Individual Cognitive Strengths and Weaknesses in Autism Spectrum Disorders: Diversity Is the Rule

Background:  The extant cognitive literature reveals little about whether individuals with ASD have a consistent cognitive profile and, if so, what it is. Logically, a range of possibilities exists, from a situation where all people with ASD have a similar pattern of cognitive strengths and difficulties to perfect heterogeneity, with no consistent cognitive profiles. Yet the comparison of group averages does not, and cannot, reveal evidence for or against these alternate hypotheses. Claims that people with ASD in general have stronger non-verbal than verbal skills have not been adequately tested in ‘high-functioning’ samples. We aimed to use a multiple single case study approach to investigate the nature and range of cognitive diversity within individuals with ASD, most of whom had normal-range intelligence.

Objectives:  Our objective was to seek evidence for one or more characteristic ASD profiles based on standard cognitive tests, which could serve as means to supporting diagnosis and as a step towards the elucidation of ASD biomarkers. The discovery of sub-groups defined by their cognitive profile would have profound implications for understanding distinct etiologies, treatment needs and prognoses.

Methods:  Participants comprised 104 children (81% male; mean age = 11.4 years, SD=2.96, range = 6.2 to 16.9), diagnosed with ASD by a team of expert clinicians on the basis of ADI-R and ADOS scores. IQ was measured using the Wechsler Intelligence Scale for Children – Fourth UK Edition (WISC-IV). All participants had fluent language and were in mainstream education; 82% had a full-scale IQ >70.

Results:   Previous literature indicated relatively strong Picture Concepts, Similarities and Matrix Reasoning characterize a ‘typical’ WISC-IV profile, in combination with relatively weak Comprehension and Symbol Search. We searched for children with this profile, defined according to Flanagan-Kaufman criteria (a standardized measure of discrepancy within WISC profiles). Just 1/104 participants had this ‘typical’ WISC-IV profile. Diverse patterns of relative strengths and difficulties were common. Only 14 children (13%) had neither strengths nor difficulties; 76 (73%) had both at least one strength and at least one difficulty. Greatest diversity was seen in verbal similarities, comprehension and coding (proportions with strengths and weaknesses differed at p<0.001). Block design, often cited as a relative strength in ASD, was not discriminating; 14 (13%) had a weakness and just 18 (17%) had a relative strength in this task.

Conclusions:  Unevenness of WISC-IV subscale scores is the norm in children with ASD. Using cognitive profiles as potential endophenotypes for distinctive biomarkers, such as characteristic patterns of neural activity, is unlikely to be productive.  A typically ‘autistic’ WISC-IV profile is rare even among high-functioning children. There is no consistent strength in any verbal or non-verbal subtest. Cognitive testing has no predictive value in the evaluation of ASD in diagnostic terms.