Language Profiles of Boys with Idiopathic Autism or Fragile X Syndrome: A Cross-Disorder Comparison

Saturday, May 19, 2012
Sheraton Hall (Sheraton Centre Toronto)
10:00 AM
A. McDuffie1, S. T. Kover2, R. J. Hagerman3 and L. Abbeduto4, (1)Psychiatry, MIND Institute University of California Davis, Sacramento, CA, (2)University of Wisconsin, Madison Waisman Center, Madison, WI, (3)Pediatrics, U.C. Davis MIND Institute, Sacramento, CA, (4)Psychiatry, MIND Institute University of California Davis, sacramento, CA

Fragile X syndrome (FXS) is the leading inherited cause of intellectual disability.  A majority of males with FXS also display symptoms of autism (e.g., poor eye contact, repetitive behaviors, social anxiety).  Language represents a core area of impairment in both autism and FXS.  Directly comparing profiles of abilities in specific language domains for boys with autism and boys with FXS can help to clarify whether cognitive impairments and symptoms of autism differentially affect language in these two neurodevelopmental disorders.


  1. Are there between-disorder differences in receptive and expressive vocabulary and grammar?
  2. Does autism severity predict receptive and expressive language performance?


Participants were boys with autism (n = 29; Mean = 7.84 years) and FXS (n = 34; Mean = 7.44 years), 4- to 10-years of age, who participated in a longitudinal study examining word learning. Nonverbal IQ scores ranged from 38 - 80. Participants with autism had significantly higher nonverbal IQs than participants with FXS, with mean IQs of 63 and 58, respectively.  Growth scores from the Leiter-R Brief IQ were used as the metric of nonverbal cognition in all analyses. Autism symptom severity was calculated from the ADOS. Severity scores ranged from 5-10 (M = 8.21) for participants with autism and 2-10 (M = 6.47) for participants with FX. This difference was significant. The Peabody Picture Vocabulary Test-4 and Expressive Vocabulary Test-2 provided measures of vocabulary.  The Test for Reception of Grammar-2 and the Sentence Completion subtest of the Comprehensive Assessment of Spoken Language provided measures of grammar. Raw scores from the language tests were used as dependent measures in all analyses.


After controlling for age and nonverbal cognition, the groups differed significantly in receptive and expressive vocabulary and receptive grammar; boys with FXS outperformed those with autism on all three measures. Nonverbal cognition was a significant covariate in each analysis; CA was not. There was a significant and negative bivariate correlation between autism severity and nonverbal cognition for boys with FXS, r(34 ) = -.44, p < .009, but not for boys with autism.

Regression analyses revealed that nonverbal cognition and diagnostic group were significant and unique predictors of receptive and expressive vocabulary. Nonverbal cognition was a unique predictor of receptive and expressive grammar. Autism severity failed to reach significance as a unique predictor for any language domain.  For receptive grammar, there was a significant interaction between Group and Autism Severity; higher severity scores were more strongly related to grammatical understanding for children with autism, R2 change  = .05, t = 2.23, p<.03, two-tailed.


Boys with FXS experience less severe language impairment than boys with autism after controlling for level of intellectual impairment. Although challenges in the social uses of language are well documented, current findings suggest that boys with autism also have especially severe impairments in acquiring the conceptual and structural aspects of language.  The negative association between autism severity and nonverbal cognition for boys with FXS suggests that severity scores may represent a different construct in FXS than in idiopathic autism.

| More