The impact of presentation rate on the effectiveness of subtitles in learning

Subtitling has received increasing scholarly attention in audiovisual translation studies. Most of the existing literature has been centred on the benefits of subtitling for learners, but not on how subtitles are presented (e.g. subtitle presentation speed, layout, etc.). The impact of the presentation rate or appearance of subtitles on learners has therefore not been explored sufficiently. Some well-explored topics in subtitling research include language learning, film comprehension and cognitive load (cf., Danan, 2004; d'Ydewalle et al., 1991; Garza, 1991; Kruger, Hefer, & Matthew, 2013; Paas et al., 2004). However, little research has been conducted to investigate the impact of presentation rate on learning, comprehension or cognitive load. Given the increasing use of subtitles in education, an investigation into this aspect has significant theoretical and applied implications, particularly at a time when technological advances have made it possible to create verbatim subtitles relatively cheap to produce. The experiment in this study has two levels (English as a foreign language, and English as first or second language) and four conditions, namely a control condition without subtitles, and three experimental conditions in which the subtitle speed has been manipulated: 1) verbatim (an average of about 17 CPS); 2) standard (an average of 15 CPS; 3) and edited (an average of 12.5 CPS). A total of 172 students form King Khalid University (Saudi Arabia) and Macquarie University (Australia) participated in this experiment. Participants were assigned to four groups: (1) a group of 42 participants watching the verbatim video; (2) a group of 40 participants watching the edited video; a group of 53 participants watching the standard video; and (4) a control group of 37 participants watching the video without subtitles. Participants were assigned randomly to the four groups. After inviting students, those who accepted to participate were asked to watch a 6-minute history video and then fill out three questionnaires: a biographical questionnaire; a cognitive load questionnaire, adopted from Brünken et al. (2003); and a comprehension test which was used as an indicator of performance. This experiment study has found that all subtitle conditions were beneficial when compared to the unsubtitled video in the case of the Arabic group. Particularly, edited subtitles had a significant impact on the results obtained for the Arabic group. In terms of English speakers, however, edited subtitles increased CL and also resulted in lower comprehension, indicating that those subtitles were not useful for this particular group. All English speakers outperformed Arabic speakers under all conditions, particularly in the unsubtitled condition where the difference was significant -- abstract.