Objectives: This study examined cognitive flexibility in implicit learning through manipulating item frequency and predictive probability in children with and without specific language impairment (SLI). The purpose of this study was to determine whether a significant difference in non-linguistic implicit learning ability would be found between children with SLI and their typically developing (TD) peers and identify a significant factor for distinguishing between SLI and TD groups. Methods: Sixteen 4- to 6-year-old children with SLI and 15 age-matched children with TD participated in this study. Accuracy and RTs were obtained via the Grammatical Judgment Task (GJT) and non-linguistic implicit learning task, where item frequencies and probability were considered simultaneously. The accuracy (%) and response time (ms) in the implicit learning task were recorded using E-Prime Software. A repeated measure of ANOVA, Pearson's product moment correlation coefficient, and discriminant analysis were used for data analysis. Results: The results showed that the SLI group had significantly lower accuracy and longer response time on the implicit learning task than the TD group; a significant difference between low frequency/low probability and the other conditions between the SLI and the TD groups was found only for accuracy; and RT in HL (conflict high frequency/low relational probability) condition was a significant factor for identifying SLI. Conclusion: When compared to peer TD children, children with SLI are less sensitive to certain rules if both frequency and probability are at lower levels. These difficulties are closely associated with their grammatical judgment performance.
- Children with language impairment
- Cognitive flexibility
- Grammatical judgment
- Implicit learning
- Joint probability