4.1 Overview
The comparative efficacy of own and provided notes may be influenced by a number of different factors. Experiment 3 investigated the impact of deliberation expectation and the deliberation process, while the final experiment in this series was designed to examine the interaction between the two note types and trial complexity. Jury-aid researchers have recognised the potential for trial complexity to have a qualifying effect on the efficacy of such aids, and a large component of the body of research examining jury-aid efficacy has manipulated trial complexity to explore this (Bourgeois et al., 1993). There is still no clear understanding, however, of how trial complexity influences or interferes with jurors’ decision-making abilities,
comprehension, and in turn with the jury-aids provided (Findlay, 2001; Horowitz, ForsterLee, & Brolly, 1996; Payne, 1976).
4.2 Trial Complexity
Trial complexity itself continues to be conceptualised in a number of different ways, and a single consistent definition remains elusive (Kramer & Kerr, 1989). Lempert (1981) suggested that criteria which may give some indication of trial complexity include trial length, degree of voluminous evidence, and complexity of legal standards. MacCoun (1987, 1989) addressed the question of trial complexity more directly and comprehensively, identifying three forms of trial complexity: dispute complexity (including the number of disputants and issues); evidence complexity (quantity and consistency of evidence, and reliability and technicality of evidence); and decision complexity (including legal complexity and complexity of inferential chains).
Heuer and Penrod (1994) determined a conceptualisation of trial complexity by asking the 228 judges participating in their field trial to provide information about numerous variables relating to trial complexity (e.g., number of witnesses, number of claims, trial duration, complexity of legal argument). A principle components
analysis with varimax rotation was performed on these responses, and revealed three components of complexity which differ slightly from previous characterisations of the construct (Lempert, 1981; MacCoun, 1987); legal complexity, evidence
complexity and length of trial.
Horowitz et al. (1996) provided yet another way to classify dimensions of trial complexity, suggesting that the components of complexity which are most immediately relevant to a juror’s task are decision complexity (the intricacies of the legal issues, and the juror’s ability to apply these), and evidence complexity.
Evidence complexity is further divided into volume of evidence to be processed (including the number of witnesses and number of plaintiffs/defendants, the number and similarities of injuries alleged), the clarity of evidence (whether the evidence clearly favours one side or is ambiguous), and the comprehensibility of the evidence itself (including the intricacy and specialisation of the language, and the difficulty of the evidentiary issues).
Horowitz et al. (1996) examined the impact of two parts of the evidence complexity dimension from this conceptualised framework of trial complexity, focusing on the effect of the number of plaintiffs and witnesses who testify (volume of evidence), and the effect of language technicality (comprehensibility of evidence) on juror decision-making. Mock-jurors viewed a toxic torts case which involved either four plaintiffs or eight plaintiffs (low vs. high information load), and which used complex legal jargon and expression, or more simple expression and lay terms (high vs. low language complexity). Information load seemed to affect jurors’ liability decisions, with those in the low information load condition assigning more blame to the defendant, in accordance with the direction of the evidence; conversely, jurors in the high information load condition assigned more blame to the plaintiffs, which was clearly counter to the evidence. Language complexity, on the other hand, affected jurors’ ability to distinguish differentially worthy plaintiffs, with less
complex language allowing more appropriate distinction between plaintiffs. Only in the low information load and low language complexity conditions, however, were jurors able to award appropriate compensation. So, high information load seemed to lead to suboptimal processing of information, such that jurors misinterpreted the direction of the evidence, and while language complexity did not affect the overall outcome, it did obscure distinctions between the plaintiffs.
4.3 Relationship Between Trial Complexity and Jury-Aids
Such research indeed suggests that particular jury-aids might be better suited to particular types of complexity. The elaboration likelihood model (Petty &
Cacioppo, 1986) and the heuristic-systematic model (Chaiken, Lieberman and Eagly, 1989) both suggest that information accessibility and message comprehensibility will impact individuals’ ability to engage central processing routes and handle
information in a systematic manner. It therefore seems logical to suggest that
different types of jury-aid will provide greater information accessibility and message comprehensibility, depending on how directly they address the kind of complexity jurors are facing. Heuer and Penrod (1994) suggested that where trial complexity concerns complex evidence, permitting jurors to ask questions may be the most useful type of assistance, whereas for lengthy trials, notetaking may improve juror comprehension and recall. Finally, where the legal issues covered in the trial are particularly complex, written judicial instructions and preinstruction on the law may be the most effective interventions. Some research has combined particular jury-aids with various types of complexity to test such hypotheses.
ForsterLee and Horowitz (1997) manipulated trial complexity by changing the extent of the defendant’s liability in a video re-enactment of a toxic torts trial in which jurors were permitted to take notes. Jurors attempting to make compensatory awards to plaintiffs were instructed either that the defendant was 70% liable (a less ambiguous condition for jurors), or 40% liable (a more ambiguous decision-making condition). While notetaking jurors showed improved performance overall, the low liability condition vitiated the effectiveness of notetaking as a cognitive aid to some extent, thus suggesting that notetaking is beneficial when jurors confront a task of moderate difficulty, but when facing more ambiguous evidence, provides little advantage.
Heuer and Penrod’s (1994) field experiment, which determined trial
complexity to consist of complexity of evidence, complexity of law, and trial length, also examined the utility of juror notetaking and question asking in trials with varying levels of each complexity dimension. Although the presence of additional aids was not manipulated by the experimenters, use of judicial commentary and summary, pattern instructions, juror orientation processes and special verdict forms was reported. Results of this study confirmed that the efficacy of particular jury-aid procedures was qualified by the type of complexity the trial was characterised by. Juror questions were found to be most beneficial to jurors in cases high in legal and evidentiary complexity, but provided little help where the trial consisted of large quantities of information. The interactions between juror notetaking and trial complexity, however, were difficult to properly assess. It was expected that
notetaking would be most beneficial where quantity of information was high, but no such result was found. Special verdict forms and orientation procedures did not produce consistent interactions in the same direction, while judge’s commentary and summary consistently produced a negative result, such that for cases high in evidence complexity, judicial commentary was perceived as being less helpful. Finally, legal complexity was found to be most challenging for jurors in cases where judges relied heavily on pattern instructions. Of course, it must be noted that Heuer and Penrod’s (1994) conclusions are based solely on jurors’ own assessments of their trial
experience, rather than objective measures of performance.
Horowitz and Bordens (2002) used a videotaped civil trial to investigate the effects of trial complexity and notetaking on compensation and damages awarded. Results indicated that six-person juries exposed to the high information load version (involving four plaintiffs) who did not take notes returned significantly greater compensation awards, but such biases towards overvaluing the claims of multiple
plaintiffs were mitigated where notetaking was permitted. However, this effect was reversed for jurors exposed to the low information load condition (involving one plaintiff); notetaking jurors awarded higher compensation than non-notetaking ones. High information load was also associated with greater variability of jurors’ punitive damage awards. The experimenters had predicted that notetaking would decrease this variability, but this hypothesis was not confirmed.
Bourgeois et al. (1993) examined the impact of allowing transcript access for jurors who were exposed to a high evidence complexity or low evidence complexity version of a medical malpractice case. Evidence complexity was manipulated by the amount of medical jargon and technical language implemented. These researchers proposed that transcript access would facilitate systematic processing, particularly when jurors were exposed to complex and technical information, and this prediction was supported, at least for jurors’ verdict scores. The trial was adapted such that systematic processing of the evidence should result in a decision in favour of the defendant; jurors confronting highly technical information without access to trial transcripts favoured the plaintiff, whereas jurors with access decided for the
defendant. Jurors exposed to the low technicality version decided for the defendant regardless of whether they were permitted to access the transcript. However, on a dependent variable measuring jurors’ evidence processing via a recognition test, no significant results were recorded.
There is, therefore, still much room for clarification on the issue of how different dimensions of trial complexity affect the benefits provided by various jury- aids (Harding, 1988), as none of the previously published research has investigated the comparative benefits of different types of jury-aids in the context of varying trial complexity type, using an objective measure of recall or comprehension. The aim of this fourth study, then, was to examine the benefits conferred by own and provided
notes under varying complexity conditions. The dimensions of complexity subjected to empirical examination in this study via different versions of the mock trial were Heuer and Penrod’s (1994) legal complexity, evidence complexity and trial length. A between subjects factorial design was employed, in which both note type and complexity type were manipulated, and juror comprehension was assessed in the same manner as the previous studies.
4.4 Hypotheses
These hypotheses are based firstly on the results of the previous experiments conducted in this series, which have firmly established the superiority of provided notes over participants’ own notes. They are also based to some extent on previous research, however the amount of directly relevant previous research is limited. The only research which has previously manipulated a dimension of trial complexity and compared jurors’ own notes with another type of jury-aid using an objective measure of performance is that conducted by Bourgeois et al. (1993), which returned
significant results with regard to jurors’ verdicts, but failed to produce any significant results relating to juror comprehension (as assessed by a recognition task).
Thus, this experiment was quite exploratory in nature, and the hypotheses were guided by logical predictions regarding which note type would increase the likelihood of systematic information processing using central routes, by enhancing information accessibility and message comprehensibility, based on the type of complexity the notes were attempting to address (Petty & Cacioppo, 1986).
Objective measures.
A complexity type by note type interaction was anticipated for the fact score dependent variable. Participants with access to provided notes were expected to achieve significantly higher fact scores than participants who took their own notes
across all complexity types, however the magnitude of the advantage conferred by provided notes was predicted to be smaller for participants in the length complexity condition. Although provided notes were still expected to constitute a more
comprehensive resource than participants’ own notes, it was thought that by
essentially increasing the information load placed on participants exposed to length complexity by adding to the overall amount of material, their relative efficacy would decrease. It was also predicted that participants in the length complexity condition would achieve significantly lower fact scores than participants in the other
complexity conditions; it was thought that the increased fact-related information load these participants were exposed to would lead to overall poorer performance.
A complexity type by note type interaction was anticipated for the law score dependent variable. With regard to the influence of note type, the pattern of results was expected to reflect that described above (the magnitude of the advantage
conferred by provided notes was predicted to be smaller for participants in the length complexity condition for the same reasons described above). On the other hand, it was also predicted that participants in the legal complexity condition would achieve significantly lower law scores overall than participants in the other complexity conditions. Although this dependent variable did not test their understanding of the additional law they were exposed to, it was thought that this exposure to additional legal principles might interfere with participants’ ability to comprehend the
fundamental legal information, leading to poorer overall performance.
Manipulation checks.
The evidence score dependent variable was a measure of participants’ comprehension of additional information contained within the evidence complexity version of the trial, thus acting as a manipulation check as well as a dependent
variable. Since those in the evidence complexity condition were the only participants exposed to this information, a complexity type by note type interaction was
anticipated, such that evidence scores for those participants were expected to be significantly higher than those for participants in the control, legal and length complexity conditions. In the latter three conditions, no differences were expected between the scores of participants according to note type, but in the evidence complexity condition, participants with provided notes were expected to achieve significantly higher evidence scores than those who took their own notes, because the provided notes would constitute a more comprehensive source of information review (Chaiken & Eagly, 1976; Bourgeois et al., 1993).
Considering that the manslaughter score dependent variable also doubled as a manipulation check, such that participants in the legal complexity condition were the only participants exposed to the information required to answer the manslaughter questions, a complexity type by note type interaction was again anticipated. Manslaughter scores for these participants were expected to be significantly higher than those for participants in the control, evidence and length complexity conditions. In the latter three conditions, no differences were expected between the scores of participants according to note type, but in the legal complexity condition, participants with provided notes were expected to achieve significantly higher manslaughter scores than those who took their own notes, because the provided notes would constitute a more comprehensive source of information review.
Although the manslaughter scenario questions presented participants with novel scenarios, successful completion required knowledge about legal concepts related to manslaughter, which only participants in the legal complexity condition had any exposure to. Thus, once again a complexity type by note type interaction was anticipated, with manslaughter scenario scores for participants in the legal
complexity condition expected to be significantly higher than those for participants in the remaining complexity conditions. In the remaining three conditions, no differences were expected between the scores of participants according to note type, but in the legal complexity condition, participants with provided notes were expected to achieve significantly higher manslaughter scenario scores than those who took their own notes, because the provided notes would constitute a more comprehensive source of information review.
Subjective ratings.
Results of the subjective measures from previous studies in the series have been inconsistent, bringing into question the utility of these particular dependent variables. Nonetheless, they were again included. A complexity type by note type interaction was anticipated for all subjective measures. Participants with access to provided notes were expected to return significantly higher verdict confidence, lower verdict difficulty and lower law difficulty ratings than participants who took their own notes. For participants in the legal complexity condition (regardless of note type), overall verdict confidence ratings were expected to be significantly lower, and overall verdict difficulty and law difficulty ratings were expected to be higher than those of participants in the remaining complexity conditions, due to the inclusion of an additional charge and subsequent additional verdict for these participants. This effect was predicted to be attenuated by the presence of provided notes, so the discrepancy between the verdict confidence, verdict difficulty and law difficulty scores of participants in the legal complexity condition and those exposed to the other complexity types was expected to be smaller for participants with provided notes than for participants making their own notes.
Utility ratings.
The contents of both the control and length complexity trials contained the type of information considered most accessible to the jurors, whereas the more scientific and legal nature of the evidence and legal complexity trials respectively meant that the information contained within them was presumably more unfamiliar and complex. Regardless of whether jurors’ notes objectively reflected this
difference, it was hypothesised that participants would perceive their own notes to be of greater utility under the former conditions than the latter ones (Bartlett, 1932), so the own notes utility ratings of participants in the control and length complexity conditions were expected to be significantly higher than those of participants in the evidence and legal complexity conditions.
Participants were expected to appreciate the resources contained within the provided notes regardless of the complexity type they were exposed to, so
participants’ perceptions of the provided notes’ utility and the subsequent ratings were not expected to differ by complexity type. However, as in previous
experiments in this series, some components of the provided notes were expected to receive significantly higher utility ratings than others, and a main effect for
component was therefore anticipated. More specifically, in line with the previous studies, it was predicted that verdict flowcharts would receive significantly lower utility ratings than each of the other provided notes components.
4.5 Method
Participants.
A total of 120 jury-eligible participants (39 males and 81 females, with a mean age of 24.2 years) completed the experiment. Participants were first, second and third year psychology students studying at the Hobart and Launceston campuses
of the University of Tasmania, who completed the experiment as part of their course requirements. All participants watched the trial DVD on a large screen in a
classroom.
Design.
This experiment employed a 4[complexity type: control, length, legal and evidence] x 2[note type: own or provided] factorial design. The dependent variables measured were the same as those assessed in the previous studies, with the addition of a further three measures. Participant comprehension of information presented in the evidence complexity condition and in the legal complexity condition was
measured by six multiple-choice questions for each. Participants’ ability to apply the specific law presented in the legal complexity condition was assessed using five scenarios with multiple-choice verdicts.
Materials.
In order to accommodate the complexity manipulation, some changes were made to the materials used in this experiment.
Mock criminal trial.
Participants in this experiment viewed one of four different versions of the mock criminal trial. The trial which was developed for use in Experiments 1, 2 and 3