One of the important cognitive skills we need in our increasingly technology driven world is the ability to learn from hard-to-understand text. Maybe you’re trying to grasp a biology textbook chapter on sexual reproduction.
Or perhaps you’re reading articles on the web to figure out how to extend your home network by repurposing an old router. (Yes, that was me last weekend. No, Saturday nights aren’t what they used to be). You can surely think of your own examples readily. A good reading strategy can make all the difference in these cases.
Self-explanation is one such reading strategy. Micki Chi and other cognitive scientists have been researching this reading strategy for a couple of decades now. It has proven quite successful. The idea is simply to pause from your reading, and explain parts of the text to yourself. Ask what it means to you. Then answer as best you can.
Yasuhiro Ozuru and colleagues conducted a study to understand how well the self-explanation reading strategy works with harder versus easier texts. They used a precise method for adjusting the text difficulty, while keeping the main points of the text the same. The idea was to adjust how easily sentences flowed from one to the next by adding or removing bridging words. This property is called “cohesion.” Their paper on the, “Contributions of self-explanation to comprehension of high- and low- cohesion texts,” was published in Discourse Processes.
Their idea was that difficult text (low cohesion) requires more active reading. Readers need to use the partial information given in the text to reason through what the author was trying to say. But readers don’t often do that on their own.
If the author has been more explicit (high cohesion), readers don’t need to reason as much. So, the self-explanation reading strategy may be even more helpful with hard texts than with clearer ones.
Yasuhiro and team tested the idea by asking college students to read a scientific passage called, “Why is there sex?” The passage discusses the evolutionary pros and cons of sexual reproduction compared to asexual reproduction.
The researchers created a hard and easy version of the passage by adjusting the cohesion. Some students got the hard version and others got the easy one.
Also, some students just read the passage in their own way. Others were told to use the self-explanation reading strategy. They were asked at certain points to explain what the reading meant to them. For this topic, they probably had a fair amount to say. They likely had their own pros and cons to share.
The study turned up some interesting results. First, the researchers further confirmed past studies showing that self-explanation is a helpful reading strategy.
Second, the quality of a person’s self-explanations also matters. If you try to pull ideas from different parts of the text into your explanation, it will be better. Trying to bring in anything you already know about the topic helps, too.
Finally, the benefit of self-explanation does appear to be greater for low-cohesion text than for high-cohesion text. It’s especially beneficial to use the self-explanation reading strategy when you are trying to comprehend a very difficult or poorly written text.
And if you’re reading a difficult biology chapter about sexual reproduction, take heart. It sure beats reading about routers.
- How to be Smart: A Simple Approach
- A Study Strategy for all Occasions: Test your Memory
- Learning about learning: Take Charge of Your Learning Strategies
- Write to Decide
- How to Learn from the Web
- Want to Be Smart? First, Know How Little You Know
- Learn to Learn by Embracing Surprise
- Memorize This! Exercises to Improve Memory Won’t Help You Learn Better
- Critical Thinking Skills: What are They and How Do I Get Them?
Image Credit: stellar
Chi, M. (1994). Eliciting self-explanations improves understanding, Cognitive Science, 18 (3), 439-477 DOI: 10.1016/0364-0213(94)90016-7
Ozuru, Y., Briner, S., Best, R., & McNamara, D. (2010). Contributions of Self-Explanation to Comprehension of High- and Low-Cohesion Texts Discourse Processes, 47 (8), 641-667 DOI: 10.1080/01638531003628809