The relationship between causal and counterfactual reasoning

This thesis represents a contribution to the study of causal and counterfactual reasoning. In six experiments, the relationship between causal selection and counterfactual reasoning and selection is directly investigated. The results support the conclusion that causal contingency information is available for both causal and counterfactual judgements, and that its availability interacts with task demands. Specifically, causal and counterfactual selections were found to depend on the specificity of the description of the outcomes (Experiments 1 to 3). Furthermore, when considering causal chains, causal and counterfactual selections correspond to probability increases and change (Experiments 4 and 5), and can be described by a model that takes those changes into account. Further evidence is offered by the analysis of causal and counterfactual conditionals. It was found that when frequency information is used, the assessments of these conditionals tend to agree (Experiments 6 and 7), as predicted by recent theories of conditional reasoning. The results are interpreted based on the main theories of reasoning available, and it is proposed that these explanations can be integrated into the larger framework of causal models.