Two are better than one: Infant language learning from video improves in the presence of peers

Abstract

Studies show that young children learn new phonemes and words from humans significantly better than from machines. However, it is not clear why learning from video is ineffective or what might be done to improve learning from a screen. The present study, conducted with 9-month-old infants, utilized a manipulation-touch screen video-which allowed infants to control presentations of foreign-language video clips. We tested the hypothesis that infant learning from a screen would be enhanced in the presence of a peer, as opposed to learning alone. Brain measures of phonetic learning and detailed analyses of interaction during learning confirm the hypothesis that social partners enhance learning, even from screens.

Keywords: infancy; language learning; phoenemic discrimination; screen media; social learning.

Fig. 1.

Examples of the individual- (A) and paired- (B) exposure sessions.

Fig. 2.

Differences between individual and paired groups were significant at both electrode sites: Fz, F(1, 29) = 6.059, P = 0.02, and Cz: F(1, 29) = 10.754, P = 0.003. Note that the ERP is plotted with negative voltage upward.

Fig. 3.

Among infants in the paired baby condition, those who were partnered with more babies showed the strongest evidence of learning. In paired sessions, infants with more partners had a greater ERP mismatch negativity, r_s = 0.68, P < 0.05, which is evidence of greater foreign language phoneme learning.