Covering the eyes or mouth of a speaker does not prevent word learning in typically developing infants

Infant gaze-following skills and selective attention to the mouth of talking faces during their first year correlate with their vocabulary growth in their second year. This correlational evidence has led to the hypothesis that these attentional strategies are key mechanisms supporting word acquisition, a process which drastically improves during toddlerhood. If so, covering the eyes or mouth region of a talking face that teaches toddlers novel word-object associations should disrupt or diminish their word learning performance. To test this relationship causally, we recorded eye gaze in 17- to 42-month-old toddlers (N = 153) while performing an audiovisual fast-mapping task in one of three conditions: (a) full face, where the speaker's face was fully visible; (b) glasses, where the eye region was covered with black opaque glasses; and (c) mask, where the mouth and nose were covered with a black surgical mask. The results showed that toddlers aged 24 months and above learned the new word-object associations and, crucially, that they did so independently of condition. Moreover, correlational evidence showed that word learning performance related with toddlers' gaze-following behavior (i.e., target-object looking and face-target shifts) but not with attention to the eyes or mouth. These findings suggest that the capacity for object word-form association from a quick audiovisual interaction emerges around 24 months and that toddlers' optimal attentional strategy at this stage involves social understanding and object exploration rather than focusing on speakers' eyes or mouth. The implications of these results are discussed, particularly in the context of naturalistic vocabulary learning environments. (PsycInfo Database Record (c) 2026 APA, all rights reserved).