Effects of captioned video on L2 speech perception in intermediate L2 learners of Spanish

Publication type
Publication status
Montero Perez, M., Simon, E., Goethals, P., & Pattemore, A.
Calico (Seattle)
View in Biblio
(externe link)


During the past decade, we have seen a surge of studies into the role of audiovisual input and captions (i.e. subtitles in the same language as the audio) for different aspects of L2 learning. Studies have, for instance, shown that audiovisual input with captions can promote vocabulary learning. However, few studies have looked into the effects of captioning on speech perception. Results of these studies suggest that watching video with captions can be helpful for speech perception (e.g., Charles & Trenkic, 2015) but this is mainly based on results for advanced L2 learners of English (Mitterer & McQueen, 2009; Wisniewska & Mora, 2020). It remains unclear if captions help speech perception in participants with lower L2 proficiency levels.

The aim of this study is to investigate the effects of watching an episode of a Spanish TV-series (75 minutes) on L2 learners' speech perception. Sixty Dutch-speaking intermediate L2 learners of Spanish were divided over captioned, uncaptioned, and control (only tests) groups. Three weeks prior to the treatment, they completed a 60-item shadowing task in Spanish which measures whether learners are able to repeat sentences after hearing them once. Participants also took a 90-item vocabulary test (i.e., Lextale-Esp) and the LLAMA D aptitude test (Meara, 2005) which predicts learners’ ability to recognize patterns in an unknown spoken language. Treatment groups watched an episode in a captioned or uncaptioned condition. During the posttest session, participants retook the shadowing task which consists of 20 treatment sentences, 20 new sentences (same TV-series as treatment), and 20 new sentences from a new series. Mixed-effects analyses will be used to analyze the effects of captions on learners' speech perception for three types of sentences. Learners' prior vocabulary knowledge and LLAMA test scores will be included in the statistical model in order to take into account their impact on L2 learners' speech perception from watching (un)captioned video.