Avoidance of emotionally arousing stimuli predicts social–perceptual impairment in Asperger's syndrome

https://doi.org/10.1016/j.neuropsychologia.2007.08.005Get rights and content

Abstract

We combined eye-tracking technology with a test of facial affect recognition and a measure of self-reported social anxiety in order to explore the aetiology of social–perceptual deficits in Asperger's syndrome (AS). Compared to controls matched for age, IQ and visual–perceptual ability, we found a group of AS adults was impaired in their recognition of fearful and sad expressions and spent significantly less time fixating the eye region of all faces. For AS subjects, but not controls, the extent of the failure to fixate the eyes predicted the degree of impairment at recognising fearful expressions. In addition, poor fear recognition and reduced fixation of the eyes were independently associated with greater levels of social anxiety in AS individuals. These findings support the hypothesis that avoidance of emotionally arousing stimuli, such as eyes, contributes to social–perceptual impairment in AS. Furthermore, our findings are consistent with theories implicating amygdala-mediated over-arousal and anxiety in the development of these social–perceptual deficits.

Introduction

Deficits in social reciprocity and communication are defining features of autistic spectrum disorders (ASD), including Asperger's syndrome (American Psychiatry Association, 1994). For example, individuals with an ASD often have significant difficulty with social interactions; they can be unaware of social norms and may struggle to interpret social cues (Volkmar et al., 1994). In the laboratory, impairments are apparent on numerous tests designed to tax social perceptive and social-cognitive abilities, especially those involving face processing. Examples include problems recognising facial identity (e.g. Boucher & Lewis, 1992; Hobson, Ouston, & Lee, 1988; Klin et al., 1999) or emotion (Baron-Cohen, Wheelwright, Hill, Raste, & Plumb, 2001a; Baron-Cohen, Wheelwright, & Jolliffe, 1997; Baron-Cohen, Wheelwright, Skinner, Martin, & Clubley, 2001b), as well as more abstract reasoning about others’ thoughts, beliefs and desires (e.g. Happe, 1994; Castelli, Frith, Happe, & Frith, 2002). Several of these neuropsychological impairments are reminiscent of those seen following iatrogenic or degenerative amygdala lesions, supporting the theory that amygdala anomalies could be involved in the aetiology of a range of symptoms characteristic of autistic spectrum disorders (Baron-Cohen et al., 2000).

A primary consequence of amygdala lesion in humans is impairment in recognising fear, and to a lesser extent sadness, anger and disgust, in facial expressions of emotion (Adolphs et al., 1999). The question of whether individuals with an ASD also show a fear recognition deficit has been a controversial one, with both positive (Howard et al., 2000, Pelphrey et al., 2002) and negative findings (Adolphs, Sears, & Piven, 2001; Castelli, 2005; Grossman, Klin, Carter, & Volkmar, 2000). However, many of these previous studies have suffered from a small sample size (for example, five ASD subjects in Pelphrey et al., 2002, seven in Adolphs et al., 2001 and 10 in Howard et al., 2000). In addition, previous investigations have not attempted to match or control within the design for group differences in non-verbal IQ or basic visual perceptive ability, factors which could feasibly affect performance on this visually based face processing task. Therefore, an initial objective of our experiment was to address these problems in a new study of facial emotion recognition in ASD, using a sample of high-functioning adults with AS who were well-matched to a control group in terms of age, verbal and non-verbal IQ and basic visual perceptive ability. This experiment comprised the first stage of our investigation.

A number of recent studies have shown that individuals with an ASD attend to faces abnormally. In particular, they spend less time fixating the eyes than control groups (Dalton et al., 2005, Klin et al., 2002; Nacewicz et al., 2006, Pelphrey et al., 2002, but see van der Geest, Kemner, Verbaten, & van Engeland, 2002 and Bar-Haim, Shulman, Lamy, & Reuveni, 2006). Using an eye-tracking paradigm with SM, a patient with bilateral amygdala lesion, Adolphs et al. (2005) has shown that SM's failure to fixate the eyes is likely to be the mechanism of her pervasive fear recognition impairment. This is consistent with other work showing that the eyes play a critical role in normal recognition of fear in facial expressions (Kohler et al., 2004; Smith, Cottrell, Gosselin, & Schyns, 2005). Considering these findings together, we predicted there would be a link between eye fixation and emotion recognition in ASD, with those individuals who spend the least time fixating the eyes having the lowest fear recognition scores. This investigation comprised the second part of our study.

A link between low eye fixation and poor fear recognition ability in AS would be consistent with theoretical accounts suggesting that reduced attendance of socially meaningful aspects of the environment, such as eyes, could impair the ability of AS individuals to accrue social knowledge, thereby leading to social–perceptual and social-cognitive deficits (Grelotti, Gauthier, & Schultz, 2002; Schultz, 2005; Schultz, Romanski, & Tsatsanis, 2000). Amygdala dysfunction has been cited as a potential cause of this reduced attendance of social stimuli—with at least two rival theories of how this could occur. In ‘hypo-active amygdala’ models, the amygdala fails to flag social stimuli as meaningful with the result that they do not receive preferential attention (Grelotti et al., 2002, Schultz, 2005, Schultz et al., 2000). In ‘hyper-active amygdala’ models, social stimuli are thought to cause an aversive over-arousal, with the result that they are actively avoided (Dalton et al., 2005, Nacewicz et al., 2006). There is some recent evidence which supports the hyper-active model: eye fixation, for example, is associated with greater amygdale activity in high-functioning autistics but not controls (Dalton et al., 2005), implying a heightened emotional response to gaze fixation in autism. However, no study has yet attempted to link a behavioural measure of the anxiety or distress experienced by ASD individuals in social situations to their level of attention to social stimuli or to their social–perceptual performance. In the final part of our study we aimed to measure participants’ self-reported social anxiety levels and to examine whether higher anxiety can predict a failure to fixate the eyes and poorer recognition of fearful faces.

In summary, we combined eye-tracking of visual scan path with a test of facial affect recognition and a measure of self-reported social anxiety in order to test the following three hypotheses: (i) when compared with well-matched groups, adults with AS, like patients with an amygdala lesion, are likely to be impaired at recognising fearful faces, (ii) within-group differences in the ability of AS adults to do this task will be correlated with the degree to which they fixate the eye region of faces, (iii) both of these factors (failure to recognise fearful faces and non-fixation of eye regions) will be associated with heightened social anxiety, as predicted by the ‘hyper-active amygdala’ model of ASD.

Section snippets

Participants

Twenty-one adults with a clinical diagnosis of AS were recruited from local ASD support groups and via advertisement in the National Autistic Society newsletter. AS diagnosis was from a UK psychiatrist or psychologist and was independently confirmed via the autism diagnostic observation schedule (ADOS, Lord et al., 2000); to be included, an individual had to score above cut-off for an autistic spectrum disorder in terms of both reciprocal social interaction and communication subscales. Ratings

Part 1—facial affect recognition

Fig. 1 summarises the Ekman–Friesen test results as a box-plot. The data were not normally distributed and could not be normalised by transformation; therefore Mann–Whitney tests are used to compare group differences. Correction for multiple comparisons was done via the Bonferroni–Holm method (Hochberg & Tamhane, 1987). AS subjects performed significantly worse than controls at recognising fearful (U = 124, p < .05, Bonferroni–Holm corrected) and sad (U = 104.5, p < .05, corrected) facial expressions.

Discussion

As predicted, individuals with AS were impaired at recognising fearful faces compared to controls. Like patients with amygdala damage, this deficit extended to other negative emotions—there was a significant effect for sad faces and a trend towards an effect for angry faces. These data replicate studies by Howard et al. (2000) and Pelphrey et al. (2002), who both found a fear recognition impairment in adults with high-functioning autism or AS, as well as trends toward an impairment for

Acknowledgement

The authors were supported by a grant from the Wellcome Trust.

References (54)

  • D.G. Amaral et al.

    The amygdala, autism and anxiety

  • American Psychiatry Association

    Diagnostic and statistical manual

    (1994)
  • S. Baron-Cohen

    How to build a baby that can read minds—Cognitive mechanisms in mindreading

    Cahiers de Psychologie Cognitive-Current Psychology of Cognition

    (1994)
  • S. Baron-Cohen et al.

    The “Reading the Mind in the Eyes” test revised version: A study with normal adults, and adults with Asperger syndrome or high-functioning autism

    Journal of Child Psychology and Psychiatry and Allied Disciplines

    (2001)
  • S. Baron-Cohen et al.

    Is there a “language of the eyes”? Evidence from normal adults, and adults with autism or Asperger syndrome

    Visual Cognition

    (1997)
  • S. Baron-Cohen et al.

    The Autism-spectrum quotient (AQ): Evidence from Asperger syndrome/high-functioning autism, males and females, scientists and mathematicians

    Journal of Autism and Developmental Disorders

    (2001)
  • Y. Bar-Haim et al.

    Attention to eyes and mouth in high-functioning children with autism

    Journal of Autism and Developmental Disorders

    (2006)
  • J. Boucher et al.

    Unfamiliar face recognition in relatively able autistic-children

    Journal of Child Psychology and Psychiatry and Allied Disciplines

    (1992)
  • F. Castelli

    Understanding emotions from standardized facial expressions in autism and normal development

    Autism

    (2005)
  • F. Castelli et al.

    Autism, Asperger syndrome and brain mechanisms for the attribution of mental states to animated shapes

    Brain

    (2002)
  • D.P. Crowne et al.

    A new scale of social desirability independent of psychopathology

    Journal of Consulting Psychology

    (1960)
  • K.M. Dalton et al.

    Gaze fixation and the neural circuitry of face processing in autism

    Nature Neuroscience

    (2005)
  • P. Ekman et al.

    Pictures of facial affect

    (1976)
  • D.J. Grelotti et al.

    Social interest and the development of cortical face specialization: What autism teaches us about face processing

    Developmental Psychobiology

    (2002)
  • J.B. Grossman et al.

    Verbal bias in recognition of facial emotions in children with Asperger syndrome

    Journal of Child Psychology and Psychiatry

    (2000)
  • F. Happe

    An advanced test of theory of mind: Understanding of story characters’ feelings and thoughts by able autistic, mentally handicapped, and normal children and adults

    Journal of Autism and Developmental Disorders

    (1994)
  • F. Happe et al.

    The neuropsychology of autism

    Brain

    (1996)
  • Cited by (226)

    View all citing articles on Scopus
    View full text