The trajectory of counterfactual simulation in development

Previous work has argued that young children do not answer counterfactual questions (e.g. “what would have happened?”) by constructing simulations of alternative possibilities in the way adults do. Here, we propose that children can engage in simulation when answering these questions, but consider different counterfactual possibilities than adults. While most previous research has relied on narrative stimuli, we use causal perception events, which are understood even in infancy. In Experiment 1, we replicate earlier findings that children struggle with counterfactual reasoning, but show that they are capable of conducting the required simulations in a prediction task. In Experiment 2, we use a novel multiple-choice method that allows us to study not only when children get it right, but also how they get it wrong. We find evidence that 4-year-olds engage in simulation, but preserve only some features of what actually happened and not others.