Research Appraisals
Evidence-based critical appraisals of the latest medical research, systematically evaluated using Oxford CEBM methodology.
Showing 2 appraisals
Proceedings of the National Academy of Sciences of the United States of America
Advancing AI negotiations: A large-scale autonomous negotiation competition.
We conducted an international AI negotiation competition in which participants designed and refined prompts for AI negotiation agents. We then facilitated over 180,000 negotiations between these agents across multiple scenarios with diverse characteristics and objectives. Our findings revealed that principles from human negotiation theory remain crucial even in AI-AI contexts. Surprisingly, warmth-a traditionally human relationship-building trait-was consistently associated with superior outcomes across all key performance metrics. Dominant agents, meanwhile, were especially effective at claiming value. Our analysis also revealed unique dynamics in AI-AI negotiations not fully explained by negotiation theory, including AI-specific technical strategies like chain-of-thought reasoning and prompt injection. When we applied natural language processing methods to the full transcripts of all negotiations, we found positivity, gratitude, and question-asking (associated with warmth) were strongly associated with reaching deals as well as objective and subjective value, whereas conversation lengths (associated with dominance) were strongly associated with impasses. The results suggest the need to establish a new theory of AI negotiation, which integrates classic negotiation theory with AI-specific negotiation theories to better understand autonomous negotiations and optimize agent performance.
7 June 2026
Read appraisal →BMJ open
Arts-based empathy education in healthcare: a critical systematic review of pedagogical mechanisms and evidence gaps
OBJECTIVES: To systematically identify and synthesise the pedagogical mechanisms through which arts-based interventions (theatre, poetry, narrative medicine) may support empathy development in healthcare learners, while critically appraising the contextual implementation factors and methodological limitations of the existing evidence base. DESIGN: Systematic review with narrative synthesis, conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. DATA SOURCES: MEDLINE (via PubMed), Embase, PsycINFO, ERIC, Cochrane Central Register of Controlled Trials, Web of Science Core Collection and CINAHL Plus, from inception to 31 August 2025. ELIGIBILITY CRITERIA FOR SELECTING STUDIES: Peer-reviewed empirical studies involving undergraduate or postgraduate healthcare students or professionals. Eligible studies reported on arts-based educational interventions designed to enhance empathy (including theatre, poetry or narrative medicine) and included quantitative or qualitative outcome measures of empathy or related constructs. No restrictions were placed on study design. DATA EXTRACTION AND SYNTHESIS: Two reviewers independently screened records, extracted data on intervention characteristics and outcomes, and assessed methodological quality using the Medical Education Research Study Quality Instrument and Critical Appraisal Skills Programme tools. A narrative synthesis was conducted to examine patterns across intervention modalities. Effect sizes (Cohen's d) were calculated for quantitative pre-post or controlled designs; meta-analysis was precluded due to high heterogeneity in study designs, populations and outcome measures. RESULTS: From 1094 records screened, 17 studies involving 835 participants (median sample size=28) met inclusion criteria. There were eight theatre-based, five poetry-based and four narrative medicine interventions. Theatre interventions, which emphasised embodied learning, showed patterns of large empathy effects (mean d=0.83). Poetry interventions, focusing on reflective practice, showed patterns of moderate effects (mean d=0.49). One study of narrative medicine demonstrated a pattern of sustained gains at 2-year follow-up (d=0.82). Key limitations of the evidence base include small sample sizes, substantial measurement heterogeneity (12 different empathy instruments), geographical bias (88% of studies from high-income, Western countries), and a predominance of pre-post study designs. CONCLUSIONS: Arts-based interventions show mechanism-specific patterns of potential benefit for empathy in healthcare learners. However, significant methodological limitations in the current evidence constrain definitive practice recommendations. Educators may consider piloting brief, facilitator-led interventions based on local resources, while acknowledging the need for more robust evidence. Future research must prioritise standardised outcome measures, rigorous controlled and longitudinal trial designs, and studies in diverse cultural contexts. PROSPERO REGISTRATION NUMBER: CRD42025116174.
24 Apr 2026
Read appraisal →