Extending counterfactual reasoning models to capture unconstrained social explanations

Published in ICML (International Conference on Machine Learning) Workshop on Counterfactuals in Minds and Machines, Honolulu 2023, 2023

Recommended citation: Stephanie Droop & Neil R. Bramley. (2023). "Extending counterfactual reasoning models to capture unconstrained social explanations." Proceedings of ICML (International Conference on Machine Learning) Workshop on Counterfactuals in Minds and Machines, Honolulu 2023.

We found people reason with counterfactuals that stay relatively close to what actually happens, and their tendency to impute unobserved factors depends on how surprising was the behaviour to be explained.

Download paper here