Skip to content

Building Empathy

Principle · All principles
Evidence · 15 claims (11 for, 4 mixed) · 26 studies (10 causal, 7 quant-synthesis, 4 review, 2 qualitative, 1 associational, 1 design, 1 theoretical), q1–q4 · 6 of 26 report an effect size · 9 claims rest on one study

Conditional relationship

A learner who forms an interpretation of another person or group from limited information may come to represent that person's perspective more accurately, and may hold more positive attitudes towards the group, after a structured activity that makes the other's point of view concrete: an account to listen to or read, a prompt to take that perspective, and a chance to compare one's own reading with what the other person says. The relationship is conditional on the outcome named (empathic concern, mental-state inference, attitude towards a group and intergroup behaviour are different outcomes), on the dose (a single passage or exercise against a sustained programme), on the target (one person against a group, and how stigmatized that group is for these learners) and on the horizon (immediately after the activity, weeks later, or in later conduct). No matching pattern page holds a conditional model of it yet.

Learners: any age (child to adult); the wiki's evidence comes from adults in laboratories, students in grades 4–6 in classrooms, and undergraduates in experiential programmes. Knowledge type: attitude-motivation and social inference, not a procedure with a correct answer. The wiki has no claim testing empathy-building as the general instructional relationship stated here; the claims below test parts of it.

Fitting the design to a situation

Ask for four facts when the brief omits them: which outcome is meant (concern for a person, accurate reading of someone's view, attitudes towards a group, or conduct); who the other person or group is, and how stigmatized they are for these learners; how much time there is (one activity or a programme over weeks); and who the learners are (children, students, or adults such as teachers who act on others).

If the brief says… Then change… Basis
The goal is better attitudes towards a particular group, in a school class Run a programme over several weeks: first-person accounts from members of the group, each followed by a discussion of their perspective and of how much the group varies; measure attitudes again weeks later Building empathy improves intergroup attitudes [+M] (grades 4–6, cluster-randomised: about a third of an SD, up to 8–13 weeks)
Only one session or a single exercise Use one concrete account, a prompt to take the person's perspective, and a comparison of readings; do not promise lasting attitude change, and re-measure later if you can Building empathy improves intergroup attitudes [~M] (a review of 418 experiments: most test light-touch interventions whose long-term impact is unclear)
The goal is behaviour (contact, cooperation, inclusion), not only attitudes Add a real, structured contact opportunity with the group, and measure the behaviour separately from attitude ratings The classroom programme raised attitudes but not registration for a contact event [~M]; Intergroup contact reduces prejudice [+W] (a meta-analysis cited in a policy report)
The plan relies on reading a story, novel extract or film alone Pair the reading with discussion and a written perspective task; do not expect a single passage to change social-cognitive performance Reading fiction gives at most a very small gain on theory-of-mind tasks [~M] (g = .15–.16; a single literary passage showed no advantage in a 792-adult replication)
The group is highly stigmatized for these learners Expect little change straight after the activity; avoid framing that asks learners to approve of the group; check again one to two weeks later For the most stigmatized group the effect was weak at once and appeared 1–2 weeks later [~M] (laboratory experiments with adults)
Learners are young children (preschool or early primary) Use role-play and pretend play with adult support, inside a sequenced social-emotional curriculum, rather than discussion of perspectives SEL programmes improve social-emotional skills [+M] (ages 2–6: d = 0.30 social, 0.54 emotional competence); Pretend play facilitates perspective taking [+W] (a review)
A school-wide or year-long programme Choose a sequenced, active, focused and explicit programme, with perspective-taking taught as one skill among several; follow outcomes over the next year SEL programs improve social-emotional skills [+M]; SEL benefits persist at follow-up [+M] (empathy is not isolated in either)
The learners are teachers or staff who discipline students Give a brief online exercise on an empathic rather than punitive mindset about discipline, with students' own accounts of being disciplined A brief empathic-discipline exercise for teachers cut suspensions [+M] (middle-school teachers; the preregistered replication's overall effect was small and marginal, larger for previously suspended students)
Higher education or adult learners in experiential programmes (practicums, service learning, study abroad) Build in one-to-one contact with the people served and a structured reflection on what the learner understood of their view Students perceive that interaction in experiential learning fosters empathy [+W] (12 interviews, self-report); One-on-one contact with community partners disrupts stereotypes [+W] (a partner's report)
Setting is online-self-paced Use recorded first-person accounts, a written summary of the speaker's view that the learner checks against the speaker's own words, and a private before-and-after rating untested proposal
The topic is politically or personally charged for these learners Separate understanding from agreement: the learner first summarises the other view in terms its holder would accept, and only then states their own position untested proposal
Success is to be judged by self-reported empathy right after the activity Add a measure not tied to the activity, taken at a delay, and an optional behavioural choice; self-report straight after an activity whose aim is obvious is open to demand untested proposal

Observation, state and explanation

Empathy is an inferred state. What a design can observe is a response under stated conditions: what a learner says another person feels or wants, how they summarise someone else's view, a rating on a feeling thermometer, a choice to take part in contact. "Empathic" is a reading of those responses, and agreement with a view is not the same response as understanding it. Self-reported empathy after an exercise designed to produce it is especially open to demand: learners know what the activity is for.

Response Plausible explanations Observation that could change the interpretation
Learner summarises another person's view accurately but rejects it with hostility Understanding without concern; identity threat raised by the activity; disagreement on substance that accurate perspective-taking does not remove Ask the learner to say what the other person would object to in their summary, and separately what they themselves disagree with; repeat with a less threatening topic.
Learner reports more warmth towards a group right after a narrative, with no later change in conduct Short-lived state change; demand from an activity whose aim is obvious; attitude change that does not reach behaviour Re-measure at a delay with a measure not tied to the activity, and record an optional behavioural choice (as Weiss et al. 2023 did with contact registration).
Learner attributes feelings to a character or interviewee that the account does not support Projection of own perspective; stereotype filling the gaps; too little information in the account Ask what in the account supports each attribution; give a second account from the same person and see whether the learner revises.
Learner says little in a structured listening round Low engagement; discomfort with disclosure or the topic; language demand of the task; reading the norms as unsafe Offer a written or paired response mode and a less personal prompt; ask privately what made the round hard.
Learner's interpretation changes after dialogue Perspective-taking; conformity to the group or facilitator; new factual information alone Ask the learner to explain what changed and why, and check the change against a private response made before the group discussion ends.

These rows are proposals from the claims below and from the old page's cautions, untested with learners. A probe can itself change the state, and none of them identifies a cause on its own; record what each one leaves unresolved.

Evidence and qualifications

  • Building empathy improves intergroup attitudes [~M]: three laboratory experiments in which inducing empathy for one member of a stigmatized group led to more positive attitudes towards the group, with the effect for the most stigmatized group weak at once and appearing 1–2 weeks later; a cluster-randomized classroom programme for grades 4–6 that combined outgroup exposure with perspective-taking discussions and raised positive affect towards outgroups by about a third of a SD up to 8–13 weeks later, without changing registration for an actual contact event; and a review of 418 prejudice-reduction experiments that finds perspective-taking the modal intervention, signs of publication bias, and that 76% of studies test light-touch interventions whose long-term impact is unclear. The classroom programme combined three mechanisms, so its effect is not attributable to empathy alone. Marked mixed: it supports an attitude change and limits it for one-off exercises and for behaviour. Sources read from abstracts; not settled by the judge.
  • Fiction reading improves empathy [~M]: a meta-analysis of 14 experiments finds reading fiction gives a small improvement on social-cognitive tasks against nonfiction or no reading (g = .15–.16, below the small-effect threshold), and a three-lab randomized replication with 792 adults found no advantage for a single literary passage on the eyes test, while lifetime fiction exposure predicted scores. The outcome is social-cognitive task performance, not empathy as the title says. A limiting claim: brief narrative exposure alone predicts little. Reading literary fiction improves theory of mind [~M] carries the same three entries and is a merge candidate.
  • Students perceive that interacting with others in experiential learning fosters empathy by enhancing perspective-taking [+W]: a thematic analysis of twelve undergraduate interviews in which students described collaboration, practicums and study abroad as enhancing perspective and fostering empathy. It records what learners reported valuing, not a measured change against a comparison. Checked by the judge on the full text.
  • SEL programs improve social emotional skills [~M]: meta-analyses of universal school SEL programmes (K–12, and ages 2–6) find gains in social-emotional skills, larger for sequenced, active, focused and explicit programmes. The skill composite includes social awareness but the claim page does not isolate empathy or perspective-taking, so it bears on structured, sustained programmes in general rather than on this relationship. Partly checked.

Keep learners, the exact activity and comparator, the outcome instrument and the horizon with each of these when transporting them. A null on one measure (contact registration, the eyes test) is not evidence that nothing changed, nor that the measures are equivalent.

The evidence by question

Each question below links a general claim that gathers the studies behind it into one table, one row per distinct study, with each study's design and effect as recorded and no averaged verdict (eval/synthesis/building-empathy.json, scripts/synthesize_claim.py). The count is of distinct studies across the claim and the narrower claims that bear on it.

Further evidence, not yet read against this model

Claims this page cited before it was rewritten as a conditional model. They are evidence about mechanisms empathy activities may rely on, but none has yet been re-read against the model above; each says how far its own sources have been checked.

Objective and learner-valued goal

The designer's objective and what the learner wants from the activity can differ, and the difference is an observation to record rather than resolve by assumption. Ask what the learner hopes to get from the work, separately from the stated objective, and do not read participation as agreement.

For example, in a nursing course a designer may aim for students to represent a patient's experience of a diagnosis accurately in an empathy interview, while a student may value getting through difficult conversations without freezing. A feeling-thermometer rating would establish neither aim. A summary the patient (or a standardized patient) confirms tests the designer's objective; the student's own account of a later conversation is evidence about theirs. Agreement between the two is something to check, not assume.

What would revise this model?

The expectation that structured perspective-taking improves attitudes towards a group should weaken if sustained, well-implemented programmes compared with defensible controls fail to show it at a delay, or if their effects disappear once the publication bias the review flags is taken into account. If the attitude change appears without any change in measured perspective-taking or empathic concern, the mechanism this page assumes is wrong, and another explanation (exposure, information, demand) fits better. If gains on ratings never reach conduct, keep the two outcomes apart rather than calling both empathy. Do not protect the model by relabelling every null as poor facilitation.

An immediate change in a rating, a delayed change in attitude, more accurate inference of what another person thinks, and changed behaviour towards others are separate claims. The present evidence does not establish which activity type, dose or facilitation produces durable change, does not test interviews, journaling or listening rounds as such, and does not say how the relationship differs by age.

Claims

Examples

  • Empathy interviews: Learners interview a peer, client, or community member using prompts focused on needs, frustrations, and lived experience.
  • Perspective journaling: After discussion or field observation, learners write from another stakeholder's viewpoint and then reflect on what they still do not know.
  • Narrative comparison: Learners compare multiple first-person or community narratives about the same issue before discussion.
  • Structured listening rounds: Small groups take turns summarizing another person's view before offering their own response.
  • Conduct empathy interviews to understand user experiences with technology

Key Sources

  • Nelsestuen, K., & Smith, J. (2020). Empathy interviews. The Learning Professional, 41(5), 59-59.
  • Sachs, M. E., Kaplan, J., & Habibi, A. (2019). Echoing the emotions of others: Empathy is related to how adults and children map emotion onto the body. Cognition & Emotion, 33(8), 1639-1654. https://doi.org/10.1080/02699931.2019.1591938
  • Setlhodi, I. I. (2018). Mentoring in the web-class for adult learners. In Handbook of research on student-centered strategies in online adult learning environments (pp. 206-220). IGI Global. https://doi.org/10.4018/978-1-5225-5085-3.ch010
Page metadata
Field Value
type principle
id building-empathy
title Building Empathy
description For a learner forming an interpretation of another person or group, structured perspective-taking with concrete accounts and comparison of interpretations may improve attitudes towards that group, a relationship qualified by outcome, dose, target and horizon.
status review
canonical True
generated by claude/unspecified on 2026-10-02
trust unverified — no one has confirmed this page's content
sources 2 — sachs-2019, setlhodi-2018

Frontmatter is the page's machine-readable record — see the Schema & Guide.