Related Research Initiatives - Strategic Behavior

An urgent issue in the medical sector is overcrowding in the emergency department (ED). In an attempt to divert less urgent patients to alternative forms of care - such as territorial health - hospitals are now publishing online and in real-time the level of crowding of ER units. In this research, we propose a model describing the patient’s choice between going to the emergency department or other territorial health services after observing symptoms of different gravity. We then test through state-of-the-art experimental methods how the information provided affects the decision about where to be treated, investigating how different information affects overcrowding. Based on the results from the model and the experimental investigation, we clarify the mechanisms behind the patients' choices and suggest some policy interventions.

Studies on repeated strategic interactions often assume immediate, accurate feedback leads to equilibrium play. We challenge this view, hypothesizing players may reinforce heuristics instead. Our experiment uses sequences of 2-person 3x3 games incorporating five common heuristics. We compare two learning settings: "overlapping strategies" where heuristic and equilibrium actions coincide, and "non-overlapping strategies" where they differ. Results show equilibrium learning is harder in overlapping treatments, as players reinforce heuristic-compatible behaviors. This effect varies across heuristics. Our findings suggest caution when interpreting choice behavior in repeated interactions: observed behavior may not reflect strategic intent, and accurate feedback can reinforce non-equilibrium strategies. We discuss implications for field studies and behavioral policy design.