Overall 3E, Empower, and Empathize scores were higher in the second AI-patient encounter, but the uncontrolled design precludes attributing this to training¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Mean overall 3E performance increased from Session 1 to Session 2 (0.403 to 0.471; one-sided p<0.001), as did Empower and Empathize, while the Be Explicit difference was not significant. → Masum Hasan 2026

Evidence¶

Masum Hasan 2026¶

Masum Hasan, Ron Epstein, Thomas Carroll, Ehsan Hoque. (2026). Scalable AI-based clinical communication training and automated assessment. https://arxiv.org/abs/2609.22517 doi:10.48550/arxiv.2609.22517

q2 · i? · design · r2

Within-participant comparison of LLM-judged 3E scores across two counterbalanced AI-patient encounters among 59 participants. The "overall score increased from 0.403 to 0.471"; Be Explicit showed no significant change (∆ = 0.014; p=0.1204).

"The overall score increased from 0.403 to 0.471 (∆ = 0.068; 95% CI, 0.038–0.098; one-sidedp <0.001). Higher scores were also observed for Empower (∆ = 0.083; 95% CI, 0.046–0.121; one-sidedp <0.001) and Empathize (∆ = 0.107; 95% CI, 0.055–0.159;p= 0.001), whereas the difference for Be Explicit was not significant"

Discussion¶

Related Claims¶

• Participants with prior communication-training experience achieved higher overall 3E scores than those without — related
• Controlled GenAI virtual-patient and communication studies reported short-term skill gains, but positive findings were not universal — a broader claim this one bears on
• HAL's text-based human-likeness advantage did not transfer to embodied audiovisual interaction, and measured 3E performance was virtually identical between models — related