Face Recognition Accuracy, Response Time and Visual Search Strategies of Adolescents with and without Autism Spectrum Disorders

Background:  The development of face recognition was theorised to continue to improve from childhood and adulthood, suggesting different face processing styles along the developmental trajectory. Although several studies have investigated face recognition ability of individuals with autism spectrum disorders (ASD), evidence of the differences between individuals with and without ASD in face recognition accuracy has been inconsistent. It is possible that the inconsistent evidence can be attributed to the differences in face recognition abilities along the developmental trajectory. However, few had investigated face recognition in ASD from a developmental perspective.

Objectives:  The current study replicated the methodology of previous studies involving children and adults with ASD. The aim of the study was to compare accuracy, response time and visual search strategies, i.e., number of fixations and fixation duration during recognition of unfamiliar faces in adolescents with and without ASD.

Methods:  The current study recruited 28 adolescents with ASD and 30 matched typically developing (TD) peers. Participants viewed of 12 pairs of face stimuli cut into puzzle pieces (encoding); followed by a face recognition phase. Half of the puzzle pieces were presented with the eyes bisected and the other half with eyes as a whole. The accuracy and response time in face recognition tasks were recorded. Measurements of visual search strategies were recorded using a remote eye tracker.

Results:  Adolescents with ASD demonstrated increased difficulty in face recognition compared to their TD counterparts, despite showing similar response time. TD adolescents showed increased accuracy when the eyes were whole but decreased when eyes were bisected. Adolescents with ASD demonstrated no differences in accuracy in the two eye conditions. Adolescents with ASD were less likely to derive effective processing using the ‘face information triangle’, as reduced scanning to the eyes but increased on other areas of the face were observed.

Conclusions:  When compared with the studies involving children and adults, this study highlighted that face recognition differences between individuals with and without ASD first appear during adolescence. This provided evidence on the importance of the consideration of face recognition abilities in ASD from a developmental perspective.