Measuring Language Competence in Children with ASD and Average Intelligence

Background: Social communication impairments are a defining feature of ASD; however, children with ASD show heterogeneity in language competence. Standardized measures of communication may not capture subtle impairments shown by children with HFA. We rely on these measures to diagnose language impairment, determine eligibility for services, define participant or diagnostic groups, and measure response to intervention.
Objectives: This study aims to identify measures of language competence that are sensitive to the milder but significant language impairments characteristic of children with HFA. We will explore whether direct and parent report measures of language competency correlate with social language use in a naturalistic playground setting.
Methods: Thirty-three boys with autism and 29 TD controls participated in a study that included direct assessment of cognitive development and language competency, parent report, and a playground interaction captured by video. Participants ranged from age 8 to 12, and were matched for age (M 9.73, SD 1.48). Intelligence was estimated using the WASI, with a mean IQ of 108.2 (SD 24.5) for the ASD group and a 120.4 (SD 13.8) for the TD group. All participants had a FSIQ of ≥ 80. The ASD group diagnosis was made by a clinical psychologist with expertise in ASD and confirmed with an ADOS. Video data from playground interactions will be coded for frequency of Verbal Initiation, Verbal Rejection, and Verbal Bouts (duration and number of verbal exchanges). Participants completed the Clinical Evaluation of Language Fundamentals - Fourth Edition (CELF - 4) subtests Concepts and Following Directions (CFD), Recalling Sentences (RS), and Formulated Sentences (FS). Parent estimates of child language facility included the Adaptive Behavior Assessment System-Second Edition (ABAS-II), the Social Communication Questionnaire (SCQ), and the Social Responsiveness Scale (SRS),
Results: Controlling for IQ, a significant multivariate main effect for group was observed for the CELF-4, F (3, 56) = 5.36, p =.003, partial eta squared = .223. Between group differences were statistically significant for the CELF-4 CFD F(1) = 15.12, p<.001) and FS F(1) = 4.77, p = 0.03), with children with ASD performing more poorly. Across groups, the Concept and Following Directions scale is moderately correlated with ABAS Communication (r =.42, p =.001) the SRS Social Communication Scale (r = -.41, p =.001) and the SCQ (r = -.39, p =.002). The Formulating Sentences scale is also moderately correlated with ABAS Communication (r = .42, p = .001), SRS Social Communication (r = -.32, p = .01), and SCQ (r = -.36, p = .004).
Conclusions: Group differences emerged between TD and HFA children on measures of ability to follow auditory instructions and verbally formulate sentences, with children with ASD showing greater signs of impairment. Parent report of communication skills in social and daily living settings were also significantly different across groups and moderately correlated with direct child assessment. Additional analysis of language data collected in naturalistic playground setting will speak to which type of measure, parent or child measures, are better predictors of language competence in a naturalistic playground situation.