Cognitive inflexibility in Japanese adolescents and adults with autism spectrum disorders

Aim: To investigate executive function in Japanese adolescents and adults with autism spectrum disorders (ASD) compared to Japanese controls.

Methods: Thirty-three individuals with ASD and 33 controls participated. The ASD and control groups' demographic variables were matched for gender (male/female: 20/13 vs 20/13), age (26.1 ± 11.5 vs 26.8 ± 9.6), years of education (13.2 ± 2.9 vs 14.2 ± 1.9), full-scale intelligence quotient (IQ) (103.0 ± 16.7 vs 103.7 ± 14.7), performance IQ (96.2 ± 16.1 vs 97.8 ± 15.0), and verbal IQ (107.9 ± 16.3 vs 107.7 ± 14.4). Participants performed the Wisconsin Card Sorting Test (WCST), which assesses the executive processes involved in problem solving and cognitive flexibility, and the Continuous Performance Test (CPT), which assesses attention and impulsivity. Symptoms were assessed by the Autism-Spectrum Quotient Japanese version (AQ-J). First, we compared the scores of the WCST between the groups using a Mann-Whitney U-test and conducted an analysis of covariance for the variables with the scores of category archives and CPT scores as covariates. Second, we analyzed the correlation between the scores of the WCST and the AQ-J in the ASD group using Pearson's r.

Results: The total errors (TE) and the percentages of perseverative errors of the Milner type (%PEM) and Nelson type (%PEN) among the TE in the ASD group were significantly worse compared with the control group (ASD vs Control, respectively: TE: 16.0 ± 6.2 vs 12.6 ± 3.5, P = 0.012; %PEM: 11.7 ± 10.7 vs 6.6 ± 8.9, P = 0.037; %PEN: 20.1 ± 14.5 vs 8.7 ± 10.4, P = 0.0011). In contrast, no significant difference was observed between the two groups in the scores of categories achieved on the WCST or the CPT. An analysis of covariance revealed significant differences between the groups in the %PEN scores (P = 0.0062) but not in the TE or the %PEM scores. These results suggest that Japanese adolescents and adults with ASD have cognitive inflexibility. Furthermore, our results suggest that Japanese adolescents and adults with ASD may have difficulties using negative feedback because perseverative errors of the Nelson type indicate persistence in choosing the incorrect reaction. By contrast, there was no significant correlation between the WCST and AQ-J scores.

Conclusion: We confirmed the presence of cognitive inflexibility in Japanese adolescents and adults with ASD. Our results also indicated that subjects with ASD may not use negative feedback effectively.