Language Profiles of Infants Showing Early Signs of Autism Spectrum Disorder (ASD)
Impairment or delays in language acquisition are among the clearest early indicators for Autism Spectrum Disorder (ASD). Typically, children understand words before they can use them (i.e., receptive language advantage). While children on the autism spectrum may have both receptive and expressive language delays or impairments, some past research suggests receptive language may be relatively more impaired (e.g., Hudry et al., 2010; 2014). However, few studies have investigated receptive-expressive language profiles in infants showing early signs of ASD.
To investigate language profiles in a unique cohort of infants showing early signs of ASD identified via community-referral, using multiple measures.
A cohort of 94 infants with multiple language assessment data were assessed (67.0% male) aged 9-16 months (M = 12.32, SD = 1.97). Parents/caregivers completed the Vineland Adaptive Behavior Scales (VABS) for functional receptive and expressive communication, and the MacArthur Bates Communication Development Inventory (MCDI, Words and Gestures form) for receptive and expressive vocabulary. Direct assessment included the Autism Observation Scale for Infants (AOSI), a measure of early signs of autism, and the Mullen Scales of Early Learning (MSEL), yielding a measure of child receptive and expressive ability.
Both receptive and expressive language domains were delayed in this cohort when compared to chronological age. Correlational analyses revealed only moderate to low agreement among the two parent-report and one direct assessment of language. No evidence of reduced receptive advantage was observed in this cohort, in that language profiles were indicative of a more normative profile. Regression analyses revealed that autism symptoms were not predictive of receptive advantage on any of the measures. However, MSEL receptive advantage was significantly predicted by higher nonverbal abilities, and MCDI receptive advantage by older chronological age. No predictive associations were apparent for the VABS receptive advantage.
Existing research on language profiles with diagnosed children and ‘high-risk’ infant siblings may not generalise to community-referred samples and/or very young infants showing early signs of ASD.