Abstract
Young adults generally recognize own-age faces more accurately than other-age faces, but less is known about the perceptual mechanisms underlying this bias. With the hypothesis that visual scanning strategies may be at the origins of the advantage for own-age faces, we monitored eye movements while young adults (N=19) performed an old/new recognition memory task with adult and infant faces. After a learning phase, where 24 faces (12 adult, 12 infant) were randomly presented (3 sec/each), participants had to recognize the familiar identities among 24 distractors. Eye-movements were recorded throughout the task. Adults used a more conservative strategy (c) in responding to adult compared to infant faces (p< .01) and made more false alarms (FA) for infant compared to adult faces (p< .02). In both the learning and the recognition phases the proportion of looking time on the eye region of interest (ROI) was greater for adult compared to infant faces (p< .02); conversely the proportion of time spent looking at the cheeks and forehead ROI was greater for infant compared to adult faces (p< .02). Dwells (transitions between ROIs) were larger for adult than for infant faces (p< .01), suggesting the use of a more dynamic scanning strategy for the former than for the latter. The proportion of looking time on the eye ROI positively correlated with response performance (d’; p< .05). Overall, results are in line with earlier demonstrations that both the familiarity of identity (Heisz & Shore, 2008) and the familiarity of face category (Goldinger, He, & Papesh, 2009; Wu, Laeng, & Magnussen, 2012; but see Blais, Jack, Scheepers, Fiset, & Caldara, 2008) affects the scanning behavior in adult participants. Critically, our correlational results are the first to indicate that the eyes are a diagnostic feature for identity recognition.
Meeting abstract presented at VSS 2015