Face Recognition In 5-Year-Olds with ASD: An Investigation of Identity, Featural and Configural Changes

Background:   Relative to typically-developing children, children with ASD exhibit impairments in face processing (e.g., Wolf et al., 2008), and attend less frequently to eyes (Klin et al. 2002). It has been suggested that an overall local bias to visual attention (Happe & Frith, 2006) and/or a general deficit in processing visual configural information might account for these face recognition impairments (Behrman et al., 2006).

Objectives:   The primary aim of the current project was to understand, relative to controls, whether children with ASD differentially recognize faces when presented with changes in their identity, featural, or configural characteristics. We also plan to examine the effect of  preferential attention toward either eye or mouth regions in these groups.

Methods:  Ten children with ASD (6 boys, mean age: 5:1) and 12 control children (all boys, mean age: 4:6) were matched on non-verbal cognitive ability (p>.10). Face processing was explored using an infrared Tobii X120 eyetracker in a Visual Paired Comparison paradigm. In the Familiarization phase, children saw a human face for 10 seconds. A blank screen and centering stimulus were followed by a Test trial showing two side-by-side faces: one face was from the Familiarization phase and the other was novel. Children saw two trials in three conditions: (1) Novel face at Test had a different Identity than the Familiar face (features and configuration of features differed), (2) the Novel face at Test had the overall identity as the Familiar face, but a key Feature was different and (3) the Novel face had the same identity and features as the Familiar face, but the Configural relationship between features was altered (e.g., the distance between the nose and mouth was shortened).

Results: A novelty preference ratio was calculated for each child in each condition. Children’s scores were included if children looked at least once to each face during test, and for a minimum of two seconds during Familiarization on both trials). Looking more than 50% toward the novel face during Test suggested recognition of the Identity, Featural, or Configural change.  One-sample t-tests compared looking during Test to chance rates of looking (.5). Results revealed that children in the control group preferred to look at the novel face in the Test trial when the novel face differed on Identity or Featural information (t(11)=3.06, p<.05 and t(8)=3.42, p<.01, respectively), but not when configural changes were the defining difference, t(11)=.41, p>.10. Children in the ASD group did not look more than expected by chance at the novel face in any condition (all ps=n.s.). However, there was a tendency toward increased looking at the novel Identity in the ASD group.

Conclusions: Although preliminary, these results appear consistent with previous research indicating differential face processing abilities in children with ASD (e.g., Wolf et al., 2008). With a larger sample size, we also will analyze our gaze tracking data to explore how children’s attention to different parts of faces (i.e., eyes v. mouth) impacts their recognition ability in the three conditions.