Causal inference in legal decision making: explanatory coherence vs. bayesian networks

Abstract

Reasoning by jurors concerning whether an accused person should be convicted of committing a crime is a kind of casual inference. Jurors need to decide whether the evidence in the case was caused by the accused's criminal action or by some other cause. This paper compares two computational models of casual inference: explanatory coherence and Bayesian networks. Both models can be applied to legal episodes such as the von Bülow trials. There are psychological and computational reasons for preferring the explanatory coherence account of legal inference. © 2004 Taylor and Francis Group, LLC.