Social functioning, systemising ability and emotion recognition in autism spectrum disorders

Background: Autism spectrum disorders (ASD) are characterised by a triad of impairments afflicting social interaction, verbal communication and imagination. Accordingly, individuals with ASD are found to have marked difficulties in empathising with others, which includes the ability to recognise emotional states. However, recent evidence suggests that alongside these deficits may exist strengths in the ability to systemise: to deal with concepts that have definitive rules, such as maths, physics and computers. This enhanced ability to systemise may also provide a compensatory mechanism in the social realm, by way of rule-based learning about emotional states. However, the relationship between severity of functioning in ASD, systematic abilities and how this may influence emotion recognition given different types of stimuli remains unclear. This study examined emotion recognition in human faces in addition to emotion in non-human faces. These non-human faces were stylised ‘caricatures’ of emotion (e.g. a smiley ‘emoticon’). We reasoned that if those with ASD were using systemising as a compensatory mechanism for emotion recognition, they may have preserved or enhanced non-human emotion recognition.

Objectives: To test the relationship between severity of functioning, systemising strengths and performance on an emotion recognition task with human and non-human stimuli in individuals with and without ASD.

Methods: 21 children with ASD and a comparable number of children without ASD aged 11-15 were recruited from local specialist and mainstream schools respectively. The groups were matched for age, verbal IQ and sex. All the children completed the short-version Social Responsiveness Scale (SRS-short), the Systemising Quotient (SQ), the Empathising Quotient (EQ) and an emotion recognition task with human and non-human stimuli.

Results: Severity of social functioning, as measured by the SRS-short and EQ, showed significant correlations with overall performance on the emotion task for both the ASD and control groups, in the direction expected. However, another finding was a significant positive correlation between SQ and emotion task scores for the ASD group. Additionally, the ASD group performed equally as well as controls on the emotion task. Further analysis showed that these results may be attributable to reduced, although non-significant, performance across all human stimuli and significantly better performance across all non-human stimuli in the ASD group. Performance was unrelated to verbal IQ.

Conclusions: Overall, the findings suggest that both severity of functioning and systemising ability in ASD is related to the ability to recognise emotions. Whilst controls show difficulties in recognition of non-human stimuli, individuals with ASD do not show the same detrimental effect. This may be explained by a reduced salience to human stimuli and/or compensatory mechanisms in processing features by the ASD group, which in turn may be related to systemising ability.