Factors influencing jury instruction comprehension: New insights from working memory Janet Randall, Samantha Bonnin, Samantha Laureano, Ashley Robbins, & Yian Xu, Northeastern University, randall@neu.edu Abstract In many states, jurors are challenged by jury instructions, which tell them how to evaluate and decide on a case. The linguistically complex language that these instructions contain can not only keep less-educated and non-native-speaking jurors from fully participating but also can lead to misinformed verdicts. Our primary question, which we have been testing in a series of studies, is: what can we do to improve juror comprehension? In our studies, subjects hear a set of six Massachusetts jury instructions and answer true/false comprehension questions about them. The subjects are randomly distributed across four conditions in a 2x2 between-subjects design. They hear either Original (O) or Plain English (P) instructions, and they listen (L) either without the text or with the text to read along (R). The Plain English versions minimize two linguistic factors known to impede understanding, passive verbs and unfamiliar legal expressions. In Study 1, which tested undergraduates (n=214), comprehension improved significantly with the addition of reading but only marginally with Plain English, as shown in Figure 1; the students' high baseline comprehension rates left little room for improvement. But college students are not representative of jurors. According to the 2013 Massachusetts Census, half of Massachusetts' adult citizens have not gone beyond 12th grade. Study 2, therefore, replicated Study 1 on a more jury-like subject pool using Amazon MTurk (n=389), with lower educational levels (27% had not gone to college). The results, in Figure 2, as expected, showed that (a) MTurkers' overall comprehension was lower than students', and (b) comprehension significantly improved for subjects who had both the "Plain English" versions and the texts to read along. Our MTurkers in Study 2, though, still did not model jurors: like the students in Study 1, they heard each instruction with its corresponding questions before going on to the next. Jurors, on the other hand, hear all the instructions grouped together with no time to process them individually. This "grouped" presentation of the instructions could tax processing and working memory (Cowan, 2010; Engle & Conway 2004) and present a serious challenge when jurors need to apply the instructions in their deliberations. This would predict that comprehension rates for "grouped" instructions will be lower than those for "ungrouped." This was our Study 3 (n=180), using new undergraduates, tested this hypothesis and it was confirmed (p<.001). As shown in Figure 3, the students hearing the "grouped" instructions, showed significant gains across the board from both (a) switching to "plain English" and (b) reading along (p<.001; p<.05, respectively), which the "ungrouped" Study 1 students did not. Having modeled more closely what jurors encounter in the courtroom, we have a more accurate picture of not only how difficult jury instructions are, but also how much more comprehensible they can become. Three simple changes: (a) switching to Plain English, (b) providing the texts to read along with the verbal instructions and (c) spacing out the instructions to avoid processing and memory overload, can engage jurors more fully in the courtroom experience and help improve their understanding. If the judiciary applies these findings to change courtroom practices, we will see better-informed verdicts and, ultimately, fairer trials. References Cowan, N. (2010). The magical mystery four: How is working memory capacity limited, and why?. Current directions in psychological science, 19(1), 51-57. Engle, R. W., & Conway, A. R. (2004). Working memory and comprehension. In Working memory and thinking (pp. 73-97). Routledge. [cid:image005.jpg@01D5FD7D.197F7360] [cid:image006.jpg@01D5FD7D.197F7360]