Abstract
Jurors in criminal trials hold immense responsibility and sexual assault trials pose particular challenges. Models of juror decision-making assert when jurors do not understand their task, they are more likely to be influenced by details that are irrelevant. There are many such details in sexual assault cases, where rape myths abound—none of which should play a role in jurors’ decisions. Question trails are aids that have been implemented to support jurors in understanding their task and making legally sound decisions. Question trails break down the jurors’ task into a series of decisions that cover the core elements of a charge and the facts that must be proven for a guilty verdict to be reached. To date, research on question trails has focused primarily on their effect on subjective aspects of jurors’ experience. In contrast, almost nothing is known about whether question trails change the way in which jurors approach their decision-making and the outcomes that arise. The fundamental goal of this thesis was to test whether—and if so, how—question trails influence the decisions that potential jurors make, with a particular focus on sexual assault cases.
In Study 1, across three experiments, we presented online participants in the United States with a vignette of sexual assault and asked them to imagine that they were acting as a juror in the trial. We then provided some participants with a question trail designed for sexual assault cases and others with the relevant legal statute. To test whether the question trail decreased the effect of irrelevant details on participants’ decisions—or reduced bias—we manipulated whether the vignettes were written to push them towards conviction or towards acquittal and asked them to return a verdict. The question trail did not exert consistent effects on participants’ decisions: in Experiment 1 the question trail reversed the bias seen in the control group, in Experiment 2 it created bias where there was none, and in Experiment 3 we saw no bias in either condition.
Despite our bias manipulation resulting in inconsistent findings across the three experiments, the data from Study 1 made it clear that participants’ verdicts tended to come down to whether or not the accused had a reasonable belief that the complainant was consenting to sexual activity. Therefore, in Study 2, we explored how potential jurors understand and apply the reasonable belief element in cases of sexual assault. In an online qualitative study with 50 participants in the United States, we developed three themes: (1) the context of hook-up culture and contextual thresholds, (2) that consent is implied until it’s not, and (3) why did they trust the accused. The results of this analysis indicated that potential jurors struggle to apply the concept of reasonable belief in two fundamental ways. First, they collapse the two core elements of the reasonable belief standard—honesty and reasonableness. Second, they struggle to conceptualise an erroneous belief in consent that is unintentional yet blameworthy (i.e., an unreasonable belief).
In Study 3, we created revised question trails designed to circumvent these two issues. The first (Experiment 4) involved language changes only, while the second (Experiment 5) employed both language and structural changes. Both revised question trails resulted in an increase in guilty verdicts, but for different reasons. In Experiment 4, the increase was driven by a shift from accepting a reasonable belief in consent to rejecting any belief in consent. In Experiment 5, the increase was driven by a shift from reasonable to unreasonable belief in consent.
The work presented in this thesis is the first to investigate the ways in which question trails shape potential jurors’ decisions. Despite finding that our basic question trail from Study 1 was unhelpful at best and harmful at worst, we saw that making even relatively minor changes to it to target misconceptions had significant effects on decision-making. We discuss the implications of our findings for those involved in research and practice, and point out that decision aids for jurors are limited by the laws on which they are designed.