Anxiety Disorders 17 (2003) 33±44
Social phobics do not see eye to eye: A visual scanpath study of emotional expression processing Kaye Horleya,b,*, Lea M. Williamsa,c, Craig Gonsalvezb, Evian Gordona,c b
a The Brain Dynamics Centre, Westmead Hospital, Westmead, NSW 2145, Australia Department of Psychology, University of Wollongong, Wollongong, NSW 2522, Australia c Department of Psychology, University of Sydney, Camperdown, NSW 2006, Australia
Received 16 January 2001; accepted 12 June 2001
Abstract Clinical observation suggests that social phobia is characterised by eye avoidance in social interaction, re¯ecting an exaggerated social sensitivity. These reports are consistent with cognitive models of social phobia that emphasize the role of interpersonal processing biases. Yet, these observations have not been veri®ed empirically, nor has the psychophysiological basis of eye avoidance been examined. This is the ®rst study to use an objective psychophysiological marker of visual attention (the visual scanpath) to examine directly how social phobia subjects process interpersonal (facial expression) stimuli. An infra-red corneal re¯ection technique was used to record visual scanpaths in response to neutral, happy and sad face stimuli in 15 subjects with social phobia, and 15 age and sexmatched normal controls. The social phobia subjects showed an avoidance of facial features, particularly the eyes, but extensive scanning of non-features, compared with the controls. These ®ndings suggest that attentional strategies for the active avoidance of salient facial features are an important marker of interpersonal cues in social phobia. Visual scanpath evidence may, therefore, have important implications for clinical intervention. # 2002 Elsevier Science Inc. All rights reserved. Keywords: Social phobia; Anxiety; Face processing; Emotion; Eye movements; Visual scanpaths
* Corresponding author. Tel.: 61-2-9845-6835; fax: 61-2-9635-7734. E-mail address:
[email protected] (K. Horley).
0887-6185/02/$ ± see front matter # 2002 Elsevier Science Inc. All rights reserved. PII: S 0 8 8 7 - 6 1 8 5 ( 0 2 ) 0 0 1 8 0 - 9
34
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
1. Introduction Social phobia is an anxiety disorder that is characterised by fear of negative evaluation in social situations. One of the most striking observations in clinical studies of social phobia is the avoidance of eye contact in social interactions È hman, 1986) that may be a consequence of these (Greist, 1995; Marks, 1969; O fears. Since Darwin (1872/1955), psycho-evolutionary research has shown that the eyes are the most fear-inducing feature in situations of social appraisal by È hman, 1986). others (O Cognitive models of anxiety (Beck & Emery, 1985) provide a useful conceptual framework for understanding the likely link between fear of negative evaluation and active avoidance of salient facial features (such as the eyes) in social phobia. These models propose that socially phobic individuals have a selective bias towards processing information that contains a potential threat to oneself, affecting normal attentional strategies and the perception of social stimuli and interactions. Evidence from cognitive paradigms, including dichotic listening and modi®ed Stroop, is consistent with the notion of biases towards potentially threatening interpersonal information in social phobia (for review, see Logan & Goetsch, 1993). Nevertheless, clinical observations of interpersonal dysfunctions in social phobia, and eye contact avoidance in particular, have not been veri®ed empirically, nor has the psychophysiological basis of eye avoidance been examined. This was the ®rst study to use an objective marker of visual attention (the visual scanpath) to examine directly how individuals with social phobia process interpersonal (facial expression) stimuli, compared to healthy subjects. Accurate social perception involves the integration of several basic components, and the ability to perceive emotional expressions in others is a particularly important component. Both the perception and expression of facial affect provides the basis for interpersonal communication. Face and facial emotion processing relies on the integrity of visuomotor and visuospatial information processing systems (Bruce & Young, 1986). It is through these systems that we acquire emotional information, and signal social involvement, empathy, control, dominance and interpersonal adequacy. Argyle (1983, p. 80) has highlighted the ``central importance'' of eye movements in social interaction, both to receive and to send information. In this study, visual scanpath recording provided information on both eye movement and eye ®xation parameters. Visual scanpath studies have shown that healthy subjects produce a regular pattern of eye movement and ®xations to face stimuli. Subjects focus in particular on the salient facial features of eyes, nose and mouth, producing scanpaths that represent an inverted triangle in shape (Mertens, Siegmund, & GruÈsser, 1993; Walker-Smith, Gale, & Findlay, 1977). Of these features, the greatest attention is usually paid to the eyes, as the most revealing source of information about emotional expression in social interactions (Lundqvist, È hman, 1999). Esteves, & O
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
35
Visual scanpaths to face stimuli have been investigated only in schizophrenia. Our group (Gordon et al., 1992; Manor et al., 1999; Williams, Loughland, Gordon, & Davidson, 1999) and others (Ishizuka, Kashiwakurra, & Oija, 1998; Streit, Wolwer, & Gaebel, 1997) have shown that patients with schizophrenia show ``restricted'' scanpaths to facial stimuli, re¯ected in few foveal ®xations, a longer ®xation duration and a reduced scanpath length. Schizophrenia subjects also display a reduced attention (few ®xations, shorter ®xation duration) to salient facial features (Gordon et al., 1992; Phillips & David, 1997; Williams et al., 1999). The isolated visual scanpath studies of Post-Traumatic Stress Disorder (PTSD) and Generalised Anxiety Disorder (GAD) have examined other threat-relevant stimuli. PTSD subjects have been found to initially ®xate on threat words more than controls, re¯ecting preferential processing of threatening material (Bryant, Harvey, Gordon, & Barry, 1995). In contrast, GAD subjects have not shown aberrant scanpaths in response to ambiguous, potentially threat-related social scenes (Freeman, Garety, & Phillips, 2000). Evidence from both clinical and cognitive studies suggests the scanpaths of individuals with social phobia (as a distinct anxiety disorder) would show a marked avoidance of salient features, particularly the eyes, but not necessarily the more global scanpath differences (e.g., in scanpath length) shown by schizophrenia subjects. We took into account reports that the avoidance of eye gaze also occurs in depression (Shea, Glenn, & Heefner, 1995), autism (Hobson & Lee, 1998), and mental retardation (Elias-Burger, Sigelman, Danley, & Burger, 1981). In this study it was proposed that individuals with social phobia are more likely to avoid the eyes of others because of fear (particularly of negative evaluation) rather than because of a generally ¯attened affect (depression) or a more fundamental cognitive disorder (autism, mental retardation). Social phobia subjects were, therefore, assessed for depression and screened for mental retardation. We predicted that subjects with social phobia would show signi®cantly less attention (fewer ®xations, shorter duration) to eye features compared to healthy controls, but that there would be no between-group differences in other scanpath parameters, such as length of scanpath. 2. Method 2.1. Subjects Fifteen subjects (eight males and seven females, mean age 38:27 years; S:D: 13:74 years; range 21±64) with generalised social phobia were recruited from hospital anxiety clinics. Diagnostic information was derived from independent clinical psychologist evaluations, based on the Structured Clinical Interview DSM-IV (First, Spitzer, Gibbon, & Williams, 1995) and two psychometric indices: the Social Phobia Anxiety Inventory (Turner, Biedel,
36
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
Dancu & Stanley, 1989), and the Fear of Negative Evaluation (Watson & Friend, 1969). Diagnosis of social phobia was con®rmed according to DSMIV (American Psychiatric Association, 1994) criteria. The degree of concurrent depression was assessed using the revised Beck Depression Inventory (BDI-II; Beck, Steer, & Brown, 1996). Nine social phobia subjects were on medication, including four on selective serotonin reuptake inhibitors (sertroline, paroxetine, ¯oxetine), three on reversible inhibitor of monoamine oxidase (aurax-moclobemide), and one each on benzodiazepine (alprazolan) and tricyclic antidepressants (clomipramine). Fifteen healthy control subjects were age (mean 38:4 years; S:D: 12:99 years; range 21±65) and sex-matched to the clinical subjects. Exclusion criteria for all subjects were a recent history of substance abuse, or past history of substance dependence, epilepsy, other neurological or psychopathological disorders, mental retardation or head injury, left-handedness, or ocular pathology (assessed using the Westmead Hospital Clinical Information Base (WHCIB); Williams, Gordon, Bahramali, Wright, & Meares, 2000) that incorporates sections from the Composite International Diagnostic Interview (CIDI) (Robins et al., 1988), and the Edinburgh Handedness Inventory (Old®eld, 1971). All subjects were asked to refrain from smoking or drinking caffeine for 3 h before the recording session, and alcohol for the preceding 24 h. 2.2. Apparatus and stimuli Face stimuli were colour-digitised photographs of the same individual depicting neutral, happy and sad expressions. These stimuli were associated with the highest inter-rater reliability scores in Mazurski and Bond's (1993) standardised series of facial affect slides. The images were equiluminant and subtended 11:3 14:9 of visual angle. Eye movements were recorded using a computerised infra-red eye gaze monitoring system (CEDRIC Mark II) linked to a host computer (486 DX4-100 MHz) and an auxillary video display unit for observing the monitored eye. Retinal and corneal re¯ections produced by the infra-red light were recorded by the camera every 20 ms to obtain ®xation points (error of resolution < :5 ). 2.3. Procedure Subjects were seated in a reclining chair in a sound attenuated laboratory. A soft head restraint minimised head movements and ensured a distance of 60 cm between the subject and the centre of the stimulus display screen. A calibration procedure for eye ®xation preceded presentation. Presentation of each stimulus, according to software developed by our group, was initiated only when the subject maintained a ®xation on a central dot area (.758) for 1 s, ensuring that the initial direction of retinal attention was controlled in all subjects. Subjects were instructed to ®xate on the centre of the screen
Table 1 Means and standard deviations (S.D.) for eye movement parameters Eye movement parameters
Control group, mean (S.D.)
l. Total number of ®xations (>200 ms) 2. Total ®xation duration (ms) 3. Median ®xation duration (ms) 4. Raw scanpath length (cedric coordinates)a 5. Fixation scanpath length (cedric coordinates) 6. Index of ®xation duration to features (eyes, nose, mouth) 7. Index of ®xations to eyes (>200 ms) 8. Index of ®xation duration to eyes versus nose versus mouth (ms) a
Happy
Sad
Neutral
Happy
Sad
19.00 (3.14)
17.00 (2.88)
18.00 (2.96)
13.25 (4.37)
15.50 (5.45)
14.83 (3.44)
8834.25 (803.47)
8687.14 (1052.11)
8755.71 (929.62)
6426.67 (2671.74)
7125.0 (1971.29)
7205.0 (1915.11)
425.00 (92.80)
440.71 (100.95)
426.43 (71.53)
355.0 (100.86)
430.0 (172.89)
431.67 (165.96)
130.57 (34.70)
147.89 (51.14)
141.96 (45.67)
207.14 (126.55)
141.96 (45.67)
213.08 (128.60)
57.28 (21.66)
41.32 (11.21)
51.56 (13.82)
36.17 (19.31)
36.83 (18.27)
40.99 (15.13)
6992.86 (1840.35)
7387.14 (1727.62)
6784.29 (1964.69)
5308.33 (2550.22)
5286.67 (3248.87)
5498.33 (3213.92)
4.14 (2.66)
3.21 (2.52)
4.56 (2.77)
2.14 (2.54)
3.5 (3.63)
2.07 (2.76)
2240.0 (1795.22)
1844.29 (1668.887)
2318.57 (1449.66)
1000.0 (1299.47)
1425.71 (1545.13)
991.43 (1436.7)
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
Neutral
Social phobia group, mean (S.D.)
1 cedric coordinate approximately .28.
37
38
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
until a picture appeared, and then to look at the picture in any manner they chose. Neutral, happy and sad faces were each presented for 10 s with an interstimulus interval of 15 s. 2.4. Data analysis A ®xation was de®ned as a set of consecutive gaze coordinates, con®ned within a diameter of 18 of visual ®eld for a minimum duration of 200 ms (Noton & Stark, 1971). Eye movement data in which there was a loss of tracking integrity were excluded. Blinks identi®ed by a loss of corneal re¯ection were automatically excluded from subsequent data analysis, as were ``off screen'' gazes. Eye movement parameters (outlined in Table 1) included both temporal (number of ®xations, total ®xation duration, median ®xation duration) and spatial (®xation scanpath length, raw scanpath length1) indices. Spatio-temporal indices were calculated to examine eye ®xations to feature versus non-feature regions. Feature areas were de®ned as eyes: 2:17 2:48 2 ; nose: 1:24 4:34 ; and mouth: 3:10 1:86 . Feature masks were individually tailored to re¯ect feature or non-feature face regions, resulting in each ®xation ``tagged'' to re¯ect the feature space in which it occurred. Prior to analysis, outlier data for each parameter, greater or less than one and a half times the interquartile range from the upper or lower quartile, were removed from analysis and the nearest value inserted (Tabachnik & Fidell, 1996, p. 69). Outlier data constituted .8% of the total control group eye movement data and 3% of social phobia data. The entire data for one subject (and their matched control) was excluded due to loss of scanpath recording. To test the prediction that social phobia would be associated with face-speci®c visual scanpath disturbances, repeated measures MANOVAs were employed, with temporal and spatial eye movement parameters as the dependent measure. The between-subjects factor was group (control vs. social phobia) and the withinsubjects factor was facial affect (with three levels: neutral, happy and sad). Two additional parallel sets of MANOVAs were conducted with spatiotemporal (feature vs. non-feature) parameters to examine group disturbances in attention to salient facial features with affect (neutral, happy, sad) as the withinsubjects factor. These analyses were ®rst conducted for combined features (eyes, nose and mouth), and subsequently for the speci®c eye region alone. 3. Results The descriptive data for both social phobia and control groups on each eye movement parameter for the three stimuli are shown in Table 1. 1 Raw scanpath length was included to ensure that group differences in ®xation scanpath length were not due simply to group differences in number of ®xations.
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
39
3.1. Facial affect MANOVA results supported indications from Table 1 that social phobia subjects exhibit a ``hyperscanning'' strategy, characterised by a reduced number of ®xations and total duration, and an increased raw scanpath length compared to controls. There was a signi®cant group by affect interaction [F
2; 48 4:12, P < :05] for number of ®xations. Planned between-group contrasts showed that this interaction was due to the signi®cantly fewer ®xations made by social phobia subjects to both neutral [F
1; 26 15:93, P < :001] and sad [F
1; 25 5:51, P < :05], but not happy faces. This pattern of reduced ®xation in the social phobia group would also account for the signi®cant group main effect for number of ®xations [F
1; 24 8:50, P < :01]. For total ®xation duration, the two-way interaction was not signi®cant, but there was a signi®cant group main effect [F
1; 24 10:60, P < :01]. This effect was due to the signi®cant reduction in total ®xation duration across all stimuli for the social phobia group compared to controls [neutral, F
1; 26 12:68, P < :001; happy, F
1; 24 6:62, P < :05; sad, F
1; 25 8:66, P < :01]. The signi®cant group by affect interaction for ®xation scanpath [F
2; 48 4:68, P < :05], was explained by the decreased scanpath length in social phobia subjects for the neutral face [F
1; 26 7:46, P < :01], but not for sad or happy (affect) stimuli. This interaction would explain the additional main effects for affect [F
2; 48 4:88, P < :05] and for group [F
1; 24 4:27, P < :05] for this parameter. For raw scanpath length the group main effect was signi®cant, [F
1; 24 5:36, P < :05], but there was no two-way interaction. In contrast to the group pattern for ®xation scanpath, the social phobia group exhibited an increased raw scanpath length compared to controls for both the neutral [F
1; 24 6:69, P < :05] and sad faces [F
1; 25 5:01, P < :05], with a similar trend for the happy face [F
1; 24 3:93, P < :06]. 3.2. Feature analysis Feature analysis (with combined eyes, nose and mouth regions) revealed a signi®cant group main effect for number of ®xations [F
1; 24 4:0, P < :05]. Planned contrasts showed that the social phobia group, compared to controls, made fewer ®xations to salient features for the neutral [F
1; 26 4:42, P < :05] and sad face [F
1; 26 4:26, P < :05], with a similar trend towards signi®cance for the happy face [F
1; 26 3:85, P < :06]. There was also a signi®cant group main effect for total ®xation duration [F
1; 22 4:77, P < :05]. Social phobia subjects showed a reduced ®xation duration compared to controls for both neutral [F
1; 26 6:31, P < :05] and happy [F
1; 26 5:46, P < :05], but not sad faces. This ®nding
40
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
Fig. 1. Example scanpaths for neutral, happy and sad faces in a social phobic subject (1a±c) and control subject (2a±c) (dot size indicates number of fixations).
most probably explains the signi®cant main effect for affect [F
3; 66 36:07, P < :001]. Analysis of the speci®c eye region revealed a signi®cant group by affect interaction [F
2; 52 4:28, P < :05]. Planned contrasts showed that this effect was explained by the social phobia subjects' comparative reduction in the number of ®xations to the eyes for the sad face in particular [F
1; 26 5:73, P < :05]. Typical scanpaths to neutral, happy and sad faces are depicted in Fig. 1a±c for a social phobia subject, and Fig. 2a±c for a control subject. 3.3. Relationship between scanpath parameters and clinical variables Chi-square analyses revealed no signi®cant associations between scanpath parameters and either medication (selective serotonin reuptake inhibitors, reversible inhibitor of monoamine oxidase, benzodiazepine, and tricyclic antidepressants) or level of depression (re¯ected in BDI-II score).
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
41
4. Discussion This study provided the ®rst direct, empirical investigation of face processing disturbances in social phobia. The results suggest that individuals with social phobia show a distinctive ``hyperscanning'' strategy for processing faces, characterised by a lack of ®xations, both in number and duration, and a markedly increased raw scanpath length. This strategy was more prominent for negative (sad) and neutral expressions, than for positive (happy) expressions. Social phobia subjects also tended to avoid ®xating on salient facial features (eyes, nose, mouth), and avoidance of the eye region was most apparent for sad faces. Of the speci®c differences that emerged for facial features, the social phobic group showed a speci®c avoidance of eyes, particularly for the sad face, re¯ected in fewer ®xations. This ®nding is in accord with previous research indicating that it is the eyes in particular that signal social threat in facial expressions (Darwin, È hman, 1986). Avoidance of the feared situation is a common coping 1872/1955; O strategy, as it reduces the threat for socially anxious individuals. Avoidance of salient features, particularly the eyes, suggests that scanning problems in social phobia have a basis in fear of potential threat contained in faces. Our ®nding that social phobia disturbances in eye avoidance were most apparent for the sad face, compared to the happy face, is consistent with the notion that fear of potential social evaluation is behind these problems. Although previous research indicates that socially anxious people construe facial expressions negatively (Pozo, Carver, Wellens, & Scheier, 1991; Winton, Clark, & Edelmann, 1995), the ®nding in this study most likely re¯ects cognisance by the social phobic group of the less intrinsic threatening aspects associated with a positive affect. Our visual scanpath ®ndings offer empirical con®rmation for the clinical observation that individuals with social phobia tend to avoid eye contact during social interaction. Cognitive models that emphasise a bias towards potentially threatening information in social phobia, may also offer an interpretative framework for these ®ndings. Salient facial features (particularly the eyes) may convey the potential threat of social evaluation to individuals with social phobia. The engagement of a hyperscanning strategy in which features are avoided, but non-features are extensively scanned, may therefore represent an adaptive coping strategy for dealing with an exaggerated sensitivity to the implied threat in the faces of others. The observation of hyperscanning in this study of social phobia was distinct from the type of de®cit revealed in eye movement studies of face stimuli in other disorders, pointing again to the likely speci®c nature of the social phobia problem. For example, patients with schizophrenia show a diametrically opposed pattern of restricted scanpaths. While the number of disorder studies is limited, these preliminary ®ndings point to a speci®c dysfunction in social phobia that is feasibly due to a distinct cause, namely the fear and hypervigilance associated with potential sources of social evaluation.The combination of long raw scanpath length and brief ®xations observed in social phobia is consistent
42
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
with the notion of hypervigilance to face stimuli and avoidance because of evaluative fears, with a consequent reliance on gestalt (holistic) rather than detailed (sequential) processing. By contrast, the restricted scanpaths and lengthy ®xations of schizophrenia subjects may re¯ect an abnormal reliance on sequential processing, due to a fundamental de®cit in forming integrated gestalts of socially meaningful stimuli (Frith, Stevens, Johnstone, Owens, & Crow, 1983; Williams et al., 1999). Hypervigilance, assessed using behavioural signs, has been cited as a common feature of anxiety disorders (Beck & Emery, 1985; Eysenck, 1992). Therefore, one might expect to observe a hyperscanning strategy in anxiety disorders other than social phobia. Yet, while not one study has used face stimuli, the previous GAD study of social scene stimuli (designed to be of concern to GAD subjects) did not observe hyperscanning (Freeman et al., 2000). Several possibilities arise from these null ®ndings. First, the anxiety disorder of social phobia may have a distinctive hyperscanning dysfunction that re¯ects a speci®c and focused fear of social evaluation, and is not apparent in other anxiety disorders. From this viewpoint, the precise expression of hypervigilance may differ across different forms of anxiety disordersÐfor example, GAD may be manifested in hypervigilance to many varied stimuli that are individually determined, rather than a focused concern with the potential threat of facial contact (but, these suggestions require future studies testing several disorders within a single study and several different types of stimuli). Second, the null results in the Freeman et al. (2000) study could have been due to methodological differences. Since numbers in the study were limited, there may or may not have been suf®cient power to reveal a similar hyperscanning strategy to our study. But, more importantly, half of the GAD subjects had a comorbid depression, and another half had a comorbid second anxiety disorder, that may have confounded the ®ndings. As depression was examined in our study and no association was found with depression, level of concurrent depression in social phobia is not likely to account for our ®ndings. Visual scanpath evidence of avoidance processing may, therefore, objectify a well-known characteristic of social phobia, and provide speci®c relevance for treatment interventions focused on interpersonal strategies. Effective interpersonal skills rely on the decoding of both verbal and non-verbal information. If individuals with social phobia engage in avoidance of signi®cant features, as suggested by their scanpath patterns, it is likely that they are further disadvantaged in social situations by a decreased ability in obtaining the signi®cant information that is necessary for accurate interpretation of social situations. These ®ndings provide the ®rst preliminary empirical evidence that social phobia is associated with speci®c visuo-cognitive disturbances in processing interpersonal stimuli, such as faces. The preliminary data suggests that these disturbances may be particularly apparent for faces depicting negative expression, and might re¯ect a hypervigilance to potential sources of negative social assessment. Evidence from neuroimaging research points to distinct neural substrates
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
43
for different facial emotions (George et al., 1995). Future studies might seek to further explore the association between social phobia and visual face processing, and the possible biological bases using both eye movement and neuroimaging techniques. The use of additional face stimuli and other anxiety disorder groups would also serve to elucidate both the stimulus and diagnostic speci®city of our preliminary ®ndings. References American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). Washington, DC: Author. Argyle, M. (1983). The psychology of interpersonal behaviour (4th ed.). Middlesex, UK: Penguin Books. Beck, A. T., & Emery, G. (1985). Anxiety disorders and phobias: a cognitive perspective. New York: Basic Books. Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory-II. San Antonio, TX: Psychological Corporation. Bruce, V., & Young, A. (1986). Understanding face recognition. British Journal of Psychology, 77, 305±327. Bryant, R. A., Harvey, A. G., Gordon, E., & Barry, R. J. (1995). Eye movement and electrodermal responses to threat stimuli in Post-Traumatic Stress Disorder. International Journal of Psychophysiology, 20, 209±213. Darwin, C. (1872/1955). The expression of the emotion in man and animals. London: John Murray. Elias-Burger, S. F., Sigelman, C. K., Danley, W. E., & Burger, D. L. (1981). Teaching interview skills to mentally retarded persons. American Journal of Mental De®ciency, 85(6), 655±657. Eysenck, M. W. (1992). Anxiety: the cognitive perspective. Hove, UK: Erlbaum. First, M. B., Spitzer, R. L., Gibbon, M., & Williams, J. B. W. (1995). Structured Clinical Interview of DSM-IV. New York: Biometrics Research, New York State Psychiatric Disorder. Freeman, D., Garety, P. A., & Phillips, M. L. (2000). An examination of hypervigilance for external threat in individuals with Generalised Anxiety Disorder and individuals with persecutory delusions using visual scanpaths. Human Experimental Psychology, 53(2), 549±567. Frith, C. D., Stevens, M., Johnstone, E. C., Owens, D. G., & Crow, T. J. (1983). Integration of schematic faces and other complex objects in schizophrenia. Journal of Nervous and Mental Disease, 171, 34±39. George, M. S., Ketter, T. A., Parekh, P. I., Horwitz, B., Herscovitch, P., & Post, R. M. (1995). Brain activity during transient sadness and happiness in healthy women. American Journal of Psychiatry, 152, 341±351. Gordon, E., Coyle, S., Anderson, J., Healey, P., Cordaro, J., Latimer, C., & Meares, R. (1992). Eye movement response to a facial stimulus in schizophrenia. Biological Psychiatry, 31, 626±629. Greist, J. H. (1995). The diagnosis of social phobia. Journal of Clinical Psychiatry, 56(Suppl. 5), 5±12. Hobson, R. P., & Lee, A. (1998). Hello and goodbye: a study of social engagement in autism. Journal of Autism & Developmental Disorders, 28(2), 117±127. Ishizuka, K., Kashiwakurra, M., & Oiji, A. (1998). Eye movements in patients with schizophrenia: visual stimuli, semantic content and psychiatric symptoms. Acta Psychiatrica Scandinavica, 97(5), 364±367. Logan, A. C., & Goetsch, V. L. (1993). Attention to external threat cues in anxiety states. Clinical Psychology Review, 13, 541±559. È hman, A. (1999). The face of wrath: critical features for conveying Lundqvist, D., Esteves, F., & O facial threat. Cognition & Emotion, 13(6), 691±711.
44
K. Horley et al. / Anxiety Disorders 17 (2003) 33±44
Manor, B. R., Gordon, E., Williams, L. M., Rennie, C. J., Bahramali, H., Latimer, C. R., Barry, R. J., & Meares, R. A. (1999). Eye movements re¯ect impaired face processing in patients with schizophrenia. Biological Psychiatry, 46(7), 963±969. Marks, I. (1969). Fear and phobias. London: Heinemann Medical Books. Mazurski, E. J., & Bond, N. W. (1993). A new series of slides depicting facial expressions of affect: a comparison with the pictures of facial affect series. Australian Journal of Psychology, 45, 41±47. Mertens, I., Siegmund, H., & GruÈsser, O.-J. (1993). Gaze motor asymmetries in the perception of faces during a memory task. Neuropsychologia, 31, 989±998. Noton, D., & Stark, I. (1971). Eye movements and visual perception. Scienti®c American, 224, 35±43. È hman, A. (1986). Face the beast and fear the animal: animal and social fears as prototypes for O evolutionary analyses of emotion. Psychophysiology, 23(2), 123±145. Old®eld, R. C. (1971). The assessment and analysis of handedness. The Edinburgh Inventory. Neuropsychologia, 9, 97±113. Phillips, M. L., & David, A. S. (1997). Abnormal visual scanpaths: a psychophysiological marker of delusions in schizophrenia. Schizophrenia Research, 29, 235±245. Pozo, C., Carver, C. S., Wellens, A. R., & Scheier, M. F. (1991). Social anxiety and social perception: construing others reactions to self. Personality and Social Psychology Bulletin, 17, 355±362. Robins, L. N., Wing, J., Wittchen, H. U., Helzer, J. E., Babor, T. F., Burke, J., Farmer, A., Jablenski, A., Pickens, R., Reiger, D. A., Sartorius, N., & Towle, L. H. (1988). The Composite International Diagnostic Interview. Archives of General Psychiatry, 45(12), 1069±1077. Shea, G. D., Glenn, D., & Heefner, A. S. (1995). Depression, interpersonal style, and communication skills. Journal of Nervous and Mental Disease, 183(7), 485±487. Streit, M., Wolwer, W., & Gaebel, W. (1997). Facial-affect recognition and visual scanning behaviour in the course of schizophrenia. Schizophrenia Research, 24, 311±317. Tabachnik, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.). New York: Harper Collins. Turner, S. M., Beidel, D. C., Dancu, C. V., & Stanley, M. A. (1989). An empirically derived inventory to measure social fears and anxiety: the Social Phobia Anxiety Inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychology, 1, 35±40. Walker-Smith, G. J., Gale, A. G., & Findlay, J. M. (1977). Eye movement strategies involved in face perception. Perception, 6, 313±326. Watson, D., & Friend, R. (1969). Measurement of social evaluative anxiety. Journal of Consulting and Clinical Psychology, 33, 448±457. Williams, L. M., Gordon, E., Bahramali, H., Wright, J., & Meares, R. (2000). Late component ERPs are associated with distinct schizophrenic syndromes. International Journal of Neuroscience, 105, 37±52. Williams, L. M., Loughland, C. M., Gordon, E., & Davidson, D. (1999). Visual scanpaths in schizophrenia: is there a de®cit in face recognition? Schizophrenia Research, 40, 189±199. Winton, E. C., Clark, D. M., & Edelmann, R. J. (1995). Social anxiety, fear of negative evaluation and the detection of negative emotion in others. Behaviour Research and Therapy, 33(2), 193±196.