One of my school leadership tasks has been to try to translate the tricky terms and insights related to cognitive science for those staff members who do not spend their time at education conferences (the vast majority). I’m thinking of terms such as “self-testing”, “self-explanation”, “dual coding” and “elaborative interrogation”, among many, many others.

“Self-explanation” might seem the easiest of those terms. As the name suggests, we explain things to ourselves and, in doing so, improve our understanding of that thing. Simple, right?

But hang on, let’s take a couple of steps back. Self-explanation means “explain things to yourself”? Is that it? And how is that different from asking “why” questions? Surely, if you’re asking why something is the case, you’re having to explain it, aren’t you?

I somehow skimmed over “self-explanation” when I first came across it, giving it an internal nod before moving on to the slightly more cryptic-sounding “dual coding”. Only when revisiting it did it strike me that I couldn’t give a very thorough account of the difference between self-explanation and elaborative interrogation, or really say much more than self-explanation meant explaining things to yourself. I decided I needed to know more. Here’s what I found.

Self-explanation is effective

The research is clear that self-explanation is useful. In their book, Understanding How We Learn, Yana Weinstein, Megan Sumeracki and Oliver Caviglioli describe a study in which students using self-explanation with physics problems showed improved understanding in a subsequent text. As the authors point out, this initial study was correlational: does self-explanation improve learning or does learning lead to improved self-explanation? A follow-up study, however, randomly assigned students into two groups: one left to study using their usual methods, the other specifically prompted to self-explain. When tested, the self-explanation group performed “significantly better than the control group”. So, good news for self-explanation.

But let’s take this beyond single studies. A meta-analysis by Bisra et al (2018) concluded that “self-explanation prompts are a potentially powerful intervention across a range of instructional conditions”. The EEF’s Metacognition and Self-Regulated Learning guidance report includes a table of techniques summarising an extensive study by Dunlosky et al (2013). The study rates self-explanation as having “moderate utility” (self-testing, in comparison, has “high utility” ).

There is, in other words, enough evidence to say that self-explanation at least shows promise as an instructional strategy.

Definitions of practice vary

Let’s turn to definitions. In a way, it is as simple as “explain things to yourself ”. However, in the research, there are a variety of ways of inducing self-explanation.

In some methods, the process of self-explanation is outlined to students prior to a study session. Others involve adding prompts to self-explain during instruction, often with a box in which students can write their thoughts.

Sometimes the prompts are generic: “Explain what this paragraph means to you.” Sometimes, they are content-specific: “Could you explain how the blood vessels work?”

The EEF’s report defines it this way: “Explaining how new information is related to known information, or explaining steps taken during problem solving.”

So it’s worth reflecting on why self-explanation appears to help and the circumstances in which we’d use it.

First, we should be clear about the features that differentiate it from elaborative interrogation. Elaborative interrogation relies on prior knowledge, whereas self-explanation tends to be used in the context of learning new information. After all, if we want to elaborate, we need something to elaborate on. We wouldn’t get very far with the question, “Why does Macbeth say life ‘is a tale told by an idiot, full of sound and fury, signifying nothing’?” if we had no prior knowledge of the play. We could, however, attempt this self-explanation prompt: “Explain what Macbeth’s statement means to you.”

For this reason, self-explanation is well suited to the acquisition, rather than embellishment, of knowledge and appears effective in helping with comprehension. If you imagine yourself doing it, you can see why: you read the first couple of paragraphs of an article and, before ploughing on, stop to explain to yourself what you’ve just read. We might usefully do the same when presented with a system diagram. The process gives you time to reflect on what you’ve learned and draw inferences. As each new piece of information is encountered, we can also link it to what we already know.

As you can see, self-explanation is a helpful way to improve the comprehension of longer pieces of information - longer stretches of text, for example. A “why” question more naturally attaches itself to a shorter fact.

So, what are the nuances about self-explanation that teachers should consider?

Although some studies suggest that self-explanation is often more effective than teacher explanation, this might make us understandably nervous. A more moderate - and likely more acceptable - suggestion is given by Bisra et al (2018): “The most powerful application of self-explanation may arise after learners have made an initial explanation and then are prompted to revise it when new information highlights gaps or errors.”

Dunlosky notes some additional concerns around the time required for self-explanation and the longevity of its effects: “Further research must establish whether these effects are durable and whether the time demands make it worthwhile.”

We should summarise, then: self-explanation shows promise, but as we’ve seen, there are different ways we might implement it, different prompts we might use and even different degrees to which we might rely on it.

We need to dig beneath the quick definition to find out what the problems are, how it’s been tested and what still needs further research. This might lead to less confidence in a strategy, but sometimes being tentative is a good thing. Now, explain that to yourself.

Jon Eaton is director of Kingsbridge Research School at Kingsbridge Community College

This article originally appeared in the 1 May 2020 issue under the headline “Explain yourself - actually, don’t”