Prediction of language abilities in children at elevated likelihood of autism: The role of self-reported and observed parenting behaviours

Parenting behaviours are thought to play an important role in children’s language development. Yet, this relationship is understudied in children at elevated likelihood (EL) of autism, whose language abilities are highly variable. Moreover, few studies examine whether associations differ between EL-children with and without autism, and most rely exclusively on observational measures despite the added value of self-report questionnaires. This study aimed to 1) examine the predictive value of self-reported and observed positive and negative parenting behaviours for children’s receptive and expressive language development, 2) explore whether these associations differ across outcome groups (i.e., no autism, broad autism phenotype, autism), and 3) investigate group differences in parenting behaviours and language outcomes. Participants were children at EL of autism, either siblings of an autistic child ( n = 68) or children born preterm ( n = 41), and their mothers. Parenting behaviours were assessed at 24 months, and language abilities and outcome diagnosis were determined at 36 months. Neither self-reported nor observed positive or negative parenting behaviours predicted language outcomes, whereas early language abilities did. Relationships did not differ significantly across outcome groups. No group differences were found in parenting behaviours, but children in the autism group showed slightly lower expressive language outcomes compared to children in the no autism group. These findings highlight the need for further research to identify which factors predict language outcomes for whom, and under which circumstances. In the meantime, they underscore the importance of early identification of language difficulties and incorporating outcome diagnosis into research on EL-children.