Research
My research asks how people rely on AI in decision-making settings: when they over-rely, when they override, and when they follow AI suggestions appropriately, and what interventions should be used to improve collaboration and decision performance. I've run studies across several settings: clinical assessment, misinformation detection, and group decision-making. Alongside those studies, I've built a framework for AI-assisted decision-making that pulls the field's findings into one structure, for researchers and practitioners who need to design or evaluate these systems, and for a general audience looking to improve their AI literacy and understanding of AI's impact.
AI-Assisted Decision-Making
A framework for AI-assisted decision-making, built from an evidence base of scientific studies. Every AI-assisted decision has three components: the Human and the Task, which characterize each decision setting, and the AI Assistant ecosystem, the only component that can be leveraged to improve the collaboration. It's public, it's a working draft, and it changes as the evidence base grows.
All publications
- AI assistance in medical decision-making: the role of recommendations and explanations in simulated clinical cases ACM Transactions on Computing for Healthcare Link
- Effects of interaction and explanation type on human-AI collaboration for fake news detection IEEE Access Link
- Large language models approach clinician performance in ESC cardiovascular risk stratification: a vignette-based benchmark study European Heart Journal - Digital Health Link
- Dissecting medical referrals in health services: role of physician professional networks BMC Health Services Research Link
- The amplifying effect of explainability in AI-assisted decision-making in groups CHI 2025: ACM Conference on Human Factors in Computing Systems Link
- Assistant robots with an agenda foster uncooperative behaviors 2025 IEEE International Conference on Robot and Human Interactive Communication (RO-MAN) Link
- Looking for cognitive bias in AI-assisted decision-making HHAI Workshops Link
- AI trust: Can explainable AI enhance warranted trust? Human Behavior and Emerging Technologies Link