Exploring the Construct of Social Attention

Background: Autism spectrum disorders often entail atypicalities in attention directed to social stimuli in the real world. However ‘social attention’ is a poorly defined term, used differently in various areas of psychology. If we are to understand the crucial role of social attentional impairments in the development of autism we must examine this construct in more detail. This examination should incorporate multiple measures of attention and types of social information and must also acknowledge individual differences within a diagnostic group.
Objectives: To examine the nature of social attention in people with and without autism, using cross-task analyses to investigate both an underlying social attention construct and individual and group differences in its manifestation.
Methods: Three experiments employing different methods (content analysis, change detection and eye-tracking), all produced measures of attention to social information (e.g. people, and particularly eyes).  Comparable measures were incorporated into an analysis of group and individual response patterns across all three tasks, assessing attention to social information in the visual domain.
Results: Though sample sizes were small, statistical techniques including principal components analysis, cluster analysis and a case study revealed an absence of relationships between different measures of social attention.

Conclusions: This report presents an important first step in investigating the complex nature of social attention. The assumption of a single social attention construct should continue to be questioned; social information varies across tasks just as attentional processes do. In future, it may be useful to consider a hierarchy of social information, each level demanding different degrees of attention. This hierarchy is likely to be flexible, varying between individuals and across situations. This conclusion has significant consequences for our understanding of social impairments in autism.