The increased use of videos for learning
Nowadays, learners encounter many of the information sources they engage with online. Teachers increasingly assign projects that require students to study a text or a video retrieved online. Indeed, since the popularisation of video-sharing platforms like YouTube, videos have become a major component of leisure time and an integral component of educational environments.
However, the efficacy of videos for learning has been questioned, especially if compared to content-equivalent texts. One issue is that today’s students perceive themselves as digital natives and may over-judge their competences in learning from digital sources. Although students may express a clear preference for digital texts over printed texts, they generally perform better in comprehension questions when reading in the print condition. The situation is complicated by the fact that most of the available instructional videos are produced in English, which means that non-native English speakers need to process information in a foreign language and/or rely on subtitles in their first language (i.e., same-language subtitles, SLS).
Learning from subtitled instructional videos
Learners process multimedia by coordinating the visual and the auditory channels to decode and integrate words and images. However, each channel has limited resources to dedicate to processing. Videos with oral texts (e.g., narrating voice) or with on-screen text (e.g., subtitled videos) involve the processing channels differently. Narrated videos are processed in the auditory-verbal channel, whereas on-screen-texted videos are processed in the visual-verbal channel.
Subtitled videos differ from static texts in several aspects. Firstly, subtitles can be described as a fleeting text on a dynamic background, which places demands on learning that are different from those entailed by static texts on stable backgrounds. Secondly, similarly to what happens when learning from videos, the reader has to manage cognitive resources across different sources of information and adjust the reading pace to the pace with which subtitles appear on the screen.
On the one hand, some studies suggest the higher effectiveness of videos over texts as learners may focus more attention and produce more affective responses to videos as compared to texts. On the other hand, the different modalities (learning from text or videos) seem equivalent in terms of cognitive processing, but it is unclear whether this equivalence extends to subtitled videos.
Comparing learning across modalities
In a study that we recently carried out, we investigated the effect of modality on students’ learning performances, by comparing a narrated video, a subtitled video, and a text-on-screen condition. The participants were undergraduate students from a University located in central Italy, randomly assigned to a modality condition: text-on-screen (n = 82), video with sound (n = 84), and video with subtitles without sound (n = 81). Students were assigned learning material on the topic of stem cells. In its original version, the learning material was a video. In the subtitles condition, we removed the audio-track from the video and included subtitles reproducing the exact content of the original audio-track, in sync with the corresponding slide. In the printed text condition, students were assigned the text transcribed from the original audio track (speech only).
Subtitles may be detrimental for long-term learning
The choice of the modality to use should be based on the specific purpose of the learning content. If the purpose is immediate surface comprehension, performances are equivalent across modalities. However, if the purpose is long-term deep learning, digital texts appear to be superior to other modalities whereas subtitled videos seem inferior to the other types of video for transfer of learning over time. Yet, educational materials available on the Web often include subtitles as they are originally produced in English, so they need to be accompanied by texts in other languages to be usable by non-English speaking students. Students must be made aware by teachers and instructors that they need to put extra effort when they try to learn from dynamic graphics and text because the simultaneous processing of the two types of representation may overload their visual channel.
The increase of the availability and frequency of the use of online education videos warrants more research, especially for non-native English speakers accessing English-produced material. Indeed, the Web Content Accessibility Guidelines 2.0 prescribe subtitles for any media with audio included to ensure a high level of accessibility (World Wide Web Consortium, 2008). Results from our study confirm the substantial equivalence of all conditions in immediate comprehension. Conversely, results confirm the disadvantage of subtitled videos for deep learning outcomes, which suggests introducing subtitled material (likely videos in L2 subtitled in L1) in a more conscious and careful manner, drawing from research conducted on EFL learners.
Editor’s note: This article has been inspired and adapted from Tarchi, C., Zaccoletti, S., Mason, L., Learning from text, video, or subtitles: A comparative analysis, Computers and Education, Volume 160, January 2021.
Christian Tarchi, PhD, Researcher in Developmental and Educational Psychology, Department of Education, Languages, Interculture, Literatures and Psychology, University of Florence, Italy