Want to help kids learn math and science? Consider the power of the self-explanation technique. Research suggests that kids can develop a deeper understanding of math and science concepts if we ask them to explain their thinking out loud. But for best results, we need to provide kids with crucial information…and ask kids the right questions.
Who can benefit? Experiments indicate that self-explanation can help students of all ages — from the toddler years on up. In this article, I will review:
the specific benefits of self-explanation, and why it can be so helpful;
which academic subjects are best-suited to the self-explanation technique;
evidence that even preschoolers and toddlers get a boost from self-explanation;
learning scenarios where self-explanation probably isn’t helpful; and
tips for reaping the benefits of self-explanation at home and in the classroom.
Observing. Listening. Tinkering. Testing. Reading. Spaced repetition.
These are some of the ways that people learn. But to deeply understand a new concept – to really master it – one of the best strategies of all is to explain it.
Explain it to yourself, or to an imaginary audience. You don’t actually need a conversation partner to benefit from this technique. The main goal is to come up with an explanation that helps you make sense of the new information (Rittle-Johnson and Loeher 2017).
Why would this enhance learning? There are probably many reasons, some of them very straightforward. For example, when we stop to explain our reasoning out loud, we may end up spending more time engaging with the concepts.
But there are additional possibilities.
Self-explanation encourages us to connect new learning to what we already know – a key strategy for long-term learning
As Bethany Rittle-Johnson and Abbey Loehr note, students asked to self-explain often refer to their prior knowledge when they explain the steps they are taking to solve a problem. This helps them integrate the new knowledge into long-term memory.
And what if the new information conflicts with students’ prior beliefs? Self-explanation is still beneficial, because it makes students notice the mismatch – which may prompt them to try to figure out what’s going on (Rittle-Johnson and Loehr 2017).
Self-explanation may also make students aware of the underlying structural features of what they are learning
This is an extremely valuable lesson, because students don’t just learn how to solve a specific problem. They may also extract general principles that they can apply to new situations (Rittle-Johnson and Loehr 2017).
How do we know that self-explanation works?
For more than two decades, researchers have tested this strategy in experiments on children and adults alike. What do these studies tell us?
As it turns out, self-explanation isn’t terribly helpful for learning about things that lack regular, organizing principles.
But when we’re trying to make sense of information that is known to follow a clear, consistent pattern, self-explanation can be an effective learning tool.
This often comes up when students are learning mathematics, logic, or computer science. It can arise when we’re learning about certain science phenomena, too — pretty much any situation where we are trying to understand how underlying principles can explain what’s going on.
For instance, take chess – a rule-based game that rewards logic and strategy.
In one experiment, people lacking any prior knowledge with chess were taught the basic rules. Then they watched the final stages of a game played by a computer, with each participant being randomly assigned to one of three conditions:
Observe only. The participant simply watches the endgame unfold.
Predict and explain. While watching the game, the participant must try to predict each move that the computer will make, and also explain – out loud – why the computer would make these moves. If a prediction turns out to be wrong, the participant must try to explain why the computer did something different.
Predict only. The participant must predict each move, but without explaining.
Afterwards, study participants had to play a series of endgames themselves. What happened?
Chess novices were more accurate in their predictions when they were also required to explain. And the predict-and-explain group performed the best when it was their turn to play (de Bruin et al 2007).
In other words, self-explanation seems to have helped players go from the specific (anticipating the best moves in one particular scenario) to the general (learning the underlying, strategic principles that apply to all games).
Similar results have been reported in other experiments – including experiments where middle and high school students have been asked to prepare for a math or science test by looking over “worked examples” (i.e., problems with accompanying solutions).
When students verbalized the reasoning behind the steps of a given solution – or explained why they believed a certain answer to be true — they performed better on subsequent tests. (Rittle-Johnson et al 2017).
But watch out: There are limitations and caveats.
For self-explanation to be helpful, we need to supply kids with the right background information, and guide them toward the correct explanations.
In addition, self-explanation isn’t the only way to develop a strong conceptual understanding. More about this below.
But – overall — weighing the evidence across more than 60 studies, investigators conclude that prompting students to self-explain can be a “potentially powerful” learning intervention (Bisra et al 2018). And it isn’t just for secondary school students.
The self-explanation technique for young children
Example: Self-explanation helps 5-year-olds detect rule-based patterns
In a study of 5-year-olds, Bethany Rittle-Johnson and her colleagues (2008) gave kids some pattern-detection problems to solve.
Each problem consisted of a sequence of 6 plastic bugs like this:
and kids were asked what comes next (e.g., a red spider).
After children answered, they were told the official solutions. Then they were asked to explain why the official answers were correct. The researchers put another group of kids through the same procedure, but without asking them to explain. Which group developed better pattern-detection abilities? When given new puzzles to solve, the “explainers” performed better.
Example: Self-explanation may help young children generate hypotheses about how things work – and encourage them to test their hypotheses
Evidence for this comes from experiments that Christine Legare conducted on children who ranged in age from 2 to 6 years.
The children were shown new toys. Some of these toys worked in predictable ways (consistently lighting up in response to a specific action). Others didn’t follow the same pattern. And when kids provided causal explanations for the non-conforming toys, they went on to play more with these toys – investigating them thoroughly, testing multiple (self-generated) hypotheses.
By contrast, when children offered non-causal explanations (like, “It wasn’t supposed to turn on”), they didn’t explore as much. It was as if coming up with a causal hypothesis inspired kids to investigate, tinker, and test (Legare 2012).
Example: Self-explanation may prompt preschoolers to think about the hidden or unseen properties of a system
Two studies support this conclusion. In both, researchers presented preschoolers (aged 3 to 5) with a new device to analyze. Then they asked the children to explain how the device worked.
Compared with kids who were merely asked to describe the device or it’s behavior, the “explainer” kids were more likely to focus on the device’s structural, causal features, and less likely to focus on superficial perceptual features (Legare and Lombrozo 2014; Walker et al 2014).
What about counter-examples? When is self-explanation less helpful?
As we’ve seen, self-explanation may be valuable because it makes children aware of what they don’t yet understand. If that’s true, then we might expect self-explanation to be less helpful when kids are already well-informed about the concepts.
And that seems to be the case. When researchers provided school-aged children with high-quality, concept-driven instruction in mathematics, kids received no added benefits from self-explanation (Rittle-Johnson 2008).
On the flip side, self-explanation can cause trouble if kids
lack enough background to come up with correct explanations, or
end up fixating on something that isn’t true.
To use this technique effectively, it’s important to provide kids with the information they will need to make new connections. It’s also important to ask students the right questions — carefully-selected, thoughtful questions that can help guide students to the right solutions.
In addition, we should steer away from problems with counter-intuitive solutions.
It isn’t realistic to expect kids to rediscover major STEM concepts on their own. On the contrary, we should expect that kids will favor misconceptions about many things (such as concepts in physics and biology) unless we explicitly debunk these ideas (Keleman 2019).
So there are limits to what kids can learn through self-explanation, and we can see how this plays out in another experiment by Bethany Rittle-Johnson.
Example: Kids fail to learn a new mathematic concept through self-explanation
In this study, Rittle-Johnson presented 85 elementary school students (grades 3 through 5) with algebraic problems like these:
3 + 4 + 8 = _ + 8
Some kids were given explicit instructions on a procedure to follow (e.g., “Add together 3+4+8, then subtract 8 from the sum…”). Others were simply asked to discover their own procedure. Neither group of kids received instruction in the underlying concept of equivalence.
Afterwards, half the kids in each group were asked to provide explanations for their solutions. The results?
The researchers found that self-explanation helped reinforce a child’s mastery of the procedures, and it helped kids apply their procedures to new problems.
But kids didn’t show an improved understanding of why the procedures worked. They weren’t more likely to understand that the equal sign means sums on both sides must be equal (Rittle-Johnson 2006).
What about the role of listeners?
We’ve seen that learners can benefit from trying to explain. Does it matter if there is an audience? It might.
In the bug experiment for 5-year-olds, Rittle-Johnson and colleagues found that self-talk helped kids learn. But kids made even bigger gains when they explained their ideas to their mothers.
Moreover, an experimental study by Brown and Kane (1988) offers intriguing hints than even 3-year-olds get a boost from trying to teach someone else.
The study worked like this: Kids were given a chance to try to solve a problem encountered by a character from a story: a man who couldn’t reach a high shelf.
If the kids were stumped, the researchers gave them the solution: There were some spare tires nearby. Stack the tires to make a stool.
Afterwards, kids were presented with a second, analogous story about a farmer who needed to stack hay bales on a tall tractor.
Could the children solve this second problem by themselves? It depended on what happened next.
Some kids were told to simply recount the story before answering.
Others were told to teach a puppet the solution.
And that simple difference had a big impact. The 3-year-olds who were asked to teach were twice as likely to solve the problem on their own.
But – overall – research on the subject of kids-playing-teacher still has a long way to go.
It’s clear that teaching others can help us learn (e.g., Annis 1983; Lachner et al 2022). In addition, it makes sense that young children in particular might benefit from a conversation partner or audience. It is motivating, and may help them stay on task.
However, it’s not clear if other students are better off teaching others, as opposed to focusing on self-explanation (Pi et al 2022). Self-explanation may have the advantage of keeping the learner focused on his or her personal struggle to understand.
Putting it all together: How can we reap the biggest benefits?
Four tips for making self-explanation an effective learning tactic
As noted above, self-explanation isn’t always helpful. Bethany Riddle-Johnson and her colleagues (2017) have identified some of the pitfalls, and offered suggestions for making self-explanation an effective learning tactic. Here are some tips based on their ideas.
1. If there are abstract concepts to learn, don’t expect kids to discover these on their own.
Give them the necessary background information. Let kids learn through direct instruction or conversation. Then you can try giving them related problems to solve — and try out the self-explanation technique.
2. Help kids develop high-quality explanations by modeling, or providing partial answers.
For example, in the case of the pattern detection bug sequence (above) you might first walk your child through an example that you solve and explain. Describe the sequence you see, and point out the repeated pattern. Then show how your answer (the next proposed bug in the sequence) fits.
For an alternate approach, you can offer kids with a partial explanation, and ask them to fill in the missing steps. In some studies, teachers have presented students with several explanations, and asked them to choose the best one.
3. Ask kids to explain why correct information is correct.
Most experiments of self-explanation have asked students to explain a correctly worked out example. If a child has come up with an incorrect solution, and doesn’t realize that, asking him or her to justify the solution may not be terribly helpful.
4. Show kids expanples of errors based on common misconceptions, and ask kids to explain why such errors are wrong.
This is different than asking kids to justify an incorrect answer. The child begins with the knowledge that something is incorrect, and attempts to explain the nature of the mistake.
Riddle-Johnson and her colleagues note that research is limited in this area. But several studies suggest that identifying and explaining flawed reasoning can help students better understand correct reasoning. It may also teach kids to avoid using flawed reasoning themselves.
More information
For a related look at self-explanation and learning, see my article about the role that gestures play in helping kids learn math, science, and the meaning of new words.
For more information about science education, visit my page, “Science for kids: How to raise a science-minded child.”
And if you are interested in early childhood stem, check out my pages about preschool math and preschool science.
References: How self-explanation and teaching others can help kids learn math and science
Annis LF. 1983. The processes and effects of peer tutoring. Human learning: Journal of Practice and Research Applications 2(1): 39-47.
Benware CA and Deci EL. 1984. Quality of learning with an active versus passive motivational set. American Educational Research Journal 21(4): 755-65.
Bisra K, Liu Q, Nesbit JC, Salimi F, and Winne PH. 2018. Inducing Self-Explanation: a Meta-Analysis. Educ Psychol Rev 30: 703–725.
Brown AL and Kane MJ. 1988. Preschool children can learn to transfer: Learning to learn and learning from example. Cognitive Psychology 20: 493-523.
de Bruin ABH, Rikers RMJP, and Schmidt HG. 2007. The Effect of Self-Explanation and Prediction on the Development of Principled Understanding of Chess in Novices. Contemporary Educational Psychology 32(2):188-205.
DeCaro MS, Rittle-Johnson B. 2012. Exploring mathematics problems prepares children to learn from instruction. J Exp Child Psychol. 113(4):552-68.
Lachner A, Hoogerheide V, van Gog T, and Renkl A. 2022. Learning-by-Teaching Without Audience Presence or Interaction: When and Why Does it Work?. Educ Psychol Rev 34: 575–607
Legare CH and Lombrozo T. 2014. Selective effects of explanation on learning during early childhood. J Exp Child Psychol. 126:198-212.
Legare C. 2012. Exploring explanation: explaining inconsistent evidence informs exploratory, hypothesis-testing behavior in young children. Child Dev. 83(1):173-85.
Legare CH, Gelman SA, and Wellman HM. 2010. Inconsistency with prior knowledge triggers children’s causal explanatory reasoning. Child Dev. 81(3):929-44.
Matthews P and Rittle-Johnson B. 2009. In pursuit of knowledge: Comparing self-explanations, concepts, and procedures as pedagogical tools J Exp Child Psychol. 104(1):1-21.
Pi Z, Zhang Y, Shi D, Guo X, and Yang J. 2022. Is self-explanation better than explaining to a fictitious student when learning from video lectures? British Journal of Educational Technology 53(6): 2012-2028
Rittle-Johnson B. 2006. Promoting transfer: effects of self-explanation and direct instruction. Child Dev. 77(1):1-15.
Rittle-Johnson B and Loehr AM. 2017. Eliciting explanations: Constraints on when self-explanation aids learning. Psychonomic Bulletin & Review 24(5): 1501–1510.
Rittle-Johnson B, Loehr A, and Durkin K. 2017. Promoting self-explanation to improve mathematics learning: A meta-analysis and instructional design principles. ZDM Mathematics Education 49: 559-611.
Rittle-Johnson B, Saylor M, Swygert KE. 2008. Learning from explaining: does it matter if mom is listening? J Exp Child Psychol. 100(3):215-24.
Walker CM, Lombrozo T, Legare CH, and Gopnik A. 2014. Explaining prompts children to privilege inductively rich properties. Cognition. 133(2):343-57.
Wong RM, Lawson MJ, and Keeves J. 2002. The effects of self-explanation training on students’ problem solving in high-school mathematics Learning and Instruction 12(2): 233-26.
Content of “How kids learn math and science” last modified 5/2024. Portions of the text are based on earlier versions of this article, written by the same author
image of butterflies and spiders by Parenting Science
image of young children writing on chalkboard by M Stock / istock