REMind did not significantly increase measured empathy, but baseline human-directed empathy correlated positively with robot-directed empathy¶

Claim · All claims
Evidence · 1 study (2 entries) · design r2 · q2 · i3 large
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Empathy scores increased by 15.7 points but the change was not statistically significant; H3(a) was not confirmed. → Elaheh Sanoubari 2026 q2 i3 Empathy toward the human character at pre-test was significantly positively correlated with empathy toward the robot avatar at post-test (Spearman's ρ = .62). → Elaheh Sanoubari 2026 (2)

Evidence¶

Elaheh Sanoubari 2026¶

Elaheh Sanoubari, Neil Fernandes, Keith Rebello, Alicia Pan, Andrew Houston, and Kerstin Dautenhahn. (2026). Play-Testing REMind: Evaluating an Educational Robot-Mediated Role-Play Game. https://arxiv.org/abs/2604.00300 doi:10.48550/arxiv.2604.00300

q2 · i? · design · r2

Paired t-test on empathy items (adapted from Pöyhönen et al.) in the play-testing study; the increase of 15.7 points was "not statistically significant, t(17) = 1.62, p = .062". No effect size is printed.

"Empathy scores increased by 15.7 points, but the change was not statistically significant, t(17) = 1.62, p = .062. H3(a) was not confirmed."

Elaheh Sanoubari 2026 (2)¶

Elaheh Sanoubari, Neil Fernandes, Keith Rebello, Alicia Pan, Andrew Houston, and Kerstin Dautenhahn. (2026). Play-Testing REMind: Evaluating an Educational Robot-Mediated Role-Play Game. https://arxiv.org/abs/2604.00300

q2 · i3 · design · r2

Spearman correlation in the same study between pre-test human-directed and post-test robot-directed empathy: "Spearman's ρ= .62, p = .006", indicating children with higher baseline empathy were more likely to empathize with the robot.

"a significant positive correlation was found between empathy toward the human character at pre-test and the robot avatar at post-test, Spearman's ρ= .62, p = .006, confirming H3(b)"

Discussion¶

Learner Variables¶

• Social Awareness and Relationship Skills — outcome: instruction changes it

Related Claims¶

• LLM-annotated EPITOME scores show the Non-Empathetic chatbot expresses significantly lower empathy on all three dimensions, while Standard and Empathetic chatbots are not significantly different — related
• During REMind, nearly all children reported feeling sad for the victim and attributed bystander inaction mainly to low confidence or fear of retaliation — related
• REMind did not significantly change children's expectations that defending would help the victim feel better — related
• Playing REMind significantly increased children's perceived self-efficacy for defending peers — related
• Children who chose to comfort the victim reported higher self-efficacy and victim-help expectations than those who chose to confront the bully (descriptive, imbalanced groups) — related
• After REMind, children's expectations that confronting a bully would stop the bullying shifted toward a calibrated midpoint — related
• Self-actualization (POI) correlates of rated counseling performance reported by Foulds could not be confirmed on replication — related