Effects of a Visual Technology on Mock Juror Decision Making

Effects of a Visual Technology on Mock Juror Decision Making JAIHYUN PARK1* and NEAL FEIGENSON2 1Baruch College-CUNY New York, NY, USA 2Quinnipiac University School of Law Hamden, CT, USA Summary: Two studies explored the effects of lawyers’ use of PowerPoint on liability judgments in a case involving statistical evidence. Participants (Study 1, N=192; Study 2A, N=180; Study 2B, N=189) watched videotaped opening statements for plaintiffs and defendant. In general, defendant’s responsibility was judged to be greater when plaintiffs used PowerPoint slides than when they did not and less when defendant used PowerPoint slides than when it did not. Furthermore, PowerPoint’s impact was greatest when its use was unequal. PowerPoint enhanced persuasion partly through central and partly through peripheral processing. In general, each party’s use of PowerPoint increased participants’ recall of that party’s evidence, which in turn increased defendant’s judged responsibility (when plaintiffs used PowerPoint) or reduced it (when defendants used PowerPoint), indicative of central processing. PowerPoint also functioned as a peripheral cue, influencing participants’ judgments of defendant’s responsibility by affecting their perceptions of the respective attorneys. Copyright © 2012 John Wiley & Sons, Ltd. PowerPoint slide shows, nearly ubiquitous in business and educational settings, are becoming increasingly common in law as well. Apparently, many lawyers assume that illustrating opening statements, witness examination, or closing arguments with PowerPoint slides makes their presentations more effective and persuasive. Reliable empirical support for this widespread assumption, however, is scanty. Despite growing efforts to explore courtroom technology generally (e.g., Center for Legal and Court Technology, 2007; Federal Judicial Center/National Institute for Trial Advocacy, 2001) and a handful of experimental studies of other digital visual technologies, such as computer animations (e.g., Dunn, Salovey, & Feigenson, 2006; for a review, see Feigenson, 2010), there have been no published controlled experimental studies of the effects of PowerPoint in legal settings. There are certainly reasons to expect that augmenting the spoken word with PowerPoint slides would improve lawyers’ ability to communicate with and persuade their audiences. Dual-process theories such as the Heuristic- Systematic Model (Chaiken, Liberman, & Eagly, 1989) and Cognitive-Experiential Self-Theory (Epstein, 1994) are especially promising as frameworks for understanding jurors’ uptake of courtroom communications because they develop the basic insight that persuasion and judgment, like cognition in general, occur through heuristic and intuitive as well as deliberate and reflective processing (Kahneman & Frederick, 2002). Thus, according to the leading dual-process theory of persuasion, the Elaboration Likelihood Model (ELM; Petty & Cacioppo, 1986; Petty & Wegener, 1999), people process messages through a central route, effortfully attending to message content, and/or through a peripheral route, expending less cognitive effort and using aspects of the message or source as heuristic cues to judgment. ELM has been employed as a theoretical framework for understanding how jurors process complicated trial infor- mation. For instance, McAuliff, Kovera, and Nunez (2009) hypothesized that mock jurors exposed to an expert witness’s research studymight engage in central processing (as indicated by their ability to differentiate an internally invalid from an internally valid study) and/or peripheral processing (as indicated by reliance on the study’s ecological validity as a heuristic cue) in judging the quality of the study and the expert’s testimony. The researchers found limited evidence of central processing (mock jurors rated study quality lower when there were missing control groups but not when there were confounds or experimenter bias) but no effect on verdicts, and they found no tendency to resort to peripheral cues in lieu of more systematic thinking about the evidence. In another study, Levett and Kovera (2009) found some evidence that jurors process expert testimony using both routes: An opposing expert who addressed the validity of the first expert’s research sensitized jurors to the validity of that research to some extent, which seems indicative of central processing, but the mere presence of an opposing expert led mock jurors to be more skeptical of all expert testimony, which seems indicative of peripheral processing. The use of PowerPoint might persuade jurors through either central or peripheral processing. With respect to central route processing, visual displays tend to be more vivid than words alone (see Bell & Loftus, 1985); by attracting and holding jurors’ attention, slide shows would be predicted to improve jurors’ retention and possibly their understanding of the information displayed. Some research on courtroom technologies other than PowerPoint, for instance, computer animations, has found limited enhancement of recall of key information (Dahir, 2005; Morell, 1999). In addition, dual coding theory (Paivio, 1971, 1986) posits that people think both visually and verbally. By offering visual stimuli, PowerPoint slides should appeal more directly to the visual processing channel and may be especially effective for people whose learning style inclines toward the visual (Dunn, 2000). Research on multimedia learning supports the efficacy of well-designed visual instruction (Mayer, 2001). Indeed, one mock juror study found that a multimedia (i.e., video plus animation) tutorial accompanying an expert witness’s testimony about basic DNA science helped participants with visual learning styles to learn the material better (Hewson & Goodman-Delahunty, 2008). The results of research on the effectiveness of PowerPoint in improving learning outcomes in educational settings, *Correspondence to: Jaihyun Park, Baruch College-CUNY, Psychology Department, Box B-8215, 55 Lexington Avenue, New York, NY 10010, USA. E-mail: [email protected] Copyright © 2012 John Wiley & Sons, Ltd. Applied Cognitive Psychology, Appl. Cognit. Psychol. 27: 235–246 (2013) Published online 6 December 2012 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/acp.2900 however, are equivocal at best. A few studies have found that students attending lectures augmented by PowerPoint slides do better on exams and other course assignments than those attending lectures supplemented by the use of an overhead projector, chalkboard, or no visual displays at all. Most studies, though, have found that ‘adding computer- generated slides to a class results in no significant differences in learning outcomes’ (Levasseur & Sawyer, 2006, p. 111), and the few studies reporting positive effects tend to be marred by confounds: In one study, for instance, students exposed to PowerPoint slides were learning entirely different material than those not exposed, whereas in several others, students who saw PowerPoint slides in class were also given copies of those slides for further study. Interestingly, more elaborate and dynamic PowerPoint slides (including animation and sound) have been shown not to enhance and even to impair learning (for a review, see Levasseur & Sawyer, 2006). Levasseur and Sawyer (2006) speculate that the general failure to uncover positive learning effects from the use of PowerPoint in the classroom may be due in part to educators’ misuse of the technology, on the one hand, by creating visual displays that trigger arousal to the point of interfering with students’ cognitive processing abilities, or on the other, by using text-only slides that fail to take advantage of PowerPoint’s capacity for effective visual communication. With respect to peripheral processing, PowerPoint slide shows might be persuasive because, for instance, jurors may infer from the lawyer’s use of PowerPoint that the lawyer is especially well prepared and competent, and jurors may then use that inference as a heuristic for the strength of the argument the lawyer is presenting (see Petty & Wegener, 1999). Real jurors are generally motivated to understand trial evidence and reach a decision they can justify, which would incline them to process trial information effortfully. If the information is sufficiently complicated, however, they may be unable to attend carefully to message content and may be more likely to rely instead on peripheral cues, such as the perceived competence or likeability of the message source, the lawyer (see Petty & Wegener, 1999). One study has found that a computer animation led mock jurors to judge the expert witness whose testimony the animation illustrated to be more credible, but there was no indication that they relied on this peripheral cue (message source credibility) in deciding liability or damages (Dahir, 2005). To date, there has been no published research on whether PowerPoint actually affects legal decision makers’ judgments about liability or guilt or, if it does, how those effects occur. In an unpublished paper using the ELM as a theoretical framework, Binder (2006) manipulated presentation mode (PowerPoint vs. large paper tablet vs. no visual aid) and Need for Cognition (Cacioppo & Petty, 1982) as a participant variable to determine whether and under what conditions PowerPoint might affect mock jurors’ perceptions of an expert witness’s credibility and their ultimate judgments of guilt or innocence. Binder (2006, study 2) found that mock jurors who saw PowerPoint text slides presenting the factors that (according to the expert’s testimony) may affect the accuracy of eyewitness testimony remembered significantly more of those factors than did those who did not see slides. Binder also found (2006, study 1) that mock jurors with low Need for Cognition who saw the text of the factors either written on a large tablet of paper or in text in PowerPoint found the expert witness more credible than did those who only heard the expert’s testimony, suggesting (according to the author) that those jurors may have used the visuals as a peripheral cue to the expert’s credibility. The use of PowerPoint had no effect on the ultimate judgment of guilt in either study, however. And the lack of difference between the PowerPoint and tablet conditions (study 1) may have been due to the use of text-only visuals that failed to exploit the more potentially persuasive features of PowerPoint (such as the display of pictures). The purpose of the present studies was to begin to address the gap between psychological research and legal practice by exploring whether and how PowerPoint slides may influ- ence mock juror decision making in a civil case. Specifically, we predicted that when either party used PowerPoint, liability judgments would be more favorable to that party. We also sought to discover whether any effects of PowerPoint on liability judgments occur through central processing (e.g., by increasing memory for and understanding of case-relevant information) or through peripheral processing (e.g., by acting as a cue to argument strength). We predicted that the effects of PowerPoint on judged liability would be mediated by both central and peripheral processing. In addition, because some within the legal community have expressed concerns about unequal access to new courtroom technologies (e.g., Commonwealth v. Serge, 2006; Galves, 2000), we designed our research to feature the differential use or lack of use of PowerPoint by both plaintiff and defendant, to determine whether any impact on judgments occurs or is especially prominent when only one side of the case uses the technol- ogy. Specifically, we predicted that the judged liability of the defendant would be the least when only the defendant used PowerPoint and the greatest when only the plaintiff used it. Finally, to enhance both ecological validity and the prospects for positive results, we designed materials utilizing more of the potentially persuasive visual capabilities of the medium. Thus, the slides used by each party included the display of graphs and charts to explain statistical evidence and modest animations to enhance visual contrasts, while avoiding the sorts of ‘bells and whistles’ that, according to the literature on educational uses of PowerPoint, might arouse participants’ attention but distract them from substan- tively processing message content. STUDY 1 Method Participants Participants were 192 undergraduates at two universities in the northeastern USA. They were recruited from the introductory psychology participant pool and received course credit for partial fulfillment of a requirement (university #1) or from the general undergraduate student body without the induce- ment of course credit (university #2). Participants were 56.1% female; by ethnicity, participants were 59.9% white, 13.5% Asian-American, 10.9% African-American, 8.3% 236 J. Park and N. Feigenson Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) Hispanic-American, and 7.4% other or missing. Results did not differ significantly between participants at the two universities, so data were collapsed across the two groups. Design The design of the study was a 2 (use of PowerPoint by plaintiffs: used vs. not used)� 2 (use of PowerPoint by defendant: used vs. not used) factorial design. All variables were manipulated between participants. Materials and procedures The study used a modified version of a real case in which a railroad company was sued by its African-American employees, who alleged racial discrimination in promotion and discipline practices. An effort was made to balance the strength of the arguments from both parties, in order to permit a more valid observation of the effectiveness of the visual presentations. A simple pretest using only the texts of the attorneys’ statements (without accompanying visual presentations) confirmed that the balancing was successful: An almost equal number of participants showed the opposite verdict preferences by judging the defendant to be liable or not liable (10 vs. 11, respectively). Four versions of videotapes were then created that contained only opening statements read by the attorneys for the respective parties (each statement took about 27minutes, so that the total presentation time was about 54minutes). For ecological validity, the plaintiffs’ attorney always spoke first. In two versions, both parties either accompanied or did not accompany their spoken statements with visual presenta- tions. In the other two versions, only one party used a visual presentation. Each visual presentation consisted of six PowerPoint slides or groups of slides (see Appendix A),1 embedded in the videotape at the points in the respective spoken statements corresponding to the content of the slides (see Appendix B). The slides were designed to illustrate important aspects of each party’s side of the case that were expected to benefit most from visual explanation or empha- sis: Four slides for each party illustrated statements about statistical evidence, while the remaining two slides for each party illustrated other sorts of claims (e.g., about anecdotal evidence and company policies). In addition, the slides for each party were balanced for aggregate length (about 2½minutes) and, insofar as possible given the different contents of the respective statements, formal visual features (e.g., background color and visual complexity). After watching the videotape, participants completed a questionnaire asking them to judge the defendant’s liability and responsibility for committing racial discrimination (dichotomous and continuous measures, respectively) and to indicate their confidence in the respective judgments. Although participants were exposed only to opening state- ments on behalf of both parties rather than a full-scale trial simulation, liability and responsibility measures were included to gauge whether the use of PowerPoint might affect ultimate judgments and not only component processes. Participants were also asked to indicate the persuasiveness of selected arguments by each party; to answer five true–false questions about statistical information presented by the attorneys (see Appendix C) (this information was conveyed by slides as well as orally in the PowerPoint conditions); to evaluate the lawyers for the respective sides (level of preparation, competence, like- ability, persuasiveness, and credibility); in the PowerPoint conditions, to evaluate selected slides used by the respective parties; and to indicate demographic information. Seven-point scales (1—not at all, 7—very much) were used for all contin- uous measures. Participants were debriefed after completing the questionnaires. Results and discussion Preliminary analyses A preliminary analysis was conducted to examine whether confidence ratings affected liability and responsibility judgments. The mean ratings were 4.87 (SD = 1.10) and 5.03 (SD = 1.16) on a 7-point scale, respectively. Each mea- sure was then dichotomized by a median split. A 2� 2� 2 analysis of variance taking the two independent variables (i.e. the use of PowerPoint by the plaintiffs and by the defen- dant) and the dichotomized confidence ratings was conducted on the responsibility measure, which indicated no main or interaction effects of confidence (all Fs< 1.20). In addition, a Chi-square analysis revealed that participants’ confidence was not associated with their liability judgments [w2(1)< 1.50]. Consequently, confidence ratings were dropped from further analyses. Univariate analyses Liability/responsibility judgments. A 2 (use of PowerPoint by plaintiffs: used vs. not used)� 2 (use of PowerPoint by defendant: used vs. not used) analysis of variance on judgments of legal responsibility (continuous measure) showed a main effect for the use of PowerPoint by plaintiffs such that the defendant railroad was judged more responsible for committing racial discrimination against its African- American employees when the plaintiffs used PowerPoint than when plaintiffs did not, F(1, 188) = 4.16, p= .043, �2 = .022 (Table 1). The main effect for the use of PowerPoint by defendant did not reach the significance level, F (1, 188) = 2.44, p = .12, �2 = .013. The interaction of the two factors on responsibility judgments was not significant, F (1, 188) = 1.39, p = .24, �2 = .007. The effect of PowerPoint was especially pronounced when only one party used it: The defendant’s judged respon- sibility was highest (M= 3.880) when the plaintiffs used a visual presentation and the defendant did not; it was lowest (M = 3.043) when the defendant used a visual presentation and the plaintiffs did not (Table 1). As predicted, a planned comparison confirmed that these means were significantly different, F(1, 188) = 6.35, p= .012, �2 = .033. The effect of PowerPoint on dichotomous liability judgments showed the same order of judged liability: The percentage of participants who found the defendant liable was the highest (48%) when the plaintiffs used visuals but the defendant did not, and the lowest (26%) when the plaintiffs did not use visuals but the defendant did. 1 Appendices can be accessed on this web address: http://www.baruch.cuny. edu/wsas/academics/psychology/jpark.htm. Visual technology and juror decision making 237 Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) http://www.baruch.cuny.edu/wsas/academics/psychology/jpark.htm http://www.baruch.cuny.edu/wsas/academics/psychology/jpark.htm Memory/understanding of trial information. Participants answered five true–false questions about statistical evidence (two questions were based on the plaintiffs’ evidence and three questions on the defendant’s evidence; see Appendix C), and a score was calculated by counting the number of correct answers for each side (ranging from 0 to 2 for the plaintiffs and from 0 to 3 for the defendant, respectively). No effect of PowerPoint use was observed on recall of the plaintiffs’ evidence. In contrast, there were main effects for the use of PowerPoint by both parties on the ability of participants to recall statistical evidence proffered by the defendant. When the defendant used PowerPoint, participants were more likely to recall correctly the defendant’s statistical evidence, M=1.276 (defendant used visuals) and M= .942 (defendant did not use visuals), F(1, 188) = 4.140, p= .04, �2 = .022. When the plaintiffs used PowerPoint, however, participants were less likely to recall correctly the defendant’s statistical evidence, M= .970 (plaintiffs used visuals) and M=1.248 (plaintiffs did not use visuals), F(1, 188) = 3.893, p= .05, �2 = .020. The interaction was not significant. Perceptions of the attorneys. A 2 (use of PowerPoint by plaintiffs: used vs. not used)� 2 (use of PowerPoint by defendant: used vs. not used) analysis of variance on judgments about the lawyers indicated main effects for the use of PowerPoint by both parties on judgments about the defendant’s lawyer. When the defendant used visuals, the defendant’s lawyer was judged better prepared [F(1, 188) = 5.76, p = .017, �2 = .03], more competent [F(1, 188) = 6.37, p = .012, �2 = .033], and more credible [F(1, 188) = 5.01, p = .026, �2 = .025] than when the defen- dant did not use them. A multivariate analysis of variance (MANOVA) on these dependent measures with the same factors revealed a main effect of the defendant’s use of PowerPoint, F(3, 186) = 2.722, p = .046, �2 = .042. This indicated that the observed main effect was consistent across the three dependent variables. No other effect was significant. When the plaintiffs used visuals, the defendant’s lawyer was judged less well prepared [F(1, 188) = 4.39, p= .038, �2 = .023] and less persuasive [F(1, 188) = 5.12, p= .025, �2 = .026] than when they did not use them. A MANOVA on the two measures showed a main effect of plaintiffs’ use of PowerPoint, F(2, 187) = 3.129, p = .046, �2 = .032. Again, the main effect was consistent across the two depen- dent variables. There were no significant effects of either party’s use of PowerPoint on perceptions of the plaintiffs’ lawyer (all Fs< 2.00). Mediational analysis Additional analyses were conducted to test the possibility that the effects of PowerPoint use on responsibility judgments were mediated by recall of case materials and/or perceptions of the attorneys, and thus whether the two routes of proces- sing in the ELM model, central and peripheral, might both characterize how decision makers use PowerPoint to think about case information. Two separate sets of mediational analyses were conducted for the two main effects of the use of PowerPoint by both sides, although we found a significant main effect only of plaintiffs’ use of PowerPoint on responsibility judgments. Mediational analyses on the nonsignificant main effect of the defendant’s use of PowerPoint on responsibility judgments were also performed because some authors have argued that a significant effect of the predictor variable on the outcome variable is not necessary for mediation to occur (Collins, Graham, & Flaherty, 1998; Kenny, Kashy, & Bolger, 1998; Shrout & Bolger, 2002). We used as mediators the measures on which the use of PowerPoint (by the respective parties) had a significant effect in univariate analyses, because the predictor variable must have an effect on mediating variables in order for mediation to occur (Baron & Kenny, 1986; Collins et al., 1998; Kenny et al., 1998; Shrout & Bolger, 2002). Some adjustments were made before mediational analyses were conducted. Although we had one eligible measure for recall (of the defendant’s statistical evidence), we had multiple eligible measures for perceptions of the defendant’s attorney in both analyses of mediation. A composite variable for perceptions of the attorney was created for each of the two sets of mediational analyses by taking the mean of the eligible measures. Specifically, perceived levels of the defendant’s Table 1. Descriptive statistics on dependent measures as a function of the use of PowerPoint by plaintiffs and by the defendant (Study 1) Use of PowerPoint by plaintiffs Used Not used Use of PowerPoint by the defendant Dependent measures Used (n= 48) Not used (n= 50) Used (n= 47) Not used (n= 47) Judged liability Continuous variable 3.792 (0.232) 3.880 (0.228) 3.043 (0.235) 3.681 (0.235) Dichotomous variable 0.396 (0.070) 0.480 (0.068) 0.255 (0.070) 0.362 (0.070) Memory of evidence From plaintiffs 1.137 (0.102) 1.120 (0.100) 1.070 (0.104) 1.083 (0.103) From defendant 1.143 (0.144) 0.797 (0.139) 1.409 (0.157) 1.087 (0.145) Perception of the defendant lawyer Well prepared 5.708 (0.159) 5.200 (0.156) 5.915 (0.161) 5.660 (0.161) Competent 5.604 (0.167) 5.100 (0.164) 5.617 (0.169) 5.277 (0.169) Persuasive 5.063 (0.204) 4.760 (0.200) 5.617 (0.206) 5.128 (0.206) Credible 4.771 (0.200) 4.600 (0.196) 5.426 (0.202) 4.702 (0.202) Note: Numbers in parentheses next to the means represent standard errors; dichotomous variable of judged liability was coded such that ‘1’ was assigned to ‘liable’ and ‘0’ was assigned to ‘not liable’ options. 238 J. Park and N. Feigenson Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) attorney’s preparedness, competence, and credibility were included to create a composite variable in testing mediation for the effect of the PowerPoint use by the defendant on responsibility judgments. In fact, the intraclass correlation coefficient for these measures was .788, which indicated a high level of consistency among the measures and justified making a composite variable. In addition, perceived levels of the defendant’s attorney’s preparedness and persuasiveness were used to create a composite variable in testing mediation for the effect of plaintiffs’ use of PowerPoint on responsibility judgments. The intraclass correlation coefficient for the measures was .698 (the bivariate correlation coefficient was r= .536, p< .001). Creating composite variables was further justified to avoid multicollinearity problems that may occur when multiple, highly correlated variables were used in the same model. In order to examine mediation and overall fit, a causal model was tested for each of the two main effects of PowerPoint use by both parties on responsibility judgments. The recall measure (of the defendant’s statistical evidence) and the composite measure of perceptions of the defendant’s attorney were employed as mediators for each model. The causal model was tested using maximum-likelihood estimation in AMOS 5.0 (Arbuckle, 2001). First, mediation for the main effect of the use of PowerPoint by the defendant on responsibility judgments was tested. We entered the defendant’s use of PowerPoint as the predictor variable. Next, the recall measure and the composite score for perceptions of the defendant’s attorney were entered as the mediators. Finally, the judged responsibility of the defendant (continuous variable) was entered as the outcome variable. The model’s overall goodness of fit was assessed by Chi-square test, comparative fit index (CFI), Tucker– Lewis coefficient (TLI), and root-mean-square-error of approximation (RMSEA). A nonsignificant Chi-square test, a low value of RMSEA, and greater values of CFI and TLI indicated a satisfactory fit of the model to the data, w(1) = 0.103, p = .854, CFI = 1.000, TLI = 1.194, and RMSEA= .000. As presented in Figure 1, the defendant’s use of PowerPoint had a significant effect on recall of the defendant’s statistical evidence (b= .167, p< .05) and perception of the defendant’s attorney (b= .204, p< .01), which in turn had a significant effect on judged responsibility of the defendant (b=�.295, p< .001, b=�.288, p< .001, respectively). The squared multiple correlations (R2) for recall, perception of the defendant’s attorney, and judged responsibility of the defendant were .028, .042, and .198, respectively. In addition, we tested mediation for the main effect of the plaintiffs’ use of PowerPoint on responsibility judg- ments. In this analysis, the use of PowerPoint by plaintiffs was entered as the predictor variable, the recall measure (of the defendant’s statistical evidence) and the composite score for perceptions of the defendant’s attorney were entered as the mediators, and the judged responsibility of the defendant (continuous variable) was entered as the outcome variable. This model’s overall goodness of fit was assessed by the same indexes as in the analysis earlier. The results indicated a good fit of the model to the data, w2(1) = 0.470, p = .493, CFI = 1.000, TLI = 1.092, and RMSEA= .000. As exhibited in Figure 2, the use of PowerPoint by plaintiffs had an impact on recall of the defendant’s statistical evidence (b=�.145, p< .05) and perceptions of the defendant’s attorney (b=�.179, p< .02), which in turn had a significant effect on the judged respon- sibility of the defendant (b=�.279, p< .001, b=�.354, p< .001, respectively). The squared multiple correlations (R2) for recall, perception of the defendant’s attorney, and judged responsibility of the defendant were .021, .032, and .240, respectively. These findings indicated that the effects of the use of PowerPoint by both parties on the judged responsibility of the defendant were mediated by the amount of accurate recall of the defendant’s statistical evidence and by altered perceptions of the defendant’s attorney. Taken together, the present findings provide some, albeit equivocal, evidence that using PowerPoint can help attorneys persuade jurors more effectively. When lawyers illustrated key portions of their oral presentations with PowerPoint slides, jurors rendered liability judgments that were more favorable to those lawyers’ clients. Results of the mediational analyses provide evidence that the effects of these visual displays on liability judgments occurred through both central processing (i.e., increased recall of case information) and peripheral processing (i.e., more favorable perceptions of the attorney using the slides or less favorable perceptions of the opposing attorney). Recall of Defendant’s Evidence Judged Responsibility of Defendant Perceptions of Defendant’s Attorney Use of PowerPoint by Defendant .167* -.288***.204** -.295*** Figure 1. Path model of the effects of the defendant’s use of PowerPoint on judged responsibility of the defendant mediated by participants’ recall of the defendant’s statistical evidence and by perceptions of the defendant’s attorney (N= 192, Study 1). Standardized path coefficients are shown in the diagram. Perceptions of the defendant’s attorney represent a composite variable from three observed variables (perceived preparedness, competence, and credibility). Residuals for the endogenous variables were not shown for simplicity. A correlation between the residuals of the two mediators was specified in the model. *p< .05. **p< .01. ***p< .001 Visual technology and juror decision making 239 Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) These results, however, were due to the effects of the use of PowerPoint by either attorney on recall of only the defen- dant’s statistical evidence and on perceptions of only the defendant’s attorney’s preparation, competence, credibility, and persuasiveness. Although the results of Study 1 generally confirmed the predictions, two issues of internal validity might be raised. First, the contents of the respective attorneys’ arguments differed. Although this preserved ecological validity—in real cases, opposing sides present different arguments—one might argue that the observed differences in liability judgments may be attributable to the relative strength of the arguments rather than the use of PowerPoint (even though the arguments were pretested and found to be evenly balanced). Second, the order of presentation was fixed in Study 1—the plaintiffs’ attorney went first in all conditions. This fixed order may have yielded the asymmetric results noted earlier regarding the effects of PowerPoint use on the recall of only the defendant’s statistical evidence and on perceptions of only the defendant’s lawyer’s competence, persuasiveness, and so on. The latter sorts of judgments in particular are likely to reflect comparative judgments; because the plaintiffs’ attorney went first, parti- cipants’ perception of the plaintiffs’ attorney served as a baseline for their evaluation of the defendant’s attorney. As the defendant’s arguments and materials were more likely to remain in the working memory of participants, a recency effect may explain why effects of the PowerPoint use were more pronounced on recall and perception measures for the defendant. Again, although this design enhanced ecological validity, it might be argued that a more thorough test of PowerPoint effects would require counterbalancing the presentation order. To address these problems, we conducted additional studies. STUDY 2 We conducted a set of two studies (i.e., Studies 2A and 2B) to eliminate the possible confounds described earlier. The two studies differed from Study 1 only in design while using substantially identical materials and procedures. In each study, PowerPoint use was manipulated for one party only; the other party did not use PowerPoint. In addition, the order of presentations was counterbalanced. It was expected that this new design would allow us to test the effects of PowerPoint on liability judgments more clearly. Furthermore, the questionnaire was changed in a few respects: The judge’s instruction was revised to provide more detailed information as to the nature of the judgment and the burden of proof (i.e., preponderance of the evidence), and more recall questions were included. Study 2A: method Participants A total of 186 participants were recruited from the introductory psychology participant pool at a university in the northeastern USA. They received course credit for partial fulfillment of a requirement. The data from six participants were dropped from further analyses because they did not follow the instructions properly. Among those who were included in the analyses, 56.7% were female; 28.9% were white, 40.0% were Asian- American, 5.0% were African-American, and 26.1% were Hispanic-American. Design The design was a 2 (use of PowerPoint by plaintiffs: used vs. not used)� 2 (order of presentation: plaintiffs first vs. defendant first) factorial design. Both factors were manipulated between participants. Materials and procedures The same materials were used as in Study 1 except for the following items: (a) the judge’s instruction was revised (see Appendix D), and (b) four true–false questions were added to the recall questions and one question was dropped, yielding a total of eight questions used to assess participants’ memory of the case-relevant information (four questions for the plaintiffs and four questions for the defendant). The additional questions can be found in Appendix C. The procedures used were the same as in Study 1. Study 2A: results and discussion Univariate analyses Liability/responsibility judgments. A 2 (use of PowerPoint by plaintiffs: used vs. not used)� 2 (order of presentation: plaintiffs first vs. defendant first) analysis of variance on Recall of Defendant’s Evidence Judged Responsibility of Defendant Perceptions of Defendant’s Attorney Use of PowerPoint by Plaintiffs -.145* -.354***-.179** -.279*** Figure 2. Path model of the effects of the plaintiffs’ use of PowerPoint on judged responsibility of the defendant mediated by participants’ recall of the defendant’s statistical evidence and by perceptions of the defendant’s attorney (N= 192, Study 1). Standardized path coeffi- cients are shown in the diagram. Perceptions of defendant’s attorney represent a composite variable from two observed variables (perceived preparedness and persuasiveness). Residuals for the endogenous variables were not shown for simplicity. A correlation between the residuals of the two mediators was specified in the model. *p< .05. **p< .01. ***p< .001 240 J. Park and N. Feigenson Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) judged legal responsibility (continuous measure) indicated a main effect for the use of PowerPoint by plaintiffs. The railroad company was found more responsible for racial discrimination against its African-American employees when the plaintiffs used PowerPoint than when they did not, F(1, 176) = 6.706, p = .01, �2 = .037 (Table 2). Order of presentation did not yield any significant difference on responsibility judgment, F(1, 176) = .228, p= .63, �2 = .001. The interaction between the two factors was not significant, F(1, 176) = .953, p = .33, �2 = .005. The effect of plaintiffs’ PowerPoint use on the dichotomous liability judgment showed a comparable pattern: The percentage of participants who found the defendant liable was 59.5% when plaintiffs used the visuals but 33.7% when the plaintiffs did not use them. Memory/understanding of trial information. Participants answered eight true–false questions about the case informa- tion (four questions were based on the plaintiffs’ evidence and the other four questions on the defendant’s evidence). A score was computed by counting the number of correct answers for each side, which could range from 0 to 4. The same 2� 2 analysis of variance on the recall of the plain- tiffs’ evidence indicated a main effect of plaintiffs’ use of PowerPoint, F(1, 176) = 11.081, p = .001, �2 = .059. When plaintiffs used PowerPoint, participants were more likely to remember correctly evidence presented by the plaintiffs’ attorney, M = 2.242 (plaintiffs used visuals) and M = 1.674 (plaintiffs did not use visuals). No other effect was significant on the same measure. No effect of PowerPoint use was found on recall of the defendant’s evidence. Perception of the attorneys. A 2 (use of PowerPoint by plaintiffs: used vs. not used)� 2 (order of presentation: plaintiffs first vs. defendant first) MANOVA was conducted on perceptions of each attorney. Of the five items, likeability was dropped from the analysis because no effect was found on this measure, and it was not significantly correlated with other measures in Study 1. The results showed a main effect of plaintiffs’ use of PowerPoint, F(4, 173) = 8.264, p< .001, �2 = .160. Results of subsequent univariate tests revealed that when the plaintiffs used PowerPoint, the plaintiffs’ attorney was judged better prepared [F(1, 176) = 19.228, p< .001, �2 = .098], more competent [F(1, 176) = 6.172, p= .014, �2 = .034], and more persuasive [F(1, 176) = 21.299, p = .001, �2 = .108] than when the plaintiffs did not use PowerPoint. The main effect of PowerPoint use on credibility did not reach the significance level, F(1, 176) = 2.056, p= .153, �2 = .012. Unlike the results in Study 1, the plaintiffs’ use of PowerPoint did not affect perceptions of the defendant’s attorney in any way (all Fs< 2.00). This indicates that the results in Study 1 likely resulted from the procedural character- istics (i.e., the fixed order of presentation); when the order was counterbalanced in this study, such effects disappeared. Mediational analysis As in Study 1, an additional analysis was conducted to test whether the effect of the plaintiffs’ use of PowerPoint on responsibility judgments was mediated by participants’ recall of case materials and/or their perceptions of the attorneys. An identical method was used to test a causal model to examine mediation and overall fit. For the perceived judgment of plaintiffs’ attorney, a composite variable was created using preparedness, competence, and persuasiveness, on which effects of the plaintiffs’ use of PowerPoint were found significant. We first entered plaintiffs’ use of PowerPoint as the predictor variable. The recall of the plaintiffs’ evidence and the composite score for perceptions of the plaintiffs’ attorney were entered as mediators. The judged responsibility of the defendant (continuous variable) was entered as the outcome variable. The model’s overall goodness of fit was assessed by multiple indexes. A nonsignificant Chi-square test, a low value of RMSEA, and high values of CFI and TLI indi- cated a satisfactory fit of the model the data, w2(1) = 0.998, p = .318, CFI = 1.000, TLI = 1.000, and RMSEA= .000. The Table 2. Descriptive statistics on dependent measures as a function of the use of PowerPoint by plaintiffs and order of presentation (Study 2A) Use of PowerPoint by plaintiffs Used Not used Order of presentation Plaintiffs first Defendant first Plaintiffs first Defendant first Dependent measures (n= 45) (n= 46) (n= 44) (n= 45) Judged liability Continuous variable 3.822 (0.242) 4.174 (0.239) 3.432 (0.245) 3.311 (0.242) Dichotomous variable 0.422 (0.073) 0.565 (0.072) 0.341 (0.074) 0.333 (0.071) Memory of evidence From plaintiffs 2.133 (0.171) 2.348 (0.168) 1.682 (0.173) 1.667 (0.169) From defendant 1.632 (0.148) 1.597 (0.143) 1.669 (0.156) 1.692 (0.161) Perception of the plaintiffs’ lawyer Well prepared 5.400 (0.164) 5.630 (0.159) 4.818 (0.166) 4.778 (0.163) Competent 5.222 (0.183) 5.370 (0.181) 4.795 (0.186) 4.889 (0.182) Persuasive 5.422 (0.176) 5.609 (0.172) 4.750 (0.177) 4.667 (0.175) Credible 5.022 (0.170) 5.109 (0.169) 4.886 (0.172) 4.756 (0.171) Note: Numbers in parentheses next to the means represent standard errors; dichotomous variable of judged liability was coded such that ‘1’ was assigned to ‘liable’ and ‘0’ was assigned to ‘not liable’ options. Visual technology and juror decision making 241 Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) plaintiffs’ use of PowerPoint had a significant effect on recall of the plaintiffs’ evidence (b= .243, p< .001) and on percep- tions of the plaintiffs’ attorney (b= .408, p< .001), which in turn had a significant effect on the judged responsibility of the defendant (b= .231, p< .001, b= .467, p< .001, respec- tively), as shown in Figure 3. These results suggest that the effects of PowerPoint use by the plaintiffs observed in Study 2A were mediated by both more accurate recall of the plaintiffs’ evidence and enhanced perceptions of the plaintiffs’ attorney. Study 2B: method Participants A total of 189 participants were recruited from the introductory psychology participant pool at a university in the northeastern USA. They received course credit for partial fulfillment of a requirement. Participants were 59.2% female; partici- pants were 28.6% white, 36.0% Asian-American, 6.9% African-American, 24.3% Hispanic-American, and 4% other or missing. Design The design was a 2 (use of PowerPoint by defendant: used vs. not used)� 2 (order of presentation: plaintiffs first vs. defendant first) factorial design. Both factors were manipulated between participants. Study 2B: results and discussion Univariate analyses Liability/responsibility judgments. A 2 (use of PowerPoint by defendant: used vs. not used)� 2 (order of presentation: plaintiffs first vs. defendant first) analysis of variance on judged legal responsibility (continuous measure) was performed. The results showed a main effect for the use of PowerPoint by the defendant. The defendant was judged less responsible for the racial discrimination against its African- American employees when the defendant used PowerPoint than when it did not, F(1, 185) = 7.448, p = .007, �2 = .039 (Table 3). Order of presentation did not have any effect on responsibility judgments, F(1, 185) = .044, p = .84, �2 = .000. The interaction between the two factors was not significant, F(1, 185) = .145, p = .70, �2 = .001. The effect Recall of Plaintiffs’ Evidence Judged Responsibility of Defendant Perceptions of Plaintiffs’ Attorney Use of PowerPoint by Plaintiffs .243*** .467***.408*** .231*** Figure 3. Path model of the effects of the plaintiffs’ use of PowerPoint on judged responsibility of the defendant mediated by participants’ recall of the plaintiffs’ statistical evidence and by perceptions of the plaintiff’s attorney (N= 180, Study 2A). Standardized path coefficients are shown in the diagram. Perceptions of plaintiffs’ attorney represent a composite variable from three observed variables (perceived preparedness, competence, and persuasiveness). Residuals for the endogenous variables were not shown for simplicity. A correlation between the residuals of the two mediators was specified in the model. ***p< .001 Table 3. Descriptive statistics on dependent measures as a function of the use of PowerPoint by the defendant and order of presentation (Study 2B) Use of PowerPoint by the defendant Used Not used Order of presentation Plaintiffs first Defendant first Plaintiffs first Defendant first Dependent measures (n= 47) (n= 48) (n= 46) (n= 48) Judged liability Continuous variable 2.723 (0.222) 2.854 (0.220) 3.413 (0.225) 3.375 (0.221) Dichotomous variable 0.234 (0.068) 0.271 (0.067) 0.370 (0.069) 0.417 (0.067) Memory of evidence From plaintiffs 1.387 (0.152) 1.429 (0.160) 1.502 (0.165) 1.443 (0.158) From defendant 2.532 (0.145) 2.229 (0.143) 2.087 (0.147) 1.938 (0.142) Perception of the defendant lawyer Well prepared 5.915 (0.172) 5.708 (0.170) 5.326 (0.174) 5.188 (0.169) Competent 5.596 (0.173) 5.604 (0.171) 5.283 (0.174) 4.979 (0.171) Persuasive 5.617 (0.191) 5.458 (0.189) 5.087 (0.193) 4.750 (0.189) Credible 5.128 (0.196) 4.979 (0.194) 4.652 (0.198) 4.583 (0.194) Note: Numbers in parentheses next to the means represent standard errors; dichotomous variable of judged liability was coded such that ‘1’ was assigned to ‘liable’ and ‘0’ was assigned to ‘not liable’ options. 242 J. Park and N. Feigenson Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) of the defendant’s PowerPoint use on the dichotomous liability judgment showed a comparable pattern: The percentage of participants who found the defendant liable was 25.3% when the defendant used the visuals but 39.4% when the defendant did not. Memory/understanding of trial information. The identical questions were used to assess the amount of recall of case information. The same 2� 2 analysis of variance on recall of the defendant’s evidence indicated a main effect of the defendant’s use of PowerPoint, F(1, 185) = 6.490, p = .012, �2 = .034. When the defendant used PowerPoint, participants were more likely to remember correctly evidence presented by the defendant’s attorney,M=2.379 (the defendant used visuals) and M=2.011 (the defendant did not use visuals). No other effect was significant on the same measure. No effect of PowerPoint use was found on recall of the plaintiffs’ evidence. Perception of the attorneys. A 2 (use of PowerPoint by defendant: used vs. not used)� 2 (order of presentation: plaintiffs first vs. defendant first) MANOVA was conducted on perceptions of each attorney. As in Study 2A, all the items were used except for likeability. We observed a main effect of the defendant’s use of PowerPoint, F(4, 182) = 3.422, p= .01, �2 = .07. Results of univariate tests indicated that when the de- fendant used PowerPoint, the defendant’s attorney was judged better prepared [F(1, 185) = 10.467, p = .001, �2 = .054], more competent [F(1, 185) = 7.421, p= .007, �2 = .039], more persuasive [F(1, 185) = 10.547, p= .001, �2 = .054], and more credible [F(1, 185) = 4.961, p= .026, �2 = .025] than when the defendant did not use PowerPoint. The defendant’s use of PowerPoint did not affect perceptions of the plaintiffs’ attorney in any way (all Fs< 1.80). As in Study 2A, coun- terbalancing the presentation order led us to observe more clearly the effects of PowerPoint use on perceptions of the attorney, as we had anticipated. Mediational analysis A causal model was tested to test whether the effect of the defendant’s use of PowerPoint on responsibility judgments was mediated by participants’ recall of case materials and/ or their perceptions of the attorneys. An identical method was used for this analysis. For the perceived judgment of plaintiffs’ attorney, a composite variable was created using preparedness, competence, persuasiveness, and credibility. The defendant’s use of PowerPoint was entered as the predictor variable. The recall of the defendant’s evidence and the composite score for perceptions of the defendant’s attorney were entered as mediators. The judged responsibility of the defendant (continuous variable) was entered as the outcome variable. The results indicated a satisfactory fit of the model to the data, w2(1) = 0.711, p = .399, CFI = 1.000, TLI = 1.022, and RMSEA= .000. As presented in Figure 4, the defendant’s use of PowerPoint had a significant effect on recall of the defendant’s evidence (b= .183, p< .05) and perceptions of the defendant’s attorney (b= .275, p< .001), which in turn had a significant effect on judged responsibility of the defendant (b=�.340, p< .001, b=�.305, p< .001, respectively). Again, the effects of PowerPoint use by the defendant were mediated by both more accurate recall of the defendant’s evidence and more positive perceptions of the defendant’s attorney. General discussion The present studies show that lawyers’ use of PowerPoint can affect legal decision makers’ liability judgments, at least in an experimental setting using simplified case materials. Our findings about the circumstances in which those effects occur and the processes by which they occur suggest both policy and theoretical implications. First, we found in Study 1 that when the plaintiffs illustrated their spoken presentation with PowerPoint slides, participants tended to hold the defendant railroad more responsible for committing racial discrimination against its African-American employees. Studies 2A and 2B replicated this finding and showed further that when either party used PowerPoint slides, mock jurors tended to judge the defen- dant’s legal responsibility in a direction favoring that party: The plaintiffs’ use of PowerPoint led mock jurors to hold the defendant more responsible; the defendant’s use of PowerPoint led them to hold the defendant less responsible. Studies 2A and 2B also showed that this effect of PowerPoint use on judged responsibility was independent of the order in which the parties made their presentations. Thus, the effect appears to be robust. It is also noteworthy that this effect was obtained using PowerPoint slides that were relatively Recall of Defendant’s Evidence Judged Responsibility of Defendant Perceptions of Defendant’s Attorney Use of PowerPoint by Defendant .183* -.305***.275** -.340*** Figure 4. Path model of the effects of the defendant’s use of PowerPoint on judged responsibility of the defendant mediated by participants’ recall of the defendant’s statistical evidence and by perceptions of the defendant’s attorney (N= 189, Study 2B). Standardized path coefficients are shown in the diagram. Perceptions of the defendant’s attorney represent a composite variable from four observed variables (perceived preparedness, competence, persuasiveness, and credibility). Residuals for the endogenous variables were not shown for simplicity. A correlation between the residuals of the two mediators was specified in the model. *p< .05. ***p< .001 Visual technology and juror decision making 243 Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) simple and few in number (see Appendix A); the impact of more extensive and dramatic visual displays (such as those that legal visual consultants often design for lawyers) on decision making in actual cases might well be greater. More interestingly, Study 1 showed that PowerPoint affected ultimate judgments of responsibility most when only one party used it: The defendant’s judged responsibility was the least when the defendant used PowerPoint but the plaintiffs did not. This differential effect was not observed in Studies 2A or 2B, most likely because in those studies, unlike Study 1, PowerPoint use was manipulated for one party only. This finding that the differential use of visual technology affects liability judgments is consistent with one study of the effects of computer animations on mock juror decision making (Dunn et al., 2006, experiment 1). In that study, the percentage of participants who rendered a verdict for the plain- tiff in a vehicular accident case was significantly greater when the plaintiff’s lawyer used an animation and the defendant’s lawyer used only a still diagram than in any other condition.2 Thus, our study, like that of Dunn and her colleagues, may lend some support to critics of new courtroom technologies who are concerned that parties’ unequal access to those tech- nologies may unfairly affect case outcomes. It is true that this concern is most pressing when the high cost of new technolo- gies puts them out of reach of one of the parties (see Galves, 2000). PowerPoint, being accessible to just about any lawyer with a personal computer, would not seem to raise this fairness concern as urgently. On the other hand, parties’ unequal resources may mean that only one side can afford to hire visual design professionals to create more persuasive PowerPoint slides, so some potential for unfairness may remain. Second, the present research sheds some light on how lawyers’ use of PowerPoint may influence legal decision makers. In Study 1, the defendant’s use of PowerPoint led participants to recall the defendant’s statistical evidence more accurately and to think better of the defendant’s lawyer’s level of preparation, competence, and credibility; in Studies 2A and 2B, each party’s use of PowerPoint both improved participants’ recall of that party’s evidence and enhanced their perceptions of that party’s lawyer. Moreover, in Study 1, participants thought most highly of the defendant lawyer’s performance when the defendant used PowerPoint but the plaintiffs did not. Importantly, we also found that the effect of both the defendant’s and the plaintiffs’ uses of PowerPoint on the defendant’s judged responsibility were mediated by the visuals’ effect on participants’ recall of the statistical evidence proffered by the party using the visuals and by participants’ perceptions of that party’s lawyer. At the same time, in Study 1, when the plaintiffs used PowerPoint, participants were more likely to hold the defendant respon- sible in part because they remembered the defendant’s statistical evidence less accurately and found that the defen- dant’s lawyer was less well prepared and less persuasive. These findings indicate that PowerPoint can be an effective tool of persuasion in court through its impact on both central and peripheral processing by legal decision makers. We found that using PowerPoint enabled attorneys on either side of the case to persuade by helping decision makers to understand trial information better, which is indicative of central proces- sing. We also found that using PowerPoint enhanced persua- sion by providing peripheral cues to the perceived argument strength. When a lawyer used PowerPoint, participants thought better of his performance, and this in turn led them to hold the defendant more responsible (if the plaintiffs used it) or less responsible (if the defendant used it). (In Study 1, this effect was observed only for the defendant’s use of PowerPoint.) And, in Study 1, when the plaintiffs used PowerPoint, participants thought less well of the defendant’s attorney’s performance, and this in turn led them to hold the defendant more responsible. This mediating effect of mock jurors’ perceptions of the message source (the lawyer) on their responsibility judgments seems indicative of peripheral processing (Petty & Wegener, 1999). That the plaintiffs’ use of PowerPoint (in Study 1) also enhanced persuasion by leading participants to recall the defendant’s statistical evidence less accurately, which led them to hold the defendant more responsible, is more difficult to interpret in ELM terms but is also suggestive of peripheral processing. We reiterate that this evidence of both central and periph- eral processing effects from PowerPoint emerged using relatively few slides of mostly simple design. Stronger effects might well be created by more extensive displays. (As noted in the introduction, however, visual displays that are too dramatic might well undermine jurors’ ability to attend to and understand message content, as might other flaws in the displays (e.g., misspelled words and poor slide design).) To be sure, jurors in actual cases would presumably be more motivated than were college students in an experimental setting to understand the evidence correctly; therefore, they would be more likely to process trial information, visual or verbal, centrally rather than peripherally (Petty & Cacioppo, 1986; Petty & Wegener, 1999), and this might reduce their reliance on visual displays as peripheral cues and/or increase reliance on them as substantive aids to decision making. On the other hand, the complexity of information in an actual case (our case materials were greatly simplified compared to the real case on which they were based) could well reduce jurors’ situational ability to process that information (Petty, Cacioppo, Strathman, & Priester, 2005), increasing their reliance on visual displays as peripheral cues.3 It may be objected that our study design did not disaggre- gate the effects of using PowerPoint from the effects of using visual displays in general. Unlike Binder (2006, study 1), for example, we did not compare presentations using PowerPoint with presentations of the same visual information in another medium (e.g., poster board). In contemporary legal practice, however, PowerPoint is so accessible and easy to use—the 2 We cannot, however, account in any principled way for a major difference between the two studies: In Dunn and her colleagues’ study, it was only when the plaintiff used courtroom technology and the defendant did not that an effect occurred; in the present study, the converse was true. We can only speculate that the discrepancy may have to do with particular features of the respective visuals in two studies or other case-specific variables not con- trolled for. 3 Researchers have recently shown particular interest, for instance, in whether jurors may use visual displays of neuroscientific evidence as periph- eral cues (e.g., McCabe & Castel, 2008; Schweitzer et al., 2011). The most extensive study to date found that neuroimages do not bias juror decision making (Schweitzer et al., 2011). Although the present study does not in- volve neuroimages, we agree that this is a fruitful area for further research. 244 J. Park and N. Feigenson Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) program is on practically every PC and therefore can be used at little to no marginal cost beyond the lawyers’ time, in contrast to the expertise and cost required to produce profes- sional-quality poster boards—that for many lawyers, the choice whether or not to illustrate one’s opening statement comes down to the choice whether to use PowerPoint or nothing at all. Studying the effects of PowerPoint versus no visual display, therefore, has considerable ecological validity. The present study is limited in several respects. Partici- pants were undergraduates and thus not representative of actual jurors in some respects. In particular, the effects of PowerPoint or other new visual technologies on legal judg- ments may be affected by differences in viewers’ ages; this would be worth examining in subsequent studies. Participants did not have the opportunity to deliberate before reaching their judgments of liability or responsibility. The stimulus materials, consisting only of videotaped opening statements by actors playing the role of lawyers for the respective parties, greatly oversimplified the ways in which information would be presented in an actual trial.4 Perhaps most importantly, our findings could have been influenced by any number of variables for which we did not control, such as particular design features of the visuals. The PowerPoint slides used by the plaintiffs and the defen- dant, respectively, were not identical—they depicted different items of trial information and did so in different ways, even though the slides were matched as closely as possible in number, overall duration, complexity of content, and graphic style in general. These discrepancies were required by differences between the substance of the respective parties’ statements and by ecological validity concerns (participants in the condition in which both sides used visuals would have been suspicious had the visuals been identical), but we cannot rule out the possibility that they may have affected the findings in Study 1. However, the results of Study 2, in which only one party’s use of PowerPoint was manipulated and the other party did not use PowerPoint, tend to show that it was the use of PowerPoint and not differences between slide content or argument content that affected recall of case- relevant information, perceptions of the respective attorneys, and judged responsibility. In addition, in order to conduct any experimental study of the very important question whether differential use of courtroom methods by the respective parties matters to legal decision making, such confounds must at some point be risked; they should be minimized but cannot be completely eliminated. Nevertheless, these studies provide a first step toward a more thorough and systematic exploration of the effects of PowerPoint on legal decision making. We found that even relatively modest visual displays can affect judgments of legal responsibility. More importantly, our results shed some light on two commonly raised concerns about the increasing use of new visual technologies in court: that visual displays may persuade not only by helping decision makers to under- stand the case better but also by offering heuristic cues to judgment and that unequal access to these visual tools of persuasion may unfairly influence legal outcomes. More research in more realistic settings is needed to determine whether these concerns are indeed well grounded and to learn more about the impact of PowerPoint and other visual technologies on the justice system. REFERENCES Arbuckle, J. L. (2001). AMOS users’ guide 5.0. Chicago: Smallwaters. Baron, R., & Kenny, D. (1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical con- siderations. Journal of Personality and Social Psychology, 51, 1173–1182. Bell, B., & Loftus, E. (1985). Vivid persuasion in the courtroom. Journal of Personality Assessment, 49, 659–664. Binder, D. (2006). The relationship between need for cognition, argument strength, and the persuasiveness of courtroom technology. Unpublished dissertation. University of Wyoming. Cacioppo, J. T., & Petty, R. E. (1982). The need for cognition. Journal of Personality and Social Psychology, 42, 116–131. Center for Legal and Court Technology. (2007). [On-line] http://www.legal- techcenter.net/index. html. Chaiken, S., Liberman, A., & Eagly, A. H. (1989). Heuristic and systematic information processing within and beyond the persuasion context. In J. S. Uleman, & J. A. Bargh (Eds), Unintended thought (pp. 212–252). New York: Guilford. Collins, L. M., Graham, J. W., & Flaherty, B. P. (1998). An alternative framework for defining mediation. Multivariate Behavioral Research, 33, 295–312. Commonwealth v. Serge, 586 Pa. 671 (2006). Dahir, V. B. (2005). The effect of presentation mode and argument quality on memory, comprehension, and decision-making: Implications for com- puter display usage and evidence strength in the courtroom. Unpublished dissertation, University of Nevada, Reno. Dunn, R. (2000). Capitalizing on college students’ learning styles: Theory, practice, and research. In R. Dunn, & S. A. Griggs (Eds), Practical approaches to using learning styles in higher education (pp. 3–18). Westport, CT: Bergin & Garvey. Dunn, M., Salovey, P., & Feigenson, N. (2006). The jury persuaded (and not): Computer animation in the courtroom. Law and Policy, 28, 228–248. Epstein, S. (1994). Integration of the cognitive and psychodynamic uncon- scious. American Psychologist, 49, 709–724. Federal Judicial Center/National Institute for Trial Advocacy. (2001). Effective use of courtroom technology. Washington, DC: Author. Feigenson, N. (2010). Visual evidence. Psychonomic Bulletin & Review, 17, 149–154. Galves, F. (2000). Where the not-so-wild things are: Computers in the court- room, the Federal Rules of Evidence, and the need for institutional reform and more judicial acceptance. Harvard Journal of Law and Technology, 13, 165–260. Hewson, L., & Goodman-Delahunty, J. (2008). Using multimedia to support jury understanding of DNA profiling evidence. Australian Journal of Forensic Sciences, 40(1), 55–64. Kahneman, D., & Frederick, S. (2002). Representativeness revisited: Attribute substitution in intuitive judgment. In T. Gilovich, D. Griffin, & D. Kahneman (Eds.), Heuristics and biases: The psychology of intuitive judgment (pp. 49–81). Cambridge, UK: Cambridge University Press. Kenny, D. A., Kashy, D. A., & Bolger, N. (1998). Data analysis in social psychology. In D. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), Handbook of social psychology (4th ed., Vol. 1, pp. 233–265). New York: McGraw-Hill. Levasseur, D., & Sawyer, K. (2006). Pedagogy meets PowerPoint: A research review of the effects of computer-generated slides in the classroom. Review of Communication, 6, 101–123. Levett, L., & Kovera, M. (2009). Psychological mediators of the effects of opposing expert testimony on juror decisions. Psychology, Public Policy, and Law, 15, 124–148. 4 The slides used in this study could, if used in an actual trial, be introduced to illustrate the lawyers’ opening statements and/or the testimony of expert witnesses; in either case, they would not be subject to the demanding authentication requirements to which computer-animated re-enactments or simulations of events are held (e.g., Commonwealth v. Serge, 2006). In this respect, showing mock jurors the slides to illustrate the spoken words of the respective parties’ attorneys, as we did, does not present significant ecolog- ical validity issues. Visual technology and juror decision making 245 Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013) http://www.legaltechcenter.net/index http://www.legaltechcenter.net/index Mayer, R. (2001).Multimedia learning. Cambridge: Cambridge University Press. McAuliff, B., Kovera, M., & Nunez, G. (2009). Can jurors recognizing missing control groups, confounds, and experimenter bias in psycholog- ical science? Law and Human Behavior, 33, 247–257. McCabe, D., & Castel, A. (2008). Seeing is believing: The effect of brain images on judgments of scientific reasoning. Cognition, 107, 343–352. Morell, L. (1999). New technology: Experimental research on the influence of computer-animated displays on jurors. Southwestern University Law Review, 28, 411–415. Paivio, A. (1971). Imagery and verbal processes. Hillsdale, NJ: Lawrence Erlbaum. Paivio, A. (1986). Mental representations: A dual coding approach. New York: Oxford University Press. Petty, R., & Cacioppo, J. (1986). The elaboration likelihood model of persuasion. In L. Berkowitz (Ed.), Advances in experimental social psychology (vol. 19, pp. 123–205). Orlando, FL: Academic Press. Petty, R., & Wegener, D. (1999). The elaboration likelihood model: Current status and controversies. In S. Chaiken, & Y. Trope (Eds.), Dual-process theories in social psychology (pp. 41–72). New York: Guilford Press. Petty, R. E., Cacioppo, J. T., Strathman, A. J., & Priester, J. R. (2005). To think or not to think: Exploring two routes to persuasion. In T. Brock, & M. Green (Eds.) Persuasion: Psychological insights and perspectives (2nd edn, pp. 81–116), Thousand Oaks, CA: Sage Pub- lications, Inc. Schweitzer , N., Saks, M., Murphy, E., Roskies , A., Sinnott- Armstrong, W., & Gaudet, L. (2011). Neuroimages as evidence in a mens rea defense: No impact. Psychology, Public Policy, and Law, 17, 357–393. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexper- imental studies: New procedures and recommendations. Psychological Methods, 7, 422–445. 246 J. Park and N. Feigenson Copyright © 2012 John Wiley & Sons, Ltd. Appl. Cognit. Psychol. 27: 235–246 (2013)