LLM-annotated empathy dimensions correlate positively with intention to follow, self-efficacy, and chatbot usefulness despite participants not preferring empathetic chatbots¶

Claim · All claims
Evidence · 1 study (2 entries) · causal r1 · q2 · i1 small
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i1 Emotional reaction correlates significantly with intention to follow (ρ = 0.156), self-efficacy (ρ = 0.139), and usefulness (ρ = 0.249). → Li Siyan 2026 q2 i1 Interpretation correlates significantly with chatbot usefulness (ρ = 0.143, p = 0.016). → Li Siyan 2026 (2)

Evidence¶

Li Siyan 2026¶

Li Siyan, Kai-Hui Liang, Shopnil Shahriar, Yilin Ye, Shiyoh Goetsu, Wei-Wei Du, Masahiro Yoshida, Tsunayuki Ohwa, Xuhai Xu, and Zhou Yu. (2026). Invisible Impact of Empathy on Behavioral Change: Isolating the Effect of Empathy in Long-term Physical Activity Coaching Chatbot Interactions. https://arxiv.org/abs/2606.26641 doi:10.48550/arxiv.2606.26641

q2 · i1 · causal · r1

Spearman correlations between LLM-annotated EPITOME empathy dimensions and check-in form items (intention to follow, self-efficacy, usefulness) across collected conversations. All three emotional-reaction correlations are small positive (ρ between 0.139 and 0.249, e.g. ρ = 0.156), so the impact is coded small.

"significant positive correlations between emotional reaction and all three self-report items(intention to follow: 𝜌= 0.156, 𝑝= 0.028, self-efficacy: 𝜌= 0.139, 𝑝= 0.033, usefulness: 𝜌= 0.249, 𝑝= 0.0001)."

Li Siyan 2026 (2)¶

Li Siyan, Kai-Hui Liang, Shopnil Shahriar, Yilin Ye, Shiyoh Goetsu, Wei-Wei Du, Masahiro Yoshida, Tsunayuki Ohwa, Xuhai Xu, and Zhou Yu. (2026). Invisible Impact of Empathy on Behavioral Change: Isolating the Effect of Empathy in Long-term Physical Activity Coaching Chatbot Interactions. https://arxiv.org/abs/2606.26641

q2 · i1 · causal · r1

Same Spearman correlation analysis: the interpretation dimension is significantly positively correlated with chatbot usefulness (ρ = 0.143, p = 0.016). The correlation is small, so impact is coded 1; results are correlational and not causal.

"Our results indicatea significant positive correlation (𝜌= 0.143, 𝑝= 0.016) between interpretation and chatbot usefulness, as well assignificant positive correlations between emotional reaction and all three self-report items"

Discussion¶

Related Claims¶

• In a six-week within-subject study, the Empathetic PA coaching chatbot improves intention to follow and self-efficacy faster over time than the Non-Empathetic condition — related
• Average check-in and weekly ratings show no significant differences between chatbot conditions, with the Non-Empathetic chatbot often rated as more engaging and useful — related
• 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
• Non-expert participants often fail to notice empathy differences between chatbot versions, so self-reported perceived empathy alone is an unreliable assessment method — related
• Agreement between chatbot preference and perceived learning attribution was low (39%) — related