In short
- Dodell-Feder and Tamir (2018) combined 14 experiments and found that reading fiction gave a small improvement on tests of social understanding. The participants were adults.
- The best-known experiment, by Kidd and Castano (2013), did not replicate when Panero and colleagues (2016) repeated it with 792 participants.
- Lecce and colleagues (2014) found that a training program built on conversations about the mind raised the scores of 9- and 10-year-olds on theory-of-mind tasks.
- Abrami and colleagues (2015) found that critical thinking can be taught and that dialogue helps. Its link to story reading is indirect.
Many school brochures say that reading stories builds empathy. The claim sounds right, and teachers want it to be true. The research gives it only partial support. This article reports the experiment that made the claim famous, the attempt to repeat it, and two meta-analyses, with what each measured and whom it studied. It then sets out what a teacher can do with a story, what research says about teaching critical thinking, and how to describe all this to parents.
Do stories build empathy in children?
Perhaps a little, and the evidence is weaker than the claim. Reading stories may help readers understand other people slightly better. The effect measured in experiments is small. The best-known experiment did not replicate. Most of the studies were done with adults. None of them measured how children treat each other, and we found none with pupils reading in a foreign language.
It helps to know what the studies measured. The experiments mostly measured theory of mind, not empathy in the everyday sense of caring how someone feels. Theory of mind is the ability to understand that other people hold beliefs and wishes that may differ from your own.
The most common test is the Reading the Mind in the Eyes Test. A person sees photographs of the eye region of faces. For each one, they choose which of four words best describes what the person is thinking or feeling. A higher score shows skill at reading expressions. It does not show that someone behaves kindly.
The famous experiment and the failed replication
Kidd and Castano (2013) reported five experiments with adults in the journal Science. Participants who read literary fiction did better on tests of theory of mind than those who read nonfiction, popular fiction or nothing at all. The authors described the improvement as temporary. The study received a great deal of attention.
Panero and colleagues (2016) tried to repeat the finding. Three independent research groups tested 792 participants. Each person was randomly assigned to read literary fiction, popular fiction or nonfiction, or to read nothing. The researchers found no significant advantage on the eyes test for literary fiction over any other condition.
They did find something else. A measure of lifetime exposure to fiction predicted scores on the eyes test in every condition. The measure asks people to pick out the names of real authors from a list.
The authors offer two likely explanations. People who read others well may be drawn to fiction. Or a lifetime of reading may slowly strengthen the skill. They add that other factors, such as verbal ability, may also play a part.
What the meta-analyses found
A meta-analysis combines the results of many studies. Two are relevant here, and they answer different questions.
Mumper and Gerrig (2017) asked whether people who read more score higher. They combined correlations between lifetime reading habits and measures of empathy and theory of mind. Fiction reading had a larger correlation with these measures than nonfiction reading. The authors describe the effects as small. The correlation with fiction was about 0.07 for empathy questionnaires and about 0.2 for theory-of-mind tasks.
A correlation cannot show cause. Readers of fiction may differ from other people before they open a book. The authors themselves call for research on causal mechanisms.
Dodell-Feder and Tamir (2018) asked the causal question. They combined 53 effect sizes from 14 experiments in which people were assigned to read fiction, nonfiction or nothing. Fiction reading led to a small, statistically significant improvement in social-cognitive performance. The authors report that the result held up in their checks and did not appear to come from publication bias.
The limits matter for a school. The samples were university students and adults recruited online, with a mean age of about 28. The authors say it is not clear whether the improvement is short-term or lasting. They recommend that future work tests whether the effect transfers to real-world social behaviour.
| Study | Who took part | What was measured | What was found |
|---|---|---|---|
| Kidd and Castano (2013) | Adults, in five experiments | Theory-of-mind tests after reading a passage | Higher scores after literary fiction than after nonfiction, popular fiction or nothing |
| Panero and colleagues (2016) | 792 participants, tested by three research groups | The eyes test after reading a passage, or nothing | No significant advantage for literary fiction |
| Mumper and Gerrig (2017) | 30 independent samples from correlational studies | Lifetime reading habits, set against empathy questionnaires and theory-of-mind tasks | Small correlations, larger for fiction than for nonfiction |
| Dodell-Feder and Tamir (2018) | 14 experiments with students and adults, mean age about 28 | Social-cognitive tests and questionnaires after reading fiction, nonfiction or nothing | A small, statistically significant improvement after fiction |
Why stories might help: the theory
The best-known explanation comes from Mar and Oatley (2008). They argue that fiction works as a simulation of the social world. A story abstracts, simplifies and compresses social experience. The reader lives through it with the characters and, in the authors' words, learns through experience.
On this view, a story lets a pupil follow a hard choice and see its outcome without paying for it. Mar and Oatley propose that this can help readers understand people who are different from themselves.
This is a theory, argued from existing research. It is not a classroom trial. It explains why an effect might exist. It does not show how large the effect is, and the experiments above suggest it is small.
What a teacher can do that has support
The reading alone has thin support. The conversation has more. Lecce and colleagues (2014) designed a training program for children aged 9 and 10. It was built on conversations about the mind, that is, about what people think and believe.
Forty-five children took the training, and 46 were in an active control group. The trained group showed significantly greater gains in theory of mind. The difference was still there two months later.
This is one study with 91 children, and it measured test performance, not behaviour. It was a training program, not a reading lesson. Applying it to story discussion is our step, not the authors'. A story is a convenient place to hold such a conversation, because it gives the class minds to talk about.
Three moves follow. Nobody has tested them as a package with pupils reading in a foreign language.
- Talk about feelings and reasons. Ask what a character feels and why they acted. 'Why do you think she said nothing?'
- Ask pupils to take a character's view. 'You are the brother. What do you know? What do you want?' Then ask the same of a second character.
- Connect the story to their own lives. 'Has this happened to you or to someone you know? What did you do?' Abrami and colleagues (2015) found that authentic or situated problems helped in the teaching of critical thinking.
Classroom example (an illustration, not a tested script)
Story moment: Dina sees that her best friend has copied her homework. She says nothing to the teacher, and she does not speak to her friend all day.
Feelings and reasons: 'How does Dina feel at lunch? Why does she say nothing to the teacher?'
The other character's view: 'You are the friend. Why did you copy? What do you think when Dina stops talking to you?'
Evidence: 'What in the story tells you that?'
Their own lives: 'Has a friend ever done something you did not like? What did you say?'
For a hesitant pupil: 'Is Dina angry, sad or afraid? Choose one and say why.'
Fit the questions to age and level
The advice here comes from classroom practice, not from the studies. Younger or lower-level pupils need concrete questions: 'Was he happy or sad? Why?' Older pupils can handle a character with two motives, or a choice with no clear right answer.
In an English lesson, a pupil may understand a character's feeling and lack the words to say it. Give sentence frames, such as 'She felt ... because ...'. Judge the idea and the English separately. Teaching Values in the EFL Classroom deals with this problem, and Discussion Questions to Draw the Moral From a Story gives more questions.
One informal sign is worth watching over a term. Early answers are often general: 'because he is bad'. Later answers may name the character's fear, hope or mistake. That is a note for you, not a measurement of empathy.
Do stories build critical thinking?
Here the evidence is about teaching, not about stories. Abrami and colleagues (2015) carried out a meta-analysis of instruction in critical thinking. They define critical thinking as purposeful judgement that results in interpretation, analysis, evaluation and inference.
The review covered 341 effect sizes from experimental and quasi-experimental studies that used standardised measures. The average effect of instruction was 0.30, and results varied from study to study. The authors conclude that there are effective strategies at all educational levels and across subjects. Three things had positive effects on critical thinking skills: the opportunity for dialogue, exposure to authentic or situated problems and examples, and mentoring.
The link to story reading is indirect. The review did not test story discussion. A story discussion can supply two of the three ingredients. It is dialogue, and a character's dilemma is a situated problem. That makes it a reasonable place to practise reasoning. It does not prove that reading stories makes pupils think critically.
The reasoning habits fit a story lesson easily:
- Ask for evidence. 'What in the story tells you that?'
- Ask for another explanation. 'Could there be a different reason?'
- Ask pupils to weigh a choice. 'Was that the best thing to do? What else could she do?'
- Let pupils disagree, and ask each side for its reasons.
How to describe this to parents without overclaiming
Parents like to hear that reading builds empathy. Tell them what the class does and what the research supports.
- Describe the activity. 'This term your child read six stories and discussed why the characters acted as they did.'
- State the aim, not a result. 'We use stories to practise seeing a situation from another person's side.'
- Quote the pupil. One sentence your child said about a character tells a parent more than a label.
- Do not promise a change in character. Avoid 'our reading program makes children more empathetic'. The studies do not show that.
- Do not score or rank empathy. The studies used research tests, not classroom marks, and a school should not invent a score.
How to Share Values Learning With Parents has message templates. The guide to values and life lessons through stories sets this evidence beside the research on moral stories and school programs.
How iRead handles this
In iRead, stories carry a moral, such as honesty, kindness, courage or perseverance. Quiz questions are generated from the story itself, and they include comprehension of the plot and the moral of the story. Values learned appear next to vocabulary gained in progress views, including the parent's view.
The core principle: A story gives the class minds to talk about. The measured effect of reading alone is small, so plan the conversation.
See it in iRead: Stories in iRead carry a moral, and dedicated quiz questions ask what the story taught. Values learned appear in progress views, including the parent's view.
See values through storiesKey takeaway
Stories may help children understand other people a little, and the research does not support a stronger claim. Use the story as a reason to talk about what characters think, feel and choose.
Frequently asked questions
Does reading fiction make children more empathetic?
The evidence is limited. Dodell-Feder and Tamir (2018) found a small positive effect of fiction on tests of social understanding across 14 experiments, and the participants were adults. No study cited here measured how children treat each other. Reading may help a little. It has not been shown to change behaviour.
Did the Kidd and Castano study on fiction and empathy replicate?
No. Kidd and Castano (2013) reported that reading literary fiction improved adults' scores on theory-of-mind tests. Panero and colleagues (2016) repeated the experiment with 792 participants across three research groups. They found no significant advantage for literary fiction over popular fiction, nonfiction or no reading.
Do stories improve critical thinking?
Not proven directly. Abrami and colleagues (2015) found that critical thinking can be taught, with dialogue and authentic problems among the things that help. A story discussion can provide both. No study cited here tested whether reading or discussing stories raises scores on a critical thinking measure.
Does story discussion build empathy in a second language?
Nobody has tested it directly, as far as we found. The studies cited here were done with people reading in their first language. In practice, pupils may understand a character and lack the English to say so. Sentence frames and questions with options help them answer.
Sources
- Mar, R. A., & Oatley, K. (2008). The function of fiction is the abstraction and simulation of social experience. Perspectives on Psychological Science, 3(3), 173-192. doi.org/10.1111/j.1745-6924.2008.00073.x
- Kidd, D. C., & Castano, E. (2013). Reading literary fiction improves theory of mind. Science, 342(6156), 377-380. doi.org/10.1126/science.1239918
- Panero, M. E., Weisberg, D. S., Black, J., Goldstein, T. R., Barnes, J. L., Brownell, H., & Winner, E. (2016). Does reading a single passage of literary fiction really improve theory of mind? An attempt at replication. Journal of Personality and Social Psychology, 111(5), e46-e54. doi.org/10.1037/pspa0000064
- Mumper, M. L., & Gerrig, R. J. (2017). Leisure reading and social cognition: A meta-analysis. Psychology of Aesthetics, Creativity, and the Arts, 11(1), 109-120. doi.org/10.1037/aca0000089
- Dodell-Feder, D., & Tamir, D. I. (2018). Fiction reading has a small positive impact on social cognition: A meta-analysis. Journal of Experimental Psychology: General, 147(11), 1713-1727. doi.org/10.1037/xge0000395
- Lecce, S., Bianco, F., Devine, R. T., Hughes, C., & Banerjee, R. (2014). Promoting theory of mind during middle childhood: A training program. Journal of Experimental Child Psychology, 126, 52-67. doi.org/10.1016/j.jecp.2014.03.002
- Abrami, P. C., Bernard, R. M., Borokhovski, E., Waddington, D. I., Wade, C. A., & Persson, T. (2015). Strategies for teaching students to think critically: A meta-analysis. Review of Educational Research, 85(2), 275-314. doi.org/10.3102/0034654314551063