Learning L2 constructions from captioned audio-visual exposure: The effect of learner-related factors

Most studies showing the benefits of caption-supported audio-visual input for second language (L2) learning have focused on vocabulary learning and content comprehension, and there is a dearth of research on the effects on grammar learning. The present study aims to contribute to filling this gap focusing on L2 grammar learning from a constructionist perspective of language. It reports an extensive classroom intervention in which learners viewed an original version TV series under two conditions, Captions and Non-Captions. The results demonstrated that the Captions group outperformed the Non-Captions group, and that individual differences, such as proficiency and working memory (WM) capacity, played an important role in learning grammar constructions from audio-visual input, while language aptitude had no effect on learning gains. Intermediate-level learners benefited most from the viewing intervention, suggesting that their proficiency level was optimally suited to the difficulty level of the series. WM and group interacted significantly, indicating that the group that did not have the support of captions relied more strongly on WM. This study extends the known benefits of audio-visual input for vocabulary and comprehension to the underexplored area of L2 constructions and suggests that grammar learning may need a substantial amount of exposure to this input. (C) 2020 Elsevier Ltd. All rights reserved.