Chloé Galinier¹ (), Justin Duncan¹, Caroline Blais¹, Daniel Fiset¹; ¹Université du Québec en Outaouais

The other-race effect refers to a recognition disadvantage for other-race compared to own-race faces. Although perceptual and social factors are well-studied, attentional mechanisms are often overlooked. Recent findings using behavioral measures (Duncan et al., VSS 2022) reveal that own-race faces, unlike other-race faces, are recognized automatically. To pinpoint this effect at the electrophysiological level, the exact same dual-task paradigm was used while recording EEG. Twenty-nine White participants completed a dual-task. In each block, they memorized a pair of White (own-race) or East-Asian (other-race) faces. They then performed a tone (Target 1; T1) categorization task followed by a delayed face (Target 2; T2) recognition task whereby they were asked whether T2 more closely matched the left or right face of the memory set. T1 and T2 presentations were separated by a stimulus onset asynchrony (SOA: 150, 300, 600, 1,200ms) to control potential conflict for central attention resources. T2 difficulty was varied by presenting either full signal faces (i.e., 0% Identity 1 - 100% Identity 2), or morphed faces (i.e., 60% Identity 1 - 40% Identity 2).Behavioral results confirmed previous findings (Duncan et al., VSS 2022), showing greater automatization for own-race versus other-race face recognition. For own-race faces, task difficulty effects diminished at shorter SOAs, suggesting cognitive slack absorption and perceptual handling of difficulty. Conversely, for other-race faces, difficulty effects remained consistent across SOAs, indicating post-perceptual processing. All electrophysiological components showed attentional modulation, with the P300 component revealing a marginal interaction between attentional modulation and race. Specifically, the P300 component, reflecting the allocation of limited-capacity attentional resources, showed higher amplitudes for own-race faces under shorter SOA conditions. These findings suggest that the P300 could be a potential locus for the behavioral effect, i.e., the automatization of own-race face recognition.