26753
Examining Complex Facial Emotion Processing in Typical Development and Autistic Traits: An Investigation of Dimensional Models of Emotion
Impairments in facial emotion recognition (FER) are commonly associated with Autism Spectrum Disorders (ASD). While longstanding research has typically shown this in clinical ASD, emerging evidence suggests that FER impairment may also extend to the broader autism phenotype including non-clinical populations with high autistic-like traits. However, the precise nature of this impairment remains equivocal. It is possible that past investigation of FER in ASD is clouded by the conceptualizations of emotion used in these studies, which rely on discrete conceptualizations or the categorization of emotions based on the valence alone. A dimensional approach, which considers emotions to arise from the intersection of independent neural systems, may provide insights into the mechanisms underlying ASD-linked FER impairment.
Objectives:
To explore the effect of autistic traits on the perception of complex facial emotions to determine what dimensions of processing may drive ASD-linked impairment in emotion processing.
Methods:
A total of 100 typically developing individuals completed the Autism Quotient (AQ) and viewed 52 complex facial emotion stimuli from the Cambridge Mind Reading Face-Voice Battery (CAMs). Participants were required to rate each stimulus based on “How you think the person is feeling” on 9-point Likert scales relating to the valence, arousal, dominance and approach dimensions. To observe the effect of autistic traits on the perception of the stimuli, correlations between AQ score and ratings on each of the four dimensions were examined. As correlations can be conducted only one dimension at a time, further analysis was subsequently conducted to calculate the absolute difference between high and low autistic trait groups within the four dimensional space whereby larger values indicated a greater degree of distance between low and high AQ groups. Individual dimensions were also examined to investigate which dimensions contributed the greatest to this difference.
Results:
Autistic traits were shown to influence the perception of emotion on all four dimensions. Correlation analysis showed that higher autistic traits were associated with more positive valence ratings for emotions labelled as ‘guarded’ and greater approachability ratings for ‘insincere’, ‘confronted’ and ‘appalled emotions’. Higher autistic traits were also associated with lower arousal ratings for emotions labelled as ‘lured’. When examining the absolute difference between high and low AQ groups within the four dimensional space, valence and arousal ratings appeared to contribute to only half of the difference between high and low autistic traits. Differences on dominance and approach dimensions appeared to account for larger differences between high and low autistic trait groups for many of the complex emotions.
Conclusions:
These findings suggest that under-examined dimensions of emotions such as dominance and approach may also underlie ASD-linked FER impairment. There is a need to consider these other underlying dimensions when examining FER impairment in clinical ASD samples to accurately capture potential underlying mechanisms.