Children with ASD have varied cognitive profiles (Goldstein et al., 2008; Szatmari et al., 1995; Zander & Dahlgren, 2010). The research on the relationship between cognitive patterns and the core symptoms of autism is more limited. One study reported a significant relationship between the magnitude of social impairment and degree of divergence between cognitive abilities in school age children whose nonverbal abilities were greater than their verbal abilities (Joseph et al., 2002). Another reported significant relationship between discrepantly high verbal and nonverbal abilities and social impairment (Black et al., 2009).
Objectives:
Ascertain frequency of scatter in global cognitive scores in children with ASD and explore the relationship between magnitude and direction of cognitive variability and (a) severity of social disability, (b) level of social ability, and (c) discrete domains of social impairment.
Methods:
The DAS-II was administered to163 children: 112 from a genetics study of ASD (Simons Simplex Consortium study) and 51 from a longitudinal study of ASD. The sample ranged in age from 7 to 17 years (mean = 10.5, s.d. = 2.83) with a mean GCA score of 93.59 (s.d. 18.88). Social functioning was measured using the ADOS, Vineland-II, and SRS.
Results:
Discrepancies greater than 1 s.d. (15-points) in either direction were found at the following frequencies: verbal and nonverbal reasoning
abilities (29.3%); verbal and spatial abilities (39.3%); nonverbal reasoning and spatial abilities (31.8%). The direction of the discrepancy was fairly equally divided in each instance. Discrepancies of greater than 2 standard deviations were less frequent: 8.5%, 12.2%, and 2.9%, respectively, but still greater than what would be expected based on the normative sample.
In children with verbal > spatial abilities, the magnitude of the discrepancy between verbal and spatial abilities was significantly correlated with ADOS calibrated severity (r = 0.54, p < 0.01) and Vineland-II socialization (r = 0.40, p < 0.05) scores. In children with verbal > nonverbal scores, the magnitude of the discrepancy between verbal and nonverbal abilities was correlated with the ADOS calibrated severity score (r = 0.56, p < 0.05) but not the Vineland socialization score. ADOS and Vineland-II scores were not significantly related to magnitude of cognitive variability in children with nonverbal strengths. Relationships between cognitive variability and domain scores on the SRS will also be reported.
Conclusions:
Many children with ASD have significantly varied cognitive profiles. Given the variability and individual differences, cognitive profiles, not just overall ability, are important to consider when conceptualizing the abilities of children with ASD. Although significant variability in cognitive profiles is found with some frequency, it is not associated with magnitude of social impairment in all children. Magnitude of social impairment was associated with degree of cognitive variability in children with stronger verbal than nonverbal/spatial skills. These results differ from prior results (Joseph et al., 2002; Black, et al., 2009), which may be due to differing measurement properties and differing sample characteristics. Overall, increased cognitive variability may be related to severity of neurocognitive disturbance in a subset of children with ASD.
See more of: Clinical Phenotype
See more of: Symptoms, Diagnosis & Phenotype