Pre-learning vocabulary before viewing captioned video: an eye-tracking study

While there is increasing evidence that learners pick up new words while viewing video with captions (i.e. L2 subtitles), little is known about the role of instructional interventions prior to viewing. This study uses eye-movement registration to investigate the effect of pre-learning vocabulary before watching a captioned video. Thirty intermediate L2 learners of French watched a captioned video that contained ten pseudo-words in a within-subject design; five pseudo-words were pre-learned, the other half only appeared in the video. We investigated the effect of pre-learning on participants' attention allocation to the pseudo-words in the captions, the captions that contained a pseudo-word and the corresponding images as well as in two vocabulary post-tests (i.e. a form and meaning recognition test). Four types of eye-metrics were used to analyse learners' attention allocation: first and total fixation duration, fixation count, and regression count. Findings indicated that pre-learning does not affect participants' fixations and time spent on the pseudo-words in the captions, the captions, and the images. However, pre-learned pseudo-words were skipped more often than unknown pseudo-words in the captions. The effect of pre-learning was more pronounced on the vocabulary post-tests. Overall, participants spent more time on test items for unknown pseudo-words.