Proceedings of the Annual Meeting of the Cognitive Science Society

Causal relationships in the real world can have diverse mechanisms with differing statistical signatures. We investigate whether people can distinguish between causes that merely change the timing of events ("when" causes) and those that bring about or prevent those events ("whether" causes). We designed experiments where the rate of an event varies over time due to one such causal influence. Events were shown in real time in Experiment 1 and as a timeline visualization in Experiment 2. Our results suggest that people are capable of identifying "when" and "whether" causes but with a distinctive pattern of confusability: People confuse Generation with Hastening; and Prevention with Delaying. We develop a Causal Abstraction from Summarizing Events (CASE) model, which explains people's judgments as mediated by their rate-change-event detection. We discuss how this line of research can be extended to study human cognition about dynamic causal influences and its relevance to real-life judgment and decision-making.