What Does Group Research Tell Us About Jury Deliberations?
There is a patent medicine quality to jury advice that can be easy to poke fun at. For as much as the rules of evidence are suffused with “folk wisdom” of dubious provenance, advice given to lawyers about juries is even worse, with a lot of widely circulated insights that do not stand up to real scrutiny.
One example that you'll still encounter is the use of Myers Briggs derived psychometrics as the basis for analyzing potential jurors. I have encountered this in jury selection literature on multiple occasions and it's a fairly good indication that the author hasn't bothered to vet any of their advice. Myers Briggs, a personality sorting methodology devised by a mystery novelist and a bank clerk back in the 1940s, has been repeatedly debunked over the last 80 years as pseudoscientific hokum. The sixteen categories that it spits out are each designed to sound like a generally flattering version of everyone and no one has ever been able to make use of the categorizations to predict behavior and even a halfway rigorous fashion.
It's not surprising that voodoo and pseudoscience take root in jury analysis. The truth is that it's hard to rigorously study juries. All social science research is inherently difficult in that people are complex autonomous actors subject to a fiendishly large set of potential influences. But jury research it's even harder in that researchers rarely have direct access to genuine trial deliberations and the sui generous nature of each case makes it difficult to sort out the relevance of individual variables. And unlike some other fields of study, there aren't large datasets to mind for trends or correlations—trials are scattered amongst 50 state courts and the federal system, virtually no uniform data is gathered, and the events themselves grow rarer by the year.
Occasionally there are efforts to subject juries to real scientific inquiry. For example, the Seventh Circuit American Jury Project in 2007 studied the impact of multiple changes to the trial process and issued thoughtful recommendations, most of which were ignored. And the state of Arizona allowed social scientists access to juries in real cases in connection with its changes to trial procedure in the 1990s. But for the most part, we're stuck with research done on mock juries, which have both selection (people willing to participate don't accurately mirror real jury pools) and seriousness (nobody thinks that real lives or money are on the line) issues. Or we can look at broader social science research, done outside the jury context and do our best to reason through how it might apply in the jury room.
What can that tell us? It is impossible to be definitive and you should distrust anyone who tries. But there are some results that you should consider when speaking to and arguing to a jury at trial.
One category of results concerns aspects of group decision making that are generally superior to individual decision making. You can think of these as areas where a group is generally better than the sum of its parts. The first such area is error detection. All people make mistakes and even smart, conscientious, people can miss here or misunderstand testimony in a courtroom. But when it comes to spotting and correcting such mistakes groups of jurors outperform even careful individuals. [1] And it's easy to see why—while everyone makes random mistakes, people don't generally make the same random mistakes as their neighbor. Similarly, groups tend to outperform individuals when it comes to the volume of factual recall. A group of people tends to recall more facts and in more detail than any individual member.
A third category where groups tend to outperform individuals is lie detection. Most individuals are terrible at this—the most frequently cited data suggests that the average person is little better than a coin flip. But groups of people, well far from perfect, do tend to achieve noticeably better results. [2] It is not entirely obvious why this is the case. One possibility is that group discussion allows for more careful scrutiny of inconsistencies in false statements. Another is that people tend to wildly overestimate their ability to detect deception and hearing competing assessments about whether a person is telling the truth may lead to be more careful analysis.
A fourth area where groups tend to make more accurate assessments than individuals is probability. Again, most individuals are shockingly bad at probabilistic reasoning, both in accurately estimating the likelihood that uncertain events occur and improperly interpreting that data. While groups are not fantastic at this either they do tend to perform better. [3]
What about areas where groups tend to underperform individuals? The notion of groupthink, that collections of otherwise intelligent people can make fundamentally flawed judgments when acting together, has been an active topic of research for decades. The most classic species of groupthink is conformity bias. People tend to dislike and avoid interpersonal conflict, particularly with strangers and particularly concerning matters in which they have no personal interest. So once an opinion is expressed in a group setting, particularly by multiple people, the bar to raising a contrary opinion goes up. This can sometimes lead to what are called cascade effects, where the order in which people in a group expressed their opinions has an outsized, even determinative effect on the group's ultimate decision.
A related phenomenon is the free rider or bystander effect. Because groups can diffuse individual feelings of responsibility and agency, many people in a group of strangers will simply decline to participate in a meaningful way. The precise numbers vary, but mock jury studies suggest that as many as 1/3 of jurors essentially don't participate in deliberations and simply sign on to whatever the rest of the group does. [4] These jurors do not necessarily view themselves as selfish or lazy, they may simply feel that their contributions are unnecessary or not meaningful.
One other species of groupthink of particular relevance to juries is polarization. When people are in a group with others that share a common belief, the intensity of that belief tends to increase. [5] This can result in the group reaching decisions that none of the original members would have on their own.
So, for example, imagine a jury where a majority of people suspect that the plaintiff may be malingering a bit with respect to his injuries. If the jurors expressed these views to each other and feel validated by the agreement, the jury may come away certain that the plaintiff is greatly exaggerating his injuries, even though none of the original jurors began with that belief. Or imagine a majority of individual jurors feel that a corporate defendant deserves to be punished for what it did and react positively to each other when they express similar views. The result may be a runaway verdict higher than any individual sure would have picked on his or her own. If you've ever seen children egg each other on this dynamic will not strike you as surprising.
A lot of work has been done at institutions to try to guard against groupthink, by imposing structural rules some group discussions and decision making designed to encourage respectful disagreement and avoid maladaptive conformity. But jury deliberations are infamously unstructured—jurors are given cryptic instructions on the law that they are to follow and a verdict sheet to fill out, but there is no Roberts rules of order for the jury room. Courts could try to improve things and there are occasional nods in that direction when it comes to things like how jury foreman are selected. But courts are reluctant to update jury procedures and until they do there is not a lot that lawyers can do on their own.
What limited control attorneys do have is largely focused on which jurors are selected and what is said to them in opening and closing statements. Nobody is immune to groupthink, everyone feels social pressure even if they do not fold to it. But some people are more willing to speak up than others. When picking a jury it is important to choose jurors whom you can convince. But it may be even more important to pick jurors who can convince others on your behalf.
[1] M.L. McCoy, N. Nunez and M.M. Dammeyer, The Effect of Jury Deliberations on Jurors’ Reasoning Skills Law and Human Behavior (1999) 23(5), pp. 557–575, 560.
[2] N. Klein and N. Epley, Group Discussion Improves Lie Detection Proceedings of the National Academicy of Sciences of the United States of America (2015) 112(24), pp. 7460–7465.
[3] B. Maciejovsky, et al., Teams Make You Smarter: How Exposure to Teams Improves Individual Decisions in Probability and Reasoning Tasks Management Science (2013) 59(6), pp. 1255–1270.
[4] Gordon, Sara, "All Together Now: Using Principles of Group Dynamics to Train Better Jurors" (2015). Scholarly Works. 896. https://scholars.law.unlv.edu/facpub/896
[5] Brian M. Barry, Judging Better Together: Understanding the Psychology of Group Decision-Making on Panel Courts and Tribunals, https://iacajournal.org/articles/10.36745/ijca.479
