Investigating the difference in comprehension between Sesotho and English subtitles : the case of full and keyword subtitles

Abstract

Research on the benefits of subtitles for learning was pioneered by Price (1983). Her study found that captions, also known as intralingual subtitles, which were primarily intended for the deaf and hearing-impaired viewers, facilitate language acquisition for viewers who were not hearingimpaired. Since this ground-breaking study, researchers have investigated the use of different types of subtitles in different educational contexts. The use of subtitles in the classroom may be particularly useful in the multilingual classrooms, where they can be used to facilitate comprehension of content. In the present study the impact of subtitles on comprehension when learning content of a specific subject was explored. The aim of this study was two-fold. First it aimed to find out how the language and style of subtitling affected comprehension. The empirical component of the study examined how Sesotho L1 and English L2 subtitles impacted the Sesotho L1 participants’ comprehension of the academic content presented in the videos. The participants each watched five videos, each presented in one of the five modes: full subtitles English (FSE), full subtitles Sesotho (FSS), keyword subtitles English (KSE), keyword subtitles Sesotho (KSS) and no subtitles (NS). The comprehension of the content was assessed by using a comprehension test at the end of viewing each video. Second, it used the SMI iViewX RED500 eye tracker to examine how the extent of subtitle reading affects comprehension. The participants’ eye movements were monitored as they viewed the experimental videos. The study considered eye-tracking measures such as the Reading Index for Dynamic Texts (RIDT) and Unique Fixation per Mean Word (UFMW). Other issues to consider when studying subtitles include how task effort and attitude towards subtitles affect participants’ performance. These two factors were not the main aims of the study, but they are included in the study because they could potentially affect comprehension and the reading of the subtitles. Through self-report questionnaires, the participants were asked to rate their perceived task effort experienced while viewing the videos and their attitudes towards subtitles in general and in the educational context. The data was analysed using the mixed-effects model. With regard to language, no significant effect was found in the differences for the performance between the presentation modes, thus the hypothesis that subtitle language would have an impact on comprehension was refuted. In terms of subtitle style, no practical effect on comprehension was found, therefore refuting the hypothesis that subtitle style would have an effect on comprehension. Despite this, the participants’ experience provides interesting insights into the study. The participants found FSE to be most helpful and KSS to be the least helpful, regardless of this, the negative correlation found between

comprehension and subtitle helpfulness for the KSS was found to be practically visible. In contrast, the content was found to be most difficult when presented with KSS and least difficult when presented with FSE. This is consistent with the responses in the attitudinal questionnaire where participants indicated that they would recommend FSE for use in the educational context. This could be attributed to the fact that the majority of the participants were familiar with English subtitles in general television viewing. Another possible explanation is that the participants perceive English as "the language of education", i.e. they are so accustomed to English being the dominant language of education in South Africa (particularly at university), that they see the FSE as the "normal" or "ordinary" subtitles to use.