What Factors Really Affect Jurors' Decisions: A Mathematical Modeling Approach to a Common Psychological Question
Abstract
A common limitation of psychology and law is that psychologists are only able to examine a few variables at a time before the analysis becomes too complex. Research has shown eyewitness testimony, DNA testimony, and expert testimony, as well as the race and gender of the plaintiff, defendant and juror all affect the outcome of a criminal case. But, the effects are not necessarily independent. This study involved 750 participants, each viewing a different combination of case information. To examine the interactions among all of the variables, a committee of artificial neural networks was used to make mock-juror guilt predictions, given a certain set of parameters. Minor differences in guilt predictions were seen when manipulating plaintiff, suspect, and juror composition, but evidence appears to be the key driver guilt predictions.