Attentional Priority for Special Interests in Autism Spectrum Disorder (ASD) and Neurotypical Passions

Background: Autism Spectrum Disorder (ASD) is characterized by impairments in social interaction. A component of this deficit is lack of expertise for face processing (Grelotti et al., 2002). For typical individuals, faces are prioritized for processing over non-social items however our research previously demonstrated that this is not the case in ASD – where equal preference for social and non-social images is seen (Remington et al., 2012).

Why is that the case? It has been suggested that the specialized face processing system is not innately assigned in typical infants but becomes specific due to strong interest and experience of faces, and therefore has the potential to become specialized for any category (Gauthier et al., 2000). It is therefore striking that autistic individuals often hold an obsessive interest in a specific category. 

Objectives: We aimed to establish whether in autistic people, their category of choice has adopted the expertise system and is afforded prioritized processing resources.

Methods: A selective-attention paradigm was used with autistic adolescents who hold obsessional interests, and neurotypical controls with similar passions/interests. Participants classified a target word based on category (e.g. for trains: ‘Steam or Electric’?). The target (e.g. ‘Bullet Train’) was presented alone or in a list of 1 or 5 non-words, thereby varying the ‘perceptual load’ (amount of potentially task relevant information) of the task. A distractor (photograph of special interest item) was presented next to the words and participants were explicitly told to ignore it. The distractor was either congruent or incongruent (same or alternative sub-category to target). Each participant performed the task with personalized stimuli related to their special interest. All pictures were non-social (i.e. did not contain faces).

By comparing RTs we can establish to what extent distractors are processed under various levels of load, allowing us to index the degree to which a particular stimulus category is prioritized in the information processing system. Load Theory (Lavie, 1995) suggests that distractor processing of non-social items only occurs if perceptual load does not exhaust processing capacity. Conversely social stimuli are afforded a ‘special saliency’ and are processed at all levels of load. Our study examined whether stimuli related to each participant’s area of interest would be processed in a similar manner to other non-social stimuli, or would show a ‘special saliency’ in ASD.

Results: Preliminary results indicate that, as with other non-social items (Lavie et al., 2003; Remington et al., 2012), neurotypical controls showed a congruency effect that was eliminated at high levels of load: greater distractor-impact under lower load levels.  Conversely, for ASD, the congruency effect did not differ across load levels. This indicates that the autistic participants processed the pictures of their special interest to the same degree under all load levels, despite the fact that this slowed their performance on the central task.

Conclusions: This research begins to elucidate a potential link between social and non-social symptoms in ASD. Moreover, the findings will have crucial practical implications for intervention programs that aim to improve social skills of individuals with ASD.