Abstract
The maze task is an online activity in which participants continually select between two words to form grammatical sentences. This study investigated whether the maze task facilitates production and perception accuracy in English for Cambodian secondary-school learners. Participants (N = 101) practiced target sentences in the maze task with either morphological or syntactic choices, whereas a control group practiced unrelated sentences. Learning within the task was assessed via error rates, response times, and the coefficient of variation across 10 rounds. Transfer was evaluated with time-pressured oral translation and grammaticality judgment tests (GJTs). Results showed evidence of proceduralization and automatization for the maze task itself, and all participants improved in both the oral tests and in perceiving morphological errors in the GJTs. However, there was no evidence that this improvement was due to the maze training, since the control group improved to the same extent as the other groups. Exploratory analyses suggested that the tests themselves resulted in learning, with the probability of morphological error perception increasing with each successive trial of the two GJTs. The findings highlight the skill specificity of procedural knowledge, as posited in skill acquisition theory, and the pedagogical potential for GJTs.
Get full access to this article
View all access options for this article.
