the response-incompatibility effectafter the switch, consistent with the assumption that positive affect increaseddistractibility from novel distracters. Nevertheless, it remains an important issuefor future research to provide more direct evidence that increased switch costsunder positive affect in the learnt irrelevance condition do actually reflect anovelty bias.

3 While in Dreisbach and Goschke's (2004) study there was no indication thatpositive affective pictures changed participants' mood, in a subsequent studyMüller, Dreisbach, Goschke, Hensch, Lesch, and Brocke (2007) a reliable positivecorrelation was obtained between self-reported positive mood and the differencebetween the switch costs in the distractibility and the perseveration conditions,suggesting that a tonic positive mood may also be associated with reducedperseveration and increased distractibility.






tasks that depend on the right frontal lobe (cf. Gray, Braver, &Raichle, 2002). A more recent study (Martin & Kerns, 2011) foundthat a positive mood slightly but reliably impaired performance ina running memory span test, in which participants were instructedto remember the last six digits in a series of random numbers. Theauthors suggested that positive mood increased spread of activa-tion from items in working memory and thereby impaired theability to maintain task-relevant information in the focus ofattention. This interpretation fits with evidence that positive affectlow in motivational intensity is associated with more widespreadactivation of remote associates in memory (Bolte et al., 2003;Topolinski & Deutsch, 2013). However, contrary to these findings, afurther recent study (Yang, Yang, & Isen, 2013) found that positivemood (induced by giving participants an unexpected gift)improved working memory performance in an operation spantask. Given the different working memory paradigms used inthese studies, further research will be needed to delineate thetask features and moderators that may account for this discre-pancy in mood effects.

More consistent findings have been obtained in a series ofstudies, which investigated the effect of positive affective stimulion maintenance and updating of task-relevant information in theAX Continuous Performance Task (AX-CPT) (Cohen, Braver, &O'Reilly, 1996). In this task, participants are presented with asequence of letters and have to execute a specific response when atarget letter (X) is preceded by a specific cue (A), but have toexecute a different response when the target letter is preceded bya different cue (BX) or when a non-target letter (Y) is presented(AY or BY) (note that in the actual task, B and Y stimuli areinstantiated by letters randomly drawn from a larger letter set,excluding letters X and A). Target trials (AX) are presented muchmore frequently than the other trials types to induce a dominanttendency to execute the target response. The AX-CPT has theadvantage that it provides specific information about the costs andbenefits of the active maintenance of task-relevant information.In particular, robust maintenance of the A cue should impairperformance on AY trials, because the cue-induced bias to executethe predominant (but incorrect) target response must be overcome.By contrast, maintenance of the B cue should improve performanceon BX trials, because the cue biases response preparation towardsthe (correct) non-target response, which should make it easier toovercome the dominant tendency to execute the target response.

If positive affect attenuates maintenance and facilitates switch-ing (as suggested by Dreisbach and Goschke (2004)), it shouldconfer a benefit specifically to AY trials (due to weaker preparationof the incorrect target response), but impair performance on BXtrials (due to a weaker bias towards the correct non-targetresponse). This hypothesis was first tested by Dreisbach (2006),who presented either positive, negative or neutral affective pic-tures at the beginning of each trial (affect was manipulatedbetween participants). The main result was that positive (com-pared to neutral and negative) pictures reduced error rates on AYtrials, but increased error rates and RTs on BX trials, especiallywhen task-irrelevant distracters were presented during the cue-target interval. This pattern of improved AY performance andimpaired BX performance is consistent with the assumption thatpositive affect attenuated cue maintenance, thereby leading toimproved performance when subjects had to execute the non-target response to the Y-probe following the A-cue, but impairingperformance when the predominant target response to the X-probe had to withhold following the B-cue.

The finding of improved performance on AY trials underpositive affect was replicated in two subsequent studies. vanWouwe, Band, and Ridderinkhof (2011) induced positive affectby presenting emotionally positive or neutral movie clips before arun of trials and found that error rates on AY trials were again

reduced under positive affect. In addition, positive affect wasassociated with a smaller target-related N2-component of theevent-related potential (ERP), presumably reflecting reduced con-flict between the predominant target response and the correctnon-target response. Likewise, Frober and Dreisbach (2014) foundthat positive affective pictures presented at the beginning of eachtrial were associated with performance benefits (reduced errorrates and faster RTs) on AY trials, compared to a different group ofparticipants who were presented neutral pictures. While thesethree studies provide consistent evidence for improved AY perfor-mance under positive affect, a further recent study by Chiew andBraver (2014) did not replicate these findings. In this study, aneutral or positive video clip was presented before the task block,and, in addition, on each trial in the positive affect block either apositive or neutral affective picture was randomly presented. Errorrates in the positive relative to the neutral affect block weregenerally reduced, but slightly increased on AY trials. Trials withpositive and neutral picture within the positive affect block did nodiffer in error rates, but AY trials with positive relative to neutralpictures were associated with a (very small) increase of RTs. Thus,if anything, positive affect was associated with a slight trendtowards increased proactive control. One possible explanationfor this discrepancy might reside in the fact that Chiew and Bravernot only presented a positive affective picture on each trial, butparticipants also watched a positive film clip prior to the AX-CPT,which may have attenuated phasic emotional responses to thetrial-by-trial positive pictures. Moreover, in contrast to Chiew andBraver (2014), Dreisbach (2006) and Frober and Dreisbach (2014)presented distractors during the cue–probe-delay, which presum-ably increased the requirement to shield task-relevant informationfrom interference and may thus have rendered the task moresusceptible to effects of reduced cue maintenance under positiveaffect. While further research will be required to elucidate whichhitherto unexplored moderator variables may account for thisdiscrepancy, in general studies using the AX-CPT yielded consis-tent evidence that positive affect improves performance specifi-cally on AY trials, presumably reflecting reduced cue maintenance.

It is instructive to interpret these findings in the context of thedistinction between proactive and reactive control proposed byBraver (2012) in his “dual mechanisms of control” framework.Proactive control denotes preparatory control processes such asthe active maintenance of task-relevant context information,which serve to prepare for an upcoming difficult task or anexpected conflict. Reactive control denotes the transient recruit-ment and rapid adjustment of cognitive control in response toconflicts and can be conceived of as a trouble-shooting mechanismthat serves to overcome competing responses and to cope with aconflict after it has been detected. Within this framework,improved performance on AY trials under positive affect can beattributed to reduced proactive control, which should reducemaintenance of the A cue and lead to weaker preparation of the(incorrect) target response, thus reducing response conflict whenthe Y probe appears.

If positive affect attenuates proactive control, this should alsoshow up in performance impairments on BX trials, because lessmaintenance of the B-cue reduces preparation of the (correct)non-target response, which should make it more difficult toovercome the bias towards the predominant (but incorrect) targetresponse. However, whereas Dreisbach (2006) did find thatpositive affect impaired performance on BX trials, van Wouweet al. (2011) obtained no evidence for an impairment of BXperformance under positive affect, nor did they find effects ofpositive affect on cue-related ERP components (the P3b andcontingent negative variation), which they considered as signa-tures of proactive control. Accordingly, these authors concludedthat positive affect had no effect on proactive control and






attributed improved performance on AY trials to enhanced reac-tive control, which helped to overcome the cue-induced biastowards the dominant but incorrect target response when unex-pectedly the Y probe appeared. At present it is difficult to resolvethis discrepancy, as the two studies used very different affectinduction methods (affective pictures on each trial vs. moodinduction via film clips before a block of trials). Moreover,Dreisbach (2006), but not van Wouwe et al. (2011), presenteddistractors during the cue–probe delay. Finally, a subtle butpotentially important detail in van Wouwe et al.'s study was theinclusion of occasional no-go probes following B cues, whichrequired participants to refrain from responding. This was doneto motivate participants to sustain attention following B cues andmay have had the effect, that they specifically maintained B-cuesto prevent premature responses, which would counteract amaintenance-reducing effect of positive affect. However, it isunlikely that these procedural differences fully account for theconflicting results, as Frober and Dreisbach (2014) also found noeffect of positive affect on BX trials, even though they used trial-by-trial picture presentation to induce positive affect, presenteddistractors during the delay, and did not include no-go trials.

While further research will be required to clarify why reducedproactive control under positive affect primarily shows up on AYtrials but is not consistently evident on BX trials of the AX-CPT, it isnoteworthy that converging evidence for reduced proactive con-trol stems from a study (Frober & Dreisbach, 2012) on the effects ofpositive affect on the usage of informative cues. The cues helpedparticipants to prepare either the next response in a spatial cueingtask or the next to-be-performed task in a task-switching para-digm. The probability that a cue contained valid information aboutthe next task was well above chance but not perfect. Thus thedifference in performance between validly compared to invalidlycued trials (the cue-validity effect) is an indicator of the degree towhich participants engaged in proactive control and made use ofthe cues to prepare for the next response or upcoming task. Acrossthree experiments, brief presentation of positive affective pictureslow in arousal was associated with a reduced cue validity effect ascompared to positive affective pictures high in arousal, with thecue validity effect in the neutral affect group lying in between thetwo positive affect groups. This indicates that positive affect low inarousal reduced cue usage and proactive control, whereas positiveaffect high in arousal appears to increase proactive control,suggesting that arousal is an important moderator of effects ofpositive affect on proactive control.

3.3. Dissociations between effects of positive affect and reward

Although the primary focus of this review is on positive affect,it is a theoretically important question how the effects of task-irrelevant positive affect relate to effects of performance-contingent reward. At first sight it might seem plausible thatpositive affective stimuli (e.g., pleasurable pictures) can be con-ceived of as conditioned reward cues due to their strong pre-experimental associations with rewarding objects or situations.Conversely, given that rewards typically arouse positive affect, itappears intuitively plausible that positive affect and rewardincentives exert similar effects on cognitive control. However,contrary to this assumption, recent evidence indicates that task-irrelevant positive affect and cues signaling the prospect of gaininga performance-contingent reward differ markedly in their effectson cognitive control. Here we do not aim to provide a compre-hensive review of the rapidly expanding literature on reward andcognitive control (cf. Chiew & Braver, 2011; Dreisbach & Fischer,2012; Zedelius et al., 2014), but focus specifically on effects ofreward incentives on the stability–flexibility balance in taskswitching and working memory studies.

In line with other recent proposals (Dreisbach & Fischer, 2012)the control dilemma framework assumes that task-irrelevantpositive affect and reward cues serve different adaptive functionsin goal-directed action. Task-irrelevant positive affect (especiallywhen low in arousal and approach motivation intensity) presum-ably constitutes a safety cue that signals that goal pursuit runssmoothly or even better than expected and therefore attenuateseffortful proactive control in favor of a more exploratory mode ofcontrol. By contrast, the primary function of reward incentives isto enhance the motivation to recruit effortful control in order toincrease the chances of gaining the reward.

Müller, Dreisbach, Goschke, Hensch, Lesch, and Brocke (2007)used Dreisbach and Goschke's set-switching paradigm to investi-gate effects of a performance-contingent monetary reward on theshielding–shifting balance. After a baseline block without reward,a cue before a subsequent reward block signaled the opportunityto gain a monetary bonus if a pre-specified performance criterionwas met. Whereas there were no differences between the switchcosts in the perseveration and distractibility conditions in controlblocks, in the reward block switch costs were reliable higher in theperseveration compared to the distractibility condition, indicatinghigher cognitive stability and reduced susceptibility to interfer-ence. Moreover, the effect of the monetary incentive was moder-ated by whether participants reported that they had investedmore effort after the reward cue or whether they had taken arelaxed attitude towards the reward and felt no need for increasedeffort. The self-reported tendency to exert more effort wasassociated with smaller switch costs in the distractibility andincreased switch costs in the perseveration condition, indicatingincreased goal shielding. By contrast, participants reporting arelaxed attitude towards the task tended to show the oppositepattern of reduced perseveration and increased distractibility (i.e.,the pattern observed by Dreisbach and Goschke (2004) undertask-irrelevant positive affect). Thus participants who consideredit sufficiently valuable to exert enhanced effort to obtain a rewardshowed greater cognitive stability and an improved filtering out ofdistracting stimuli.

In line with these findings, it has been shown that performancein a delayed response task, in which participants had to maintainfive words during a delay period and which presumably alsorequired filtering out irrelevant cognitive contents, was reliablyimproved by cues signaling the opportunity to obtain either asecondary (monetary) or a primary (liquid) reward (Beck, Locke,Savine, Jimura, & Braver, 2010). Likewise, in a study using the AX-CPT (Locke & Braver, 2008), offering participants a monetary bonusfor each trial on which their response time was faster than theirmedian response time in a baseline blocks led to markedly fasterresponse times compared to the baseline block. Importantly, errorrates were not increased in the reward block, except on AY trials,suggesting that the prospect of a monetary reward increased cuemaintenance and proactive control. Consistent with this finding,three further recent studies (Chiew & Braver, 2013, 2014; Frober &Dreisbach, 2014) likewise showed that in conditions with anopportunity to gain a performance-contingent monetary reward,error rates were selectively increased on AY trials, indicatingincreased cue maintenance and enhanced proactive control.4

In conclusion, there is consistent evidence that the prospect ofgaining a performance-dependent reward increases proactivecontrol, as indicated by improved maintenance and shielding oftask-relevant information. This raises the question whether rewardcues always increase cognitive stability or whether reward cues

4 These studies yielded several further interesting findings, e.g., with respect topupil dilation as an index of proactive control, and with respect to transient (trial-by-trial) vs. sustained (block-wise) reward effects, which are beyond the scope ofthe present review, which primarily focuses on positive affect.






may also increase cognitive flexibility when a task requiresintentional set switching (cf. Aarts, van Holstein, & Cools, 2011).Relevant evidence stems from task-switching studies which exam-ined whether switch costs can be reduced by increasing partici-pants' motivation to engage in active advance preparation for anew task. While it has been controversially debated whetherswitch costs reflect structural limitations of the cognitive systemwhich may not be amenable to motivational manipulations (cf. DeJong, 2000; Nieuwenhuis & Monsell, 2002; Verbruggen, Liefooghe,Vandierendonck, & Demanet, 2007), a number of recent studiesconsistently showed that reward incentives can substantiallyreduce switch costs, especially when participants have sufficienttime to prepare in advance for a task (Kleinsorge & Rinkenauer,2012; Savine, Beck, Edwards, Chiew, & Braver, 2010). Interestingly,in a study in which reward magnitude was varied from trial to trial(Shen & Chun, 2011), switch costs were specifically reducedfollowing cues signaling the prospect to obtain a reward thatwas higher than on the previous trial, suggesting that especiallytransient increases in reward expectancy increased cognitiveflexibility.5

Moreover, there is evidence that reward incentives can pro-mote disengagement from to-be-abandoned task-sets as indicatedby modulations of the so-called backward inhibition effect (Jiang &Xu, 2014). Backward inhibition (Koch et al., 2010; Mayr & Keele,2000) denotes the observation that RTs are increased whenparticipants in a sequence of tasks must switch back to a taskthey performed two trials before (ABA), compared to whenswitching to a new task (CBA). This N-1 repetition cost in ABAsequences is attributed to the fact that after a task switch thepreviously executed task-set is inhibited to prevent proactiveinterference. Backward inhibition thus indicates how rapidlyparticipants disengage from a to-be-abandoned task and can beconsidered a signature of cognitive flexibility. In Jiang and Xu's(2014) experiment, on a random selection of trials a reward cueinformed participants that they could receive a monetary bonusbased on their performance. The backward inhibition effect (i.e.,the N-2 repetition cost) was reliably larger when the task on trialN-1 was rewarded, compared to when trial N-1 was not rewarded,indicating that the prospect of receiving a reward for task N-1increased inhibition (and thereby reduced perseveration) of taskN-2. Conversely, rewarding the task on trial N-2 eliminated thebackward inhibition effect, suggesting that the rewarded N-2 taskwas activated more strongly and/or received less inhibition aftercompletion (note that a rewarded task on trial N-2 was alwaysfollowed by a non-rewarded task on trial N-1).

In summary, reward incentives have markedly different effectsthan task-irrelevant positive affective stimuli. Findings from task-switching, working memory and AX-CPT studies provide consis-tent evidence that reward cues have a primarily motivationaleffect and enhance participants' willingness to recruit effortfulproactive control in order to increase the likelihood of gainingreward by improving task performance. In line with this conclu-sion, brain imaging studies have revealed that the opportunity togain a performance-contingent reward was associated with

increased activation in a network of brain regions involved incognitive control functions. fMRI studies of working memory (Becket al., 2010; Pochon, 2002), context updating and maintenance(Jimura, Locke, & Braver, 2010; Kouneiher, Charron, & Koechlin,2009; Locke & Braver, 2008; Savine et al., 2010), and task switch-ing (Kouneiher et al., 2009; Savine et al., 2010) showed thatperformance-contingent reward induced increased activation inregions of the anterior and dorsolateral PFC, the inferior frontaljunction, the parietal cortex, and the anterior cingulate cortex,which are involved in active task preparation and goal mainte-nance, response inhibition, and performance monitoring. More-over, some of these studies showed that the opportunity to obtaina reward was associated with a shift in the activation dynamics inthe lateral PFC from a transient to a mostly tonic mode, especiallyin participants showing high reward sensitivity (Jimura et al.,2010). In addition, Beck et al. (2010) found that on trials of adelayed response task, on which participants could gain aperformance-contingent liquid reward, there was an additionalshift in the activation dynamics in the PFC from primarily target-related activation (indicating reactive control) to primarily cue-related activation (indicating proactive control). Of note, effects ofreward incentives on brain activation were often found in regionsthat overlapped with or were adjacent to regions involved in taskprocessing or regions sensitive to an actual increase of taskdifficulty (e.g., Pochon, 2002; Taylor et al., 2004). This suggeststhat reward incentives led to an enhanced recruitment of brainsystems mediating successful task performance especially underdemanding conditions.

3.4. Interim summary

The studies reviewed in the previous sections have providedevidence that task-irrelevant positive affect exerts a modulatoryinfluence on the mode of cognitive control. In cognitive setswitching studies positive affect low in approach motivationintensity has been shown to shift the shielding–shifting balancetowards increased flexibility, as indicated by reduced persevera-tion at the cost of heightened distractibility (as we discussed inFootnote 2 this interpretation presupposes that increased switchcosts in the learnt irrelevance condition of Dreisbach andGoschke's set switching paradigm reflect increased interferencefrom novel distracters rather than enhanced inhibition of theprevious distracter before the switch). Likewise, low motivation-intensive positive (compared to negative) affect has been shown toreduce goal shielding as indicated by increased between-taskcrosstalk in a dual tasking context. In contrast, positive affect highin approach motivation intensity had the opposite effect and wasassociated with increased perseveration and reduced distractibil-ity (Liu & Wang, 2014), in line with the motivational dimensionalmodel of affect and with findings from studies of affectivemodulations of the scope of attention (Harmon-Jones et al., 2012).

With respect to working memory, studies that examined effectsof tonic positive mood yielded mixed findings, including evidencefor impaired, enhanced, as well as both improved and impairedperformance for verbal and spatial working memory, respectively.These discrepancies may partly be due to the different workingmemory tasks used in these studies, which differ in the demandsthey put on specific component processes (e.g., passive mainte-nance vs. continuous updating). More consistent findings havebeen obtained with the AX-CPT, in which positive affect (inducedby affective pictures presented on each trial or by affective filmclips presented before a run of trials) has repeatedly been shownto improve performance selectively on AY trials, suggesting thatpositive affect reduced cue maintenance and attenuated proactivecontrol. However, while reduced proactive control should alsoshow up in impaired performance on BX trials, there is no

5 Interestingly, in this study increased reward expectancy was associated withthe shortest RTs on task-switch trials, but increased RTs on task-repeat trials,compared to trials on which reward magnitude stayed constant. According to theauthors, lowest switch costs may not have coexisted with fastest task-repeat RTs inthe same condition because of a tradeoff between flexibility and stability. That is,on trials on which participants were most flexible they may have been lessprepared to repeat a just performed task. Thus, larger than expected rewardsincreased flexible preparation for different tasks, thereby facilitating switchingbetween task-sets, but they also reduced perseveration of the previous task-set,thereby increasing RTs on task repeat trials. This provides converging evidence forthe idea that cognitive stability and flexibility constitute antagonistic controlrequirements.






consistent evidence for a detrimental effect of positive affect on BXtrials.

In marked contrast to the effects of task-irrelevant positiveaffect, studies examining effects of reward incentives consistentlyshowed that cues signaling the prospect of gaining a performance-contingent reward increase proactive control. Importantly, whileincreased proactive control shows up in enhanced cognitivestability in tasks requiring robust maintenance and filtering outof distracting information, a reward-induced increase of proactivecontrol can also show up in enhanced flexibility when tasksrequire intentional set switching, presumably by promoting theusage of task-cues to prepare in advance for an upcoming task.6

4. Computational and neurobiological mechanismsunderlying effects of positive affect on cognitive control

Whereas neuroimaging studies have yielded evidence that effectsof performance-contingent reward incentives on the recruitment ofcognitive control are mediated by sustained recruitment of brainnetworks involved in cognitive control (Beck et al., 2010; Kouneiheret al., 2009; Locke & Braver, 2008; Pochon, 2002; Savine et al., 2010;Taylor et al., 2004), relatively little is known about the computa-tional and neural mechanisms that mediate effects of task-irrelevantpositive affect on cognitive control. Within our control dilemmaframework we proposed that positive affect may modulate thestability–flexibility-balance by influencing the setting of meta-control parameters, in particular, the updating threshold thatdetermines the stability of goal representations in working memory(Doya, 2008; Goschke, 2013). On a neurobiological level, meta-control parameters may be instantiated by the action of neuromo-dulators such as dopamine, serotonin and noradrenaline, whichmodulate neuronal processing in spatially distributed brain areasand thus constitute a plausible biological substrate for the regulationof global control parameters (Doya, 2002). In particular, animalresearch, pharmacological and genetic imaging studies, and studiesof neurodegenerative diseases provide strong evidence that dopa-mine (DA) in the PFC and basal ganglia plays a central role inmodulating the balance between cognitive stability and flexibility (e.g., Armbruster, Ueltzhöffer, Basten, & Fiebach, 2012; Colzato,Waszak, Nieuwenhuis, Posthuma, & Hommel, 2010; Cools,Sheridan, Jacobs, & D'Esposito, 2007; Cools, Stefanova, Barker,Robbins, & Owen, 2002; Stelzel, Basten, Montag, Reuter, & Fiebach,2010; van Holstein et al., 2011) (for reviews see Cools, 2008; Cools &D'Esposito, 2011; Durstewitz & Seamans, 2002; Floresco, 2013;Klanker, Feenstra, & Denys, 2013; Robbins & Arnsten, 2009).

4.1. The dopamine hypothesis of positive affect

In light of the prominent role of DA in the regulation ofcognitive control and working memory maintenance it has beenproposed that (some of the) effects of positive affect on cognitivecontrol may be mediated by influences of DA on neural processingin the PFC and basal ganglia. Notably, Ashby et al. (1999), cf. Ashby

et al. (2003) proposed that moderate increases of positive affectare associated with increased DA release in midbrain areas such asthe ventral tegmental area (VTA), and that ascending DA projec-tions from the VTA to the PFC mediate effects of positive affect onworking memory, whereas DA projections to the anterior cingulatecortex (ACC) and basal ganglia mediate effects of positive affect onexecutive attention and cognitive set switching. It should be notedthat Ashby et al. were careful to make clear that their hypothesiswas not meant to imply that DA mediates the pleasant feelingsassociated with positive affect. In fact, there is now strongevidence that midbrain DA systems most likely do not mediatepleasurable feelings (“liking”), but play a specific role in incentivemotivation (“wanting”) (Berridge & Kringelbach, 2013) and thecomputation of reward prediction errors (Schultz, 2013).

Since its initial formulation, the DA hypothesis of positive affecthas gained great popularity, as is documented by the fact that inmany studies of the past decade effects of positive affect oncognitive flexibility have been interpreted with explicit referenceto the role of DA. However, this popularity notwithstanding, itmust be noted that there is currently no direct evidence thateffects of positive affect are in fact mediated by effects of DA onprefrontal-striatal brain circuits. Rather, support for the DAhypothesis stems almost exclusively from studies showing thatpositive affect and increased DA release have – at least undercertain conditions – similar effects on cognitive control, fromwhich it is inferred that effects of positive affect may partly bemediated by dopaminergic activity.

Moreover, research in the past two decades has revealed thateffects of DA on cognitive control are highly complex and depend –

in sometimes opposite ways – on a multitude of factors, includingdifferences in baseline DA levels, target regions (e.g., PFC or basalganglia), receptor types, and whether DA effects are phasic or tonic(for reviews see Beaulieu & Gainetdinov, 2011; Cools & D'Esposito,2011; Floresco, 2013; Robbins & Arnsten, 2009). In particular, thereis evidence that DA in the PFC and basal ganglia play different oreven opponent roles in stable working memory maintenance andflexible set switching, respectively (Cools & D'Esposito, 2011; Hazy,Frank, & Oreilly, 2006; O'Reilly & Frank, 2006; van Schouwenburg,Aarts, & Cools, 2010). Thus, to evaluate the DA hypothesis ofpositive affect one must distinguish between different specificmechanisms by which DA may mediate effects of positive affect onthe stability–flexibility balance.

We propose three distinct yet interacting mechanisms bywhich effects of positive affect could in principle be mediated.These mechanisms refer to (1) the role of DA in the PFC inmodulating the stability of working memory representations,(2) the role of phasic midbrain DA activity as a gating signal forworking memory updating, or (3) the role of tonic DA levels in thebasal ganglia in the modulation of flexible set switching. We firstdiscuss briefly whether positive affect is in fact associated withincreased activity in midbrain and mesolimbic DA systems andthen turn to the question whether and how each of the threedopaminergic mechanisms might in principle account for some ofthe effects of positive affect on the stability–flexibility balance.

4.2. Is positive affect associated with increased dopaminergicactivity?

A critical assumption of the DA hypothesis is that positive affectis associated with increased DA release in midbrain areas and thatascending projections from these areas to the basal ganglia, PFC,and ACC mediate the effects of positive affect on cognitive control.This assumption derives its plausibility from the fact that theinduction of positive affect in the laboratory often involvesunexpected rewards (e.g., giving participants an unanticipatedgift) or the presentation of reward-related stimuli (e.g., funny

6 In this section we only discussed studies on the effects of reward cues thatsignaled the prospect of gaining a reward prior to a task. It should be noted thatrewards following successful task performance appear to have very different effectsthan reward cues presented prior to task performance. While reward cues before atask were associated with an increased focusing of attention, reward delivery aftertask performance broadened the scope of attention (Gable & Harmon-Jones, 2011).Likewise, presentation of positive pictures signaling the receipt of a monetaryreward for successful performance reduced conflict-induced shielding of task-setsin a task-switching paradigm (Braem et al., 2013). Reward following goal attain-ment presumably signals that successful goal pursuit does not require recruitmentof effortful control and thus induces a more relaxed or exploratory mode ofprocessing (cf. Carver, 2003).






films; pleasurable pictures). It is well-established from single cellrecordings in animals that unexpected rewards or cues signalingan upcoming reward elicit phasic burst activity of DA neurons inthe VTA, which is assumed to code a positive reward predictionerror that signals that an outcome is better than expected (forreviews see Salamone & Correa, 2012; Schultz, 2013). Consistentwith these findings, human neuroimaging studies have shown thatunexpected rewards elicit increased fMRI signals in midbrain DAareas (for recent meta-analyses see Diekhof, Kaps, Falkai, & Gruber,2012; Sescousse, Caldu, Segura, & Dreher, 2013). Based on suchfindings, the DA hypothesis of positive affect assumes that (task-irrelevant) positive affective stimuli likewise elicit DA release inreward-related brain areas. Support for this assumption stemsfrom fMRI studies, which showed that pleasant emotional stimulisuch as beautiful faces (Aharon et al., 2001), pictures of romanticcouples or erotic pictures (Hamann, Herman, Nolan, & Wallen,2004; Sabatinelli, Bradley, Lang, Costa, & Versace, 2007), picturesof loved ones (Aron et al., 2005), beautiful paintings (Ishizu & Zeki,2011) and pleasant music (Montag, Reuter, & Axmacher, 2011)elicit activation in brain structures involved in the processing orprediction of reward, including the VTA, nucleus accumbens,caudate nucleus, and medial PFC. Moreover, activation in themesolimbic DA system has been shown to correlate positivelywith the subjective valence of pleasant pictures (Gerdes et al.,2010; Schlochtermeier et al., 2013), imagined scenes (Costa, Lang,Sabatinelli, Versace, & Bradley, 2010), affective sentences(Colibazzi et al., 2010), and liked music (Blood & Zatorre, 2001)(see also Knutson, Katovich, & Suri, 2014). Of note, a recent studyin which positron emission tomography (PET) study was used tomeasure DA D2 receptor binding potential revealed that peakpleasure experiences while listening to music were associatedwith increased DA release in the striatum (caudate, putamen, andnucleus accumbens) (Salimpoor, Benovoy, Larcher, Dagher, &Zatorre, 2011).

While these findings suggest a link between positive affect andDA release, several important caveats must be kept in mind. First,meta-analyses of neuroimaging studies show that positive affect isneither exclusively nor specifically associated with activation inmidbrain or mesolimbic DA systems, but elicits activation in broadnetwork of subcortical as well as cortical brain structures (includ-ing the PFC, cingulate gyrus, temporal lobe, amygdala, and occipi-tal cortex) (Lindquist, Wager, Kober, Bliss-Moreau, & Barrett, 2012;Vytal & Hamann, 2010). Secondly, some of the regions involved inreward processing (e.g., the nucleus accumbens) are also activatedby unpleasant stimuli (Carretie et al., 2009; Leknes & Tracey, 2008)and cellular-level studies indicate that the mesolimbic DA systemgenerates both appetitive and aversive prediction error signals(Brooks & Berns, 2013). Thirdly, the reported studies do not allowdifferentiating between tonic and phasic DA activity and whetherDA effects are mediated by D1 and D2 receptor types.

4.3. Can effects of positive affect on cognitive control be accountedfor by dopaminergic modulations of prefrontal-striatal brain systems?

While evidence that positive affective stimuli elicit activity inmidbrain and mesolimbic DA systems is a necessary precondition forthe DA hypothesis of positive affect, it is not sufficient for concludingthat effects of positive affect on cognitive control are in fact mediatedby dopaminergic modulations of the PFC and/or basal ganglia. Todate evidence for this hypothesis stems almost exclusively fromfindings supporting the weaker prediction that (some of) the effectsof increased DA release on cognitive control mimic (some of) theeffects of positive affect on the stability–flexibility balance. Below wereview evidence for similar effects of positive affect and DA withrespect to DA in the PFC and working memory maintenance, the roleof phasic DA activity in the VTA as a gating signal for working

memory updating, and the role of tonic DA levels in the basal gangliain set switching.

4.3.1. Dopaminergic modulation of cognitive stability in theprefrontal cortex: regulation of the updating threshold

The PFC, which is critically involved in cognitive control, goalmaintenance, and top-down biasing of attention and responseselection, contains a high concentration of DA receptors and isstrongly influenced by ascending modulatory inputs from mid-brain DA systems. Evidence from single cell recordings in animalsas well as pharmacological and genetic neuroimaging studiesindicates that DA levels in the PFC are related in an invertedU-shape manner to active working memory maintenance (forreviews see Arnsten, 2009; Cools & D'Esposito, 2011; Floresco,2013; Robbins & Roberts, 2007). An optimal (medium) DA levelrenders active representations resistant against interference (pos-sibly by enhancing the signal-to-noise ratio of neural processing(Cohen & Servan-Schreiber, 1992)), but also reduces the capacityfor flexible updating and switching. Conversely, too low or high DAlevels are assumed to lead to fragile working memory representa-tions, which can easily be updated but are prone to interference(Durstewitz & Seamans, 2002; Durstewitz, Seamans, & Sejnowski,2000a; O'Reilly, 2006; Seamans & Yang, 2004; van Schouwenburget al., 2010). Durstewitz and Seamans (2008) proposed a dual-statemodel of the PFC, according to which the PFC can operate in twocontrol modes: a DA D1 receptor dominated state characterized bya high barriers among different attractor states, which promotesrobust maintenance but impairs shifting; a DA D2 receptordominated state characterized by a low barriers between attractorstates, which promotes flexible switching but increases interfer-ence. The balance between DA D1 and DA D2 dominated statesthus constitutes a possible neurobiological implementation of theupdating threshold (cf. Section 2.2) as a critical meta-controlparameter regulating the flexibility–stability balance. Accordingly,one may speculate that one possible mechanism by which positiveaffect might increase cognitive flexibility is by shifting the balancebetween these opponent states towards a D2 receptor dominatedstate. Conversely, increased goal shielding (as induced byperformance-contingent reward incentives) could be mediatedby a shift towards a D1 dominated state (cf. Frober & Dreisbach,2014). However, as there is currently no direct evidence for theseassumptions, further research will be required to obtain evidencethat effects of positive affect are actually mediated by changes inthe balance between DA D1 and DA D2 dominated states. More-over, the suggested account raises the question why positive affectand performance-contingent reward incentives – if both areassociated with increased DA release in midbrain areas thoughtto modulate processing in the PFC – have different or evenopposite effects on the flexibility–stability balance.

4.3.2. Dopamine gating hypothesisIn order to explain when and how working memory is updated,

a number of recent theories postulate a specific gating mechanism,which regulates whether or not afferent information gains accessto the PFC. In the absence of a gating signal working memoryrepresentations are maintained and shielded from distraction,whereas when a gating signal is triggered by novel, salient, orunexpected reward-predicting stimuli, the gate to working mem-ory is opened such that novel information gains access to workingmemory and eventually replaces the currently maintained goal ortask-set. According to one specific version of this hypothesis(Braver & Cohen, 2000; D'Ardenne et al., 2012; Montague,Hyman, & Cohen, 2004) the gating signal is implemented on aneural level by phasic burst activity of DA neurons in the VTA. Suchactivity is typically elicited by unexpected rewards or reward-cues






and has been assumed to encode a reward prediction error signal(Schultz, 2013). According to the gating hypothesis, this phasic DAsignal serves at the same time as a gating signal that triggers theencoding of afferent information into working memory.7 To datethe most direct evidence for the DA gating hypothesis wasobtained in a study using a modified AX-CPT (D'Ardenne et al.,2012), which showed that the encoding of task-relevant contextcues was associated with increased fMRI activity in the VTA.Moreover, this activity correlated positively with behavioral per-formance as well as with activity in a region of the right DLPFC,which the authors had previously shown to be causally involved incue maintenance with the help of transcranial magnetic stimula-tion. These findings are consistent with the idea that a phasicmidbrain DA signal triggers the encoding of new context repre-sentations into the PFC and thereby induces an updating ofworking memory.

To account for effects of positive affect or reward on setswitching and working memory updating, one must considerpossible interactions between the gating mechanisms and theupdating threshold. First, if the updating threshold is relativelyhigh, reward cues may promote the selective gating of task-relevant information into working memory in tasks such as theAX-CPT. According to Braver's (2012) dual modes of controlframework a phasic DA gating signal is a precondition for encodingtask-relevant information into working memory, such that thisinformation can be maintained across time. Thus, if a reward cuethat appears immediately before the context cue in the AX-CPTelicits phasic midbrain DA activity that functions as a gating signal,this should promote the encoding of the context cue and itssubsequent maintenance in working memory, provided the cuefollows in sufficiently close temporal proximity to the reward cuethat triggered the gating signal.

By contrast, a low updating threshold that corresponds to low“barriers” between different attractor states in the PFC should havethe effect that weaker gating signals suffice to open the gate toworking memory, thus rendering gating less selective. As aconsequence, switching between states should become easier,but novel or salient task-irrelevant stimuli should also have ahigher likelihood of gaining access to working memory, therebyincreasing distractibility. This could explain why reduced cuemaintenance in the AX-CPT task under positive affect was espe-cially pronounced when distractors were presented during thedelay, and why positive affective stimuli increased switch costs inDreisbach and Goschke's set switching paradigm when participanthad to filter out novel distracters on the switch trial.

4.3.3. The role of DA in the basal ganglia in the modulation ofcognitive flexibility

Increasing evidence indicates that, complementary to the role ofprefrontal DA in stabilizing working memory representations, DA inthe basal ganglia is involved in cognitive set switching (Cools, 2008;Cools & D'Esposito, 2011; Frank, Loughry, & O'Reilly, 2001; O'Reilly &Frank, 2006; van Schouwenburg et al., 2010). Extending earlier workon the functions of the basal ganglia in the selective gating of motoractions (e.g., Redgrave, Prescott, & Gurney, 1999), several recentmodels assume that working memory updating and set switchingdepend on a delicate balance of excitatory and inhibitory effects ofDA on D1 and D2 receptors in opponent direct (“Go”) and indirect

(“NoGo”) thalamo-cortico-striatal pathways, respectively (Frank etal., 2001; Hazy et al., 2006; Hazy, Frank, & O'Reilly, 2007; Herd,Banich, & O'Reilly, 2006; O'Reilly et al., 2010). More specifically,activity of DA D1 neurons in the Go pathway of the dorsal striatum isassumed to facilitate updating of working memory representationsin the PFC by inhibiting the tonic inhibitory influence of thesubstantia nigra pars reticulate on the thalamus, thereby disinhibit-ing the inflow of afferent information into the PFC and facilitatingupdating of working memory representations.8 Activity of DA D2neurons in the NoGo pathway is assumed to counteract updatingand to raise the threshold for cognitive set switching by increasinginhibition of the thalamus and suppressing afferent inflow into thePFC (cf. Frank & O'Reilly, 2006).

Disregarding much of the complexity of these models, ofparticular relevance for the DA hypothesis of positive affect isthe assumption that the net effect of increased tonic striatal DAlevels is a “go bias” that facilitates the gating of new informationinto working memory and promotes attentional or cognitive setswitching. By contrast, the net effect of reduced DA levels is a“nogo bias” that prevents updating (Frank & Fossella, 2011; Hazyet al., 2006; O'Reilly & Frank, 2006; van Schouwenburg et al.,2010). Consistent with this assumption, evidence from pharmaco-logical and genetic imaging studies indicates that increased DAlevels in the basal ganglia promote updating and set shifting(Cools et al., 2007; Samanez-Larkin et al., 2013; Stelzel et al.,2010; van Holstein et al., 2011; van Schouwenburg, den Ouden, &Cools, 2010). Moreover, BOLD responses in the basal ganglia havebeen shown to predict moment-to-moment fluctuations of cogni-tive flexibility in task-switching (Leber, Turk-Browne, & Chun,2008) and working memory updating (McNab & Klingberg,2008). Most interestingly, there is evidence that DA in the PFCand the basal ganglia mediate opponent processes, with increasedprefrontal DA promoting stability and increased striatal DA pro-moting flexibility (Cools & D'Esposito, 2011). For instance, whileDA depletion in the PFC induced increased distractibility, DA lossin the striatum was associated with reduced distractibility (Croftset al., 2001). Conversely, DA depletion in the striatum impairedworking memory while leaving certain forms of attentional setshifting intact (Collins, Wilkinson, Everitt, Robbins, & Roberts,2000), whereas administration of the D2 receptor antagonistsulpiride in humans impaired task switching while protectingagainst simpler (but not more complex) types of distraction(Mehta, Manes, Magnolfi, Sahakian, & Robbins, 2004). Finally,Parkinson's disease patients off medication suffering from DAdepletion in the dorsal striatum showed increased task-switchcosts (Cools, Barker, Sahakian, & Robbins, 2001a, 2001b) butenhanced resistance to distraction in a delayed response task(Cools, Miyakawa, Sheridan, & D'Esposito, 2010).

These findings parallel the opposite effects of positive affect onperseveration and distractibility reported in the set-switchingstudies reviewed above (Dreisbach & Goschke, 2004; Liu andWang, 2014). Converging evidence that increased DA in the basalganglia may have effects that mimic the effects of positive affecton set switching stems from studies, which examined individualdifferences in spontaneous eye-blink rates (EBR) as an allegedmarker of DA levels in the striatum.9 These studies found reliable

7 As the gating signal serves simultaneously as a prediction error signal thatdrives reinforcement learning (Lee, Seo, & Jung, 2012), the gating hypothesis alsoprovides an account of how the system can learn when to update: each time areward cue is gated into working memory, the association between the cue, thegating signal, and the goal that leads to reward will be strengthened. Thus, thelikelihood that future encounters with the same cue will again elicit a gating signalwill be increased (Braver & Cohen, 2000; Montague et al., 2004).

8 Note that the assumption in Frank and O'Reilly's (2006) model that activity inthe Go pathway triggers updating is similar to the DA gating hypothesis by Braverand Cohen (2000), D'Ardenne et al. (2012) and Montague et al. (2004) discussed inthe previous section. The main difference is that in the Frank and O'Reilly modelphasic burst activity of midbrain DA neurons serves only as a learning signal forwhen to gate, whereas the gating mechanisms itself is implemented by the basalganglia.

9 The use of spontaneous EBR as a marker for striatal DA levels is supportedby studies showing that administration of DA agonists increased spontaneousEBR in monkeys (Jutkiewicz, 2004; Karson, 1983; Kleven & Koek, 1996), rats






and consistent associations between the spontaneous EBR andperformance in Dreisbach and Goschke's (2004) set-switchingparadigm. Strikingly, across three independent studies from ourand other labs (Dreisbach et al., 2005; Müller, Dreisbach, Brocke,Lesch, Strobel, & Goschke, 2007; Tharp & Pickering, 2011), sponta-neous EBR was positively correlated with the switch cost in thedistractibility condition, whereas it was negatively related to theswitch cost in the perseveration condition. Likewise, when parti-cipants were classified according to whether their EBR was aboveor below the sample median, the high (but not the low) EBR groupshowed the identical pattern of reduced perseveration andincreased distractibility that has previously been observed afterthe induction of positive affect (Dreisbach & Goschke, 2004; Liu &Wang, 2014). The finding that participants with high blink rateswere better at switching attention to a new target category, butshowed increased switch costs when having to ignore noveldistractors is consistent with the assumption that high EBR isassociated with increased DA levels in the striatum, which shouldreduce the threshold for cognitive set shifting and bias attentiontowards novel or salient stimuli. Of note, a recent study(Chermahini & Hommel, 2012) provided initial evidence that aninduced positive mood did indeed reliably increase both sponta-neous EBR as well as cognitive flexibility in a divergent thinkingtask (especially in individuals showing a below median baselineEBR, presumably indicating low striatal DA levels). While thesefindings should be replicated before drawing overly strong con-clusions, they provide important initial support for the assumedlink between positive affect and increased striatal DA levels asindexed by individual differences in the spontaneous EBR (see alsoColzato, Slagter, Spapé, & Hommel, 2008).

4.4. Interim summary

According to the DA hypothesis of positive affect, effects ofpositive affect on the balance between cognitive stability andflexibility are partly mediated by dopaminergic modulations ofneural processing in prefrontal-striatal brain circuits. We refinedthis hypothesis by proposing three distinct mechanisms by whichDA may mediate effects of positive affect and reward. According tothe first proposal, reduced maintenance and increased distractibilityobserved under positive affect may reflect a lowered updatingthreshold, possibly reflecting a shift towards a DA D2 receptordominated state in the PFC. A low updating threshold correspondsto low barriers between different attractor states in the PFC andleads to unselective gating of novel or salient information intoworking memory, thereby promoting flexible set shifting andworking memory updating, but also increasing distractibility. Analternative – not mutually incompatible – account capitalizes onrecent computational models of the basal ganglia, according towhich an increased striatal DA level has the net effect of facilitatingset switching and working memory updating. This hypothesis

receives indirect support from the finding, that a high spontaneousEBR (as an alleged marker of increased tonic DA in the striatum)was associated precisely with the pattern of reduced perseverationand increased distractibility that has been found under positiveaffect of low approach motivation intensity.

Apart from its possible role in positive effect, DA may also playa role in effects of performance-contingent reward. According tothe DA gating hypothesis (Braver & Cohen, 2000; D'Ardenne et al.,2012; Montague et al., 2004) cues signaling the prospect to gain aperformance-contingent reward presumably trigger a phasic DAsignal that is assumed to open the gate to working memory. Thiscould explain why presenting a reward cue at the beginning of atrial in the AX-CPT enhances the selective gating of the subsequentcontext cue into working and thereby promotes its subsequentmaintenance (Braver, 2012).10 However, while different versions ofthe DA hypothesis can in principle account for some of the effectsof positive affect, it must also be clearly stated that currently thesehypotheses rest exclusively on indirect evidence for similar effectsof increased DA and positive affect on the stability–flexibilitybalance.

5. Discussion

In contrast to a large body of research on effects of positiveaffect on perception, attention, and problem-solving, the study ofhow positive affect influences complementary modes of cognitivecontrol is still in its infancy. Nevertheless, as this review attests,the past decade has witnessed a rapid increase of evidence thatpositive affect modulates the balance between cognitive stabilityand flexibility in the context of set switching and working memorymaintenance vs. updating. This research has revealed a much moredifferentiated picture as was anticipated in the initial studies adecade ago. Accordingly, simple hypothesis relating positive affectto a general increase of cognitive flexibility are increasinglyreplaced by more sophisticated theories that attempt to accountfor dissociations between effects of task-irrelevant positive affectand performance-contingent reward and take into account mod-erator variables such as motivational intensity and arousal. In thissection we summarize main conclusions, point to unresolvedquestion, and outline perspectives for future research.

5.1. General conclusions

5.1.1. Effects of positive affect on cognitive control modesFrom the control dilemma framework we had derived the

prediction that positive affect should be associated with bothbenefits and costs depending on the processing demands of aparticular task. The results of the reviewed task-switching andworking memory studies are generally consistent with this pre-diction and indicate that task-irrelevant positive affect (especiallywhen low in approach motivation intensity) shifts the shielding–shifting balance from robust maintenance and proactive controltowards flexible updating and reactive control. This was reflectedin the observation that positive affect (1) reduced perseveration of

(footnote continued)(Kaminer, Powers, Horn, Hui, & Evinger, 2011) and humans (Blin, Masson, Azulay,Fondarai, & Serratrice, 1990), whereas DA receptor antagonists reduced blink rates(Jutkiewicz, 2004; Kaminer et al., 2011; Lawrence, Redmond, Elsworth, Taylor, &Roth, 1991) and blocked an agonist-induced increase in blink rate (Elsworth et al.,1991). Moreover, neurotoxic lesions of the substantia nigra and its DA projectionssignificantly reduced blink rates in monkeys (Lawrence & Redmond, 1991), and thiseffect was reversed by administration of a DA D1 receptor agonist (Taylor et al.,1991). Parkinson patients with a loss of DA cells in the substantia nigra show areduction in EBR (Deuschl & Goddemeier, 1998; Karson, Burns, Lewitt, Foster, &Newman, 1984; Karson, Lewitt, Calne, & Wyatt, 1982), which can be counteractedby dopaminergic treatment (Agostino et al., 2008; Karson et al., 1982, 1984). Thereis thus converging evidence that spontaneous EBR may be useful marker of DAlevels in the striatum (but see van der Post, de Waal, de Kam, Cohen, & van Gerven,2004), although the roles of DA D1 and DA D2 receptors in the regulation of EBRremain unclear.

10 Moreover, there is evidence that DA in the striatum may be involved inmediating effects of performance-contingent reward on task-switching perfor-mance. In a study of genetic variation in the DAT1 gene, which is assumed tomodulate baseline DA levels in the striatum (Aarts et al., 2010), carriers of the 9-repeat polymorphism of the DAT1 gene with presumably high striatal DA levels notonly showed greater activity in the ventromedial striatum in response to rewardcues, but also a greater reduction of task-switch costs and greater switch-relatedactivity in the dorsomedial striatum following cues signaling high relative to lowreward, when compared to 10-repeat homozygotes with presumably low striatalDA levels. This suggests that effects of anticipated reward on task switching maypartly be mediated by increased DA levels in the striatum.






cognitive sets at the cost of increased interference from noveldistracters in set switching studies, (2) attenuated goal shieldingand increased between task crosstalk in dual tasking, and(3) impaired maintenance of task-relevant context information inthe AX-CPT. These findings are consistent with studies that haveexamined complementary benefits and costs of positive affect inother domains of cognitive processing and showed that positiveaffect improved performance in tasks requiring a broadened scopeof attention and access to remote semantic associates (Bolte et al.,2003; Rowe et al., 2007; Topolinski & Deutsch, 2013), but impairedperformance in tasks requiring focused attention and filtering outof distracting information (Rowe et al., 2007; Schmitz, De Rosa, &Anderson, 2009). However, our review also revealed a number ofempirical inconsistencies that will be discussed below.

5.1.2. Effects of reward on cognitive control modesA second assumption of our control dilemma framework holds

that the effects of positive affect on cognitive control modesdepend on the adaptive function of emotional signals for goal-directed action. In line with other recent proposals (Chiew &Braver, 2011; Dreisbach & Fischer, 2012), we suggested that task-irrelevant positive affect (especially when low in motivationalintensity and arousal) signals a safe environment and/or that goalpursuit runs better than expected, thus promoting a moreexploratory and flexible but also distractible control mode, whichprimarily relies on reactive control to cope with unexpectedconflicts (cf. Bolte & Goschke, 2010; Carver, 2003; Frober &Dreisbach, 2014). By contrast, positive affect elicited by theprospect of gaining a performance-contingent reward has aprimarily motivational function and promotes the recruitment ofeffortful control. Consistent with this assumption, there is con-sistent evidence that reward incentives increase effortful proactivecontrol, which shows up in improved maintenance of task-relevant context information and filtering out of distracters, butcan also improve set-switching due to enhanced usage of task-cues in order to prepare in advance for a new task.

The dissociations between positive affect and performance-contingent reward in the reviewed set switching and workingmemory studies are generally consistent with evidence from otherdomains of cognitive control (for reviews see Chiew & Braver,2011; Dreisbach & Fischer, 2012). For instance, while task-irrelevant positive affect or performance-noncontingent rewardhave been shown to attenuate error monitoring (Wiswede, Münte,Goschke, & Rüsseler, 2009; Wiswede, Münte, Kramer, & Rüsseler,2009) and counteract conflict-adaptation effects (e.g., vanSteenbergen, Band, & Hommel, 2009, 2010), indicating a morerelaxed control mode, performance-contingent reward had theopposite effect and increased conflict adaptation followingrewarded trials (e.g., Braem, Verguts, Roggeman, & Notebaert,2013; Sturmer, Nigbur, Schacht, & Sommer, 2011), indicatingenhanced conflict-induced goal shielding.

Like positive affect, reward incentives were associated withcomplementary costs and benefits. For instance, in the AX-CPTtask increased cue maintenance induced by reward cues generallyimproved performance, but led to increased error rates selectivelyon AY trials, which required inhibition of the cue-induced biastowards the dominant target response (Chiew & Braver, 2014;Frober & Dreisbach, 2014). Likewise, the opportunity to gain aperformance-contingent reward facilitated disengagement from aprevious task-set, but also increased switch cost when participantshad to switch back to the previously inhibited task (Jiang & Xu,2014). Finally, while the expectancy of a reward increased flex-ibility as indicated by faster RTs on task-switch trials, it increasedRTs on task-repeat trials, possibly reflecting a smaller repetitionbenefit due to reduced cognitive stability (Shen & Chun, 2011).

5.2. Open questions and directions for future research

Our review also revealed a number of empirical inconsistenciesand unresolved theoretical questions, which pose importantchallenges for future research on emotional modulations oncognitive control.

5.2.1. Empirical inconsistencies and replication failuresAs we already discussed in detail, while most studies using the

AX-CPT yielded evidence that positive affect with low approachmotivation intensity attenuates proactive control, contrary to whatone should expect there was no consistent evidence for impairedperformance on BX trials under positive affect. Moreover, incontrast to AX-CPT studies using phasic affect induction viaaffective pictures, studies examining effects of tonic positive moodon working memory yielded less consistent findings, includingevidence for impaired, improved, or both impaired and improvedperformance depending on the modality of the to-be-maintainedinformation. Given the different working memory tasks used(N-back, running memory span, operation span), further researchwill be required to account for these discrepancies. Finally, itshould be noted that there have been failures to find effects ofpositive affect on the AX-CPT (Chiew & Braver, 2014) and on taskswitching performance (Eber, 2010). This parallels recent failuresto replicate seemingly well-established effects of positive affect onthe scope of attention (Bruyneel et al., 2013; Martin & Kerns, 2011)and underscores the importance of regular and systematicattempts to replicate theoretically important findings.

5.2.2. Moderator variablesOur review identified a number of moderators of effects of

positive affect. Most obviously, effects of positive affective signalsdepend critically on their adaptive function in goal-directed actionand, in particular, whether or not positive affect is task-relevantand contingent on performance. In light of the observed dissocia-tions between task-irrelevant positive affect and performance-contingent reward incentives it is a central aim for future researchto systematically compare emotional and motivational effects ofpositive affect on a broader range of cognitive control processesand to examine by which underlying mechanism the two types ofeffects are mediated (cf. Chiew & Braver, 2011). The approachtaken in some recent studies to directly compare emotional andmotivational modulations of cognitive control within the sametasks appears particularly fruitful in this respect. Moreover, com-bining such designs with functional neuroimaging methodsshould help to elucidate commonalities and differences in theneural systems which underlie emotional and motivational mod-ulations of cognitive control modes.

Secondly, there is initial evidence that effects of positiveaffective stimuli on set switching are moderated by their approachmotivation intensity in similar ways as has previously beenreported for modulations of the scope of visual attention. In linewith the motivational dimensional model of affect (Harmon-Joneset al., 2012), positive affect low vs. high in approach motivationintensity had in fact opposite effects on perseveration and dis-tractibility in set switching (Liu & Wang, 2014). It remains to beexamined whether motivational intensity moderates effects ofpositive affect on the flexibility–stability balance in similar waysin other tasks such as the AX-CPT. Moreover, it will be animportant question for future studies how the effects of positiveaffect high in approach motivation intensity differ depending onwhether the affective stimuli are related to the current task goal(as is the case for performance-contingent reward incentives) orwhether they are unrelated to the current task (as is the casewhen, for instance, food pictures are presented in a set switching






task). According to the motivational dimensional model of affect(Harmon-Jones et al., 2012) both types of high approach motiva-tion stimuli should narrow the scope of attention and therebyenhance goal shielding, irrespective of whether they are related tothe current task goal. In contrast, the reviewed findings suggestthat task-related reward incentives primarily increase participants'motivation to recruit effortful control strategies to optimize taskperformance, which may lead to enhanced goal shielding or tofacilitated goal switching, depending on task requirements. It thusremains an interesting question for further research whetherreward incentives may even induce a broadened scope of attentionand more exploratory mode of control if this was required forsuccessful task performance.

Thirdly, the role of arousal in affective modulations of theflexibility–stability balance remains to be investigated more thor-oughly. Although several of the reviewed studies reported reliableeffects of positive emotional valence even when controlling forarousal levels, there is also initial evidence that positive affect lowin arousal reduced, but positive affect high in arousal increasedproactive control (Frober & Dreisbach, 2012). This finding fits witha neurobiological theory of the role of the locus coeruleus–norepinephrine (LC–NE) system in arousal (Aston-Jones & Cohen,2005). According to this theory, the LC–NE system can operate in atonic or a phasic mode, and the balance between these modes isassumed to modulate the tradeoff between exploration andexploitation. In particular, phasic activation of the LC–NE system,as elicited by salient and arousing stimuli, is thought to induce anexploitative control mode, which serves to optimize task perfor-mance by facilitating task-relevant responses and by enhancingthe focusing of attention on task-relevant stimuli. Thus arousingpictures may have increased proactive control by triggering aphasic mode of the LC–NE system. In this respect it is also worthmentioning that it has recently been found that switch costs in avoluntary task-switching paradigm were increased after emotion-ally arousing (compared to neutral) pictures, irrespective of emo-tional valence (Demanet, Liefooghe, & Verbruggen, 2011). Theauthors suggested that phasic modulations of the noradrenergicsystem by highly arousing stimuli strengthen stimulus–responsebindings of an active task-set, thereby raising the threshold for anendogenously initiated switch to a different task (Verguts &Notebaert, 2009). From a more general perspective, one theoreti-cally important aim for future research will be to elucidate moresystematically the relation between arousal and motivationalintensity. In particular, given that it has long been assumed thatheightened arousal narrows the focus of attention (Easterbrook,1959), and given that affective stimuli high in approach motivationintensity are often also more arousing than stimuli low inmotivational intensity, it remains to be investigated whethermotivational intensity and arousal exert dissociable effects oncognitive control or reflect a common underlying mechanism.

Finally, we lack studies investigating commonalities and differ-ences between the effects of tonic moods and transient emotionalresponses on cognitive control. While there is clear evidence thatbriefly presented affective stimuli modulate cognitive processingparameters even if they do not induce enduring subjective moodchanges, and sometimes even if they are not consciously perceived(for a review see Friedman & Förster, 2010), it remains animportant question for future research how transient affectiveresponses and tonic moods differ in their effects on cognitivecontrol parameters. For instance, it is frequently assumed thatemotions serve an informational role (e.g., about the success ofgoal pursuit or unexpected obstacles) and that individuals use thisinformation to adjust processing strategies accordingly (Clore,Gasper, Garvin, & Forgas, 2001; Clore & Huntsinger, 2007;Schwarz & Clore, 1983). However, it remains an open questionwhether affective states must be consciously perceived in order to

serve this function, or whether phasic emotional responses toaffective stimuli may induce automatic adjustments of cognitivecontrol modes even if they are neither accompanied by changes inconscious emotions nor mediated by deliberate reflection aboutthe informational content of an affective response.

5.2.3. Computational and neural mechanisms underlying effects ofpositive affect on cognitive control

The second main aim of this review was to discuss possiblecomputational and neural mechanisms underlying influences ofpositive affect on cognitive control modes. To this end we criticallyreviewed the popular hypothesis that effects of positive affect onthe balance between cognitive stability and flexibility may partlybe mediated by dopaminergic modulations of neural processing inprefrontal-striatal brain circuits (Ashby et al., 1999; cf. Ashby et al.,2003). In the light of recent theories of the role of DA in cognitivecontrol and working memory, we distinguished different mechan-isms by which DA could mediate effects of positive affect. Onehypothesis assumes that positive affect low in approach motiva-tion intensity is associated with a lowered updating threshold,possibly reflecting a shift from a DA D1 to a DA D2 receptordominated control mode in the PFC. This could account both forreduced maintenance and perseveration (due to low barriersbetween attractor states in the PFC) as well as increased distract-ibility (due to less selective gating of novel information into thePFC). According to a second – not mutually incompatible –

hypothesis, positive affect is associated with increased DA levelsin the basal ganglia, which according to some recent computa-tional models should have the net effect of facilitating set switch-ing and working memory updating.

However, although the DA hypothesis has great heuristic valueand can in principle account for some of the effects of positiveaffect and reward, it also raises several unresolved questions. First,as we already noted, there is currently no direct evidence thateffects of positive affect are in fact mediated by dopaminergicmodulations of prefrontal-striatal brain circuits. Thus a centralchallenge for future research will be to go beyond demonstrationsthat effects of positive affect mimic assumed effects of prefrontaland/or striatal DA and to obtain more direct evidence for amediating role of DA. This will require studies that investigate(1) whether positive affective stimuli do indeed elicit fMRI activityin midbrain and mesolimbic DA regions and that show (2) that thisactivity does in fact predict moment-to-moment fluctuations ofcognitive flexibility as measured in set switching or workingmemory updating tasks. A challenge for such more rigorous testsof the DA hypothesis results from the fact that, due to thecomplexity of the underlying neural systems, it is difficult toderive specific and falsifiable predictions without making a largenumber of ancillary assumptions. For instance, effects of DA oncognitive control modes depend on baseline DA levels, whetheractivity is tonic or phasic, whether DA effects are mediated by D1or D2 type receptors, and whether DA exerts its effects in the PFCand/or in direct and indirect basal ganglia pathways (Cools &D'Esposito, 2011; Floresco, 2013). While it is beyond the scope ofany single study to control or measure all possibly critical factors,this complexity underscores the importance of combining manip-ulations of positive affect with additional measures of, forinstance, individual differences in baseline DA levels. This couldbe achieved either by using indirect markers for baseline DA levelssuch as working memory capacity (Cools & D'Esposito, 2011) orspontaneous EBR (Chermahini & Hommel, 2012; Dreisbach et al.,2005), by using PET to measure DA synthesis capacity or bindingpotential, or by examining genetic polymorphisms known tomoderate baseline DA levels. Equally important, to test specificpredictions derived from the DA hypothesis requires theoretically






well-grounded behavioral tasks that allow assessing both the costsand benefits of complementary control modes as indicators foraffect-induced changes in meta-control parameters. Moreover, astasks used to assess cognitive control are usually not process-purebut involve several component processes, effects of positive affectshould ideally be tested using more than one task. This allowsderiving latent variable scores and testing effects of affectivemanipulations on the level of latent constructs (e.g., shifting;updating) rather than at the level of individual tasks (Miyake etal., 2000).

Finally, it should be noted that is unlikely that effects of positiveaffect on meta-control parameters and complementary controlmodes are mediated exclusively by dopaminergic systems. Rather,DA systems are known to interact with other neurotransmitterssuch as serotonin and noradrenaline, which are likewise involvedin the regulation of emotions, arousal, and cognitive control(Boureau & Dayan, 2011; Cools, Nakamura, & Daw, 2011; Doya,2008). While we restricted our review to the DA hypothesis ofpositive affect, a realistic theory will have to take into accountinteractions between different neurotransmitter systems.

5.2.4. The need for integrating positive affect and reward intocomputational models of cognitive control

On a theoretical level, assumptions about effects of positiveaffect and reward on cognitive control modes need to be inte-grated more closely into computational models of cognitive con-trol and its modulation by DA (Frank et al., 2001; Hazy et al., 2006,2007; Herd et al., 2006, 2014; O'Reilly et al., 2010) (for an earlyattempt in this direction see Ashby et al., 2004). This is of centralimportance for at least three reasons. First, predictions derivedfrom the DA hypothesis often depend on ancillary assumptionsabout complex interactions between multiple factors, the implica-tions of which are difficult to spell out in solely verbal models. Asan example, consider the hypothesis that increased distractibilityunder positive affect may reflect the fact that a low updatingthreshold in the PFC leads to less selective gating of novelinformation into the PFC, such that not only task-relevant stimuli,but also distracting information gains easy access to workingmemory. To turn this hypothesis into testable predictions requiresspecifying additional assumptions about the assumed gatingmechanisms. For instance, what is the temporal dynamics of thegating mechanism, i.e., how long does the gate to workingmemory stay open after its opening has been triggered by a phasicDA signal? Do all stimuli within this time window gain access toworking memory with equal probability? How does access toworking memory depend on stimulus features like salience,novelty, or task-relevance? Depending on how one answers thesequestions, different predictions follow with respect to how posi-tive affect will influence the gating of distracting information intoworking memory, and computational models require that suchancillary assumptions are made explicit.

Secondly, while it may appear intuitively plausible that cogni-tive flexibility and stability constitute antagonistic adaptive con-straints, explicit computational models help to explore in detailunder which conditions and in what ways tradeoffs betweenstability and flexibility show up in performance. A nice exampleis the recent neural network modeling study by Herd et al. (2014)we mentioned in the introduction, which showed that increasingthe stability of goal representations in a simulated PFC had indeedopposite effects on the network's capability to inhibit unwantedresponses vs. to flexibly switch between tasks.

Finally, computational modeling will help to specify moreprecisely which processing mechanisms are affected by positiveaffect in a given behavioral task. For instance, although at a verballevel it appears intuitively plausible that reduced error rates on AY

trials of the AX-CPT task under positive affect reflect reducedmaintenance of the A-cue, it is also conceivable that positive affectincreased the ability to rapidly overcome the cue-induced biastowards the target response once the Y-probe appeared. Explicittask models require one to specify explicitly which componentprocesses are assumed to be influenced by positive affect and thusallow deriving more specific predictions to be tested in novelexperiments.

5.2.5. Cognitive stability and flexibility at different levels of the goalhierarchy

To date emotional modulations of cognitive flexibility havebeen investigated mostly in relatively simple task-switching andworking memory paradigms. However, many daily tasks involvehierarchic structures with goals and sub-goals at different levels ofabstraction (e.g., from writing a paper to formulating the nextsentence to pressing a particular key on the notebook). Such tasksoften require being flexible and stable at the same time as when,for instance, one must maintain a superordinate goal (e.g., writingan article) while flexibly switching between means to achieve thegoal (e.g., trying different formulations to optimally express anidea). Multilevel tasks can thus impose conflicting constraints onthe optimal balance of flexibility and stability at different levelsin the goal hierarchy. If a specific means fails to attain a goal,premature abandoning of the goal may prevent one from dis-covering that the goal could have been achieved with differentmeans; conversely, rigid goal maintenance despite repeated fail-ures with different means carries the risk of investing resources ina goal that is in fact unattainable. An interesting initial step toinvestigate emotional modulations at different task levels wasmade in a recent task-switching study (Marien, Aarts, & Custers,2012), which showed that positive affective pictures reducedswitch costs when participants were induced to focus on the taskgoal (categorizing letters), whereas positive affect increasedswitch costs when participants represented the task in terms ofmeans (pressing specific keys). Of note, a recent neuroimagingstudy indicates that the distinction between levels of abstractionin goal hierarchies is also relevant for neurobiological mechanismsmediating the flexible updating of cognitive sets. In a modified AX-CPT task that required updating on different levels of contextualinformation, the basal ganglia were specifically involved in updat-ing higher-level contextual information and gating of new task-sets into the PFC, but not in updating of lower-level contextualinformation that required shifting attention between elements of atask-set within working memory (Nee & Brown, 2013). In light ofthese findings, an important goal for future research will be toinvestigate how positive affect and reward influence the flexibil-ity–stability balance at different levels of a goal hierarchy.

5.3. Closing remarks

In this review we restricted our discussion to the role of positiveaffect in the modulation of the shielding–shifting dilemma. Ourguiding hypothesis was that emotions are associated with differentsettings of meta-control parameters and that positive affect, inparticular, modulates the updating threshold that determines howefficiently goals and task-sets are shielded from distraction. How-ever, the control dilemma framework specifies a number of addi-tional control dilemmas, which involve different adaptive tradeoffsbetween complementary control modes and are linked to differentmeta-control parameters (Goschke, 2003, 2013). These include theattention breadth, which regulates the balance between goal-directed focusing of attention and stimulus-driven capture ofattention by potentially significant stimuli outside the currenttask-induced focus of attention; the temporal discounting rate, which






determines how steeply delayed reward is discounted; the degree ofnoise in neural representations which modulates the balancebetween the exploitation of learnt knowledge and the explorationof novel options; and the learning rate, which determines howrapidly associations between cues, actions, and rewards are chan-ged in the light of new experiences. A particular configuration ofsuch control parameters constitutes a global control state that ischaracterized by a particular mode of interaction between large-scale brain systems serving complementary control functions (e.g.,goal-directed vs. stimulus-driven attention; exploration vs. exploi-tation; Barrett & Satpute, 2013; Bressler & Menon, 2010). This raisesthe intriguing question how configurations of meta-control para-meters and the associated global control states are influenced bypositive affect and reward. On a neurobiological level this impliesthat it is highly oversimplified to relate single neuromodulators tospecific control parameters (as we did in this review with respect toDA). Rather, understanding emotional modulations of cognitivecontrol modes will require investigating interactive effects ofneuromodulators on configurations of meta-control parameters(Doya, 2008; Rogers, 2011). In the long run, insights into interac-tions between emotion, motivation, and global control modes mayalso shed light on impairments of cognitive control in mentaldisorders, which are often characterized by dysfunctional meta-control parameter settings and aberrant interactions betweenemotional and cognitive control networks (Goschke, 2014; Hasler,2012; Menon, 2011).

Acknowledgments

The preparation of this chapter and the authors' research onemotion and cognitive control are supported by the GermanResearch Foundation (DFG grants SFB 940/1 2013 and SFB 940/12014).

References

Aarts, E., Roelofs, A., Franke, B., Rijpkema, M., Fernandez, G., Helmich, R. C., et al.(2010). Striatal dopamine mediates the interface between motivational andcognitive control in humans: Evidence from genetic imaging. Neuropsycho-pharmacology, 35(9), 1943–1951. http://dx.doi.org/10.1038/npp.2010.68.

Aarts, E., van Holstein, M., & Cools, R. (2011). Striatal dopamine and the interfacebetween motivation and cognition. Frontiers in Psychology, 2, 163. http://dx.doi.org/10.3389/fpsyg.2011.00163.

Agostino, R., Bologna, M., Dinapoli, L., Gregori, B., Fabbrini, G., Accornero, N., et al.(2008). Voluntary, spontaneous, and reflex blinking in Parkinson's disease.Movement Disorders, 23(5), 669–675. http://dx.doi.org/10.1002/mds.21887.

Aharon, I., Etcoff, N., Ariely, D., Chabris, C. F., O'Connor, E., & Breiter, H. C. (2001).Beautiful faces have variable reward value: fMRI and behavioral evidence.Neuron, 32(3), 537–551.

Allport, D. A., Styles, E. A., & Hsieh, S. (1994). Shifting intentional set: Exploring thedynamic control of tasks. In: C. Umilta, & M. Moscovitch (Eds.), Attention andperformance XV: Conscious and nonconscious information processing (pp. 421–452). Cambridge, MA, USA: MIT Press.

Altmann, E. M., & Gray, W. D. (2008). An integrated model of cognitive control intask switching. Psychological Review, 115(3), 602–639. http://dx.doi.org/10.1037/0033-295x.115.3.602.

Armbruster, D. J. N., Ueltzhöffer, K., Basten, U., & Fiebach, C. J. (2012). Prefrontalcortical mechanisms underlying individual differences in cognitive flexibilityand stability. Journal of Cognitive Neuroscience, 24(12), 2385–2399.

Arnsten, A. F. T. (2009). Stress signalling pathways that impair prefrontal cortexstructure and function. Nature Reviews Neuroscience, 10(6), 410–422. http://dx.doi.org/10.1038/nrn2648.

Aron, A., Fisher, H., Mashek, D. J., Strong, G., Li, H., & Brown, L. L. (2005). Reward,motivation, and emotion systems associated with early-stage intense romanticlove. Journal of Neurophysiology, 94(1), 327–337. http://dx.doi.org/10.1152/jn.00838.2004.

Ashby, F. G., Isen, A. M., & Turken, U. (1999). A neuropsychological theory of positiveaffect and its influence on cognition. Psychological Review, 106(3), 529–550.

Ashby, F. G., Valentin, V. V., & Turken, A. U. (2003). The effects of positive affect andarousal on working memory and executive attention: Neurobiology andcomputational models. In: S. Moore, & M. Oaksford (Eds.), Emotional cognition:from brain to behavior. Amsterdam: John Benjamins.

Aston-Jones, G., & Cohen, J. D. (2005). An integrative theory of locus coeruleus –

norepinephrine function: Adaptive gain and optimal performance. AnnualReview of Neuroscience, 28, 403–450. http://dx.doi.org/10.1146/annurev.neuro.28.061604.135709.

Banich, M. T. (2009). Executive function. The search for an integrated account.Current Directions in Psychological Science, 18, 89–94.

Banich, M. T., Mackiewicz, K. L., Depue, B. E., Whitmer, A. J., Miller, G. A., & Heller, W.(2009). Cognitive control mechanisms, emotion and memory: A neural perspec-tive with implications for psychopathology. Neuroscience & BiobehavioralReviews, 33(5), 613–630. http://dx.doi.org/10.1016/j.neubiorev.2008.09.010.

Barrett, L. F., & Satpute, A. B. (2013). Large-scale brain networks in affective andsocial neuroscience: Towards an integrative functional architecture of the brain.Current Opinion in Neurobiology, 23(3), 361–372. http://dx.doi.org/10.1016/j.conb.2012.12.012.

Baumann, N., & Kuhl, J. (2005). Positive affect and flexibility: Overcoming theprecedence of global over local processing of visual information. Motivation andEmotion, 29(2), 123–134. http://dx.doi.org/10.1007/s11031-005-7957-1.

Beaulieu, J. M., & Gainetdinov, R. R. (2011). The physiology, signaling, andpharmacology of dopamine receptors. Pharmacological Reviews, 63(1), 182–217.http://dx.doi.org/10.1124/pr.110.002642.

Beck, S. M., Locke, H. S., Savine, A. C., Jimura, K., & Braver, T. S. (2010). Primary andsecondary rewards differentially modulate neural activity dynamics duringworking memory. PLoS One, 5(2), e9251. http://dx.doi.org/10.1371/journal.pone.0009251.

Berridge, K. C., & Kringelbach, M. L. (2013). Neuroscience of affect: Brain mechan-isms of pleasure and displeasure. Current Opinion in Neurobiology, 23(3),294–303. http://dx.doi.org/10.1016/j.conb.2013.01.017.

Blin, O., Masson, G., Azulay, J. P., Fondarai, J., & Serratrice, G. (1990). Apomorphine-induced blinking and yawning in healthy-volunteers. British Journal of ClinicalPharmacology, 30(5), 769–773.

Blood, A. J., & Zatorre, R. J. (2001). Intensely pleasurable responses to musiccorrelate with activity in brain regions implicated in reward and emotion.Proceedings of the National Academy of Sciences, 98(20), 11818–11823. http://dx.doi.org/10.1073/pnas.191355898.

Bolte, A., & Goschke, T. (2010). Thinking and emotion: Affective modulation ofcognitive processing modes. In: B. Glatzeder, V. Goel, & A. V. Müller (Eds.), Onthinking, Vol. II: Towards a theory of thinking. Heidelberg: Springer.

Bolte, A., Goschke, T., & Kuhl, J. (2003). Emotion and intuition: Effects of positiveand negative mood on implicit judgments of semantic coherence. PsychologicalScience, 14(5), 416–421.

Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., & Cohen, J. D. (2001).Conflict monitoring and cognitive control. Psychological Review, 108(3),624–652.

Boureau, Y. L., & Dayan, P. (2011). Opponency revisited: Competition and coopera-tion between dopamine and serotonin. Neuropsychopharmacology: OfficialPublication of the American College of Neuropsychopharmacology, 36(1), 74–97.http://dx.doi.org/10.1038/npp.2010.151.

Braem, S., King, J. A., Korb, F. M., Krebs, R. M., Notebaert, W., & Egner, T. (2013).Affective Modulation of Cognitive Control is Determined by Performance-Contingency and Mediated by Ventromedial Prefrontal and Cingulate Cortex.Journal of Neuroscience, 33(43), 16961–16970. http://dx.doi.org/10.1523/jneur-osci.1208-13.2013.

Braem, S., Verguts, T., Roggeman, C., & Notebaert, W. (2013). Reward modulatesadaptations to conflict. Cognition, 125(2), 324–332. http://dx.doi.org/10.1016/j.cognition.2012.07.015.

Braver, T. S. (2012). The variable nature of cognitive control: A dual mechanismsframework. Trends in Cognitive Sciences, 16(2), 106–113. http://dx.doi.org/10.1016/j.tics.2011.12.010.

Braver, T. S., & Cohen, J. D. (2000). On the control of control: The role of dopaminein regulating prefrontal function and working memory. In: S. Monsell, &J. Driver (Eds.), Control of cognitive processes: Attention and performance XVIII(pp. 713–738). Cambridge, MA: MIT Press.

Bressler, S. L., & Menon, V. (2010). Large-scale brain networks in cognition:Emerging methods and principles. Trends in Cognitive Sciences, 14(6), 277–290.http://dx.doi.org/10.1016/j.tics.2010.04.004.

Brooks, A. M., & Berns, G. S. (2013). Aversive stimuli and loss in the mesocortico-limbic dopamine system. Trends in Cognitive Sciences, 17(6), 281–286. http://dx.doi.org/10.1016/j.tics.2013.04.001.

Brown, J. W., Reynolds, J. R., & Braver, T. S. (2007). A computational model offractionated conflict-control mechanisms in task-switching. Cognitive Psychol-ogy, 55(1), 37–85.

Bruyneel, L., van Steenbergen, H., Hommel, B., Band, G. P. H., De Raedt, R., & Koster,E. H. W. (2013). Happy but still focused: Failures to find evidence for a mood-induced widening of visual attention. Psychological Research – PsychologischeForschung, 77(3), 320–332. http://dx.doi.org/10.1007/s00426-012-0432-1.

Carretie, L., Rios, M., de la Gandara, B. S., Tapia, M., Albert, J., Lopez-Martin, S., &Alvarez-Linera, J. (2009). The striatum beyond reward: Caudate respondsintensely to unpleasant pictures. Neuroscience, 164(4), 1615–1622. http://dx.doi.org/10.1016/j.neuroscience.2009.09.031.

Carver, C. S. (2003). Pleasure as a sign you can attend to something else: Placingpositive feelings within a general model of affect. Cognition and Emotion, 17(2),241–261.

Chermahini, S. A., & Hommel, B. (2012). More creative through positive mood? Noteveryone! Frontiers in Human Neuroscience, 6, http://dx.doi.org/10.3389/fnhum.2012.00319.



http://dx.doi.org/10.1038/npp.2010.68



http://dx.doi.org/10.3389/fpsyg.2011.00163




http://dx.doi.org/10.1002/mds.21887



http://refhub.elsevier.com/S0028-3932(14)00235-8/sbref4







http://dx.doi.org/10.1037/0033-295x.115.3.602

http://dx.doi.org/10.1037/0033-295x.115.3.602

http://dx.doi.org/10.1037/0033-295x.115.3.602

http://dx.doi.org/10.1037/0033-295x.115.3.602




http://dx.doi.org/10.1038/nrn2648




http://dx.doi.org/10.1152/jn.00838.2004

http://dx.doi.org/10.1152/jn.00838.2004

http://dx.doi.org/10.1152/jn.00838.2004

http://dx.doi.org/10.1152/jn.00838.2004







http://dx.doi.org/10.1146/annurev.neuro.28.061604.135709






http://dx.doi.org/10.1016/j.neubiorev.2008.09.010



http://dx.doi.org/10.1016/j.conb.2012.12.012




http://dx.doi.org/10.1007/s11031-005-7957-1

http://dx.doi.org/10.1007/s11031-005-7957-1

http://dx.doi.org/10.1007/s11031-005-7957-1

http://dx.doi.org/10.1124/pr.110.002642

http://dx.doi.org/10.1124/pr.110.002642

http://dx.doi.org/10.1124/pr.110.002642

http://dx.doi.org/10.1371/journal.pone.0009251










http://dx.doi.org/10.1073/pnas.191355898
















http://dx.doi.org/10.1523/jneurosci.1208-13.2013




http://dx.doi.org/10.1016/j.cognition.2012.07.015




http://dx.doi.org/10.1016/j.tics.2011.12.010


















http://dx.doi.org/10.1007/s00426-012-0432-1

http://dx.doi.org/10.1007/s00426-012-0432-1

http://dx.doi.org/10.1007/s00426-012-0432-1

http://dx.doi.org/10.1016/j.neuroscience.2009.09.031







http://dx.doi.org/10.3389/fnhum.2012.00319







Chiew, K. S., & Braver, T. S. (2011). Positive affect versus reward: Emotional andmotivational influences on cognitive control. Frontiers in Psychology, 2, 279.http://dx.doi.org/10.3389/fpsyg.2011.00279.

Chiew, K. S., & Braver, T. S. (2013). Temporal dynamics of motivation-cognitivecontrol interactions revealed by high-resolution pupillometry. Frontiers inPsychology, 4, 15. http://dx.doi.org/10.3389/fpsyg.2013.00015.

Chiew, K. S., & Braver, T. S. (2014). Dissociable influences of reward motivation andpositive emotion on cognitive control. Cognitive, Affective, & Behavioral Neu-roscience, http://dx.doi.org/10.3758/s13415-014-0280-0.

Clore, G. L., Gasper, K., Garvin, E., & Forgas, J. P. (2001). Affect as information.Handbook of affect and social cognition (pp. 121–144)Mahwah, NJ, US: LawrenceErlbaum Associates Publishers.

Clore, G. L., & Huntsinger, J. R. (2007). How emotions inform judgment and regulatethought. Trends in Cognitive Sciences, 11(9), 393–399. http://dx.doi.org/10.1016/j.tics.2007.08.005.

Cohen, J. D., Braver, T. S., & O'Reilly, R. C. (1996). A computational approach toprefrontal cortex, cognitive control and schizophrenia: Recent developmentsand current challenges. Philosophical Transactions of the Royal Society of London.Series B: Biological Sciences, 351(1346), 1515–1527. http://dx.doi.org/10.1098/rstb.1996.0138.

Cohen, J. D., McClure, S. M., & Yu, A. J. (2007). Should I stay or should I go? How thehuman brain manages the trade-off between exploitation and exploration.Philosophical Transactions of the Royal Society B: Biological Sciences, 362(1481),933–942. http://dx.doi.org/10.1098/rstb.2007.2098.

Cohen, J. D., & Servan-Schreiber, D. (1992). Context, cortex, and dopamine: Aconnectionist approach to behavior and biology in schizophrenia. PsychologicalReview, 99(1), 45–77. http://dx.doi.org/10.1037/0033-295x.99.1.45.

Colibazzi, T., Posner, J., Wang, Z., Gorman, D., Gerber, A., Yu, S., et al. (2010). Neuralsystems subserving valence and arousal during the experience of inducedemotions. Emotion, 10(3), 377–389. http://dx.doi.org/10.1037/a0018484.

Collins, P., Wilkinson, L. S., Everitt, B. J., Robbins, T. W., & Roberts, A. C. (2000). Theeffect of dopamine depletion from the caudate nucleus of the commonmarmoset (Callithrix jacchus) on tests of prefrontal cognitive function. Beha-vioral Neuroscience, 114(1), 3–17.

Colzato, L. S., Slagter, H. A., Spapé, M. M. A., & Hommel, B. (2008). Blinks of the eyepredict blinks of the mind. Neuropsychologia, 46(13), 3179–3183. http://dx.doi.org/10.1016/j.neuropsychologia.2008.07.006.

Colzato, L. S., Waszak, F., Nieuwenhuis, S., Posthuma, D., & Hommel, B. (2010). Theflexible mind is associated with the catechol-O-methyltransferase (COMT)Val158Met polymorphism: Evidence for a role of dopamine in the control oftask-switching. Neuropsychologia, 48(9), 2764–2768. http://dx.doi.org/10.1016/j.neuropsychologia.2010.04.023.

Cools, R. (2008). Role of dopamine in the motivational and cognitive control of behavior.Neuroscientist, 14(4), 381–395. http://dx.doi.org/10.1177/1073858408317009.

Cools, R., Barker, R. A., Sahakian, B. J., & Robbins, T. W. (2001a). Enhanced orimpaired cognitive function in Parkinson's disease as a function of dopaminer-gic medication and task demands. Cerebral Cortex, 11(12), 1136–1143. http://dx.doi.org/10.1093/cercor/11.12.1136.

Cools, R., Barker, R. A., Sahakian, B. J., & Robbins, T. W. (2001b). Mechanisms ofcognitive set flexibility in Parkinson's disease. Brain: A Journal of Neurology, 124(12), 2503–2512. http://dx.doi.org/10.1093/brain/124.12.2503.

Cools, R., & D'Esposito, M. (2011). Inverted-U-shaped dopamine actions on humanworking memory and cognitive control. Biological Psychiatry, 69(12), e113–-e125. http://dx.doi.org/10.1016/j.biopsych.2011.03.028.

Cools, R., Miyakawa, A., Sheridan, M., & D'Esposito, M. (2010). Enhanced frontalfunction in Parkinson's disease. Brain, 133(Pt 1), 225–233. http://dx.doi.org/10.1093/brain/awp301.

Cools, R., Nakamura, K., & Daw, N. D. (2011). Serotonin and dopamine: Unifyingaffective, activational, and decision functions. Neuropsychopharmacology: Offi-cial Publication of the American College of Neuropsychopharmacology, 36(1),98–113. http://dx.doi.org/10.1038/npp.2010.121.

Cools, R., Sheridan, M., Jacobs, E., & D'Esposito, M. (2007). Impulsive personalitypredicts dopamine-dependent changes in frontostriatal activity during compo-nent processes of working memory. Journal of Neuroscience, 27(20), 5506–5514.http://dx.doi.org/10.1523/jneurosci.0601-07.2007.

Cools, R., Stefanova, E., Barker, R. A., Robbins, T. W., & Owen, A. M. (2002).Dopaminergic modulation of high-level cognition in Parkinson's disease: Therole of the prefrontal cortex revealed by PET. Brain: A Journal of Neurology, 125(3), 584–594. http://dx.doi.org/10.1093/brain/awf052.

Costa, V. D., Lang, P. J., Sabatinelli, D., Versace, F., & Bradley, M. M. (2010). Emotionalimagery: Assessing pleasure and arousal in the brain's reward circuitry. HumanBrain Mapping, 31(9), 1446–1457. http://dx.doi.org/10.1002/hbm.20948.

Crofts, H. S., Dalley, J. W., Van Denderen, J. C. M., Everitt, B. J., Robbins, T. W., &Roberts, A. C. (2001). Differential effects of 6-OHDA lesions of the frontal cortexand caudate nucleus on the ability to acquire an attentional set. Cerebral Cortex,11(11), 1015–1026.

D'Ardenne, K., Eshel, N., Luka, J., Lenartowicz, A., Nystrom, L. E., & Cohen, J. D. (2012).Role of prefrontal cortex and the midbrain dopamine system inworking memoryupdating. Proceedings of the National Academy of Science of the United States ofAmerica, 109(49), 19900–19909. http://dx.doi.org/10.1073/pnas.1116727109.

De Baene, W., Kuhn, S., & Brass, M. (2012). Challenging a decade of brain researchon task switching: Brain activation in the task-switching paradigm reflectsadaptation rather than reconfiguration of task sets. Human Brain Mapping, 33(3), 639–651. http://dx.doi.org/10.1002/hbm.21234.

De Jong, R. (2000). An intention-activation account of residual switch costs. In:S. Monsell, & J. Driver (Eds.), Attention and performance XVIII: Control of cognitiveprocesses (pp. 357–376). Cambrdige, MA: MIT Press.

Demanet, J., Liefooghe, B., & Verbruggen, F. (2011). Valence, arousal and cognitivecontrol: A voluntary task switching study. Frontiers in Psychology, 2, http://dx.doi.org/10.3389/fpsyg.2011.00336.

Desimone, R., & Duncan, J. (1995). Neural mechanisms of selective visual attention.Annual Review of Neuroscience, 18, 193–222.

Deuschl, G., & Goddemeier, C. (1998). Spontaneous and reflex activity of facialmuscles in dystonia, Parkinson's disease, and in normal subjects. Journal ofNeurology, Neurosurgery & Psychiatry, 64(3), 320–324. http://dx.doi.org/10.1136/jnnp.64.3.320.

Diekhof, E. K., Kaps, L., Falkai, P., & Gruber, O. (2012). The role of the human ventralstriatum and the medial orbitofrontal cortex in the representation of rewardmagnitude – An activation likelihood estimation meta-analysis of neuroimagingstudies of passive reward expectancy and outcome processing. Neuropsychologia,50(7), 1252–1266. http://dx.doi.org/10.1016/j.neuropsychologia.2012.02.007.

Doya, K. (2002). Metalearning and neuromodulation. Neural Networks, 15(4–6),495–506. http://dx.doi.org/10.1016/s0893-6080(02)00044-8.

Doya, K. (2008). Modulators of decision making. Nature Neuroscience, 11(4),410–416. http://dx.doi.org/10.1038/nn2077.

Dreisbach, G. (2006). How positive affect modulates cognitive control: The costsand benefits of reduced maintenance capability. Brain and Cognition, 60(1),11–19.

Dreisbach, G., & Fischer, R. (2012). The role of affect and reward in the conflict-triggered adjustment of cognitive control. Frontiers in Human Neuroscience, 6,342. http://dx.doi.org/10.3389/fnhum.2012.00342.

Dreisbach, G., & Goschke, T. (2004). How positive affect modulates cognitivecontrol: Reduced perseveration at the cost of increased distractibility. Journalof Experimental Psychology: Learning Memory and Cognition, 30(2), 343–353.http://dx.doi.org/10.1037/0278-7393.30.2.343.

Dreisbach, G., Müller, J., Goschke, T., Strobel, A., Schulze, K., Lesch, K. P., et al. (2005).Dopamine and cognitive control: The influence of spontaneous eyeblink rateand dopamine gene polymorphisms on perseveration and distractibility.Behavioral Neuroscience, 119(2), 483–490. http://dx.doi.org/10.1037/0735-7044.119.2.483.

Durstewitz, D., & Seamans, J. K. (2002). The computational role of dopamine D1receptors in working memory. Neural Networks, 15(4–6), 561–572. http://dx.doi.org/10.1016/s0893-6080(02)00049-7.

Durstewitz, D., & Seamans, J. K. (2008). The dual-state theory of prefrontal cortexdopamine function with relevance to catechol-o-methyltransferase genotypesand schizophrenia. Biological Psychiatry, 64(9), 739–749. http://dx.doi.org/10.1016/j.biopsych.2008.05.015.

Durstewitz, D., Seamans, J. K., & Sejnowski, T. J. (2000a). Dopamine-mediatedstabilization of delay-period activity in a network model of prefrontal cortex.Journal of Neurophysiology, 83(3), 1733–1750.

Durstewitz, D., Seamans, J. K., & Sejnowski, T. J. (2000b). Neurocomputationalmodels of working memory. Nature Neuroscience, 3(Suppl.), S1184–S1191. http://dx.doi.org/10.1038/81460.

Easterbrook, J. A. (1959). The effect of emotion on cue utilization and theorganization of behavior. Psychological Review, 66, 187–201.

Eber, J. (2010). Affective modulation of executive functions: does it happen? (Disserta-tion). Universität Jena.

Egner, T. (2007). Congruency sequence effects and cognitive control. Cognitive,Affective & Behavioral Neuroscience, 7(4), 380–390.

Ellis, H. C., Moore, B. A., Varner, L. J., Ottaway, S. A., et al. (1997). Depressed mood,task organization, cognitive interference, and memory: Irrelevant thoughtspredict recall performance. Journal of Social Behavior & Personality, 12(2),453–470.

Elsworth, J. D., Lawrence, M. S., Roth, R. H., Taylor, J. R., Mailman, R. B., Nichols, D. E.,et al. (1991). D(1) and D(2) dopamine-receptors independently regulatespontaenous blink rate in the vervet monkey. Journal of Pharmacology andExperimental Therapeutics, 259(2), 595–606.

Fiedler, K., Martin, L. L., & Clore, G. L. (2001). Affective states trigger processes ofassimilation and accommodation. Theories of mood and cognition: A user's guide-book (pp. 85–98)Mahwah, NJ, US: Lawrence Erlbaum Associates Publishers.

Fischer, R., Dreisbach, G., & Goschke, T. (2008). Context-sensitive adjustments ofcognitive control: Conflict-adaptation effects are modulated by processingdemands of the ongoing task. Journal of Experimental Psychology: Learning,Memory, and Cognition, 34(3), 712–718. http://dx.doi.org/10.1037/0278-7393.34.3.712.

Floresco, S. B. (2013). Prefrontal dopamine and behavioral flexibility: Shifting froman “inverted-U” toward a family of functions. Frontiers in Neuroscience, 7, 62.http://dx.doi.org/10.3389/fnins.2013.00062.

Frank, M. J., & Fossella, J. A. (2011). Neurogenetics and pharmacology of learning,motivation, and cognition. Neuropsychopharmacology, 36(1), 133–152. http://dx.doi.org/10.1038/npp.2010.96.

Frank, M. J., Loughry, B., & O'Reilly, R. C. (2001). Interactions between frontal cortexand basal ganglia in working memory: A computational model. Cognitive,Affective & Behavioral Neuroscience, 1(2), 137–160.

Frank, M. J., & O'Reilly, R. C. (2006). A mechanistic account of striatal dopaminefunction in human cognition: Psychopharmacological studies with cabergolineand haloperidol. Behavioral Neuroscience, 120(3), 497–517. http://dx.doi.org/10.1037/0735-7044.120.3.497.









http://dx.doi.org/10.3758/s13415-014-0280-0

http://dx.doi.org/10.3758/s13415-014-0280-0

http://dx.doi.org/10.3758/s13415-014-0280-0








http://dx.doi.org/10.1098/rstb.1996.0138







http://dx.doi.org/10.1037/0033-295x.99.1.45

http://dx.doi.org/10.1037/0033-295x.99.1.45

http://dx.doi.org/10.1037/0033-295x.99.1.45

http://dx.doi.org/10.1037/a0018484

http://dx.doi.org/10.1037/a0018484

http://dx.doi.org/10.1037/a0018484













http://dx.doi.org/10.1177/1073858408317009

http://dx.doi.org/10.1177/1073858408317009

http://dx.doi.org/10.1177/1073858408317009

http://dx.doi.org/10.1093/cercor/11.12.1136




http://dx.doi.org/10.1093/brain/124.12.2503



http://dx.doi.org/10.1016/j.biopsych.2011.03.028



http://dx.doi.org/10.1093/brain/awp301










http://dx.doi.org/10.1093/brain/awf052



http://dx.doi.org/10.1002/hbm.20948






















http://dx.doi.org/10.1136/jnnp.64.3.320







http://dx.doi.org/10.1016/s0893-6080(02)00044-8

http://dx.doi.org/10.1016/s0893-6080(02)00044-8

http://dx.doi.org/10.1016/s0893-6080(02)00044-8

http://dx.doi.org/10.1038/nn2077









http://dx.doi.org/10.1037/0278-7393.30.2.343

http://dx.doi.org/10.1037/0278-7393.30.2.343

http://dx.doi.org/10.1037/0278-7393.30.2.343

http://dx.doi.org/10.1037/0735-7044.119.2.483

http://dx.doi.org/10.1037/0735-7044.119.2.483

http://dx.doi.org/10.1037/0735-7044.119.2.483

http://dx.doi.org/10.1037/0735-7044.119.2.483

http://dx.doi.org/10.1016/s0893-6080(02)00049-7

http://dx.doi.org/10.1016/s0893-6080(02)00049-7

http://dx.doi.org/10.1016/s0893-6080(02)00049-7

http://dx.doi.org/10.1016/s0893-6080(02)00049-7








http://dx.doi.org/10.1038/81460

http://dx.doi.org/10.1038/81460

http://dx.doi.org/10.1038/81460

http://dx.doi.org/10.1038/81460




















http://dx.doi.org/10.1037/0278-7393.34.3.712

http://dx.doi.org/10.1037/0278-7393.34.3.712

http://dx.doi.org/10.1037/0278-7393.34.3.712

http://dx.doi.org/10.1037/0278-7393.34.3.712

http://dx.doi.org/10.3389/fnins.2013.00062











http://dx.doi.org/10.1037/0735-7044.120.3.497

http://dx.doi.org/10.1037/0735-7044.120.3.497

http://dx.doi.org/10.1037/0735-7044.120.3.497

http://dx.doi.org/10.1037/0735-7044.120.3.497




Fredrickson, B. L. (2013). Positive emotions broaden and build. In: D. Patricia, &P. Ashby (Eds.), Advances in experimental social psychology, Vol. 47 (pp. 1–53).Academic Press, San Diego, CA.

Fredrickson, B. L., & Branigan, C. (2005). Positive emotions broaden the scope ofattention and thought-action repertoires. Cognition & Emotion, 19(3), 313–332.

Friedman, N. P., Miyake, A., Young, S. E., DeFries, J. C., Corley, R. P., & Hewitt, J. K.(2008). Individual differences in executive functions are almost entirely geneticin origin. Journal of Experimental Psychology: General, 137(2), 201–225. http://dx.doi.org/10.1037/0096-3445.137.2.201.

Friedman, R. S., & Förster, J. (2010). Implicit affective cues and attentional tuning:An integrative review. Psychological Bulletin, 136(5), 875–893. http://dx.doi.org/10.1037/a0020495.

Frober, K., & Dreisbach, G. (2012). How positive affect modulates proactive control:Reduced usage of informative cues under positive affect with low arousal.Frontiers in Psychology, 3, 265. http://dx.doi.org/10.3389/fpsyg.2012.00265.

Frober, K., & Dreisbach, G. (2014). The differential influences of positive affect, randomreward, and performance-contingent reward on cognitive control. Cognitive,Affective, & Behavioral Neuroscience, http://dx.doi.org/10.3758/s13415-014-0259-x.

Gable, P. A., & Harmon-Jones, E. (2011). Attentional consequences of pregoal andpostgoal positive affects. Emotion, 11(6), 1358–1367. http://dx.doi.org/10.1037/a0025611.

Gasper, K. (2003). When necessity is the mother of invention: Mood and problemsolving. Journal of Experimental Social Psychology, 39(3), 248–262. http://dx.doi.org/10.1016/s0022-1031(03)00023-4.

Gasper, K. (2004). Do you see what I see? Affect and visual information processing.Cognition & Emotion, 18(3), 405–421.

Gasper, K., & Clore, G. L. (2002). Attending to the big picture: Mood and global versuslocal processing of visual information. Psychological Science, 13(1), 34–40.

Gazzaley, A., & Nobre, A. C. (2012). Top-down modulation: Bridging selectiveattention and working memory. Trends in Cognitive Sciences, 16(2), 129–135.http://dx.doi.org/10.1016/j.tics.2011.11.014.

Gerdes, A. B., Wieser, M. J., Muhlberger, A., Weyers, P., Alpers, G. W., Plichta, M. M.,et al. (2010). Brain activations to emotional pictures are differentially associatedwith valence and arousal ratings. Frontiers in Human Neuroscience, 4, 175. http://dx.doi.org/10.3389/fnhum.2010.00175.

Gollwitzer, P. M., & Bayer, U. (1999). Deliberative versus implemental mindsets inthe control of action. In: S. Chaiken, & Y. Trope (Eds.), Dual-process theories insocial psychology (pp. 403–422). New York, NY: Guilford Press.

Goschke, T. (1996). Gedächtnis und Emotion: Affektive Bedingungen des Einprä-gens, Behaltens und Vergessens. In: D. Albert, & K.-H. Stapf (Eds.), Enzyklopädieder Psychologie Serie II, Band 4: Gedächtnis (pp. 605–694). Göttingen: Hogrefe.

Goschke, T. (2000). Intentional reconfiguration and involuntary persistence in taskset switching. In: S. Monsell, & J. Driver (Eds.), Attention and performance XVIII:Control of cognitive processes (pp. 331–355). Cambridge, MA: MIT Press.

Goschke, T. (2003). Voluntary action and cognitive control from a cognitive neu-roscience perspective. In: S. Maasen, W. Prinz, & G. Roth (Eds.), Voluntary action:Brains, minds, and sociality (pp. 49–85). New York, NY: Oxford University Press.

Goschke, T. (2013). Volition in action: Intentions, control dilemmas and thedynamic regulation of cognitive intentional control. In: W. Prinz, A. Beisert, &A. Herwig (Eds.), Action science: Foundations of an emerging discipline (pp. 409–434). Cambridge, MA: MIT Press.

Goschke, T. (2014). Dysfunctions of decision-making and cognitive control astransdiagnostic mechanisms of mental disorders: Advances, gaps, and needsin current research. International Journal of Methods in Psychiatric Research, 23(S1), 41–57. http://dx.doi.org/10.1002/mpr.1410.

Goschke, T., & Dreisbach, G. (2008). Conflict-triggered goal shielding: Responseconflicts attenuate background monitoring for prospective memory cues.Psychological Science, 19(1), 25–32.

Gratton, G., Coles, M. G. H., & Donchin, E. (1992). Optimizing the use of information:Strategic control of activation of responses. Journal of Experimental Psychology:General, 121(4), 480–506.

Gray, J. R. (2001). Emotional modulation of cognitive control: Approach-withdrawalstates double-dissociate spatial from verbal two-back task performance. Journalof Experimental Psychology: General, 130(3), 436–452.

Gray, J. R., Braver, T. S., & Raichle, M. E. (2002). Integration of emotion and cognitionin the lateral prefrontal cortex. Proceedings of the National Academy of Sciences,99(6), 4115–4120. http://dx.doi.org/10.1073/pnas.062381899.

Gruber, O., Diekhof, E. K., Kirchenbauer, L., & Goschke, T. (2010). A neural system forevaluating the behavioural relevance of salient events outside the current focusof attention. Brain Research, 1351, 212–221. http://dx.doi.org/10.1016/j.brainres.2010.06.056.

Gruber, O., & Goschke, T. (2004). Executive control emerging from dynamicinteractions between brain systems mediating language, working memoryand attentional processes. Acta Psychologica, 115(2), 105–121.

Gruber, O., Karch, S., Schlueter, E. K., Falkai, P., & Goschke, T. (2006). Neuralmechanisms of advance preparation in task switching. Neuroimage, 31(2),887–895. http://dx.doi.org/10.1016/j.neuroimage.2005.12.043.

Gruber, O., Melcher, T., Diekhof, E. K., Karch, S., Falkai, P., & Goschke, T. (2009). Brainmechanisms associated with background monitoring of the environment forpotentially significant sensory events. Brain and Cognition, 69(3), 559–564.http://dx.doi.org/10.1016/j.bandc.2008.11.008.

Hamann, S. (2012). Mapping discrete and dimensional emotions onto the brain:Controversies and consensus. Trends in Cognitive Sciences, 16(9), 458–466. http://dx.doi.org/10.1016/j.tics.2012.07.006.

Hamann, S., Herman, R. A., Nolan, C. L., & Wallen, K. (2004). Men and women differin amygdala response to visual sexual stimuli. Nature Neuroscience, 7(4),411–416. http://dx.doi.org/10.1038/nn1208.

Harmon-Jones, E., Gable, P. A., & Price, T. F. (2012). The influence of affective statesvarying in motivational intensity on cognitive scope. Frontiers in IntegrativeNeuroscience, 6, 73. http://dx.doi.org/10.3389/fnint.2012.00073.

Harmon-Jones, E., Gable, P. A., & Price, T. F. (2013). Does negative affect alwaysnarrow and positive affect always broaden the mind? Considering the influenceof motivational intensity on cognitive scope. Current Directions in PsychologicalScience, 22(4), 301–307. http://dx.doi.org/10.1177/0963721413481353.

Hasler, G. (2012). Can the neuroeconomics revolution revolutionize psychiatry?Neuroscience and Biobehavioral Reviews, 36(1), 64–78. http://dx.doi.org/10.1016/j.neubiorev.2011.04.011.

Hazy, T. E., Frank, M., & Oreilly, R. (2006). Banishing the homunculus: Makingworking memory work. Neuroscience, 139(1), 105–118. http://dx.doi.org/10.1016/j.neuroscience.2005.04.067.

Hazy, T. E., Frank, M. J., & O'Reilly, R. C. (2007). Towards an executive without ahomunculus: Computational models of the prefrontal cortex/basal gangliasystem. Philosophical Transactions of the Royal Society B: Biological Sciences,362(1485), 1601–1613. http://dx.doi.org/10.1098/rstb.2007.2055.

Herd, S. A., Banich, M. T., & O'Reilly, R. C. (2006). Neural mechanisms of cognitivecontrol: An integrative model of Stroop task performance and fMRI data.Journal of Cognitive Neuroscience, 18(1), 22–32.

Herd, S. A., O'Reilly, R. C., Hazy, T. E., Chatham, C. H., Brant, A. M., & Friedman, N. P.(2014). A neural network model of individual differences in task switchingabil-ities. Neuropsychologia, http://dx.doi.org/10.1016/j.neuropsychologia.2014.04.014.

Hofmann, W., Schmeichel, B. J., & Baddeley, A. D. (2012). Executive functions andself-regulation. Trends in Cognitive Sciences, 16(3), 174–180. http://dx.doi.org/10.1016/j.tics.2012.01.006.

Hommel, B. (1998). Automatic stimulus-response translation in dual-task perfor-mance. Journal of Experimental Psychology: Human Perception and Performance,24(5), 1368–1384.

Hommel, B., Proctor, R. W., & Vu, K. P. (2004). A feature-integration account ofsequential effects in the Simon task. Psychological Research, 68(1), 1–17. http://dx.doi.org/10.1007/s00426-003-0132-y.

Huntsinger, J. R. (2013). Does emotion directly tune the scope of attention? CurrentDirections in Psychological Science, 22(4), 265–270. http://dx.doi.org/10.1177/0963721413480364.

Isen, A. M. (2007). Positive affect, cognitive flexibility, and self-control. In: Y. Shoda,D. Cervone, & G. Downey (Eds.), Persons in context: Building a science of theindividual (pp. 130–147). New York, NY, US: Guilford Press.

Isen, A. M., Dalgleish, T., & Power, M. J. (1999). Positive affect. Handbook of cognitionand emotion (pp. 521–539)New York, NY, US: John Wiley & Sons Ltd.

Isen, A. M., Daubman, K. A., & Nowicki, G. P. (1987). Positive affect facilitatescreative problem-solving. Journal of Personality and Social Psychology, 52(6),1122–1131.

Isen, A. M., Johnson, M. M., Mertz, E., & Robinson, G. F. (1985). The influence ofpositive affect on the unusualness of word associations. Journal of Personalityand Social Psychology, 48(6), 1413–1426.

Ishizu, T., & Zeki, S. (2011). Toward a brain-based theory of beauty. PLoS One, 6(7),e21852. http://dx.doi.org/10.1371/journal.pone.0021852.

Jiang, H., & Xu, B. (2014). Reward enhances backward inhibition in task switching.Journal of Cognitive Psychology, 26(2), 178–186. http://dx.doi.org/10.1080/20445911.2013.878717.

Jimura, K., Locke, H. S., & Braver, T. S. (2010). Prefrontal cortex mediation ofcognitive enhancement in rewarding motivational contexts. Proceedings of theNational Academy of Sciences of the United States of America, 107(19), 8871–-8876. http://dx.doi.org/10.1073/pnas.1002007107.

Jutkiewicz, E. M. (2004). Effects of dopamine D1 ligands on eye blinking inmonkeys: Efficacy, antagonism, and D1/D2 interactions. Journal of Pharmacol-ogy and Experimental Therapeutics, 311(3), 1008–1015. http://dx.doi.org/10.1124/jpet.104.071092.

Kaminer, J., Powers, A. S., Horn, K. G., Hui, C., & Evinger, C. (2011). Characterizing thespontaneous blink generator: An animal model. Journal of Neuroscience, 31(31),11256–11267. http://dx.doi.org/10.1523/JNEUROSCI.6218-10.2011.

Karson, C. N. (1983). Spontaneous eye-blink rates and dopaminergic systems. Brain,106, 643–653. http://dx.doi.org/10.1093/brain/106.3.643 (SEP).

Karson, C. N., Burns, R. S., Lewitt, P. A., Foster, N. L., & Newman, R. P. (1984). Blinkrates and disorders of movement. Neurology, 34(5), 677–678.

Karson, C. N., Lewitt, P. A., Calne, D. B., & Wyatt, R. J. (1982). Blink rates inParkinsonisms. Annals of Neurology, 12(6), 580–583. http://dx.doi.org/10.1002/ana.410120614.

Kerns, J. G., Cohen, J. D., MacDonald, A. W., III, Cho, R. Y., Stenger, V. A., & Carter, C. S.(2004). Anterior cingulate conflict monitoring and adjustments in control.Science, 303(5660), 1023–1026.

Kiesel, A., Steinhauser, M., Wendt, M., Falkenstein, M., Jost, K., Philipp, A. M., et al.(2010). Control and interference in task switching – A review. PsychologicalBulletin, 136(5), 849–874. http://dx.doi.org/10.1037/a0019842.

Klanker, M., Feenstra, M., & Denys, D. (2013). Dopaminergic control of cognitiveflexibility in humans and animals. Frontiers in Neuroscience, 7, 201. http://dx.doi.org/10.3389/fnins.2013.00201.

Kleinsorge, T., & Rinkenauer, G. (2012). Effects of monetary incentives on taskswitching. Experimental Psychology, 59(4), 216–226. http://dx.doi.org/10.1027/1618-3169/a000146.









http://dx.doi.org/10.1037/0096-3445.137.2.201

http://dx.doi.org/10.1037/0096-3445.137.2.201

http://dx.doi.org/10.1037/0096-3445.137.2.201

http://dx.doi.org/10.1037/0096-3445.137.2.201

http://dx.doi.org/10.1037/a0020495

http://dx.doi.org/10.1037/a0020495

http://dx.doi.org/10.1037/a0020495

http://dx.doi.org/10.1037/a0020495




http://dx.doi.org/10.3758/s13415-014-0259-x

http://dx.doi.org/10.3758/s13415-014-0259-x

http://dx.doi.org/10.3758/s13415-014-0259-x

http://dx.doi.org/10.1037/a0025611

http://dx.doi.org/10.1037/a0025611

http://dx.doi.org/10.1037/a0025611

http://dx.doi.org/10.1037/a0025611

http://dx.doi.org/10.1016/s0022-1031(03)00023-4

http://dx.doi.org/10.1016/s0022-1031(03)00023-4

http://dx.doi.org/10.1016/s0022-1031(03)00023-4

http://dx.doi.org/10.1016/s0022-1031(03)00023-4





























http://dx.doi.org/10.1002/mpr.1410















http://dx.doi.org/10.1016/j.brainres.2010.06.056







http://dx.doi.org/10.1016/j.neuroimage.2005.12.043



http://dx.doi.org/10.1016/j.bandc.2008.11.008










http://dx.doi.org/10.3389/fnint.2012.00073



http://dx.doi.org/10.1177/0963721413481353

http://dx.doi.org/10.1177/0963721413481353

http://dx.doi.org/10.1177/0963721413481353

























http://dx.doi.org/10.1007/s00426-003-0132-y

http://dx.doi.org/10.1007/s00426-003-0132-y

http://dx.doi.org/10.1007/s00426-003-0132-y

http://dx.doi.org/10.1007/s00426-003-0132-y

http://dx.doi.org/10.1177/0963721413480364

http://dx.doi.org/10.1177/0963721413480364

http://dx.doi.org/10.1177/0963721413480364

http://dx.doi.org/10.1177/0963721413480364
















http://dx.doi.org/10.1080/20445911.2013.878717

http://dx.doi.org/10.1080/20445911.2013.878717

http://dx.doi.org/10.1080/20445911.2013.878717

http://dx.doi.org/10.1080/20445911.2013.878717




http://dx.doi.org/10.1124/jpet.104.071092




http://dx.doi.org/10.1523/JNEUROSCI.6218-10.2011








http://dx.doi.org/10.1002/ana.410120614







http://dx.doi.org/10.1037/a0019842

http://dx.doi.org/10.1037/a0019842

http://dx.doi.org/10.1037/a0019842





http://dx.doi.org/10.1027/1618-3169/a000146

http://dx.doi.org/10.1027/1618-3169/a000146

http://dx.doi.org/10.1027/1618-3169/a000146

http://dx.doi.org/10.1027/1618-3169/a000146




Kleven, M. S., & Koek, W. (1996). Differential effects of direct and indirect dopamineagonists on eye blink rate in cynomolgus monkeys. Journal of Pharmacology andExperimental Therapeutics, 279(3), 1211–1219.

Knutson, B., & Greer, S. M. (2008). Anticipatory affect: Neural correlates andconsequences for choice. Philosophical Transactions of the Royal Society ofLondon. Series B: Biological Sciences, 363(1511), 3771–3786. http://dx.doi.org/10.1098/rstb.2008.0155.

Knutson, B., Katovich, K., & Suri, G. (2014). Inferring affect from fMRI data. Trends inCognitive Sciences, 18(8), 422–428. http://dx.doi.org/10.1016/j.tics.2014.04.006.

Koch, I., Gade, M., Schuch, S., & Philipp, A. M. (2010). The role of inhibition in taskswitching: A review. Psychonomic Bulletin & Review, 17(1), 1–14.

Kouneiher, F., Charron, S., & Koechlin, E. (2009). Motivation and cognitive control inthe human prefrontal cortex. Nature Neuroscience, 12(7), 939–945. http://dx.doi.org/10.1038/nn.2321.

Kuhl, J. (1985). Volitional mediators of cognitive-behavior consistency: Self-regulatory processes and actions versus state orientation. In: J. Kuhl, &J. Beckmann (Eds.), Action control: From cognition to behavior (pp. 101–128).Berlin, Heidelberg, New York, Tokyo: Springer.

Kuhl, J. (2000). A functional-design approach to motivation and self-regulation: Thedynamics of personality systems and interactions. In: M. Boekaerts, P.R. Pintrich, & M. Zeidner (Eds.), Handbook of self-regulation (pp. 111–169). SanDiego, CA: Academic Press.

Kuhl, J., & Goschke, T. (1994). A theory of action control: Mental subsystems, modesof control, and volitional conflict-resolution strategies. In: J. Kuhl, &J. Beckmann (Eds.), Volition and personality: Action versus state orientation.Göttingen/Toronto: Hogrefe.

Kuhl, J., & Kazen, M. (1999). Volitional facilitation of difficult intentions: Jointactivation of intention memory and positive affect removes Stroop interference.Journal of Experimental Psychology: General, 128(3), 382–399.

Lawrence, M. S., & Redmond, D. E. (1991). MPTP lesions and dopaminergic drugsalter eye blink rate in African-green monkeys. Pharmacology Biochemistry andBehavior, 38(4), 869–874. http://dx.doi.org/10.1016/0091-3057(91)90255-z.

Lawrence, M. S., Redmond, D. E., Elsworth, J. D., Taylor, J. R., & Roth, R. H. (1991). TheD(1) receptor antagonist, SCH-23390, induces signs of Parkinsonism in African-green monkeys. Life Sciences, 49(25), PL229–PL234.

Leber, A. B., Turk-Browne, N. B., & Chun, M. M. (2008). Neural predictors ofmoment-to-moment fluctuations in cognitive flexibility. Proceedings of theNational Academy of Sciences of the United States of America, 105(36), 13592–-13597. http://dx.doi.org/10.1073/pnas.0805423105.

LeDoux, J. (2012). Rethinking the emotional brain. Neuron, 73(4), 653–676. http://dx.doi.org/10.1016/j.neuron.2012.02.004.

Lee, D., Seo, H., & Jung, M. W. (2012). Neural basis of reinforcement learning anddecision making. Annual Review of Neuroscience, 35, 287–308. http://dx.doi.org/10.1146/annurev-neuro-062111-150512.

Leknes, S., & Tracey, I. (2008). A common neurobiology for pain and pleasure.Nature Reviews Neuroscience, 9(4), 314–320.

Lindquist, K. A., Wager, T. D., Kober, H., Bliss-Moreau, E., & Barrett, L. F. (2012). Thebrain basis of emotion: A meta-analytic review. Behavioral and Brain Sciences,35(3), 121–143. http://dx.doi.org/10.1017/S0140525�11000446.

Liu, Y., & Wang, Z. (2014). Positive affect and cognitive control: Approach-motivationintensity influences the balance between cognitive flexibility and stability. Psycho-logical Science, 25(5), 1116–1123. http://dx.doi.org/10.1177/0956797614525213.

Locke, H. S., & Braver, T. S. (2008). Motivational influences on cognitive control:Behavior, brain activation, and individual differences. Cognitive, Affective, &Behavioral Neuroscience, 8(1), 99–112. http://dx.doi.org/10.3758/cabn.8.1.99.

Logan, G. D., & Schulkind, M. D. (2000). Parallel memory retrieval in dual-tasksituations: I. Semantic memory. Journal of Experimental Psychology: HumanPerception & Performance, 26(3), 1072–1090.

Mansouri, F. A., Tanaka, K., & Buckley, M. J. (2009). Conflict-induced behaviouraladjustment: A clue to the executive functions of the prefrontal cortex. NatureReviews Neuroscience, 10(2), 141–152. http://dx.doi.org/10.1038/nrn2538.

Marien, H., Aarts, H., & Custers, R. (2012). Being flexible or rigid in goal-directedbehavior: When positive affect implicitly motivates the pursuit of goals ormeans. Journal of Experimental Social Psychology, 48(1), 277–283. http://dx.doi.org/10.1016/j.jesp.2011.08.013.

Mars, R. B., Sallet, J., Rushworth, M., & Yeung, N. (2011). Neural basis of motivationaland cognitive control. Cambrdige, MA: MIT Press.

Martin, E. A., & Kerns, J. G. (2011). The influence of positive mood on differentaspects of cognitive control. Cognition & Emotion, 25(2), 265–279. http://dx.doi.org/10.1080/02699931.2010.491652.

Mayr, U., Awh, E., & Laurey, P. (2003). Conflict adaptation effects in the absence ofexecutive control. Nature Neuroscience, 6(5), 450–452.

Mayr, U., & Keele, S. W. (2000). Changing internal constraints on action: The role ofbackward inhibition. Journal of Experimental Psychology: General, 129(1), 4–26.http://dx.doi.org/10.1037//0096-3445.129.1.4.

McNab, F., & Klingberg, T. (2008). Prefrontal cortex and basal ganglia control accessto working memory. Nature Neuroscience, 11(1), 103–107. http://dx.doi.org/10.1038/nn2024.

Mehta, M. A., Manes, F. F., Magnolfi, G., Sahakian, B. J., & Robbins, T. W. (2004).Impaired set-shifting and dissociable effects on tests of spatial workingmemory following the dopamine D2 receptor antagonist sulpiride in humanvolunteers. Psychopharmacology (Berlin), 176(3–4), 331–342. http://dx.doi.org/10.1007/s00213-004-1899-2.

Meiran, N. (2010). Task switching: Mechanisms underlying rigid vs. flexible self-control. In: R. Hassin, K. Ochsner, & Y. Trope (Eds.), Self control in society, mind,and brain (pp. 202–220). Oxford: Oxford University Press.

Meiran, N., Chorev, Z., & Sapir, A. (2000). Component processes in task switching.Cognitive Psychology, 41(3), 211–253. http://dx.doi.org/10.1006/cogp.2000.0736.

Meiran, N., Hsieh, S., & Dimov, E. (2010). Resolving task rule incongruence duringtask switching by competitor rule suppression. Journal of Experimental Psychol-ogy: Learning Memory and Cognition, 36(4), 992–1002. http://dx.doi.org/10.1037/a0019761.

Menon, V. (2011). Large-scale brain networks and psychopathology: A unifyingtriple network model. Trends in Cognitive Sciences, 15(10), 483–506. http://dx.doi.org/10.1016/j.tics.2011.08.003.

Miller, E. K., & Cohen, J. D. (2001). An integrative theory of prefrontal cortexfunction. Annual Review of Neuroscience, 24, 167–202.

Mitchell, R. L. C., & Phillips, L. H. (2007). The psychological, neurochemical andfunctional neuroanatomical mediators of the effects of positive and negativemood on executive functions. Neuropsychologia, 45(4), 617–629.

Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D.(2000). The unity and diversity of executive functions and their contributions tocomplex ‘frontal lobe’ tasks: A latent variable analysis. Cognitive Psychology, 41(1), 49–100. http://dx.doi.org/10.1006/cogp.1999.0734.

Monsell, S. (2003). Task switching. Trends in Cognitive Sciences, 7(3), 134–140(S1364661303000287).

Montag, C., Reuter, M., & Axmacher, N. (2011). How one's favorite song activates thereward circuitry of the brain: Personality matters! Behavioural Brain Research,225(2), 511–514. http://dx.doi.org/10.1016/j.bbr.2011.08.012.

Montague, P. R., Hyman, S. E., & Cohen, J. D. (2004). Computational roles fordopamine in behavioralbehavioural control. Nature, 431(7010), 760–767. http://dx.doi.org/10.1038/nature03015.

Mueller, S. C. (2011). The influence of emotion on cognitive control: Relevance fordevelopment and adolescent psychopathology. Frontiers in Psychology, 2, 327.http://dx.doi.org/10.3389/fpsyg.2011.00327.

Müller, J., Dreisbach, G., Brocke, B., Lesch, K. P., Strobel, A., & Goschke, T. (2007).Dopamine and cognitive control: The influence of spontaneous eyeblink rate,DRD4 exon III polymorphism and gender on flexibility in set-shifting. BrainResearch, 1131(1), 155–162. http://dx.doi.org/10.1016/j.brainres.2006.11.002.

Müller, J., Dreisbach, G., Goschke, T., Hensch, T., Lesch, K.-P., & Brocke, B. (2007).Dopamine and cognitive control: The prospect of monetary gains influences thebalance between flexibility and stability in a set-shifting paradigm. EuropeanJournal of Neuroscience, 26(12), 3661–3668. http://dx.doi.org/10.1111/j.1460-9568.2007.05790.x.

Nee, D. E., & Brown, J. W. (2013). Dissociable frontal-striatal and frontal-parietalnetworks involved in updating hierarchical contexts in working memory.Cerebral Cortex, 23(9), 2146–2158. http://dx.doi.org/10.1093/cercor/bhs194.

Nieuwenhuis, S., & Monsell, S. (2002). Residual costs in task switching: Testing thefailure-to-engage hypothesis. Psychonomic Bulletin & Review, 9(1), 86–92.

O'Reilly, R. C. (2006). Biologically based computational models of high-levelcognition. Science, 314(5796), 91–94. http://dx.doi.org/10.1126/science.1127242.

O'Reilly, R. C., & Frank, M. J. (2006). Making working memory work: A computa-tional model of learning in the prefrontal cortex and basal ganglia. NeuralComputation, 18(2), 283–328. http://dx.doi.org/10.1162/089976606775093909.

O'Reilly, R. C., Herd, S. A., & Pauli, W. M. (2010). Computational models of cognitivecontrol. Current Opinion in Neurobiology, 20(2), 257–261. http://dx.doi.org/10.1016/j.conb.2010.01.008.

Oaksford, M., Morris, F., Grainger, B., & Williams, J. M. G. (1996). Mood, reasoning, andcentral executive processes. Journal of Experimental Psychology: Learning, Memory,and Cognition, 22(2), 476–492. http://dx.doi.org/10.1037/0278-7393.22.2.476.

Pessoa, L. (2008). On the relationship between emotion and cognition. NatureReviews Neuroscience, 9, 148–158.

Pessoa, L. (2009). How do emotion and motivation direct executive control? Trends inCognitive Sciences, 13(4), 160–166. http://dx.doi.org/10.1016/j.tics.2009.01.006.

Phillips, L. H., Bull, R., Adams, E., & Fraser, L. (2002). Positive mood and executivefunction: Evidence from Stroop and fluency tasks. Emotion, 2(1), 12–22. http://dx.doi.org/10.1037//1528-3542.2.1.12.

Pochon, J. B. (2002). The neural system that bridges reward and cognition inhumans: An fMRI study. Proceedings of the National Academy of Sciences, 99(8),5669–5674. http://dx.doi.org/10.1073/pnas.082111099.

Ravizza, S. M., & Carter, C. S. (2008). Shifting set about task switching: Behavioraland neural evidence for distinct forms of cognitive flexibility. Neuropsychologia,46(12), 2924–2935. http://dx.doi.org/10.1016/j.neuropsychologia.2008.06.006.

Ray, R. D., & Zald, D. H. (2012). Anatomical insights into the interaction of emotionand cognition in the prefrontal cortex. Neuroscience and Biobehavioral Reviews,36(1), 479–501. http://dx.doi.org/10.1016/j.neubiorev.2011.08.005.

Redgrave, P., Prescott, T. J., & Gurney, K. (1999). The basal ganglia: A vertebratesolution to the selection problem? Neuroscience, 89, 1009–1023.

Robbins, T. W., & Arnsten, A. F. T. (2009). The neuropsychopharmacology of fronto-executive function: Monoaminergic modulation. Annual Review of Neuroscience,32, 267–287. http://dx.doi.org/10.1146/annurev.neuro.051508.135535.

Robbins, T. W., & Roberts, A. C. (2007). Differential regulation of fronto-executivefunction by the monoamines and acetylcholine. Cerebral Cortex, 17(Suppl. 1),S151–S160. http://dx.doi.org/10.1093/cercor/bhm066.

Rogers, R. D. (2011). The roles of dopamine and serotonin in decision making:Evidence from pharmacological experiments in humans. Neuropsychopharma-cology, 36(1), 114–132. http://dx.doi.org/10.1038/npp.2010.165.

Rogers, R. D., & Monsell, S. (1995). Costs of a predictable switch between simplecognitive tasks. Journal of Experimental Psychology: General, 124(2), 207–231.

Rolls, E. T. (2010). Attractor networks. WIREs Cognitive Science, 1, 119–134.Rowe, G., Hirsh, J. B., & Anderson, A. K. (2007). Positive affect increases the breadth

of attentional selection. Proceedings of the National Academy of Sciences of the
















http://dx.doi.org/10.1038/nn.2321



















http://dx.doi.org/10.1016/0091-3057(91)90255-z

http://dx.doi.org/10.1016/0091-3057(91)90255-z

http://dx.doi.org/10.1016/0091-3057(91)90255-z







http://dx.doi.org/10.1016/j.neuron.2012.02.004




http://dx.doi.org/10.1146/annurev-neuro-062111-150512






http://dx.doi.org/10.1017/S0140525×11000446

http://dx.doi.org/10.1017/S0140525×11000446

http://dx.doi.org/10.1017/S0140525×11000446

http://dx.doi.org/10.1017/S0140525×11000446

http://dx.doi.org/10.1177/0956797614525213

http://dx.doi.org/10.1177/0956797614525213

http://dx.doi.org/10.1177/0956797614525213

http://dx.doi.org/10.3758/cabn.8.1.99










http://dx.doi.org/10.1016/j.jesp.2011.08.013






http://dx.doi.org/10.1080/02699931.2010.491652

http://dx.doi.org/10.1080/02699931.2010.491652

http://dx.doi.org/10.1080/02699931.2010.491652

http://dx.doi.org/10.1080/02699931.2010.491652



http://dx.doi.org/10.1037//0096-3445.129.1.4

http://dx.doi.org/10.1037//0096-3445.129.1.4

http://dx.doi.org/10.1037//0096-3445.129.1.4





http://dx.doi.org/10.1007/s00213-004-1899-2

http://dx.doi.org/10.1007/s00213-004-1899-2

http://dx.doi.org/10.1007/s00213-004-1899-2

http://dx.doi.org/10.1007/s00213-004-1899-2




http://dx.doi.org/10.1006/cogp.2000.0736



http://dx.doi.org/10.1037/a0019761

http://dx.doi.org/10.1037/a0019761

http://dx.doi.org/10.1037/a0019761

http://dx.doi.org/10.1037/a0019761















http://dx.doi.org/10.1016/j.bbr.2011.08.012



http://dx.doi.org/10.1038/nature03015










http://dx.doi.org/10.1111/j.1460-9568.2007.05790.x

http://dx.doi.org/10.1111/j.1460-9568.2007.05790.x

http://dx.doi.org/10.1111/j.1460-9568.2007.05790.x

http://dx.doi.org/10.1111/j.1460-9568.2007.05790.x

http://dx.doi.org/10.1093/cercor/bhs194






http://dx.doi.org/10.1126/science.1127242



http://dx.doi.org/10.1162/089976606775093909

http://dx.doi.org/10.1162/089976606775093909

http://dx.doi.org/10.1162/089976606775093909





http://dx.doi.org/10.1037/0278-7393.22.2.476

http://dx.doi.org/10.1037/0278-7393.22.2.476

http://dx.doi.org/10.1037/0278-7393.22.2.476






http://dx.doi.org/10.1037//1528-3542.2.1.12

http://dx.doi.org/10.1037//1528-3542.2.1.12

http://dx.doi.org/10.1037//1528-3542.2.1.12

http://dx.doi.org/10.1037//1528-3542.2.1.12












http://dx.doi.org/10.1146/annurev.neuro.051508.135535



http://dx.doi.org/10.1093/cercor/bhm066












United States of America, 104(1), 383–388. http://dx.doi.org/10.1073/pnas.0605198104.

Ruge, H., Jamadar, S., Zimmermann, U., & Karayanidis, F. (2013). The many faces ofpreparatory control in task switching: Reviewing a decade of fMRI research.Human Brain Mapping, 34(1), 12–35. http://dx.doi.org/10.1002/hbm.21420.

Sabatinelli, D., Bradley, M. M., Lang, P. J., Costa, V. D., & Versace, F. (2007). Pleasurerather than salience activates human nucleus accumbens and medial prefrontalcortex. Journal of Neurophysiology, 98(3), 1374–1379. http://dx.doi.org/10.1152/jn.00230.2007.

Sakai, K. (2008). Task set and prefrontal cortex. Annual Review of Neuroscience, 31,219–245. http://dx.doi.org/10.1146/annurev.neuro.31.060407.125642.

Salamone, J. D., & Correa, M. (2012). The mysterious motivational functionsof mesolimbic dopamine. Neuron, 76(3), 470–485. http://dx.doi.org/10.1016/j.neuron.2012.10.021.

Salimpoor, V. N., Benovoy, M., Larcher, K., Dagher, A., & Zatorre, R. J. (2011).Anatomically distinct dopamine release during anticipation and experience ofpeak emotion to music. Nature Neuroscience, 14(2), 257–262. http://dx.doi.org/10.1038/nn.2726.

Samanez-Larkin, G. R., Buckholtz, J. W., Cowan, R. L., Woodward, N. D., Li, R., Ansari, M.S., et al. (2013). A thalamocorticostriatal dopamine network for psychostimulant-enhanced human cognitive flexibility. Biological Psychiatry, 74(2), 99–105. http://dx.doi.org/10.1016/j.biopsych.2012.10.032.

Savine, A. C., Beck, S. M., Edwards, B. G., Chiew, K. S., & Braver, T. S (2010). Enhancementof cognitive control by approach and avoidance motivational states. Cognition &Emotion, 24(2), 338–356. http://dx.doi.org/10.1080/02699930903381564.

Scherbaum, S., Dshemuchadse, M., & Kalis, A. (2008). Making decisions with acontinuous mind. Cognitive, Affective, & Behavioral Neuroscience, 8(4), 454–474.http://dx.doi.org/10.3758/cabn.8.4.454.

Scherbaum, S., Dshemuchadse, M., Ruge, H., & Goschke, T. (2012). Dynamic goalstates: Adjusting cognitive control without conflict monitoring. Neuroimage, 63(1), 126–136. http://dx.doi.org/10.1016/j.neuroimage.2012.06.021.

Scherbaum, S., Fischer, R., Dshemuchadse, M., & Goschke, T. (2011). The dynamics ofcognitive control: Evidence for within trial conflict adaptation from frequencytagged EEG. Psychophysiology, 48(5), 591–600. http://dx.doi.org/10.1111/j.1469-8986.2010.01137.x.

Schlochtermeier, L. H., Kuchinke, L., Pehrs, C., Urton, K., Kappelhoff, H., & Jacobs, A.M. (2013). Emotional picture and word processing: An FMRI study on effects ofstimulus complexity. PLoS One, 8(2), e55619. http://dx.doi.org/10.1371/journal.pone.0055619.

Schmitz, T. W., De Rosa, E., & Anderson, A. K. (2009). Opposing influences ofaffective state valence on visual cortical encoding. Journal of Neuroscience, 29(22), 7199–7207. http://dx.doi.org/10.1523/jneurosci.5387-08.2009.

Schultz, W. (2013). Updating dopamine reward signals. Current Opinion in Neuro-biology, 23(2), 229–238. http://dx.doi.org/10.1016/j.conb.2012.11.012.

Schwarz, N., & Clore, G. L. (1983). Mood, misattribution, and judgments of well-being: Informative and directive functions of affective states. Journal ofPersonality and Social Psychology, 45, 513–523.

Seamans, J., & Yang, C. (2004). The principal features and mechanisms of dopaminemodulation in the prefrontal cortex. Progress in Neurobiology, 74(1), 1–58. http://dx.doi.org/10.1016/j.pneurobio.2004.05.006.

Seibert, P., & Ellis, H. (1991). Irrelevant thoughts, emotional mood states, andcognitive task performance. Memory & Cognition, 19(5), 507–513. http://dx.doi.org/10.3758/BF03199574.

Sescousse, G., Caldu, X., Segura, B., & Dreher, J. C. (2013). Processing of primary andsecondary rewards: A quantitative meta-analysis and review of human func-tional neuroimaging studies. Neuroscience & Biobehavioral Reviews, 37(4),681–696. http://dx.doi.org/10.1016/j.neubiorev.2013.02.002.

Shen, Y. J., & Chun, M. M. (2011). Increases in rewards promote flexible behavior.Attention Perception & Psychophysics, 73(3), 938–952. http://dx.doi.org/10.3758/s13414-010-0065-7.

Spies, K., Hesse, F. W., & Hummitzsch, C. (1996). Mood and capacity in Baddeley'smodel of human memory. Zeitschrift für Psychologie, 204(4), 367–381.

Stelzel, C., Basten, U., Montag, C., Reuter, M., & Fiebach, C. J. (2010). Frontostriatalinvolvement in task switching depends on genetic differences in d2 receptordensity. The Journal of Neuroscience: The Official Journal of the Society forNeuroscience, 30(42), 14205–14212. http://dx.doi.org/10.1523/JNEUROSCI.1062-10.2010.

Stürmer, B., Leuthold, H., Soetens, E., Schröter, H., & Sommer, W. (2002). Controlover location-based response activation in the Simon task: Behavioral andelectrophysiological evidence. Journal of Experimental Psychology: Human Per-ception and Performance, 28(6), 1345–1363. http://dx.doi.org/10.1037//0096-1523.28.6.1345.

Sturmer, B., Nigbur, R., Schacht, A., & Sommer, W. (2011). Reward and punishmenteffects on error processing and conflict control. Frontiers in Psychology, 2, 335.http://dx.doi.org/10.3389/fpsyg.2011.00335.

Taylor, J. R., Lawrence, M. S., Redmond, D. E., Elsworth, J. D., Roth, R. H., Nichols, D.E., et al. (1991). Dihydrexidine, a full dopamine-D1 agonist, reduces MPTP-induced parkinsonism in monkeys. European Journal of Pharmacology, 199(3),389–391. http://dx.doi.org/10.1016/0014-2999(91)90508-n.

Taylor, S. F., Welsh, R. C., Wager, T. D., Phan, K. L., Fitzgerald, K. D., & Gehring, W. J. (2004).A functional neuroimaging study of motivation and executive function. Neuroimage,21(3), 1045–1054. http://dx.doi.org/10.1016/j.neuroimage.2003.10.032.

Tharp, I. J., & Pickering, A. D. (2011). Individual differences in cognitive-flexibility:The influence of spontaneous eyeblink rate, trait psychoticism and workingmemory on attentional set-shifting. Brain and Cognition, 75(2), 119–125. http://dx.doi.org/10.1016/j.bandc.2010.10.010.

Topolinski, S., & Deutsch, R. (2013). Phasic affective modulation of semanticpriming. Journal of Experimental Psychology: Learning Memory and Cognition,39(2), 414–436. http://dx.doi.org/10.1037/a0028879.

van der Post, J., de Waal, P. P., de Kam, M. L., Cohen, A. F., & van Gerven, J. M. A.(2004). No evidence of the usefulness of eye blinking as a marker for centraldopaminergic activity. Journal of Psychopharmacology, 18(1), 109–114. http://dx.doi.org/10.1177/0269881104042832.

van Holstein, M., Aarts, E., Schaaf, M., Geurts, D. M., Verkes, R., Franke, B., et al.(2011). Human cognitive flexibility depends on dopamine D2 receptor signal-ing. Psychopharmacology, 218(3), 567–578. http://dx.doi.org/10.1007/s00213-011-2340-2.

van Schouwenburg, M. R., Aarts, E., & Cools, R. (2010). Dopaminergic modulation ofcognitive control: Distinct roles for the prefrontal cortex and basal ganglia.Current Pharmacological Design, 16(18), 2026–2032.

van Schouwenburg, M. R., den Ouden, H. E. M., & Cools, R. (2010). The human basalganglia modulate frontal-posterior connectivity during attention shifting.Journal of Neuroscience, 30(29), 9910–9918. http://dx.doi.org/10.1523/jneur-osci.1111-10.2010.

van Steenbergen, H., Band, G. P. H., & Hommel, B. (2009). Reward counteractsconflict adaptation: Evidence for a role of affect in executive control. Psycho-logical Science, 20(12), 1473–1477.

van Steenbergen, H., Band, G. P. H., & Hommel, B. (2010). In the mood foradaptation: How affect regulates conflict-driven control. Psychological Science,21(11), 1629–1634. http://dx.doi.org/10.1177/0956797610385951.

van Wouwe, N. C., Band, G. P. H., & Ridderinkhof, K. R. (2011). Positive affectmodulates flexibility and evaluative control. Journal of Cognitive Neuroscience,23(3), 524–539.

Vandierendonck, A., Liefooghe, B., & Verbruggen, F. (2010). Task switching: Inter-play of reconfiguration and interference control. Psychological Bulletin, 136(4),601–626. http://dx.doi.org/10.1037/a0019791.

Verbruggen, F., Liefooghe, B., Vandierendonck, A., & Demanet, J. (2007). Short cuepresentations encourage advance task preparation: A recipe to diminish theresidual switch cost. Journal of Experimental Psychology: Learning, memory, &Cognition, 33(2), 342–356. http://dx.doi.org/10.1037/0278-7393.33.2.342.

Verguts, T., & Notebaert, W. (2009). Adaptation by binding: A learning account ofcognitive control. Trends in Cognitive Sciences, 13(6), 252–257. http://dx.doi.org/10.1016/j.tics.2009.02.007.

Vytal, K., & Hamann, S. (2010). Neuroimaging support for discrete neural correlatesof basic emotions: A voxel-based meta-analysis. Journal of Cognitive Neu-roscience, 22(12), 2864–2885.

Wiswede, D., Münte, T. F., Goschke, T., & Rüsseler, J. (2009). Modulation of the error-related negativity by induction of short-term negative affect. Neuropsychologia,47(1), 83–90. http://dx.doi.org/10.1016/j.neuropsychologia.2008.08.016.

Wiswede, D., Münte, T. F., Kramer, U. M., & Rüsseler, J. (2009). Embodied emotionmodulates neural signature of performance monitoring. PLoS One, 4(6), e5754.http://dx.doi.org/10.1371/journal.pone.0005754.

Yang, H., Yang, S., & Isen, A. M. (2013). Positive affect improves working memory:Implications for controlled cognitive processing. Cognition & Emotion, 27(3),474–482. http://dx.doi.org/10.1080/02699931.2012.713325.

Yeung, N., Nystrom, L. E., Aronson, J. A., & Cohen, J. D. (2006). Between-taskcompetition and cognitive control in task switching. Journal of Neuroscience, 26(5), 1429–1438.

Zedelius, C. M., Veling, H., Custers, R., Bijleveld, E., Chiew, K. S., & Aarts, H. (2014). Anew perspective on human reward research: How consciously and uncon-sciously perceived reward information influences performance. Cognitive, Affec-tive, & Behavioral Neuroscience, 14(2), 493–508. http://dx.doi.org/10.3758/s13415-013-0241-z.

Zwosta, K., Hommel, B., Goschke, T., & Fischer, R. (2013). Mood states determine thedegree of task shielding in dual-task performance. Cognition & Emotion, 27(6),1142–1152. http://dx.doi.org/10.1080/02699931.2013.772047.










http://dx.doi.org/10.1152/jn.00230.2007

http://dx.doi.org/10.1152/jn.00230.2007

http://dx.doi.org/10.1152/jn.00230.2007

http://dx.doi.org/10.1152/jn.00230.2007
















http://dx.doi.org/10.1080/02699930903381564

http://dx.doi.org/10.1080/02699930903381564

http://dx.doi.org/10.1080/02699930903381564







http://dx.doi.org/10.1111/j.1469-8986.2010.01137.x

http://dx.doi.org/10.1111/j.1469-8986.2010.01137.x

http://dx.doi.org/10.1111/j.1469-8986.2010.01137.x

http://dx.doi.org/10.1111/j.1469-8986.2010.01137.x














http://dx.doi.org/10.1016/j.pneurobio.2004.05.006




http://dx.doi.org/10.3758/BF03199574

http://dx.doi.org/10.3758/BF03199574

http://dx.doi.org/10.3758/BF03199574

http://dx.doi.org/10.3758/BF03199574




http://dx.doi.org/10.3758/s13414-010-0065-7

http://dx.doi.org/10.3758/s13414-010-0065-7

http://dx.doi.org/10.3758/s13414-010-0065-7

http://dx.doi.org/10.3758/s13414-010-0065-7







http://dx.doi.org/10.1037//0096-1523.28.6.1345

http://dx.doi.org/10.1037//0096-1523.28.6.1345

http://dx.doi.org/10.1037//0096-1523.28.6.1345

http://dx.doi.org/10.1037//0096-1523.28.6.1345




http://dx.doi.org/10.1016/0014-2999(91)90508-n

http://dx.doi.org/10.1016/0014-2999(91)90508-n

http://dx.doi.org/10.1016/0014-2999(91)90508-n








http://dx.doi.org/10.1037/a0028879

http://dx.doi.org/10.1037/a0028879

http://dx.doi.org/10.1037/a0028879

http://dx.doi.org/10.1177/0269881104042832

http://dx.doi.org/10.1177/0269881104042832

http://dx.doi.org/10.1177/0269881104042832

http://dx.doi.org/10.1177/0269881104042832

http://dx.doi.org/10.1007/s00213-011-2340-2

http://dx.doi.org/10.1007/s00213-011-2340-2

http://dx.doi.org/10.1007/s00213-011-2340-2

http://dx.doi.org/10.1007/s00213-011-2340-2











http://dx.doi.org/10.1177/0956797610385951

http://dx.doi.org/10.1177/0956797610385951

http://dx.doi.org/10.1177/0956797610385951




http://dx.doi.org/10.1037/a0019791

http://dx.doi.org/10.1037/a0019791

http://dx.doi.org/10.1037/a0019791

http://dx.doi.org/10.1037/0278-7393.33.2.342

http://dx.doi.org/10.1037/0278-7393.33.2.342

http://dx.doi.org/10.1037/0278-7393.33.2.342














http://dx.doi.org/10.1080/02699931.2012.713325

http://dx.doi.org/10.1080/02699931.2012.713325

http://dx.doi.org/10.1080/02699931.2012.713325




http://dx.doi.org/10.3758/s13415-013-0241-z

http://dx.doi.org/10.3758/s13415-013-0241-z

http://dx.doi.org/10.3758/s13415-013-0241-z

http://dx.doi.org/10.3758/s13415-013-0241-z

http://dx.doi.org/10.1080/02699931.2013.772047

http://dx.doi.org/10.1080/02699931.2013.772047

http://dx.doi.org/10.1080/02699931.2013.772047




Documents

Emotional modulation of control dilemmas: The role of positive affect, reward, and dopamine in cognitive stability and flexibility