Gesture Development and Maternal Responsiveness between 12-24 Months

Background: Pre-linguistic infants use gestures to communicate. Children’s early gestures also predict subsequent language learning (Rowe & Goldin-Meadow, 2009). One potential mechanism, through which gesture may facilitate language learning is that an infant’s gestures elicit responses from communicative partners, which, in turn, help develop the infant’s language skills (Tamis-LeMonda et al., 2001). For example, when the infant points to a cookie, her mother responds by saying, “You want the cookie,” providing the label for the object. The response then helps the infant to learn the word “cookie.” Despite the important roles gestures play in language learning, surprisingly few studies have directly examined the relations between infant gesture production, maternal responsiveness, and language outcomes (Iverson & Goldin-Meadow, 2005). Here, we studied such relations, specifically in infant siblings of children with ASD, who show deficits in gestural communication (Winder et al., 2012).

Objectives: To investigate whether (1) infant siblings of children with ASD (high-risk; HRA) exhibit differences from low risk infants (low-risk; LRC) in gesture production between 12 and 24 months, (2) mothers of high- and low-risk infants differ in the proportion of infant gestures they respond to, and (3) early gestures, maternal responsiveness, and language outcomes are related.

Methods: 70 mother-child dyads were videotaped in the lab during a 10-minute free-play interaction at 12, 18, and 24 months. The ADOS and best clinical estimate at 18-36 months were used to determine ASD diagnosis (+) or lack thereof (-) and classify infants into outcome groups: HRA+: n = 17; HRA-: n = 25; LRC: n = 29. The MSEL was administered at 36 months, and raw scores from Expressive and Receptive Language subscales were combined to calculate language outcomes. Gestures were reliably coded from videotaped sessions following Özçalışkan and Goldin-Meadow (2009). Maternal contingent responses to infants’ gestures were coded and defined as verbal or nonverbal behaviors that occurred within one utterance of each gesture and involved translation, repetition, or expansion of the gesture.

Results: (1) Kruskal-Wallis tests revealed significant differences between groups in the total number of gestures produced at 12 and 18 months (Figure 1). Follow-up pairwise comparisons showed that HRA+ and LRC groups produced significantly fewer gestures, compared to HRA- infants at 12 months. At 18 months, HRA+ and HRA- groups produced significantly fewer gestures than the LRC group. (2) Kruskal-Wallis tests indicated no significant group differences in the proportions of maternal responses to infant gestures between 12 and 24 months (Figure 2). (3) Using Pearson correlations, we found maternal responses at 12 months were significantly negatively related to infants’ language scores at 36 months for the HRA+ group, but were positively trending in the HRA- and LRC groups.

Conclusions: Despite the differences in early gesture production in infant siblings, their mothers provided similar contingent responses to infants’ gestures, as did mothers of low-risk infants. Understanding both infant and parent behavior in high-risk dyads may shed light on the language learning process of this population and have implications for early intervention practices.