Abstract
This study expands upon research into task repetition effects by exploring the effect of task modeling between the performances of the same oral narrative task and its extension to a new task. Seventy-one advanced beginner English as a foreign language (EFL) learners were divided into three groups: task repetition with oral modeling (TR + OM), task repetition with written modeling (TR + WM), and task repetition with no modeling (TR). All groups enacted another oral narrative task (a new task of the same type), three days apart. Participants’ oral narrative task performances were analyzed in terms of complexity, accuracy and fluency. Results revealed that TR + WM was more effective than the TR + OM, and both were significantly better than the control group in leading to subordination complexity in the repeated task and the new task. Whereas the percentage of error-free clauses remained unchanged over time, the accurate verb forms increased in TR + WM’s repeated task, but declined in the new task. Fluency in terms of articulation speed and mid-clause silent pauses was improved and maintained in the new task in all groups, with both the experimental groups particularly the TR + OM outperforming the TR regarding the significant reduction of repair in the repeated and new task performances. In conclusion, these significant developments induced by the modeling conditions speak to the strength of models in providing and extending linguistic features beyond learners’ current repertoires. Pedagogical implications of these findings are discussed.
