Review Paper

Atypical scanpaths in schizophrenia: Evidenceof a trait- or state-dependent phenomenon?

Sara A. Beedie, PhD; David M. St. Clair, MD, PhD; Philip J. Benson, PhD

Beedie Benson — School of Psychology, College of Life Sciences and Medicine, University of Aberdeen, King's College;St. Clair — School of Medicine and Dentistry, Division of Applied Medicine, University of Aberdeen, Clinical Research Centre,Royal Cornhill Hospital, Aberdeen, Scotland

The development of trait markers of schizophrenia would represent an important advance in understanding the genetic architecture ofthe disease. To date, no candidate markers have satisfied all of the trait marker criteria, and many are not specific to the schizophreniaspectrum. Abnormalities in visual scanpaths are frequently reported in patients with schizophrenia and are emerging as a novel candi-date for a schizophrenia biomarker. Here we review the suitability of scanpath measures as a target for trait marker research in schizo-phrenia. Papers reporting scanpath patterns in patients with schizophrenia were identified by PubMed and Google Scholar searches andby scanning reference lists in relevant articles. Search terms included "schizophrenia," "psychosis," "scanpath," "scan path," "fixation,""saccade" and "eye movement." Scanpath abnormalities afford impressive sensitivity and specificity and appear largely independent ofpsychotropic medications. Scanpaths may demonstrate some fluctuation with symptomatology and may be useful in illuminating illnessstate or subtypes. However, there is evidence that viewing behaviours remain atypical regardless of symptom remission and may be pre-sent in unaffected relatives of individuals with schizophrenia. This research is in its early stages, and further investigation regarding pat-terns ot inheritance is required. Our findings support scanpath measures as a favourable topic for further investigation as a trait marker.

Introduction

Schizophrenia is a highly heterogeneous and complex disor-der. Diagnosis of the disease is currently based on clinical cri-teria and lacks objective tests. Although there is much supportfor a genetic component in vulnerability to the illness, and anumber of chromosome loci and candidate genes have beenrecognized, much about the genetic basis of schizophrenia re-mains unknown (for a recent review, see Williams and col-leagvies'). Several genes are likely to contribute to susceptibil-ity, and patterns of inheritance are intricate and irregular.Further, the natural boundaries that may distinguish schizo-phrenia from other psychotic and affective disorders, such asbipolar disorder, are unclear.' Case-control studies of theunderlying neuropathology or genetics of schizophrenia arelimited by the inherent heterogeneity of the disease and thepossibility that clinically similar phenotypes may also have dif-ferent etiologic causes. The development of objective criteriaand external markers would help to validate external bound-aries that represent the underlying pathology of the disease.

One approach to furthering our understanding of the gen-etic architecture of schizophrenia involves the study of endo-phenotypes.' Endophenotypes are measurable characteristics,such as neuropsychologic or psychophysiologic traits, that areassociated with genetic vulnerability for the disease. Thesecharacteristics are believed to reflect the action of genes asso-ciated with the illness, regardless of whether diagnosablepathology develops. The study of endophenotypes confersseveral benetits to genetic research in schizophrenia. First, ifthe phenotype is defined solely on the basis of the presence orabsence of a schizophrenia diagnosis in genetic studies, false-negatives will occur where individuals carry the genetic vul-nerability but do not display symptoms, and sensitivity willbe reduced.* Second, the current classification of mental illnessusing psychiatric diagnoses does not lend itself to the dissec-tion of diseases on the basis of their complex genetic architec-ture.' For example, the term "schizophrenia" may compriseseveral different disorders, each with its own genetic and non-genetic influences. Genetic analysis is made more difficult bythe likely polygenic natiore of schizophrenia.' Endophenotypes

Correspondence to: Dr. S.A. Beedie, School of Psychology, College of Life Sciences and Medicine, University of Aberdeen, William GuildBIdg., King's College, Aberdeen, Scotland, AB24 3FX; s.beedie@abdn.ac.uk

J Psychiatry Neurosci 2011 ;36(3)150-64.

Submitted Nov. 27, 2009; Revised June 14, Aug. 20. 2010; Accepted Sept. 1, 2010.

DOI; 10.1503/jpn.090169

© 2011 Canadian Medical Association

150 J Psychiatry Neurosci 2011 ;36(3)

Atypical I

help to dissect the disease phenotype into more biologicallyhomogeneous subgroups of individuals, facilitating considera-tion of genefic underpinnings.'" Although alternafive termssuch as "vulnerability marker," "trait marker" and "bio-marker" have previously been used interchangeably with theterm "endophenotype," recent authoritafive reviews haveclarified the disfincfions between these concepts.'

Researchers have debated and refined the endophenotypeconcept in psychiatry. It is broadly accepted that a number ofcore criteria should be met. These include associafion with thedisease under invesfigation, stability over time, relative in-dependence from clinical state, evidence that the characterisficis under genefic control and cosegregation in families withprobands and other affected family members.'"'" Such meas-ures must also show good psychometric or neurometric prop-erties. Ideally, tests should be quick and inexpensive to ad-minister to allow for high-powered studies. The originalendophenotype approach assesses "intermediate phenotypes"that are proposed to consfitute causal links between genes anddisease. However, commentators have caufioned that not allputafive endophenotypes meeting the above criteria will rep-resent traits lying on the pathway between genes and disease.For example, a trait may comprise an epiphenomenon of a riskgene or may be associated with different genes than those pre-disposing to the disease." Such traits may sfill confer substan-tial value in identifying more genetically homogeneousgroups. Establishing that a trait truly lies on the gene-<üseasepathway may require longitudinal research and even interven-tion studies in addifion to fulfilling the basic endophenotypecharacterisfics." Given that putative markers fulfilling the cri-teria of an endophenotype may provide valuable insight into adisease even when they have not been shown to consfitute acausal link between genes and overt expression of the disease,in the present context we adopt the more operafionally neutralterm "trait marker" to refer to stable, heritable biologicalprocesses or abnormalifies that are presumed to indicate dis-ease vulnerability without the implicafion that the marker ne-cessarily represents a causal link between gene and disorder.

A broad range of psychophysiologic measures have beenproposed as candidate trait markers in schizophrenia (for areview, see Javitt and colleagues"). Examples include smoothpursuit eye movements,'- P300 event-related potentials," sen-sory gating'' and antisaccade performance." However, nomeasure to date has met the essenfial criteria required of anunequivocal marker of vulnerability for schizophrenia. Forexample, a deficit in smooth pursuit eye movements is one ofthe most promising potential markers. But there is evidenceof heterogeneity in pursuit ability in both individuals withschizophrenia and healthy groups. Tracking deficits are notlimited to nor uniformly present in schizophrenia. The pro-portion of individuals with schizophrenia presenting im-paired tracking ranges from 12"/., to 96%, with most reportsFalling between 50"/<> and 86%, compared with l%-8% in con-trols.' Lack of diagnosfic specificity is also a common limita-tion of endophenotype measures proposed to date.

A recently emerged candidate marker is the visual scanpath,the pattern of eye movements generated during free explor-ation of a visual scene. A number of researchers have pro-

posed that scanpath measures may consfitute a trait marker ofvulnerability for schizophrenia, but few authors to date havereported the current state of scanpath research in this area. In-vesfigafions of the suitability of scanpaths as a marker are intheir infancy compared with other more tradifional measures.The purpose of this review was to draw together existingknowledge on state and trait characterisfics of scanpath abnor-malifies in individuals with schizophrenia within the frame-work of the basic concepts widely accepted to be among theopfimal characteristics for a trait marker. Papers reportingscanpath patterns in patients with schizophrenia were idenfi-fied by PubMed and Google Scholar searches and by scanningreference lists in relevant articles. Search terms included"schizophrenia," "psychosis," "scanpath," "scan path," "fixa-fion," "saccade" and "eye movement." We summarize find-ings reladng to the association of scanpath deficits with schizo-phrenia, stability of these deficits over fime and in parallelwith fluctuafions in disease state as well as evidence for gen-efic control of scanpath dysfuncfion in individuals with schizo-phrenia, as illusfi-ated by family and genefic linkage studies.

Scanpaths as potential trait markersin individuals with schizophrenia

Eye movements provide a directly observable measure of vis-ual orienting and attenfional biases.'"'" During visual explor-afion, patterns of visual scanning (scanpaths) are formed bysuccessive periods of steady gaze (fixafions) and interveningrapid movements (saccades).'" Fixations allow salient areas ofthe scene to be concentrated on the fovea, providing the vis-ual system with high-acuity informafion, while less detailedinformafion is collected by parafoveal and peripheral refinalfields. Although it is possible to dissociate the point of gazefrom the direcfion of visual attenfion," under natural viewingcondifions they are usually closely linked.'"" In healthy ob-servers, the patterns of movements produced by the saccadicsystem are not random. Rather, saccades and fixafions aresubject to the influence of a number of variables, includingthe physical and semanfic properfies of the image and the de-mands of the viewing task.^'"'

A variety of related measures are commonly used to studyscanpaths. These include fixafion and saccade frequency, fixa-fion durafion and saccade amplitude. Fixafion measures re-flect periods of detailed informafion acquisifion and planningof subsequent distal fixations. Properties of ocular fixationsmay be influenced by difficulties in attenfional disengage-ment or speed of informafion processing. Saccade measuresmay reflect the integrity of saccadic programming, the abilityto execute saccadic sequences and fine oculomotor control.Clearly though, in any given measurement epoch, such vari-ables will be highly interrelated, and an irregularity in anyone stage of visual explorafion is likely to be reflected by anumber of variables. Therefore, the interpretation of eyemovements arguably may be more valid when a number ofmeasures are considered together, parficularly when interpre-tafion is to be made within fi-ameworks of cognifive funcfions.

Experiments in human and nonhuman primates have iso-lated a fronto-parietal cortical complex that is active during

I Psychiatry Neurosci 2011;36(3) 151

Beedie et al.

pursuit, intentional and reflexive saccades and scanpath for-mation. The frontal eye field (FEF) area receives extensive in-puts from the extrastriate visual cortex and is involved in trig-gering and maintenance of saccadic and pursuit eyemovements.'"- Frontal eye field activity coincides with trans-formation of cognitive decisions into motor signals to fixate aparticular image location." Some FEF activity is correlatedwith the frequency of saccadic eye movements.' The decisionsand motivation for preparing and performing these saccadesare controlled by the posterior cingulate cortex (the cingulateeye field). The parietal eye fields manage shifts in visual atten-tion'' and influence saccade target selection. Neural activityhere dictates the latency of these reflexive, visually guided sac-cades. The supplementary eye fields receive motion and spa-tial information about image features via the dorsal visualpathway and prepare motor sequences for successive sac-cades. Saccade inhibition during fixation, short-term spatialmemory and saccade prediction are orchestrated by the dorso-lateral prefrontal cortex (DLPFC). The development, mati^rationand integrity of this cortical network are essential for triggering

saccades generated by the superior colliculi and brain stem.Atypical scanpaths in individuals with schizophrenia have

been observed in response to a broad range of visual stimuliand in different tasks. In general, the literature has associateda restricted style of scanning with the illness, characterizedby fewer fixations and saccades, increased average fixationdurations, smaller saccades and shorter scanpath length com-pared with healthy viewers.'" '" In contrast to the relativelyextensive or holistic scanning strategies of healthy viewers,scanpaths in individuals with schizophrenia also tend to beless spatially dispersed, with less attention paid to percep-tually and semantically informative areas.* "^="*-"*'»' ' ''- -*''Several researchers have employed scanpath techniques toexamine face processing or emotion processing in individualswith schizophrenia, and a substantial proportion of reports ofscanpath behaviour in this population are therefore based onfindings from face stimuli presented centrally on a display.However, replication of similar scanning patterns with stimulidevoid of social content suggests that the phenomenon is notspecific to faces or social scenes. Figure 1 illustrates fixation

Healthy viewer Viewer 1 with schizophrenia Viewer 2 with schizophrenia

Fig. 1: Fixation heat maps for representative scanpaths of a healthy viewer and 2 Individuals with schizophrenia. Red areas denote greaterdensity of fixations. Image brightness and contrast have been reduced for clarity. Source: data reported in references 70-72.

152 } Psychiatry Neurosci 2011,-36(3)

Atypical scanpaths in schizophrenia

patterns in individuals with schizophrenia and an unaffectedcontrol viewer.

A central role of frontal lobe pathologies in abnormal scan-ning is supported by neuroimaging findings,'''"" similaritiesbetween scanpaths of patients with schizophrenia and indi-viduals with frontal lobe lesions,*"^'' correlations with neuro-psychological performance* and an association with negativesymptoms, which in turn are linked with prefrontal dysfunc-tion.^ Scanpath abnormalities may represent an epiphenom-enon of a more basic dysfimction in oculomotor control thatis assessed by more established putative trait markers in indi-viduals with schizophrenia. The possibility that the plethoraof eye movement abnormalities arise from a common coredeficit is worthy of further investigation. However, findingsrecently published by our group as conference abstracts sug-gest that scanpath and tracking deficits occur independentlyin patients with schizophrenia and nonclinical viewers. " Thispreliminary work has also found associations between scan-path measures and neuropsychologic performance in indi-viduals with schizophrenia that were consistent with a role ofDLPFC dysfunction in restricted scanning.'" Previous studieshave reported that although the DLPFC is implicated insmooth pursuit performance,"'* "" lesions in this area impairperformance on saccadic tasks but do not disturb smoothpursuit performance."'"

Methodologically, scanpaths confer a number of advan-tages over many alternative measures. Scanpath recording isinexpensive, objective and noninvasive and is not reliant onparticipants' compliance with complex task instructions.Scanpath abnormalities have been proposed as a putativemarker for schizophrenia,""* •' '• '•"•' " but whether scanpathdeficits are a state- or trait-related phenomenon remains con-tentious. State markers of psychiatric disease are characteris-tics that appear and disappear with clinical episodes. Traitmarkers are phenomena that are present regardless of varia-tion in clinical state.

Association of scanpath abnormalitieswith schizophrenia

Reliability and effect sizes

Scanpath abnormalities have been widely replicated inschizophrenia research, and results show high levels of con-sistency across studies, especially with regards to fixationnumber, saccade amplitude and scanpath length measures.Reports of null findings are infrequent. The robustness of thescanpath phenomenon is highlighted by the variations intask, stimuli and recording methods. Table 1 and Table 2summarize published findings in individuals with schizo-phrenia and nonclinical comparison viewers.

We calculated effect sizes for all studies in Table 1 andTable 2 reporting group means and standard deviations forany of the fixation or saccade tneasures of interest (fre-quency and duration of fixations and saccades, amplitude ofsaccades and total scanpath length) for individuals withschizophrenia and nonclinical comparison groups. The28 studies meeting these criteria and included in the effect

size calculations are denoted by asterisks in Table 1. Whendata for more than 1 condition were reported, such as datafor individual stimuli, schizophrenia subgroups or test ses-sions, individual effect sizes were calculated for each condi-tion, resulting in 202 individual effect sizes. A summary ofeffect sizes for individual measures is presented in Table 3.Medium to large effect sizes were found for each of the5 scanpath measures, with a mean Hedges' unbiased popu-lation estimate (g*) ranging from 0.98 (fixation frequency) to0.64 (fixation duration). Standard deviations for the mean ef-fect sizes suggest that a high variation across studies is alsopresent. To date, studies of scanpath dysfunction in patientswith schizophrenia have varied considerably in both taskand stimulus conditions and methods of data analysis. Taskselection can alter the cognitive and neural mechanisms im-plicated in viewing behaviour (e.g., free-viewing tasks com-pared with visual search), and it is vital to consider such is-sues during study design. Further sources of variation arisein the operationalization of dependent measures betweenstudies. For example, some studies define scanpath length asthe sum of the distance between successive fixations, where-as other studies measure the true path of the point of gaze.Methodologie variation may not only impact on the poten-tial for generalization across studies but may constitute animportant consideration in the accurate interpretation ofviewing behaviours and perceptual experience. Stimulussize and interest need to be sufficient to elicit eye move-ments and visual exploration."' It is important to provide ad-equate reason to make fixations and saccades around the im-age to allow researchers to make inferences about cognitiveprocesses involved during integration of perceptual infor-mation. If scanpath measures are to provide a potential tar-get for trait marker research, it will be critical to considerstandardization of practices. The magnitude of the prelim-inary effect sizes here support the potential value of scan-path measures as a discriminatory tool, and further analysisof the sources of variation across studies will help to buildan optimized protocol for the identification of the scanpathdeficit in future studies.

Sensitivity and specificity

Numerous trait markers may be associated with a given dis-order. As each trait is hoped to reflect the actions of a morediscrete genetic profile than the disease phenotype, eachmarker may represent only part of the risk for that disorder.It may be argued, therefore, that weak associations betweendisorder and marker should not automatically result in dis-counting of the candidate trait measure." Nevertheless, at-tempts to use scanpath parameters to distinguish individ-uals with schizophrenia from other clinical samples havedemonstrated promising levels of sensitivity and specificity,beyond those of i"nore traditional candidate markers, such assmooth pursuit deficits.' Sensitivity denotes the proportionof individuals correctly identified as belonging to the af-fected group (true positives). Specificity refers to the propor-tion of individuals correctly identified as not belonging tothe affected group (true negatives). Kojima and colleagues'"

J Psychiatry Neurosci 2011;36(3) 153

Beedle et al.

reported responsive search score (RSS; a measure of the spa-tial distribution of visual attention) to provide 71%-80%sensitivity and 80%-88% specificity in distinguishingindividuals with schizophrenia from those with unipolardepression, epilepsy and amphetamine psychosis and non-clinical comparison groups. Matsushima and colleagues"'found similar efficiency (76.7% sensitivity and 81.4% speci-ficity) using discriminant analysis to separate outpatientswith schizophrenia from individuals with depression.

methamphetamine psychosis (MAP), anxiety disorders, tem-poral lobe epilepsy or frontal lobe lesions and healthy con-trols. Predictor variables in the discriminant function wereRSS and fixation frequency. Applying this same discrim-inant function, Kojima and colleagues'" found even higherlevels of sensitivity (89.0%) and specificity (86.7%) in groupsof individuals with schizophrenia and depression. Scanpathsmay be a valuable addition to more traditional eye move-ment trait markers in discriminating between patients with

Table 1: Methods of studies examining scanpath behaviour in individuals with schizophrenia (part 1 of 2)

Study

Benson at al.**

Bestelmeyer et al.'"*

De Wiide et al."*

Gaebel et al."Green et al."*

Green et al.''*

Hori et al.*"*

Kojima et al."

Kojima et al.'"*

Kojima et al."*

Kojima et al."*

Kojima et al."*Kurachi et al."

Leonards etal."

Loughlandet al."*

Loughiand et al."*

Loughland et al."*

Manor etal."*Matsukawaetal."

Matsushima et al."*

Matsushima et al."'*

Mikami et al.°^*

Minassian et ai."*

Moriya et al."

Schizophrenia group,no. and type

11 paranoid

22

50

2024

11 deluded8 nondeluded

37

105

25

80

145

2912

8 paranoid1 disorganized9 undifferentiated

65

65

63

251 5 • -

20

30

30

38

24

Comparison group

22 healthy controls6 cannabis-induced psychosis

37 healthy controls19 bipolar disorder36 healthy controls23 siblings^20 healthy controls26 healthy controls

22 healthy controls

36 healthy controls

50 healthy controls30 unipolar depression28 amphetamine psychosis50 epilepsy25 healthy controls25 unipolar depression10 obsessive-compulsive disorder50 healthy controls25 methamphetamine psychosis21 temporal lobe epilepsy (L)12 temporal lobe epilepsy (R)

124 healthy controls116 depression23 healthy controls

12 healthy controls28 healthy controls

61 healthy controls52 affective disorder

61 healthy controls

61 healthy controls37 first-degree relative^

25 healthy controls19 healthy controls20 systemic lupus erythematosus20 healthy controls18 frontal lobe lesion10 healthy controls10 unipolar depression10 methamphetamine psychosis10 alcohol psychosis10 anxiety disorder10 temporal lobe epilepsy10 frontal lobe lesion30 healthy controls48 methamphetamine psychosis30 healthy controls

20 healthy controls

Stimuli

Faces, landscapes, fractals

Faces, fractals, landscapes, noise

Images from Thematic ApperceptionTestDrawings depicting proverbs

Faces (context-free and context-embedded)Faces

Rorschach stimuli

S-shaped figures

S-shaped figures

S-shaped figures*

S-shaped figures

S-shaped figuresPicture completion test of the WAISFaces, fractals, landscapes

Faces, degraded faces

Faces

Faces, degraded faces

Neutral face. Rey Complex FigureS-shaped figures

S-shaped figure

S-shaped figures

S-shaped figures

Rorschach stimuli

Human figures, S-shaped figures

Taskt

Free-viewing

Free-viewing

Thematic Apperception Test

Free-viewing

"Decide what the person is feeling orthinking""Think about how the person seemsto be feeling"Eye movements recorded duringfree response period in RorschachtestS-shaped figures procedure§

S-shaped figures procedure§

S-shaped figures procedure§

S-shaped figures procedure§

S-shaped figures procedure§WAIS picture completion test

(i) Face recognition(ii) Affect recognition

Affect recognition

(i) Face recognition(ii) Affect recognition

S-shaped figures procedure§

S-shaped figures procedure§

S-shaped figures procedure§

S-shaped figures procedure§

Viewing for subsequent Rorschachresponse periodFree-viewing

154 J Psychiatry Neurosci 2011;36(3)

Atypical scanpaths in schizophrenia

schizophrenia and healthy controls. Recent findings haveshown that a multivariate eye movement phenotype com-bining scanpath with tracking and fixation stability meas-ures may accurately distinguish cases from controls at a rateapproaching 100%, far beyond the accuracy rate of any sin-gle traditional psychophysiologic or neuropsychologic meas-yj.g 71,72 jj^jg Yvork is currently in preparation for publicationby our research group.

Despite the practical significance of the current classifica-

tion system of psychiatric diagnoses, it is widely acknow-ledged that these hypothetical categorizations are unlikely toreflect the natural boundaries of the disorders. One of thebenefits of the endophenotype approach is that externalmarkers may be used to validate and refine these bound-aries. Trait markers may thus be shared across disorders, re-flective of shared genetic pathways or common biologicalbases. Equally, however, it is important to develop markerswith sufficient specificity to support genetic analysis for a

Table 1 : Methods of studies examining scanpath behaviour in individuals with schizophrenia (part 2 of 2)

Study

Nishiura et al."*

Obayashi et al."*

Obayashi et ai."'

Philips and David"

Phillips and David'"

Phillips and David"Phillips and David™

Phillips and David"

Philips and David"'*

Philips and David™*

Phillips et al."

Phillips et al.'"

Quirk and Strauss"*

Rosse et al ."

Ryuetal.''*Schwartz et al ."Streit et al."

Takahashi et al."

Takahashi et al.'"*

Tsunoda et al."Tsunoda et al."*Williams et al."*

Williams etal."*

Xia et al."

Schizophrenia group,no. and type

28 paranoid16 nonparanoid

27

18 paranoid7 disorganized2 undifferentiated1 catatonic7 deluded7 nondeiuded7 deluded7 minimally deluded8

12 pers. delusions10 nonpers. delusions

7 deluded7 nondeiuded6 deluded5 minimally deluded

19 pers. delusions8 nonpers. delusions

19 pers. delusions8 nonpers. delusions

20

16

6016 .16

38

23

391132

63

28

14

Comparison group

72 healthy controls

20 healthy controls

29 healthy controls

10 healthy controls

10 healthy controls

9 healthy controls10 healthy controls

10 healthy controls

9 healthy controls

17 healthy controls

18 healthy controls

10 addiction recovery patients

38 cocaine users

30 healthy controls

10 healthy controls18 healthy controls

37 parentst47 siblings^43 healthy controls23 siblings^

32 healthy controls60 healthy controls

28 healthy controls

41 healthy controls23 parents^

Stimuli

Faces with emotion-congruentsound, circles

Geometric figijres (from BentonVisual Retention Test)S-shaped figures

Single faces, face pairs

Single faces, face pairs

Neutral faces, chimeric faces

Scenes depicting neutral, ambiguousor threatening activity

Single faces and face pairs (fromRecognition Memory Test)

Single faces and face pairs (fromRecognition Memory Test)

Scene showing neutral, ambiguousor threatening activitiesScene showing neutral, ambiguousor threatening activities

Emotional imagesFaces with direct and averted gaze

Circles, faces symbols, landscapeUpright and inverted facesFaces

S-shaped figures

S-shaped figures

S-shaped figuresBenton Visual Retention TestFaces, degraded faces

Faces

S-shaped figures

Taskt

Judge whether current picture differsfrom previous

Benfon Visual Retention Test

S-shaped figures procedure§ 2testsessions 6 months apart

Face recognition

Face recognition

Neutral face: state whether face ispleasantChimeric faces: determine facialexpressionRecognition task

Recognition task

(i) Free-viewing (ii) Asked to attendto threatening areas

Free-viewing

(i) Affect recognition(ii) Gaze direction discrimination

Varied by stimulusAffect recognitionAffect recognition2 test sessions 4 weeks apart

S-shaped figures procedure§

S-shaped figures procedure§

S-shaped figures procedure§Viewing for stimulus reproductionFree-viewing for later recognition

Viewing for subsequent affectrecognition

S-shaped figures procedure§

(L) = left-side spike focus; nonpers. = nonpersecutory; pers. = persecutory; (R) = right-side spike focus: WAIS = Weschier Adult Intelligence Scale.'Study included in effect size calculations.f in most of these studies, it is not clear whether participants were aware of affect recognition tasks, reproduction tasks, etc., before viewing or whether they completed a free-viewing taskand were then informed of the additional task.ÍFirst-degree relatives, parents and siblings were ail unaffected relatives of patients with schizophrenia.§ln these tasks, participants are shown an S-shaped figure (target figure) and asked to state whether the figure differs from a similar figure presented previously. The question "Are thereany other differences?" is then repeated until the participant states that no further differences are present. Eye movements are recorded during each viewing period and duringquestioning.TlBoth schizophrenia and schizotypal disorder.

J Psychiatry Neurosci 2011;36(3) 155

Beedie et al.

.*:

T3

I

CIc'c

CL ¿r s- CL

z < < z

ois

QoV

ce ce Ü cez z CO z

ical

CEZ

0. 0

xati

z LLs? s?

Ma-cooV

O

n:S

o

xati

JS)

11 ^ -

Ilsco

r

IIÜ

aw:

ce

.>• ce ce CE ce ce ce< z z z z z z

"5 ô

-s ~ > VÜ ce ce CE ce ce

ont

o

oCO

Os:

O

ic J

ois

8Vü ceCO z

-oís

cont

i

VO

aw:

ce

cooVüCO

ion:

ixat

LJ_

-Oís

cont

i

Vüeo

aw:

ce

onti

VOC/)

ion:

ixat

LL

COOV

oCO

ion:

ixat

LL

-oís

cont

i

V

OCO

aw:

ce

0) ^ Jíí•o I o o

II V

o oeo eo ce ce en

-5

8IIO

rols

í co

nt

üeo

cez

cez

cez

Ï

10

rols

; co

nt

üCO

p

text

-ei

c0ü

rols

= co

nt

ü ceCO z

8 8 8 8 ^ 8 ^ 8 8A A II II Ä II ^ A II

O O O O O Ü O O Ü

e o e o e o e o ü c o ü c o e o

O O O O O g O g O O- E - v c c p - S - E

cez

cfl

roi

8V

0CO

V0eo

0

. c0

m

rol

c00V

üen

cez

cez

2

8V

0CO

Ö

8V

0eo

ont

i

0

V

üCO

c/l0

nti

00V

üCO

ce ce ce ce ce X cez z z: z z z z

Vo ceCO z

cez

ce ce ce ne ce

0C0

V

0 ûcCO z

0c

8V

CO

e"c8V

0CO

Ü ce ce ce ce CE CEeo z z z z z z

o

8 8 8 8 1 < 8 B 8 8 8Ü O O O O Ü O Ü Üc o c o c o c o o c o o e o e o

Í ?, » i! -s( l î CD CD ' S • "

O Cl) uj ro ^CÛ CD Û CD O

0 0 0 o o

0 0 0 tñ " oo ü o —- o üV V V ° * " V

CO CO CO Z -í^ (N CO

a•5

CD CO

E Eo o 'o 'o o oX ^ ^ i : ^ ^

ce o ce ce oZ l/l Z Z O}

Itrol

sco

n

Aüeo

itrol

sco

n

Aüeo

itrol

sco

n

AüCO

0

3gni

ti

i(D

roLL

trois

con

V

CO

8

" 0CD

3 =

trois

con

A0

< CO

1

con

II

üeo

ce ce ce cr ce

ce CE ce ce

V) (f)

o o

8V

oCO

0eo0

•àLJ_

CO

CO > *

^ceoez

con

V0eo

CE

z

con

V0eo

cez

ro .H- ^ - * -* - 03 tu QJ^ - O - D - D

CO

gj

tO

en13i5

ro'Sro

CO

B

!fi

tal.

£CO

"S

net

CO

ssi

156 J Psychiatry Neurosci 2011;36(3)

Atypical scanpaths in schizophrenia

oCO

ë

not

but

ÜCO

?

'S'So

oc

not

but

oen?•O

"O

o

(0 CO CQ (Q CO

2 2 9 2 y

CDü

CO

¡ling

ce E

tno

• ^

ooV

Üz <n en

VÜen

:les:

Ü

8CO

DlC

en

Iro:

O

II

o CO

ocoÜVo c c c c c c c c c c i i i rc o z z z z z ^ z :

ce cez z

o

con

VOCO

couVÜCO

1 - ce ce ce ce ce ce

ce (Cz z

c c c c c e c e c c c c c e c e-z.-z.-z.-z.-z.-z.-z.-z.

oCO

g oCO C O

.}=. -O — "*-

' ^ gS-'DCO CD

C C LL c < <

Q- Q. Q.>• Œ S ' £:•< 2 < <

O 'Sen .î

CO o

2:'D S0 CD ¡o

piQ-r en1 § gE 1 5

_ Q) ^CO CL —Ü en CO

• f i l< c c

ce oe ce oe oe oe cez z z z z z z

. y¿¿ \j m (Ji (Ji

•5dZ

ce

eu

iSen

ilin

(O

O

V

oCO

CO

cñ

yb

ai O

CO ^

*~ 8.E II

O CO

o

8V

oCO

o•g

8V

oen

OCO

"S-o

oCO' D

lude

CDQ

CO

s

ont

oV

Üen

TD

"DD

ndel

o

O

ntr

ooII

oCO

COO

o

ntr

ooII

oCO

c/)

o c e c e o e ö j o ü oc n z z z Q z c o c o

'cB ^

T3 :2 5 P P T3• > > > > > - ^

•= CO Í? Í? 5 iPQ C3CO CO

^ ^ ro ™ ™ ™ ™

O O û _ Q . Û . U _ Û _ u .

t/l UiO O

o ou oII V

c e ü o c e i r i r i rz c o c o z z z z

oCO

o Üen CO

c e c c c e c e c e c c c e c cz z z z z z z z

ce ce

'(Ö

iillip

s et

SCO

lirk

and

O

ue

tal.

'

cc

al.

'S

hwar

tz

üCO

ä .

'cö

%

cñ

CO

'S

kaha

sh

CO

h-

C .5

»'S cCO "O . £

m to

! ü! CO

en J ^ .; 0 ot i= CO

CO

z^C/D

roi

' c

8A

cr. Üz CO

roi

c

8A

ÜCO

CO

zUî

roi

C

8V

ce Üz CO

en

roi

coÜA

oe Üz en

roi

couA

OCO

w

roi

coCJII

ÜCO

i "o

8V

QCO

o

8V

ÜCO

wo"c

8V

üen

mo' coV

üCO

I I

t

'Neurosci 2011;36(3)

Beedie et al.

particular illness.* Atypical visual scanning in itself is notspecific to the schizophrenia spectrum. Scanpath abnormal-ities have been demonstrated in individuals with otherpsychiatric disorders, including social phobia," "" autism,"" ""bipolar affective disorder,''" attention-deficit/hyperactivitydisorder (ADHD),"" generalized anxiety disorder," obsessive-compulsive disorder'"'"'* and Alzheimer disease."* Import-antly, these abnormalities differ from those found in indi-viduals with schizophrenia, suggesting that scanpath dys-function as seen in schizophrenia may have value as adisorder-specific marker.

Other psychotic disordersDifferential scanpath patterns in patients with other forms ofpsychosis suggest that restricted scanpath behaviour inthose with schizophrenia is not simply the result of psycho-sis itself, tior does it itievitably lead to such symptoms. Forexample, patients with MAP show symptoms highly similarto those in patients with schizophrenia, including paranoidpsychosis with persecutory delusions and hallucinations. Al-though patients with MAP and those with schizophrenia donot differ in terms of fixation frequency or scanpath length,the spatial distribution of attention is substantially less im-paired in those with MAP. Similarly, individuals with sys-temic lupus erythematosus, manifestations of which includeschizophreniform psychosis, show more widespread disper-sion of fixations than patients with schizophrenia," and pa-tients with cannabis-induced psychosis reveal flxation clus-tering and restricted scanning that is more pronounced thanthat associated with schizophrenia.*' Investigation of saccadeand smooth pursuit performance in chronic ketamine use,which is associated with psychotic symptoms and cognitivedeficits similar to those of schizophrenia-spectrum disor-ders, has also suggested that ketamine administration is nota suitable model for oculomotor deficits in individuals withschizophrenia.'"^

Developmental disordersAutism is litiked with an erratic, disorganized pattern of vis-ual exploration.'"^ As in schizophrenia, individuals withautism demonstrate impairments in face processing and so-cial functioning. Some similarities in scanning styles havebeen fotind between schizophrenia and autism groups. Dur-ing face viewing, both individuals with schizophrenia andthose with autism assign less attention to the eyes and attendmore to the mouth and nonface areas than do healthygroups"""-''«'"''"^"'""" (although a different study found no dif-ference between autism and control groups in the proportionof time spent viewing the mouth and eye regions'"'). Alsomirroring viewing patterns in individuals with schizophrenia,adolescents and young adults with autism have demonstrateddiminished visual attention to social contextual informationwhen judging mental state.*'"" Despite these similarities inscanning styles, individuals with autism do not show the de-viant temporal parameters that are characteristic of scanningbehaviours in patients with schizophrenia."'-""

Reduced attention to salient scene regions in individualswith schizophrenia is also more extreme than that recorded

in people with ADHD."" Attention-deficit/hyperactivity dis-order has been linked with a more extensive visual scanningstyle, characterized by increased scanpath lengths comparedwith healthy groups.'" Therefore, patients with schizophre-nia and those with ADHD may show distinguishable scan-path behaviours despite the presence of social functioningand attention deficits in both groups.

Affective disordersBipolar affective disorder (BPAD) almost certainly shares acommon genetic basis with schizophrenia spectrum disor-ders.-"'"' Individuals with BPAD produce exploratory eyemovements similar to those of patients with schizophrenia,but BPAD is nevertheless distinguishable as an illness usingcertain measures. Loughland and colleagues" investigatedscanpath formation during face viewing in individuals withbipolar and unipolar depression. Whereas individuals withschizophrenia had scanning behaviour that was atypical inboth spatial and temporal domains, patients with affectivedisorders differed from controls in reduced attention to facialfeatures." A second study involving patients with bipolardisorder found this group to occupy an intermediate positionbetween healthy observers and viewers with schizophreniawith regards to a range of temporal scanpath variables dur-ing free-viewing." This second study found that spatial dis-tribution of fixations differed significantly between patientswith bipolar disorder and healthy controls for fractals, noisepatterns, landscapes and faces. Compared with healthyviewers, individuals with unipolar depression demonstratediminished fixation frequency and decreased cognitivesearch scores (an index of attention to predetermined re-gions of interest during a figure comparison task). However,in contrast with the scanning behaviours of patients withschizophrenia, unipolar depression is also associated withnormal saccade amplitude and RSS (an index of the overallfixation dispersion)."'"'

Anxiety disordersPatients with generalized anxiety disorder fixate a similarnumber of image regions as controls during scene viewing."This viewing pattern contrasts that in patients with schizo-phrenia, who allocate attention to a restricted spatial area. In-dividuals with generalized social phobia demonstrate a formof hyperscanning to faces, characterized by increased scan-path length compared with healthy viewers.'"" Responsivesearch scores (indexing attentional distribution during a fig-ure comparison task) and fixation frequency are also re-ported to be lower in individuals with schizophrenia than inpatients with obsessive-compulsive disorder.-^

Alzheimer dementiaIndividuals with Alzheimer dementia have been report-ed to show increased fixation duration and smaller sac-cade amplitude and decreased attention to salient re-gions during clock reading compared with controls.""'However, data directly comparing scanpath behavioursof schizophrenia groups and individuals with Alzheimerdisease or other forms of dementia are not yet available.

158 J Psychiatry Neutosci 2011|36(3)

Atypical scanpaths in schizophrenia

Stability and relation with clinical state

Association with clinical symptoms

In an early study, Gaebel and colleagues"' found that re-stricted scanning behaviour was related to negative symp-toms in schizophrenia, whereas positive symptoms werelinked with extensive scanning styles (or hyperscanning). Al-though not unequivocal, more recent studies have reportedthat scanpath variables correlate with measures of negativeclinical symptoms. Two studies have reported significant cor-relations with the Positive and Negative Syndrome Scale(PANSS)"^ negative symptom subscore with small-to-medium effect sizes for fixation frequency (R^ = 0.22 and0.08) and duration (R^ = 0.14 and 0.11), saccade amplitude(R2 = 0.17 and 0.16) and scanpath length {R^ = 0.21)."" Al-though effects of larger magnitude (B? for significant correla-tions ranging from 0.20 to 0.32) were found for fixation fre-quency and scanpath length in a third study, patterns werenot consistent across stimuli.' ^ One report found no correl-ation of scanpath measures with PANSS scores, with theexception of a small correlation with scanpath length(R^ = 0.09).'= Studies using individual Scale for the Assess-ment of Negative Symptoms (SANS" ) subscales have foundthat restricted scanning can be associated with dimensions ofavolition/apathy*'"'"' and affective flattening or blunting.*'''"'Restricted scanning is less often associated with alogia andanhedonia subscales. " There have also been reports of asso-ciations between restricted scanning and Brief PsychiatricRating Scale (BPRS"") subscales of blunted affect/"*""^ motorretardation,*' emotional withdrawal**"* and unusual thoughtcontent.* However, null findings have also been reported ineach instance.*'* No single eye movement measure or clinicalstate measure has consistently linked scanpath dysfunctionwith negative symptoms. Other studies have found no rela-tion between scanpath measures and negative symptoms."'^'*'Meta-analysis is also problematic since most of these studiesonly report effect sizes for significant correlations.

The exception to these inconsistent findings is the associa-tion of negative symptoms with RSS, which indexes the dis-tribution of visual attention during a visual comparison task.The RSS has been repeatedly linked with measures ofblunted affect and emotional withdrawal with moderate ef-

Table 3: Scanpath measures in individuals with schizophreniacompared with nonclinical groups across 28 studies

Measure

Fixation frequencyFixation durationSaccade frequency

Saccade amplitudeScanpath length

Overall

Hedges' g*

0.980.640.730.740.770.81

SD

0.540.520.410.510.480.53

95% CI

0.85-1.110.48-0.800.26-1.19

0.56-0.920.65-0.89

0.74-0.88

No.t

68403

3061

202

CI = confidence interval; SD = standard deviatie)n.'Studies included in the calculation of effect sizes are marked with an asterisk in Table 1.tNumber of comparisons for which effect sizes were calculated. When data wereavailable for more than 1 condition within a study (e.g.. for individual stimuii.schizophrenia subgroups or test sessions), effect sizes were calcuiated independentlyfor each condition.

feet sizes (blunted affect: mean R = 0.30; emotional with-drawal: mean R = 0.28).**-"' In studies using RSS, the vieweris presented with a simple geometric shape and asked to de-scribe in what way the figure differs from a similar stimuluspresented previously. After the viewer's response, the experi-menter asks whether there are any further differences, andRSS is measured during the viewer's response period at thisstage. Given that the viewer has already answered the initialquestion, the degree of visual exploration during the RSS per-iod is likely to reflect an interpersonal component and anability to engage in conscious, internally motivated visual ex-ploration that is not required under the free-viewing condi-tions of many other studies.

In contrast to the suggestion by Gaebel and colleagues'"that positive symptoms are associated with an extensivescanning style, findings suggest that scanpath dysfunctionoccurs relatively independently of positive symptomatology.Studies have found no evidence of association between scan-path measures and positive symptom scores, either with re-gard to PANSS subscales,--'"" individual BPRS items*'*" orBPRS composite scores for positive symptom items.*'" Onlyone study since that of Gaebel and colleagues has reportedsignificant correlations between extensive hyperscanning andpositive symptom scores.** Positive associations linking fixa-tion frequency with somatic concern (R- = 0.16) and excite-ment (R2 = 0.22) on the BPRS have not been replicated. In asecond study, a negative correlation between PANSS positivesymptom scores and scanpath length linking increased posi-tive symptoms with more restricted scanning also revealedonly a small effect (R^ = 0.09) and was seen for only 1 of3 stimuli. Similarly, a finding of more restricted distributionof visual attention in a final study was seen for only 1 of4 stimuli. Such findings suggest that scanpath dysfunction isnot associated with positive symptomatology. Additional re-ports have failed to find evidence of extensive scanning inpatients with schizophrenia* or have linked groups with pre-dominantly positive symptoms with a range of viewingstyles from restricted to extensive scanning behaviours."' Onepossibility is that that restricted scanning is more specificallylinked with delusional symptoms," at least with respect toface viewing, although scanpaths still remain atypical in pa-tients with only minimal delusional symptoms.

Investigations of associations between scanpath anomaliesand symptom clusters identified by factor analytic methodshave found the candidate marker to be relatively independentof symptom profiles. Williams and colleagues' examined as-sociations of scanpath behaviours with symptoms of psycho-motor poverty, assessing blunted affect, social withdrawaland poor rapport, disorganization, tapping positive and nega-tive aspects of thought disorder and reality distortion, whichdenotes delusions and grandeur."" Williams and colleagues™suggested that scanpath behaviour showed only minimal as-sociations with symptom factors and that most patients maybe characterized by restricted scanning styles. Only subtle as-sociations between symptom dimensions and scanpath vari-ables were found in other studies by the same group.' ' "

Association with particular symptoms does not constitute ashortcoming in the use of scanpaths as a trait marker. In fact.

I Psychiatry Neurosci 2011;36(3)

Beedie et al.

links between symptom dimensions and particular scanpathstyles may suggest possible lines for investigation in attempt-ing to partition schizophrenia into natural subtypes. How-ever, the question of whether scanpath abnormalifies remainconstant during changes in symptom severity is central to as-certaining whether eye movement measures comprise stateor trait markers. Three studies have asserted that scanpathidiosyncrasies do not remain stable over time. Phillips andDavid"'"" reported that viewing abnormalities diminishedwith improvements in delusional symptoms. Individualswith schizophrenia have been found to be significantly moreimpaired in visual exploration during inpatient treatmentperiods than during outpatient treatment in cross-sectionalresearch. " Importantly, neither of these studies demonstrateda complete recovery of "normal" scanning behaviour in re-mitted groups. Therefore, whereas changes in symptomseverity may generate some fluctuation in scanpaths, it ap-pears that at least some degree of scanpath deficit remains,regardless of clinical state.

Eye movements generally show no associafion with globalsymptom ratings in individuals with schizophrenia in correl-ational analyses.''''''''"''"' Moreover, other studies have re-vealed stability of scanpath abnormalities, regardless ofsymptom changes. Streit and colleagues"^ found that idiosyn-crasies of scanpath length, fixation duration and feature se-lecfion in patients with schizophrenia persisted across testingsessions 4 weeks apart, regardless of a decline in symptomseverity. Posifive, but not negafive, symptoms declined be-tween testing sessions, supporting a lack of associationbetween restricted scanning and positive symptoms. In an-other study, patients demonstrated relatively stable eyemovement behaviours over an 8-month period despite par-tial remission of both posifive and negafive symptoms."' Incontrast, patients in this study who showed no reducfion insymptoms across the course of the research demonstrated areduction in saccade amplitude between testing sessions."'The authors argued that exploratory eye movements are in-fluenced by both trait and mental state factors. That eyemovement idiosyncrasies do not improve with reductions insymptoms may support the use of scanpath parameters as avulnerability marker, whereas saccade amplitude may modelillness chronicity in patients with schizophrenia.

Association with other clinical features

Although Obayashi and colleagues"' suggested that saccadeamplitude may model chronicity in patients with schizophre-nia, other reports of the relation between scanpath patternsand illness duration are inconsistent. Evidence for a lack ofassociation of scanpath variables with illness duration hasbeen documented,"-"""'" although Manor and colleagues"found illness duration to correlate with scanpath length.Williams and colleagues"" reported significant correlationsbetween illness duration and scanpath variables in patientswith schizophrenia, with longer duration associated withdiminished scanning, although patterns of association wereinconsistent across sfimuli. Those few studies assessing otherclinical characterisfics in relafion to scanning have found no

association between scanpath variables and number of cliti-ical episodes,"' duration of current stay in hospital"' or totalnumber of hospital admissions.'*

A further question is whether atypical scanning is a conse-quence of pharmacologie treatment. Evidence of a subtle rela-tion between neurolepfic dose and restricted scanpath behav-iour in patients with schizophrenia has been reported in2 studies. Williams and colleagues"" found chlorpromazineequivalent dosage to correlate with fixafion frequency, but notwith fixafion durafion, distance between fixations or overallscanpath length. Hori and colleagues"" foimd a negafive cor-relation between fixation frequency and neurolepfic dailydose, although this relation became notisignificant when theycontrolled for negative symptom ratings in the analysis.

In contrast, numerous studies have found no significantassociations between scanpath variables and neurolepticdosage in patients with schizophrenia.""-"'""""'"'"'^'"''""" Al-though many of these reports are based on post hoc analyses,the findings are supported by comparisons of neuroleptic-naiVe patients with patients receiving regular neuroleptictreatment in both cross-secfional and longitudinal study de-signs. In 2 studies, the scanpaths of individuals with schizo-phrenia receiving neuroleptics did not differ from those ofnever-medicated pafients (cross secfional design,"" longitud-inal design"'""). Other researchers have found no influence ofmedication type (typical v. atypical tteuroleptics" "') or dura-tion'" on scanpath variables. Evidence that antipsychoticmedicafions may reduce abnormalifies in viewing behaviouralso suggests that restricted scanning is not a consequence ofneurolepfic treatment."""" In some instances, correlations be-tween neurolepfics and eye movements may be due to higherneurolepfic doses being associated with more pronouncedposifive symptoms (and thus a tendency toward a more ex-tensive scanning style)."'

Few studies have directly assessed the effects of specificneuroleptic medications on scanpath behaviours in patientswith schizophrenia. A study examining face viewing in out-pafients with schizophrenia receiving either risperidone orhaloperidol found that the 2 treatment groups did not differon fixation frequency, overall fixation durafion or scanpathlength."" However, patients treated with haloperidol showedreduced attention to facial features compared with healthyviewers. The authors argued that the spatial distribufion ofattention may be subject to medicafion effects, whereas tem-poral variables are not.

Genetic control

Compared with other putative candidate trait markers cur-rently tmder invesfigafion, few studies have examined scan-path anomalies in relafives of schizophrenia probands. Evenfewer studies have looked at genetic linkage. However, evi-dence of linkage of exploratory eye movement behaviours(specifically number of fixations) to chromosome 22qn.2-ql2.1 has been reported."' Chromosome 22q is associated withseveral candidate genes for schizophrenia, and microdelefionsin this region have been linked with increased risk for the dis-ease. Further, characteristics of restricted scanpath behaviours

160 J Psychiatry Neurosci 2011;36(3)

Atypical scanpaths in schizophrenia

have been demonstrated in healthy relatives of individualswith schizophrenia. Such findings support the use of scanpathmeasures as a marker of genetic liability for schizophrenia."-Parents of schizophrenia probands have been found todemonstrate reductions in fixation frequency, scanpath lengthand indices of the distribution of visual attention comparedwith nonrelative groups, suggesting that viewing behavioursmay be at least partly under genetic control."" Loughland andcolleagues"* examined face-viewing patterns in patients,healthy first-degree relatives of patients with schizophreniaand unrelated controls. Relatives' viewing behaviour was un-usual, with temporal measures of fixation patterns falling be-tween those of controls and the atypical patterns seen inprobands. Relatives also showed an atypical distribution offixations to face stimuli that was even more extreme than thatobserved in the schizophrenia group. Similarly, healthy sib-lings of schizophrenia probands have been found to occupyan intermediate position between patients with schizophreniaand healthy controls with regards to fixation frequency andspatial dispersion of fixations."

Findings to the contrary have been reported in a study byde Wilde and colleagues."" They found that healthy siblingsof individuals with schizophrenia did not differ from age-matched controls on several scanpath measures, includingscanpath length, fixation number and fixation duration, dur-ing viewing of complex scenes. Shorter scanpath length wasfound in patients with schizophrenia but not siblings, and theauthors suggested that scanpath length is not a suitable can-didate for a vulnerability marker of schizophrenia. Previousreports of .scanpath abnormalifies in healthy sibling groupswere argued to be attributable to a failure of previous studiesto match sibling and nonsibling groups on age. Also of note,however, is that the study by de Wilde and colleagues"' failedto find the prominent scanpath abnormalities in their groupof schizophrenia patients that may be expected on the basisof numerous previous studies. Specifically, patients did notdiffer from healthy controls with regards to frequency of fixa-tions or average duration of fixation. Sampling biases orinsensitivity of this particular protocol in identifying re-stricted scanning behaviours may explain these differencesbetween studies. Another possibility is that scanpath behav-iours in relatives of patients with schizophrenia may behighly heterogeneous.

Summary and future directions

At this early stage in research, measures of visual scanpathformatioti appear to constitute an appealing opportunity fortrait marker investigation in individuals with schizophrenia.Although the findings are still far from clear, scanpaths arebeginning to show promise as a candidate trait marker, atleast in those areas studied to date.

Atypical scanning behaviours are a widely replicated andapparently robust finding in individuals with schizophreniaand are not specific to face stimuli. Scanpath measures showexciting levels of sensitivity and specificity in distinguishingcase from control groups and may address the problem ofpoor diagnostic specificity posed by other candidate trait

markers. In parallel with this, current findings on scanpaths inindividuals with BPAD suggest that these measures may alsoinclude sufficient information to illuminate overlap in geneticvulnerability or clinical features between related disorders.

There is some suggestion that restricted scanning may belinked with the negative symptoms of schizophrenia, or alter-natively with delusional symptoms, at least with regard to faceviewing. This may point to the possibility of scanpaths as aspecific marker for particular symptomatologies that may helpto stratify subtypes or indicate state. Some studies have re-ported state-related changes in scanpath behaviour in patientswith schizophrenia. However, there is evidence that viewingbehaviours still remain atypical to some extent despite symp-tom remission, supporting the argument that scanpath deficitsare likely to be a constant feature in these individuals.

Genetic control of scanpath abnormalities is supported byinitial findings of linkage analysis and relative studies.

Additional research is clearly merited to further define thestate and trait characteristics of atypical scanning in individ-uals with schizophrenia, particularly with regard to the herit-ability and genetic control of the deficit.

The present review has not discussed cosegregation ofscanpath abnormalities in the families of patients with schizo-phrenia, as studies in this area have not yet been reported.Studies of scanpath behaviours in relatives of individualswith schizophrenia are not entirely unequivocal. Substantiveinvestigation of familial patterns of scanpath abnormalitieswill be necessary before such measures may be defined astrait markers.

Standardization in scanpath methodology across studies isrequired. Although scanpath deficits appear reasonably ro-bust to detect differences in task and stimuli between researchgroups, variance in the selection and defitiition of dependentmeasures to date create difficulties in comparability betweenstudies. Such investigation should also provide further infor-mation on the psychometric properties of scanpath dysfunc-tion such as test-retest statistics, which are not yet available,and separate meta-analyses will be usefitl in delineating theimpact in sources of variation in methodology across studies.

Arguably, the most useful biological markers should bewell understood mechanisms that will augment understand-ing of the disease.'" However, little is understood about theunderlying etiology of atypical scanning in individuals withschizophrenia at present. Patterns of visual exploration mayreflect a more complex interplay of neurobiologie mechan-isms and cognitive functions compared with the more dis-crete mechanisms tapped by tasks such as smooth pursuittracking or inliibition of the P50 auditory evoked response.""That is not to say, however, that the impaired mechanism re-flected by the restricted scanning phenomenon must also beaffected by such variables. It may be that restricted scanningrepresents an anomaly in a low-level oculomotor or visuo-cognitive mechanism. A number of lines of inquiry havepointed to a role of the frontal lobes, and scanpath abnor-malities appear to comprise a generalized deficit in visualbehaviour. Additional impacts of cognitive and emotionaldeficits on scanpath behaviours have also been demon-strated in face viewing studies. Further exploration of the

J Psychiatry Neurosci 2011;36(3) 161

Beedie et al.

underlying neuropathologic and cognitive etiologies of scan-path abnormalities will clarify how scanpath abnormalitiescan further inform our understanding of schizophrenia. Al-though it could be suggested that scanpath deficits reflectprefrontal control mechanisms that are common to otherpsychiatric disorders, the relative specificity of restrictedscanning styles to schizophrenia suggests that this is not thecase. Clearly, further consideration of the cognitive andneurologic pathologies underlying the scanpath deficit is es-sential to build a biological explanation of the specific physio-logic mechanisms assessed here. Deficits in the processing ofreal-world visual information, as measured by the scanpathparadigm, may be highly informative with regard to cogni-tive or neurologic function and subjective experience in indi-viduals with schizophrenia. Indeed, this phenomenon ap-pears to merit further consideration in its own right, besidesany potential role in genetics or novel intervention studies.

Understanding potential association with specific geneswill require further linkage studies or candidate gene de-signs. More work is required in relatives of individuals withschizophrenia. Furthermore, understanding whether thosedeficits tapped by scanpath methodologies represent a causallink between genetics and diagnosable pathology is likely torequire longitudinal work and intervention studies.

Limitations

neuroleptic medication. It is possible that scanpath dysfunc-tion may represent a stable, heritable trait marker of diseasevulnerability, which is exaggerated with deterioration in clin-ical state. Further, evidence for a strong association withschizophrenia and high levels of sensitivity and specificitywarrants further investigation of scanpath behaviours as atrait marker for schizophrenia and suggests that such meas-ures may constitute a promising avenue of research to under-stand the pathophysiology of the disease. At the least, scan-paths may address the limited specificity of other moretraditional biomarker targets and serve as an invaluable ad-dition to alternative measures, such as smooth pursuit eyemovements or sensory gating indices, to substantially im-prove the diagnostic specificity of psychophysiologic assess-ment in trait marker research.

Acknowledgements: This review is partly based on a cioctoral disser-tation completed by Dr. Beedie under the supervision of Dr. Bensonand Professor St. Clair. During the preparation of this manuscript,salary support for Dr. Beedie was provicled by the Chief Scientist Of-fice (CZB-4-734, awarded to Dr. Benson) and previously by the Euro-pean Commission (SGENE, awarded to Professor St. Clair).

Competing interests: As above.

Contributors: Drs. Beedie and Benson acquired the data and wrotethe article. All authors designed the review, analyzed the data, re-viewed the article and approved its publication.

This review is unable to give a definitive answer as to whichtasks (e.g., feature search, free viewing) or stimuli are themost appropriate for measuring scanpath dysfunction in in-dividuals with schizophrenia or whether it matters. Mitigat-ing cognitive and unknown perceptual disturbances may yetinfluence conformation. Until further work is done with fam-ilies of schizophrenia probands, it is difficult to assess howsimilar unaffected family members' scanpaths are to those ofpatients. Whereas there is evidence to suggest that simplespatiotemporal measures of scanpaths during viewing cap-ture some sense of the abnormality, it is not yet clear whetherone of these measures, a combination of these measures or al-ternative higher order statistics will be best suited to discrim-inate between the scanpaths of individuals with schizophre-nia and those of individuals with other illnesses andnonclinical comparison groups. Nevertheless, scanpaths areunarguably affected in individuals with schizophrenia.

Conclusion

Although a number of questions are still to be addressed, eyemovements during scene inspection are emerging as a pow-erful and informative discriminatory tool in the study ofschizophrenia. Some studies have reported state-related fluc-tuation in the degree of scanpath dysfunction present in indi-viduals with schizophrenia. However, evidence of the deñcitin healthy relatives of individuals with schizophrenia andfindings that some degree of dysfunction remains regardlessof symptom remission also suggest that restricted scanningmay pre-empt the clinical manifestation of the disease. Scan-path dysfunction does not appear to be a consequence of

References

1. Williams HJ, Owen MJ, O'Donovan MC. New findings from gen-etic association studies of schizophrenia. / Hum Genet 2009;54:9-14.

2. Lichtenstein P, Yip B, Björk C, et al. Common genetic determinants ofschizophrenia anci bipolar disc:)rder in Swedish families: a popLilation-based study. Lincet 2009;373:234-9.

3. Gottesman II, CoLild TD. The endophenotype concept in psychiatry:etymology and strategic intentions. Am / Psychiatry 2003;160:636-45.

4. Clementz BA, Grove WM, Iacono WG, et al. Smooth-pursuit eyemovement dysfunction and liability for schizophrenia: implica-tions for genetic modeling. / Abnorm Psychol 1992;101:117-29.

5. Calkins ME, Iacono WG, Ones DS. Eye movement dysfunction infirst-degree relatives of patients with schizophrenia: a meta-analyticevaluation of candidate endophenotypes. Brain Cogn 2008;68:436-61.

6. Cadenhead KS, Braff DL. Endophenotyping schizotypy: a preludeto genetic studies within the schizophrenia spectrum. Schizophr Res2002;54:47-57.

7. Ritsner M, Gottesman I. Where do we stand in the quest for neuro-psychiatrie biomarkers and endophenotypes and what next? In:Ritsner M, editor. The handbook of neuropsychiatrie biomarkers,endophenotypes and genes: neuropsychological endophenotypes andbiomarkers. New York (NY): Springer; 2009. p. 3-21.

8. Walters JT, Owen MJ. Endophenotypes in psychiatric genetics. MolPsychiatry 2007;n:SS6-90.

9. Cannon TD, Keller MC. Endophenotypes in the genetic analyses ofmental disorders. Amni Rev Clin Psychot 2006;2:267-90.

10. Turetsky BI, Calkins ME, Light GA, et al. Neurophysiologicalendophenotypes of schizophrenia: the viability of selected candi-date measures. Schizophr Bull 2007;33:69-94.

11. Javitt DC, Spencer KM, Thaker GK, et al. Neurophysiological bio-markers for drug development in schizophrenia. Nat Rev DrugDiscov 2008;7:68-83.

12. Calkins ME, Iacono WG. Eye movement dysfunction in schizo-phrenia: a heritable characteristic for enhancing phenotype defini-tion. Am I Med Genet 2000;97:72-6.

13. Blackwood DH, St Clair DM, Muir WJ, et al. Auditory P300 andeye tracking dysfunction in schizophrenic pedigrees. Arch GenPsychiatry 1991;48:899-909.

162 J Fsychiai :011;36(3)

Atypical scanpaths in schizophrenia

14. Myles-Worslcy M, Ord L, Blailes F, et al. P50 sensory gating inadeilescents from a pacific island isolate with elevated risk forschizophrenia. Biol Psydiiatry 2004;55:663-7.

15. Clementz BA, McDowell JE, Zisook S. Saccadic system functioningamong schizophrenia patients and their first-degree relatives.] Abnorm Psyclwl 1994;103;277-87.

16. Freeman D, Garety P, Phillips M. An examination of hypervigi-lance for external threat in individuals with generalized anxietydisorder and individuals with persecutory delusions using visualscan paths. Q / Exp Psi/dio/ A 2000;53;549-67.

17. Mogg K, Bradley BP, Field M, et al. Eye movements to smoking-related pictures in smokers: relatie-inship between attentional bi-ases and implicit and explicit measures of stimulus valence. Aiidic-tioii 2003;98:825-36.

18. Notem D, Stark I. Eye movements and visual perception. Sei Am1971;224:35-43.

19. Posner MI. Orienting of attention. Q / Exp Psychol 1980;32:3-25.20. Corbetta M, Akbudak E, Conturo T, et al. A commem netweirk of func-

tional areasfor attention and eye movements. Neuron 1998;21:761-73.21. Findlay JM, Walker R. A model of saccade generation based on

parallel processing and competitive inhibition. Bdinv Brain Sei1999;22:661-721.

22. Antes JR. The time course of picture viewing. / Exp Psyeliol 1974;103:62-70.

23. Henderson JM, Hollingworth A. High-level scene perception.Aniiii Rei< Psi/dio/1999;50:243-71.

24. Henderson JM, Human gaze control during real-world scene per-ception. Treniis Cogn Sei 2003;7:498-501.

25. Just MA, Carpenter PA. Eye fixations and cognitive processes.Cognit Psyeiioi 1976;8:441-80,

26. Krieger G, Rentschier I, Hauske G, et al. Object and scene analysisby saccadic eye movements: an investigation with higher-orderstatistics, Spal Vis 2000;13:20M4.

27. Lofthus G, Macworth N. Cognitive determinants of fixation locationduring picture viewing. / Exp Psyehol Hum Pereept Perform 1978;4:565-72,

28. Molnar F, About the role of visual exploration in aesthetics. In:Day HI, editor. Advanees in intrinsie motivation and aestheties. NewYork (NY): Plenum Press; 1981

29. Rayner K, Eye movements in reading and visual information pro-cessing; 20 years of research, Psyehol Bull 1998;124:372-422,

30. Yarbus A. Ei/e moz>ements and vision. New York (NY): PlenumPress; 1967,

31. Gaymard B, Ploner C, Rivaud-Pechoux S, et al. The frontal eyefield is involved in spatial short-term memory but not in reflexivesaccade inhibition. Exp Brain Res 1999;129:288-301.

32. Fukushima K. Frontal cortical control of smooth-pursuit. CtirrOpin Neurohiol 2003;13:647-54.

33. Burman DD, Segraves MA. Primate frontal eye field activity duringnatural scanning eye movements. / Neuroph\/sioi 1994;71:1266-71.

34. Schmidt D, Krause B, Wiss P, et al. Visuospatial working memoryand changes of the point of view in 3D space. Neuroinm<^c 2007;36:955-68.

35. Mûri RM, Nyffeler T. Neurophysiology and neuroanatomy of re-flexive and volitional saccades as revealed by lesion studies withneurological patients and transcranial magnetic stimulation (TMS).Brain Co¿;i¡ 2008;68:284-92.

36. Benson PJ, Leonards U, Lothian RM, et al. Visual scanpaths infirst-episode schizophrenia and cannabis-induced psychosis./ Psyeliiatry Neurosei 2007;32:267-74.

37. Bestelmeyer PE, Tatler BVV, Phillips LH, et al. Global visual scan-ning abnormalities in schizophrenia and bipolar disorder, SehizophrKi's2006;87:212-22,

38. Green MJ, Williams LM, Davidson D. Visual scanpaths to threat-related faces in deluded schizophrenia. Psychiatry Res 2003;119:271-85,

39. Green MJ, Waldron JH, Simpson I, et al. Visual processing of scKialcontext during mental state perception in schizophrenia, / PsyehiatryNeurosei 2008;33:34-42.

40. Hori Y, Fukuzako H, Sugimoto Y, et al. Eye movements during theRorschach test in schizophrenia, Psyehiatrxj Clin Neurosei 20O2;56:4O9-18.

41. Obayashi S, Matsushima E, Okubo Y, et al. Relationship betweenexploratory eye movements and clinical course in schizophrenicpatients, Eur Areh Psyehiatry Ciin Neurosei 2001;251:211-6,

42. Obayashi S, Matsushima E, Ando H, et al. Exploratory eye movementsduring the Benton Visual Retention Test: characteristics of visual be-

haviour ij-i schizophrei-iia, Psyehiatri/ din Neurosei 2003;57:409-15,43. Takahashi S, Ohtsuki T, Yu S, et al. Significant linkage to chromo-

some 22q for exploratory eye movement dysfunction in schizo-phrenia. Am / Med Genet B Neuropsyehiatr Genet 2003;123B:27-32.

44. Takahashi S, Tanabe E, Yara K, et al. Impairment of exploratory eyemovement in schizophrenia patients and their siblings, PsyehiatryClin Neurosei 2008;62:487-93,

45. Kojima T, Matsushima E, Ohta K, et al. Stability of exploratory eyemovements as a marker of schizophrenia — a WHO multi-centerstudy, Seliizop'hr Res 2001;52:203-13,

46. Kojima T, Matsushima E, Ando K, et al. Exploratory eye move-ments and neuropsychological tests in schizophrenic patients.Sehizophr Bull 1992;18:85-94.

47. Kojima T, Matsushima E, Nakajima K, et al. Eye movements inacute, chronic, and remitted schizophrenics. Biol Psychiatry 1990;27:975-89.

48. Kojima T, Potkin S, Kharazmi M, et al. Limited eye movement pat-terns in chrome schizophremc patients. Psyehiatry Res 1989;28:307-14.

49. Kojima T, Ando K, Ando H, et al. Eye movements as a marker ofschizophrenia. Sehizophr Res 1988;l:178-9,

50. Phillips M, David A, Understanding delusions in schizophreniausing visual scan paths, Biol Psyeliiatry 1996;39:549-50,

51. Loughland CM, Williams LM, Gordon E, Schizophrenia and affectivedisorder show different visual scanning behaviour for faces: A traitversus state-based distinction? Biol Psyehiatry 2002;52:338-48,

52. Loughland CM, Williams LM, Gordon E. Visual scanpaths to posi-tive and negative facial emotions in an outpatient schizophreniasample, Sehizophr Res 2002;55:159-70.

53. Manor BR, Gordon E, Williams LM, et al. Eye movements reflect im-paired face processing in patients with schizophrenia, Biol Psyehiatry1999;46:963-9,

54. Matsushima E, Kojima T, Ohbayashi S, et al. Exploratory eyemovements in schizophrenic patients and patients with frontallobe lesions, Eur Areh Psyehiatry Clin Neurosei 1992;241:210-4,

55. Mikami T, Naruse N, Fukura Y, et al. Determining vulerability toschizophrenia in methamphetamine psychosis using exploratoryeye movements. Psyehiatry Clin Neurosei 2003;57:433-40.

56. Minassian A, Granholm È, Verney S, et al. Visual scanning deficitsin schizophrenia and their relationship to executive functioningimpairment, Sehizophr Re.-; 2005;74:69-79,

57. Nishiura S, Morita K, Kurarkake K, et al. Characteristics of left andright scanning in schizophrenia patients using exploratory eyemovements: comparison with healthy subjects. Psychiatry ClinNeurosei 2007;6\:487-94.

58. Rosse RB, Schwartz BL, Johri S, et al. Visual scanning of faces correl-ates with sd-iizophrenia symptomatology. Prog NeuropsyehopliarinaeolBiol Psyehiatry 1998;22:971-9.

59. Ryu H, Morita K, Shoji Y, et al. Abnormal exploratory eye move-ments in schizophrenic patients vs healthy controls. Acta NeurolScand 2001;104:369-76.

60. de Wilde O, Bour L, Dingemans P, et al. Visual scan paths inyoung patients with schizophrenia, healthy siblings and controls.Sehizophr Rt-s 2007;89:362-3.

61. Leonards U, Duhoux S, Rey-Bellet P, et al. Do disturbances in socialinteraction influence visual scene exploration in psychotic patients?[abstract] Perception 2002;31(Suppl);182.

62. Loughland CM, Williams LM, Harris AW, Visual scanpath dys-function in first-degree relatives of schizophrenia probands: Evi-dence for a vulnerability marker? Sehizopin- Res 2004;67:ll-2L

63. Matsukawa Y, Takahashi S, Aoki M, et al. Patients with systemiclupus erythematosus show a normal responsive search score inexploratory eye movement analysis: comparison with schizophrenia,Ann Rheum Dis 2002;61:748-50.

64. Moriya H, Anod K, Kojima T, et al. Eye movements during percep-tion of pictures in chronic schizophrenia, Psyehiatry Clin Neurosei1972;26:189-99,

65. Phillips ML, David AS. Visual scan paths are abnormal in deludedschizophrenics. Neuropsyeliologia 1997;35:99-105.

66. Phillips ML, Senior C, David AS. Perception of threat in schizo-phrenics with persecutory delusions: an investigation using visualscan paths. Psyehol Med 2000;30:157-67.

67. Williams LM, Loughland CM, Green MJ, et al. Emotion perceptionin schizophrenia: an eye movement study comparing the effective-ness of risperidone vs. haloperidol. Psyehiatry Res 2003;120:13-27.

68. Williams LM, Loughland CM, Gordon E, et al. Visual scanpaths in

J Psychiatry Neurosci 2011;36(3) 163

Beedie et al.

schizophrenia: Is there a deficit in face recognition? Schizophr Res1999;40:189-99.

69. Xia M, Takahashi S, Tanabe E, et al. Eye movement studies onschizophrenics and their parents. Eur Neuropsychopharmacol 1996;6(Suppl 3):64.

70. Beedie SA, St.Clair DM, Rtijescu DP, et al. Smooth pursuit and vi-sual scanpaths: Related or independent deficits in schizophrenia?Schizophr Res 2010;l 17:247-8.

71. Benson P, Beedie SA, Giegling I, et al. A classification model ofschizophrenia using a multivariate eye movement phenotype [ab-stract]. Eur Psyehiatry 2010;5(Suppl l):6n.

72. Benson PJ, Beedie SA, Giegling I, et al. Multivariate eye movementpsychophysiology accurately differentiates schizophrenia casesfrom unaffected controls. Schizophr Res 2010;l 17:248-9.

73. Tsunoda M, Kurachi M, Yuasa S, et al. Scanning eye movementsinschizophrenic patients: relationship to clinical symptoms and re-gional cerebral blood flow using 123I-IMP SPECT. Schizophr Res1992;7:159-68.

74. Tsunoda M, Kawasaki Y, Matsui M, et al. Relationship between ex-ploratory eye movements and brain morphology in schizophreniaspectrum patients. Eur Arch Psychiatry Clin Neurosci 2005;255:104-10.

75. Luria A, Karpov B, Yarbuss À. Disturbances of active visual per-ception with lesions of the frontal lobes. Cortex l%6;2:202-12.

76. Beedie S, St.Clair D, Rujescu D, et al. Frontal brain function andvisual exploration of natural scenes in schizophrenia [abstract].Eur Psychiatry 201Ü;25(Suppl l):1104.

77. Roberts ]K. [Jrain structure and function in the schizophrenias: aneurobehavioural approach. Psyciiiatr I Univ Ott 1983;8:67-80.

78. Schmid A, Rees G, Frith C, et al. An fMRI study of anticipationand learning in smooth pursuit eye movements in humans.Aíi'iírori'fwrf 2001;12:1409-14.

79. Pierrot-Deseilligny C, Mûri R, Nyffeler T, et al. Tlie role of the humandorsolateral prefrontal cortex in ocular motor behavior. Ann N Y AcaiiSt-í2005;1039;239-51.

80. Burke MR, Barnes GR. Brain and behaviour: a task-dependent eyemovement study. Ccreb Cortex 20Ü8;18:126-35.

81. Gooding EXT, lacono WG, Hanson DR. Smooth pursuit and saccadiceye movement performance in a prefrontal leukotomy patient./ Psychiatry Neurosci 1999;24:462-7.

82. Heide W, Kurzidim K, Kompf D. Deficits of smooth pursuit eyemovements after frontal and parietal lesions. Brain 1996;119:1951-69.

83. Gaebel W, Ulrich G, Frick K. Visuomotor performance of schizo-phrenic patients and normal controls in a picture viewing task.Biol Psychiatry 1987;22:1227-37.

84. Kurachi M, Matsui M, Kiba K, et al. Limited visual search on theWAIS picture completion test in patients with schizophrenia. SchizophrRes 1994;12:75-80.

85. Matsushima E, Kojima T, Ohta K, et al. Exploratory eye movementdysfunction in patients with schizophrenia: possibility as a dis-criminator for schizophrenia. / Psychiatr Res 1998;32:289-95.

86. Phillips ML, David AS. A cognitive neuropsychiatrie approach tothe study of delusions in schizophrenia using visual scan paths.Schizophr Res 1996;18:216-7.

87. Phillips ML, David AS. Abnormal perception of simple andchimeric facial stimuli in schizophrenia: investigation of under-lying attentional mechanisms using visual scan paths. SchizophrRes 1997;24:120-l.

88. Phillips ML, David AS. Attention to threat in schizophrenia: invesH-gation of the cognitive processes underlying paranoia using visualscan paths. Schiza^thr Res 1997;24:121.

89. Phillips ML, David AS. Viewing strategies for simple and chimericfaces: an investigation of perceptual bias in normals and schizo-phrenic patíents using visual scan paths. Brain Cogn 1997;35:225-38.

90. Phillips ML, David AS. Abnormal visual scan paths: a psycho-physiological marker of delusions in schizophrenia. Schizophr Res1998;29:235-45.

91. Phillips ML, Senior C, David AS. Investigation of the cognitiveprocesses underlying paranoia using visual scan paths. SchizophrRes 1998;29:61.

92. Quirk SW, Strauss ME. Visual exploration of emotion eliciting im-

ages by patients with schizophrenia. / Nero Ment Dis 2001 ;189:757-65.93. Schwartz BL, Rosse RB, Johri S, et al. Visual scanning of facial ex-

pressions in schizophrenia. / Neuropsychiatre Clin Neurosci 1999;11:103-6.

94. Streit M, Wölwer W, Gaebel W. Facial-affect recognition and visualscanning behaviour in the course of schizophrenia. Schizophr Res1997;24:311-7.

95. Beedie S, Shephard E, Giegling I, et al. Visual scanpaths as a gener-alised deficit in schizopluenia ¡abstract]. Perception 2010;30(Suppl):34.

96. Adler LE, Freedman R, Ross RG, et al. Elementary phenotypes in theneurobiological and genetic study of schizophrenia. Biol Psychiatry1999:46:8-18.

97. Horley K, Williams L, Gonsalvez C, et al. Face to face: visual scan-path evidence for abnormal processing of facial expressions in socialphobia. Psychiatry Res 2004;127:43-53.

98. Horley K, Williams L, Gonsalvez C, et al. Social phobies do not seeeye to eye: a visual scanpath study of emotional expression pro-cessing. / Anxiety Disord 2003;17:33-44.

99. Dalton KM, Nacewicz BM, Johnstone T, et al. Gaze fixation and theneural circuitry oí face processing in autism. Nat Neurosci 2005;8:519-26.

100. Klin A, Jones W, Schultz R, et al. Defining and quantifying the socialphenotype in autism. Am / Psychiatry 2002;l59:895-908.

101. Klin A, Jones W, Schultz R, et al. Visual fixation patterns duringviewing of naturalistic siKial situatic>ns as predictors of social compe-tence in individuals with autism. Arch Gen Psyclnatry 20()2;59:809-16.

102. Pelphry KA, Sasson NJ, Reznick JS, et al. Visual scanning of facesin autism. / Autism Dev Disord 2002;32:249-61.

103. Rutherford MD, Towns AM. Scan path differences and similaritiesduring emotion perception in those with and without autism spec-trum disorders. / Autism Dev Disord 2008;38:137l-8l.

104. Karatekiii C, Asarnow J. Exploratory eye movements to pictures inchildhocid-onset schizophrenia and attention-deficit hyperactivitydisorder. / Abnorm Chiid Psychoi 1999;27:35-49.

105. Shephard S, Beedie SA, Kuriakose J, et al. Perseverative eye move-ments in obsessive-compulsive disorder and schizophrenia [abstract].PtTa'pi/o/! 2010;39(Suppl):37.

106. Mosimann UP, Felblinger J, Ballinari P, et al. Visual explorationbehaviour during clock reading in Alzheimer's disease. Bruin 2004;127:431-8.

107. Morgan GJ, Huddy V, Lipton M, et al. Is persistent ketamine use avalid mc^del of the cognitive and oculomcitor deficits in schizo-phrenia? Biol Psychiatry 2009;65:1099-102.

108. Neumann D, Spezio M, Piven J, et al. Looking you in the mouth:abnormal gaze in autism resulting from impaired top-down mod-ulation of visual attention. Soc Cogn Affect Neurosci 2006;l:194-202.

109. Spezio ML, Adolphs R, Hurley RS, et al. Analysis of face gaze inautism using "Bubbles." Neiiropsycholoifia 2007;45:144-5I.

110. van der Geest JN, Kemner C, Camfferman G, et al. Looking at im-ages with human figures: comparison between autistic and normalchildren. / Autism Dev Disord 2002;32:69-75.

111. Marsh PJ, Williams LM. ADHD and schizophrenia phenomenol-ogy: Visual scanpaths to emotional faces as a potential psycho-physiological marker? Neurosci Biobehav Rev 2006;30:651-65.

112. Craddock N, O'Donovan MG, Owen MJ. Psychosis genetics: mod-eling the relationship between schizophrenia, bipolar disorder,and mixed (or "schizoaffective") psychoses. Schiznphr Bull 2009;35:482-90.

113. International Schizophrenia Consortium; Purcell SM, Wray NR,Stone JL, et al. Common polygenic variation contributes to risk ofschizophrenia and bipolar disorder. Nature 2009;460:748-52.

114. Kay SR, Fiszbein A, Opler LA. The positive and negative syndromescale (PANSS) for schizophrenia. Schizophr Butt 1987;13:261-76.

115. Andreasen NC. The Scale for the Assessment of Negative Symptoms(SANS): conceptual and theoretical foundations. Br j Psychiatry Suppl1989;(7):40-58.

116. Overall JE, Gorham DR. The Brief Psychiatric Rating Scale. Psi/cholRi-pl962;10:790-812.

117. Liddle PF. The symptoms of chronic schizohrenia. A re-examinationof the positive-negative dichotomy. Br / Psychiatry 1987;151:145-51.

164 _ J Psychiatry Neurosci 2011;36(3)

Copyright of Journal of Psychiatry & Neuroscience is the property of Canadian Medical Association and its

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.