The Incongruence Principle of Evidence

Evidence law assumes that the meaning and value of information at trial is equal to the meaning and value of the same information in the real world. This premise underlies evidence policy, judicial applications of evidence law, and instructions to jurors for evaluating evidence. However, it is incorrect, and the law's failure to recognize this hinders its aims of accuracy and equality.In this article, I draw on fields outside of law-including Bayesian inference and cognitive psychology-to develop a model of evidence that describes how jurors combine new evidence with prior beliefs (or "priors") to make inferences and judgments. I apply this model to derive a principle that I refer to as the incongruence principle of evidence. It states that the informational value of evidence at trial is not equal to the informational value of the same evidence in the real world. I show that, contrary to standard assumptions in evidence law, the trial setting degrades the value of evidence. It does this, on the one hand, by magnifying the influence of a juror's priors-including biases based on the race, gender, appearance, and other background characteristics of trial participants-and on the other hand, by employing a unique decisional framework that is susceptible to an inferential problem called "overfitting," which can lead to false-positive judgments when combined with biased and influential priors.Finally, I show that the incongruence principle carries important implications for evidence law, including for the rule against character evidence, the hearsay rule, and impeachment evidence. I argue that recognizing the role of the incongruence principle in a juror's interpretation of evidence can help lawmakers to structure evidence law in a way that better achieves the goals of accuracy and equality.