Group Differences in Feature Scanning While Learning Novel Faces

Background:  

Past research has suggested that the way typical individuals look at a familiar face is measurably different from how they look at a novel face, as measured by eye gaze patterns. It has also been suggested that individuals with autism spectrum disorder (ASD) do not show differential eye gaze patterns when looking at novel versus familiar faces such as that of a family member.

Objectives:  

To investigate group differences between individuals with ASD and control participants across 17 exposures as novel faces (and as a control novel houses) become familiar.

Methods:  

Twelve participants with high-functioning ASD or Asperger’s (9 males; Mean age = 28.08 years, SD= 6.29) and 16 typical participants (14 males, Mean age = 27.44, SD = 6.76) passively viewed 17 unique images of 6 individuals and 6 houses while eye gaze information was collected via eye tracking technology. Specifically we measured changes in the number of fixations and total fixation duration within two areas of interest (eyes and mouth for faces; upper and lower feature for houses). Participants completed separate blocks of upright and inverted faces and houses.

Results:  

Both groups showed evidence of learning for both faces and houses; eye gaze patterns for both groups changed systematically with increased exposures. Interestingly, the effect of exposures was not significantly different between groups demonstrating that the process of learning novel faces and houses was similar in both groups.  Analysis of mean number of fixations and total fixation duration per exposure revealed significant group differences:  the participants with ASD showed no differences in eye gaze patterns for the eyes and mouth areas of the face in both upright and inverted faces. In contrast, the typical group showed a focus on eyes compared to the mouth and this difference was more evident in the inverted faces compared to the upright faces. There were no group differences in the effects of time and mean gaze patterns for upright or inverted houses, indicating that group differences in learning complex stimuli are specific to social stimuli.

Conclusions:  

The process of novel faces becoming familiar appears to be similar in individuals with ASD and typical individuals.  The areas of the faces that individuals focus on differ between groups and this difference is even more evident for inverted faces, for which learning is a more complex social cognitive task. These group differences are specific to complex social stimuli.