Dear reader,
This is an story about me going down a research rabbit hole to prove a point. I hope it helps you be more sceptical and curious about the information you come across online - even from very cool people.
Zoe’s video about science-based learning techniques
A couple of weeks ago1, Zoe.languages released a video taking insights from the handbook "The Cambridge Handbook of Multimedia Learning", and adapting them for language learning. She’s working in research and learning languages, so as you can guess, this is totally my kind of content.
I’ll do a little recap of the principles she presents, but if you’d like to watch for yourself, here are 10 minutes of nerdy content for you:
I thoroughly enjoyed the video, except for one specific sentence that left me wondering. Let’s do a quick recap of the principles mentioned according to the Cambridge handbook of multimedia learning (3rd edition):
The Pre-training Principle
People learn more deeply from a multimedia message when they are familiar with the names and characteristics of the main concepts[1].
The Redundancy Principle
Involves the effect of processing redundant information in multimedia learning, focusing on managing instructional redundancy to enhance learning effectiveness[2].
The Signaling Principle
Refers to the enhancement of learning from a multimedia message by adding cues that guide attention to relevant elements or highlight the organization of essential material[3].
The Segmenting Principle
People learn more deeply when a multimedia message is presented in learner-paced segments rather than as a continuous unit, supported by substantial experimental evidence[4].
The Self-explanation Principle
Encompasses self-explaining as a generative activity to enhance learning, including the utterances of self-explanations and the cognitive processes involved in self-explaining[5].
Throughout the video, she gives examples related to language learning. When it comes about the Redundancy Principle, the example mentioned is the following: if you listen to a language while reading the subtitles, it will split your attention and you will retain less information.
This goes completely against my intuition and experience, so I decided to cross-reference a few researches. I’m not a learning expert, and my research could be biased as well, but here’s what I found.
Research in favour of watching with subtitles
Support for Non-Native Speakers
Subtitles can support non-native speakers by providing textual context to auditory information, catering to diverse audiences and different levels of language proficiency[6].
Vocabulary Learning
Studies show that the use of subtitles can aid in vocabulary learning. For instance, one study found that groups exposed to L1 (first language) and dual subtitles performed better in terms of vocabulary learning compared to a group with L2 (second language) captions only[7].
Dual-Subtitled Viewing
Extending the Involvement Load Hypothesis to audiovisual input, a study examined the effects of dual-subtitled viewing on vocabulary learning, implying that having both closed captions and translated subtitles could be beneficial for vocabulary acquisition[8].
Signaling and Segmenting
Subtitles can act as a signaling tool to highlight important phrases or words, and segment sentences into meaningful chunks, which aligns with the Signaling and Segmenting Principles we previously mentioned. (cf previous points)
I also found a research paper doubting the effectiveness of using subtitles. It doesn’t show reduced nor enhanced effectiveness.
Ambiguity in Effectiveness
A meta-analysis indicated that while subtitles have received considerable attention in language-learning environments, conclusive findings regarding their effectiveness in enhancing language learning have yet to be reached[9].
Is there a contradiction?
OK but why doesn’t the Redundancy Principle apply to subtitles? It suggests that repeating identical information could increase cognitive load, potentially hindering the learning process.
Here’s a theory. It has been proved that audio and visual information are processed differently when we memorise knowledge. I heard about it first in the course the Learning Brain. An easy experiment would be trying to focus on audio while looking at unrelated pictures, and vice versa. It wouldn’t hinder your comprehension of the subject.
However, the Redundancy Principle argues that if you’re trying to focus on two visual elements displaying the same information (for example a chart with an explanatory text), it will affect your understanding negatively.
I’m not trying to say whether someone is right or wrong, but this is an interesting example of how research results can be interpreted in different ways. It can be misleading if you don’t cross-reference with other papers and look at the whole picture. I may have overlooked something as well, I’m just doing amateur research out of curiosity!
What do you think about this subtitle story? What does your intuition tell you? Did I miss something?
I hope I piqued your curiosity and showed you what the scientific method can look like. Hopefully I don’t take this long to write next week’s entry.
Best,
Lou
✍️ Dear language learner
Learn your target language with useful sentences! Here are a few to translate to test yourself. If they’re relevant to you, you can add them to your notes or spaced-repetition system.
🌱 Beginner
Science rocks.
🌿 Intermediate
I always do my research.
🌳 Advanced
You should always cite your references to be credible.
Thank you for reading “I heard from Lou”. It can be hard to find people who share your passion for learning. If you have friends who might like this letter, let them know.
This letter was supposed to get published last week, but for some reason I struggled to finish writing it. I get why writing about scientific subjects can be difficult. More respect to Zoe.
A lot of the difference in the research comes down to what the researchers feel the learning goal is. From a big-picture, 30,000-foot view, not using subtitles is going to give you a better understanding of how a language "feels." E.g., if you consistently watch videos without subtitles you'll eventually understand the rhythm and flow much, much better.
But. In a lot of the pro-subtitle research you share, the researchers are working with a specific learning goal in mind, like improving vocabulary or pronunciation. Subtitles works really well for the first skill, not so much for the second.
So, in practical terms, what this means is that a learner should use subtitles, sometimes, dependin on what they want to learn. Here's a plan I like to use: I sit down with a student and watch an episode of Friends. We do this cold, no prep, no explanation, etc. We use subtitles. Then, after the episode, we discuss who the characters are, what their relationships are, and so on. We then build a plan where, over the course of months, we watch specific scenes. We will often watch a scene several times. The first time with no subtitles, the next time with subtitles, highlighting certain words or phrases, the next time with dual subtitles where we compare how things are expressed in English vs. their native language, and finally, once more without any subtitles, so they can see how their comprehension and understanding has evolved.
I'm going on too long, but the point is, as you are already aware - should you use subtitles? Sure, sometimes.
This reminds me of a study my professor mentioned where adult ESL learners who watched English sitcoms with English subtitles learned up to 80% more than those who used subtitles in their native language. I personally try to keep Japanese subtitles on as much as possible, but sometimes, when I just want to relax and enjoy, I'll turn on dual subtitles and focus mostly on the English ones.