Many studies supported that children with autism spectrum disorder (ASD) have worse executive functions (EFs) when compared to typically developmental (TD) children in many domains, such as planning, flexibility, inhibition, and self-monitoring. The current study aims to use an adapted version of the computerized tower test to investigate the EFs of children with ASD. Furthermore, the researcher also assessed children's EFs-related behaviors in their schools using a teacher-filled behavior rating inventory of executive function, 2nd edition (BRIEF-2).

61 Children aged 7 to 12 years old (M = 9.23) were included in the current study. 29 of them were in the ASD group, and 31 of them were in the TD group. All participants conducted an adapted computerized tower test. All participants' teachers completed BRIEF-2 to investigate their EFs-related behaviors in their schools.

The results indicated that there were no significant differences in the tower test between the ASD group and TD group in all indexes. Therefore, it implied that the current indexes might not be sensitive enough to distinguish whether a child has ASD or not. In addition, we further investigate the correlations between the tower test and the teacher-filled BRIEF-2. We found the different patterns in the two groups. In the ASD group, we found that the task-monitor index was positively correlated with total-number-of-rule-violations, total-complete-time, and total-rule-violations-per-item-ratio. The task-monitor index was negatively correlated with total-achievement-score, implying that poorer ability to monitor tasks leads to a longer completion time, more rule violations, and a lower total achievement score. Moreover, we also found a high correlation between the organization-of-materials in BRIEF-2 and total-complete-time in the tower test, suggesting the long problem-solving time in ASD groups is highly related to the disability of keeping working space ordered. In addition, we found that the shift index is positively correlated with total-complete-time and total-rule-violations-per-item-ratio. Hence, it indicates that those with poor flexibility in solving problems tend to need more time to complete tasks and violate more rules in ASD groups. In the TD group, we only found the correlation effects were significant between inhibition and self-monitor in the BRIEF-2 and the total-rule-violations-per-item-ratio in the tower test. It suggested that individuals with behavioral regulation problems, such as impulse control and monitoring problems are more likely to make rule violations. The result indicated that behavioral regulations play a more significant role in the TD group. In contrast, cognitive and emotional regulations are more critical in ASD children.

Our findings found no significant difference in the computerized tower test between children with and without ASD, suggesting that the current indexed might not be sensitive enough to differentiate children with or without ASD. However, the results of the correlation between the tower test and teacher-filled BRIEF-2 showed that different patterns might be the cause of the EF performances between the two groups, indicating that there might be a different domain of EFs the children used in the tower test between two groups.

Therefore, further research could focus on developing new indexes in the Tower test and finding the EF mechanism of ASD children with different approaches.