Proceedings of the Annual Meeting of the Cognitive Science Society

Causal knowledge is key to making effective decisions, yet little is known about how we combine new causal informa-tion with what we already know. This scenario, with a mix of prior beliefs and new information is common to manysettings, and is pervasive in health decisions. We specifically examine how decision-making with causal models differs inabstract decisions versus those more reminiscent of daily life, and how new information interacts with people’s perceivedknowledge about the decision-making domains. We found that while people can successfully use causal models to answerabstract questions, causal models can lead to worse choices in everyday decisions, especially when people believe theyknow a lot about the domain (Experiment 1). We then used an IOED task to determine if showing people how little theyactually understand about a domain may improve the use of causal models in decision-making (Experiment 2).