More is not always better: Audiovisual speech hampers infants' face recognition

Abstract

By 9 months, infants recognize own-race faces from static images, but struggle when familiarized with naturalistic audiovisual (AV) talking faces, despite daily exposure to dynamic, speaking faces. This study examined whether AV speech or AV mouth chewing motion disrupts face recognition and whether selective attention to the mouth or cognitive load mediates this effect. We recorded the eye gaze of 10-month-olds (N = 133) and 20-month-olds (N = 125) during a Visual Paired Comparison task in three conditions: Talking (AV, speech, dynamic), Chewing (AV, no speech, dynamic), and Static (no AV, no speech, static). Infants focused more on the mouth in Chewing, the eyes in Static, and equally on both in Talking. However, recognition occurred only in the Static and Chewing conditions, independently of age. These findings suggest that AV speech, not just motion, increases cognitive load and disrupts face memorization regardless of selective attention, impacting early face processing.