Abstract
Infant scanning patterns within a face predict recognition of own-race faces in infants (Gaither et al., 2012). The present study sought to determine if infants' visual exploration strategies relate to their ability to discriminate faces within own- and other-race groups. Eye tracking measures were recorded while 6-9 month old infants completed a visual paired comparison task that measured discrimination of faces within four racial groups (Caucasian, Asian, African American, and Hispanic). For each trial, infants were familiarized to a single face from one of the groups presented simultaneously on the left and right side of the screen for 15s (Figure 1). Next, the familiarized face appeared alongside a novel same-race face for 10s. Novelty preference was calculated by dividing the duration fixating the novel face by the total duration fixating either face. The number of visual transitions during familiarization within each of the two images (eye to mouth region (1)) and between identical areas across the two images (eye to eye region (2) and mouth to mouth region (3); Figure 1) were calculated. We regressed novelty preference on participant race (Caucasian or biracial), age group (6 or 9 months), visual transitions within the face, eye to eye transitions, and mouth to mouth transitions. The model for Caucasian faces was significant, F(5,13)= 5.13, p=0.02. Number of transitions between the eye regions of each face predicted the novelty preference at test (p = 0.007). As novelty preference increased, the number of eye to eye transitions decreased (Figure 2). The model for other-race faces was not significant (see Fig.2). These results suggest that infants use different visual strategies when comparing own- and other-race faces, and that comparisons between the eye regions of two identical faces predict own-race face discrimination.
Meeting abstract presented at VSS 2017