Examining the Relation Between Walking and Receptive Language in Infants at Heightened Risk for Autism Spectrum Disorder

Background:  The onset of walking appears to be a point of inflection for communicative development in typically developing infants (e.g. Clearfield, 2011).  Notably, research shows the acquisition of this milestone is accompanied by an increase in receptive language, independent of age (e.g. Walle & Campos, 2013).  The present study examined whether this relation may be disrupted in the infant siblings of children with autism spectrum disorder (ASD; High Risk; HR)—particularly among infants who go on to receive an ASD diagnosis—given that there is ample evidence of substantial variability in language and motor development in HR infants (e.g. Gamliel, Yirmiya & Sigman, 2007; Iverson & Wozniak, 2007).

Objectives:  To examine changes in the Words Understood measure of the MacArthur-Bates Communicative Development Inventory (CDI; Fenson et al., 1993) longitudinally across the transition to walking in infants with no family history of ASD (Low Risk; LR) and three groups of HR infants: No Diagnosis, Language Delay, and ASD.

Methods:  Participants included 81 HR infants who were visited at home monthly from 5-14 months and a comparison group of 25 LR infants. This study focused on CDI data beginning 4 months prior to infants’ walk onset and ending 3 months after walk onset.  To establish walk onset, parents recorded infants’ motor milestones, and an experimenter confirmed them at each visit.  At 36 months, 15 HR infants received an ASD diagnosis (HR-ASD). Twenty-six HR infants with language delays were identified (HR-LD).  Forty HR infants did not meet ASD or LD criteria (No Diagnosis; HR-ND). 

Results:  A two-level piecewise Hierarchical Linear Model was used to examine change in receptive language (RL) over time within individuals, and change across individuals within outcome classifications. The piecewise model allowed us to estimate growth in RL over the entire trajectory (piece 1; all time points) as well as change in growth from the final crawling visit forward (piece 2).  Results revealed that the slope of growth following walk onset significantly explained variance in receptive language, even when variance explained by the overall slope (piece 1) was accounted for.  This replicates previous findings that walk onset marks an inflection point in receptive language.

To examine group differences, dummy variables for outcome classification were included on the intercept and slopes for HR-ND, HR-LD, and HR-ASD infants, with a LR reference group. To control for age, age at walk onset was included in the model.  Results revealed that HR-ASD infants displayed marginally slower growth than the LR group overall (piece 1), and significantly less change in growth following walk onset (piece 2). While the HR-ND group did not differ from the LR group, the HR-LD group demonstrated marginally reduced change in growth following walk onset compared to their LR peers. 

Conclusions:  For infants with no diagnosis—regardless of risk status— walk onset corresponded to increased RL growth.  However, this pattern was significantly attenuated for HR-ASD infants.  One possible explanation is that in typical development, the ability to walk influences underlying mechanisms of language acquisition, but this relation may diverge in ASD.