2. The Empirical Evidence
This chapter examines the empirical evidence about the effectiveness of the eight different techniques identified in chapter 1, namely trial transcripts, juror note-taking, audio-visual and digital presentation methods, juror questions, plain language directions, pre-instruction, written directions and structured decision aids. In the discussion that follows, the terms "community jurors" and "student jurors" are used to refer to participants in mock jury studies. These terms are explained in Annex 1.
2.1 Key Findings
The empirical evidence suggests that the most effective ways of enhancing juror memory and understanding are juror note-taking, pre-instruction, plain language directions and the use of written directions and structured decision aids (routes to verdict). Each of the methods targets different issues (some improve memory, some improve understanding and application of legal tests) and there is evidence to suggest that they are best used in combination, rather than as alternatives.
- Juror note-taking: there is a substantial body of evidence from good quality empirical studies to suggest that taking notes improves juror memory for the evidence. Recent studies have suggested that providing jurors with trial-ordered notebooks (structured notebooks which help jurors to organise their notes) is particularly beneficial in this respect and may assist jurors who are not skilled at note-taking.
- Pre-instruction: there is a considerable body of evidence from good quality empirical studies to suggest that pre-instructing jurors on the substantive legal issues in the case improves comprehension and memory for the evidence. Despite concerns that it might cause jurors to reach their verdict decisions prematurely, there is no evidence that this is the case.
- Plain language directions: there is a vast body of evidence, including a considerable amount from good quality empirical studies, to suggest that simplifying jury directions can improve juror comprehension of legal concepts. Almost all of this stems from studies undertaken in the US, where jury directions tend to be relatively complex, but the general principle that plain language directions assist jurors applies to any jurisdiction, including Scotland.
- Written directions and routes to verdict: There is a substantial evidence base to show that providing directions in writing improves memory and simple comprehension - in other words, they help jurors to remember and re-state those directions - but the evidence on whether they help jurors to gain a deeper understanding of directions is more equivocal. There is a developing evidence base relating to structured decision aids (routes to verdict), which are a more recent innovation. The evidence that does exist (particularly from the better designed studies) suggests that these are more effective than written directions in improving applied comprehension - jurors' ability to correctly apply legal tests to the evidence. Oral directions should be tailored to the route to verdict provided, otherwise there is a danger that jurors ignore the route to verdict.
2.2 Trial Transcripts
There is only a limited body of evidence on the effectiveness of trial transcripts and it is not especially convincing. One relatively realistic mock jury study found that a trial transcript was helpful in assisting jurors to remember the evidence led in the trial, but that jurors' own notes did so equally well. The reason for this may be that a full transcript - especially one provided in paper copy - is difficult to navigate. This difficulty will be much greater in the context of a real trial, which will be considerably longer than any simulation. Producing a full transcript quickly after the conclusion of the trial, so as not to delay jury deliberations, also poses considerable challenges.
This section examines the empirical evidence relating to providing trial transcripts - a written transcript of all of the evidence that was led - to assist jurors during their deliberations. The obvious advantage of this is that it means that jurors do not have to rely entirely on their memory for the evidence and, if disputes about what particular witnesses said arise during deliberations, these can be settled by looking at the written account. There are, however, potential difficulties. It may not be easy to produce a written transcript quickly and there would be costs involved. In a lengthy trial the transcript could be substantial and difficult for jurors to navigate and this could lengthen deliberations. This could be addressed by providing an index or perhaps a searchable electronic version - but either of these options adds time and cost. There is also the issue of what to do about evidence that was led but was later ruled to be inadmissible (which may be more difficult for a juror to ignore if it is recorded in the transcript) or exchanges that happen outside the presence of a jury. Removing these passages would of course be possible, but would add to the time taken to produce the record, again potentially delaying deliberations. 
2.2.1 Mock Juror Studies
There are only two empirical studies that have attempted to evaluate the effect of providing a full trial transcript to jurors.  Horowitz and ForsterLee's study involved 195 community mock jurors, who watched a one hour videotaped reconstruction of a civil trial, where the roles were played by actors.  The trial was designed so that there were four plaintiffs, whose claims could be legally differentiated. If jurors correctly applied the directions to the evidence, there was a 'correct' legal answer in terms of who should be compensated and to what degree. After watching the trial, participants deliberated in groups of five or six for up to 30 minutes, before returning their verdicts, and individually completed questionnaires that assessed their recall of the evidence. Some of the mock jurors were given a transcript of the evidence to take into deliberations and use when completing the recollection test.
The researchers found that jurors who had the transcript performed significantly better on the recall test than those who did not, but that they did not outperform the jurors who took their own notes.  Access to a trial transcript did not, however, help juries to reach the 'correct' legal answer - unlike the note-taking juries, who were better able to do this.  The researchers conclude that while there may be some benefits to transcript access, the benefits of note-taking were greater. They speculate that this may have been because the transcript was a lengthy document that would have been difficult to navigate, whereas a juror's own notes would have been more focused.  It may also be that jurors find it easier to navigate a document that they have themselves created.
Kelly's study involved 92 undergraduate psychology student mock jurors, who watched a 55 minute trial video based on a real criminal trial, before completing a multiple choice test of their recall for the evidence (as well as questions designed to test their comprehension of the law).  One group of participants was given a trial transcript, and another group was given a package of materials which contained a chronology of events and other materials such as a verdict flowchart and written directions on the relevant law. A third group was given both. The groups with the package performed significantly better on fact recall than those who only had the transcript.  This suggests that transcripts do not confer any advantages over more simplified written materials such as a timeline which was used in this case. That said, the experiment did only involve a small number of undergraduate students, which limits the reliance that can be placed on it.
2.3 Juror Note-taking
There is a considerable body of empirical evidence, from both mock juror studies and field studies, pointing to juror note-taking being beneficial in terms of memory for the evidence, and to note-taking being more helpful in this respect than providing a full trial transcript. The mock juror studies vary in terms of their realism - a real trial would be considerably longer than even the most realistic experiment - but if anything this probably under-estimates the benefits of note-taking. Most studies involved civil trials, but there is no reason to think that note-taking would be any less helpful in the context of a criminal trial. These findings are particular pertinent in the Scottish context, where trial judges do not routinely summarise the evidence, so jurors need to rely more heavily on their memory than they might have to in some other jurisdictions.
There is a concern that has arisen, especially in field studies, that jurors are not all skilled at taking notes and that those with better note-taking skills may dominate deliberations. This concern could be addressed at least to an extent by the provision of trial ordered notebooks ( TONs) - structured notebooks that help jurors to organise their notes. The two empirical studies that have tested TONs both found that jurors who used them performed better in terms of their recollection of the evidence than those who took freestyle notes.
This section considers the empirical evidence relating to the effectiveness of juror note-taking. The most obvious advantage of this is that it may improve juror memory for the evidence that was led during the trial (and perhaps also the legal directions, although providing these in writing is a more obvious way of addressing that issue). If note taking does improve juror memory of the evidence this may be particularly useful in the Scottish context, where (unlike in some other jurisdictions) trial judges do not routinely summarise the evidence for juries at the end of the trial. Note-taking could also have secondary advantages in that note-taking jurors may feel more involved in the trial and therefore more satisfied with their experience of jury service. 
There are, however, a number of possible difficulties. Jurors may vary in their ability to take written notes and those who produce the best notes might dominate deliberations. Notes may be inaccurate and may be given undue weight simply because they are in written form. Jurors might take notes enthusiastically at the start of the trial, but become jaded as the trial progresses.  Note-taking might also be distracting - either to the note-taker, who may not pay sufficient attention to the demeanour of witnesses, or to other jurors.  On the other hand, note-taking might equally assist in sustaining juror concentration by preventing their attention from wandering away from proceedings. 
There is an extensive body of empirical research on juror note-taking which comprises both mock jury and field experiments.
2.3.1 Mock Jury Studies
There is a large volume of mock jury studies, the vast majority of which point in the same direction - note-taking has a positive effect on memory for the evidence. This effect seems to occur even when jurors do not have access to their notes after making them.  The benefits of juror note-taking are greatest, however, when jurors are supplied with materials that help them organise their notes (generally referred to as a trial-ordered notebook). Trial-ordered notebooks are discussed towards the end of this section.
One of the earliest studies of juror note-taking was undertaken by Rosenhan et al,  whose experiment involved 144 student mock jurors, who watched a 75 minute trial video filmed in a court room. Half were permitted to take notes, half were not. Jurors completed questionnaires designed to test their memory for factual issues and the note taking group were permitted access to their notes when completing the questionnaires. Note-takers scored significantly higher on recall than jurors who did not take notes.  The study was not especially realistic, however. Aside from its use of student jurors, the videotaped 'trial' consisted only of opening speeches (although they did last for 75 minutes). That said, the benefits of note-taking may be enhanced in a longer, more complex trial. The argument against this would be that jurors lose interest in note-taking over the course of a longer trial. However, there is some, albeit limited, evidence from field studies to suggest that this does not happen.  The value of the Rosenhan et al study also lies in the fact that the researchers analysed the notes that the mock jurors made and found that the volume of notes taken and the degree to which the notes were well organised were both significantly related to recall.  This suggests that there may be a role for some sort of organisational aid, such as trial-ordered notebooks, which are discussed at the end of this section.
Subsequent studies have been near unanimous in replicating the relationship between note-taking and memory. Fitzgerald, for example, found that note-takers were able to recall significantly more relevant details about a civil liability trial than jurors who did not take notes.  The mock jurors were a mixture of college students, university support staff and older adults (aged between 55 and 75) recruited from the community. The trial reconstruction was a relatively realistic one, lasting for two hours and filmed using professional actors working from a script adapted from a real trial. The note-taking benefits were apparent across the range of participants. More recently, in a study involving 144 mock jurors who were either university students or staff, Thorley found that note-taking significantly enhanced recall of trial information.  Although the simulated trial he used was only 30 minutes long, as noted above it is likely that the benefit in a longer trial would be even greater. Thorley also found that asking note-takers to spend ten minutes following the trial reading through and reviewing their notes enhanced recall still further. 
All of the studies discussed thus far have permitted mock jurors to access their notes after the conclusion of the trial. There have, however, been a series of studies that show that the beneficial effect of note-taking occurs even when participants do not have access to their notes. The most extensive of these was undertaken by Horowitz and Bordens and involved 576 community jurors who watched one of two versions of a civil trial - a relatively simple one that lasted for 58 minutes or a more complex one that lasted for 72 minutes.  After watching the trial, the mock jurors were divided into six or twelve person juries (the main focus of the experiment was to examine the impact of jury size on information recall) and deliberated for up to 75 minutes before returning a verdict. Half of the juries were composed of jurors who had been permitted to take notes, half were composed of jurors who had not. After their jury had returned a verdict, the notes were collected and individual jurors were asked to recall as much information about the trial as possible. Jurors who were permitted to take notes recalled significantly more relevant facts than those who were not, despite not having access to their notes when completing the recall task.  The beneficial effect of note-taking was stronger in the twelve person juries than in the six person juries,  which might suggest the existence of a pooling effect whereby information that is missed by some jurors is recorded by others.
A number of other mock jury studies have found that note-taking enhances juror memory for the evidence even when jurors do not have access to their notes.  ForsterLee et al suggest that this is because the primary benefit of note-taking occurs at the encoding stage - in other words it is the process of making notes itself that is beneficial, perhaps because it enhances concentration and encourages information processing.  They attempted to demonstrate this in a study that compared the recall of note-taking jurors with and without access to their notes. In the study, 192 community mock jurors listened to a two hour audiotape of a civil trial reconstruction. Note-takers outperformed those who did not take notes on a fact recollection task. However, the difference was only significant in the group who were permitted access to their notes when completing the task,  which suggests that even if there is some benefit to be had from note-taking at the encoding stage, this benefit is further enhanced if jurors are allowed to consult their notes after the trial is over.
This is confirmed by two other studies that have compared the two conditions, both of which found that while note-taking in itself improved juror memory for trial facts, jurors who had access to their notes while attempting to recall information outperformed jurors who took notes but did not have access to them.  In reality, it is unlikely that note-taking jurors would have their notes removed prior to deliberation - this may result in considerable juror dissatisfaction - and the evidence suggests that if this were done, it would reduce the benefit of note-taking.
Two mock jury studies - both conducted relatively recently - have tested the use of trial-ordered notebooks ( TONs). A TON is a structured notebook which aims to help juries organise their note-taking. The first was conducted by Hope et al, whose TON contained sections (each of 2 sides of A4 paper) for each witness's evidence in chief and cross-examination, the opening statements, the closing statements and the trial judge's charge.  It was a small-scale study, involving only 58 community jurors who were randomly assigned to a group that either took freestyle notes, took notes in a TON or did not take notes. Participants watched a 35 minute mock trial video before completing a questionnaire to test their memory for details of the trial. They were permitted to refer to their notes in doing so. The group with the TON performed significantly better than the freestyle note group, whose performance was only slightly better than the group who did not take notes.  (Interestingly, when participants were asked to self-report their ability to recall details of the trial, there was no significant difference between the three groups.  ) The researchers also analysed the content of the notes that jurors took and found that the TON group recorded significantly more correct trial details (and significantly more legally relevant trial details) than the freestyle note-takers. 
The second study was undertaken by Thorley et al and involved 130 mock jurors who were either students or ex-students. Participants watched a 30 minute mock trial video. Jurors either took freestyle notes, used a TON (adapted only very slightly from the one used by Hope et al) or did not take notes. They were then asked to write down as much information as they could remember about the evidence and complete an exercise in recognition, where they were presented with statements about the evidence and asked if they were true or false. The note-taking groups were divided, with half having access to their notes and half having their notes removed after watching the trial video. The researchers analysed the content of the notes and found that a significantly higher volume of correct information was recorded by the TON group compared to the freestyle note-takers.  The notes made by the TON note-takers were also significantly more accurate.  In the free recall exercise, both the freestyle note-taking and TON groups performed significantly better than the group who did not take notes, but the best performance was in the TON group who had access to their notes while completing the tasks.  There was little difference in performance on the recognition task between the groups, with all groups performing well. However, the researchers suggest that this may have been because the recognition test statements were quite detailed and may have provided powerful retrieval cues for jurors who did not take notes.  It might also have been because the trial - at 30 minutes - was relatively short and the evidence was still very fresh in the minds of the participants when they completed the test.
Finally, a study undertaken by Dann et al examined mock juror satisfaction with note-taking.  The study involved 480 community mock jurors who watched a 70 minute videotaped trial filmed in a real courtroom using legal professionals to play the parts of the judge and lawyers. The jurors deliberated in groups of eight with no time limit and deliberations were recorded. The study did not focus on the effect of note-taking on memory, but instead examined its effect on comprehension of DNA evidence and juror satisfaction. Note-taking did not improve comprehension,  which is unsurprising, but 92 per cent of those who took notes reported that they found it to be helpful and this figure was far higher than for any of the other innovations that were tested.  Dann et al also found that jurors in three quarters of the note-taking juries made express reference to their notes in the course of deliberations and only one of these references was inaccurate (the juror had recorded the wrong last name of a witness). 
2.3.2 Field Studies
The first and most extensive field experiment was undertaken by Heuer and Penrod. Here the researchers tested the effectiveness of a number of different techniques, of which note-taking was one, in a mixture of real civil and criminal trials in Wisconsin. A total of 29 judges agreed to randomly assign their trials so that jurors were either not allowed to take notes or were provided with writing materials and were encouraged to do so. Data was collected from 67 trials and questionnaires completed by 550 individual jurors, 260 of whom were in the note-taking condition (although only 172 of these actually took notes). The researchers found that while jurors appeared to take their note-taking task seriously (as evidenced by the content of the notes, which were examined by the researchers),  the jurors who took notes did not perform any better on multiple choice tests of memory and understanding than those who did not.  However, jurors were permitted to take the questionnaire home with them after the trial had concluded and mail it back to the researchers - the average time between the trial being completed and the questionnaire being completed was 2.1 days  - and they did not have access to their notes when completing it, so the failure to show any beneficial effect of note-taking may simply be due to this.
A further field experiment by Heuer and Penrod, this time involving 160 trials across 33 US states, attempted to address this deficiency by ensuring that jurors completed questionnaires before leaving the courtroom.  However, this study suffered from a different limitation, in that jurors were simply asked how well they thought they remembered the evidence. There was no significant difference between note takers and those who did not take notes in their answers to this question,  but the lack of any objective assessment of memory means that this finding is of limited value, especially as we know that jurors over-estimate their ability to remember evidence.  One further point that it is worth noting from the experiment, however, is that it provides some limited evidence that jurors do not tire of making notes over the course of the trial. Heuer and Penrod found that jurors made proportionally more notes on the defence case (which is always presented last) than on the prosecution case. 
A smaller scale study, undertaken by Flango, involved only four trials (two criminal, two civil), in which two of the juries were permitted to take notes and two were not. Jurors completed questionnaires after the conclusion of the trial. The majority of jurors who took notes reported that they were useful as a memory aid and that they felt that they were able to participate more effectively in deliberations because of their note-taking,  but like Heuer and Penrod's second study the use of a self-reporting measure limits the value of this finding.
These are the only field experiments that have attempted to test the impact of note-taking by real jurors against a 'control' group of real jurors who were not permitted to take notes. There have, however, been two further field experiments in the US that have surveyed jurors about their experiences of taking notes. Sand and Reiss looked at the impact of a number of different procedures (of which note-taking was one) in real criminal and civil trials. Note-taking was tested by six judges in 32 trials (14 of which were criminal), in which jurors were advised they could take notes and were given materials to do so.  In all 14 of the criminal cases, the judges reported that they were "pleased" with the experiment,  with two stating that they thought that note-taking aided the jurors in recalling facts and in keeping track of the exhibits.  Three judges and three lawyers observed that note-taking appeared to increase juror attentiveness and that, "far from proving to be a distraction, it instead raised the jury's interest in the trial".  All of these observations are, however, of limited value, as they were based purely on the perception of the judges and lawyers involved, rather than any objective measure. Only twelve jurors were surveyed, with seven of these reporting that the notes they took served as a useful memory aid. 
Cohen and Cohen examined the impact of note-taking in 60 trials (a mixture of civil and criminal) as part of a wider jury reform project undertaken in ten courts in Tennessee in 2001-2002. Questionnaires were completed by 54 judges, 103 lawyers and 669 jurors.  Jurors reported "overwhelmingly" that the notes were helpful to them during deliberations and "most" jurors did not find note-taking distracting.  Judges were unanimous in stating that the note-taking "did not appear to distract jurors from the trial" and were generally enthusiastic about the practice.  The lawyers were also "virtually in agreement" that the note-taking was helpful and not disruptive.  As a result of the project, the Tennessee Rules of Civil and Criminal Procedure were amended to specifically authorise jurors to take notes and to use them during deliberations. 
There are a number of other studies involving real jurors that have collected information about juror experience of note-taking as part of a broader study. In research undertaken for the New Zealand Law Commission, researchers were given permission in 48 jury trials (all involving criminal cases) to interview jurors after the conclusion of the trial. Of the 234 jurors who reported taking notes, 83 per cent stated that they referred to them during deliberations and 94 per cent stated that they found them useful as a memory aid.  However, respondents also reported a number of negative issues. Some stated that they felt restricted by their lack of experience of note-taking and that they did not know how to go about the task. Others stated that jurors with sparse or no notes tended to contribute less to deliberations. 
Juror experience of note-taking also arose in a study undertaken by Chesterman et al,  who interviewed 175 jurors who had sat on real criminal trials in New South Wales between 1997 and 2000. Only a small minority of the jurors commented on note-taking (which was not the primary focus of the study), but two jurors from separate trials commented that inconsistencies between jurors' notes became a source of disagreement during deliberations and that jurors who took notes on the same trial sometimes had different accounts of the evidence.  One juror recounted that a fellow juror not only took 17 exercise books of notes, but also wanted to work through every book during deliberations. 
Finally, Matthews et al interviewed 361 real jurors who had sat on trials in six English criminal courts as part of a wider project to examine juror satisfaction. The researchers note that a small number of jurors commented that it was difficult to keep up with evidence while taking notes and some jurors stated that they "had a problem knowing what to write down, and how much to write down". 
2.4 Audio-visual and Digital Presentation Methods
There is some evidence that visual images can improve juror memory and comprehension for both evidence and legal directions, although none of the relevant studies have used especially realistic research methods. There is also evidence that the use of visual images by the prosecution can bias jurors towards conviction, so caution must be taken in this respect.
The evidence base in terms of the use of digital technology - specifically, providing information to jurors on tablet computers - is very limited and no firm conclusions can yet be drawn about whether this has any beneficial or negative effects.
This section of the review examines the empirical evidence on audio-visual presentation methods and the use of digital technology to convey information to jurors. The focus is on studies that have evaluated the effect of presentation methods on memory or understanding - whether for evidence or for legal directions. There is some overlap here with studies that have evaluated the use of written directions and routes to verdict - both of which might be regarded as visual presentation methods - but these are considered separately.  The focus of this section is on images that are used to accompany either legal directions or scientific evidence, and on the use of digital technology such as tablets.
2.4.1 Audio-visual Presentation Methods
It has been suggested that jurors are more likely to retain information if it is accompanied by visual images and that it may also assist with comprehension of complex concepts.  This might be because jurors are more likely to pay attention to an interesting presentation format, but it might also be because pictures can facilitate learning when they accompany text. The evidence base in terms of the effect of audio-visual presentation methods on juries is, however, a small one and it has found mixed results.
Two studies have examined the use of visual images in the presentation of scientific evidence. Goodman-Delahunty and Hewson tested the impact of an expert tutorial on DNA evidence, which was presented to 470 community mock jurors either verbally or accompanied by multi-media images (3D images that modelled the structure of DNA, alongside graphics that illustrated key points). A further group did not receive the expert tutorial. The comprehension scores of the two groups who received the tutorial were significantly higher than the group that did not, with the group who saw the tutorial accompanied by the visual images achieving the highest comprehension scores (a marginally significant difference compared to the group who only heard the tutorial verbally).  The research methods used, however, were not the most realistic, meaning that the results must be treated with caution. The simulated trial involved a 35 minute audio-tape accompanied by either photos of the speaker or (in the multi-media condition) photos of the speaker and visual images.
Positive results were also found by Morell, in a study that involved 126 student mock jurors who watched videotapes (all identical in their length and verbal content) in which expert testimony was accompanied by (a) no visual aids, (b) diagrams, (c) computer animation, or (d) diagrams and computer animation. Each version of the video lasted for six minutes and after watching it participants were asked to write freely all they could remember about what was said. Morell reports that participants in the two computer animation groups recalled information significantly more accurately and in more detail than participants who did not view the computer animation. The diagram alone made no significant difference.  The length of the testimony and the fact that it was not presented in the context of a trial mean that the experiment lacks external validity. That said, given that the computer animation had a beneficial effect on memory even for a short six minute video, it is possible that in the context of a real trial (which would be far longer and where the challenges of memory are likely to be much greater) the effect would have been even more pronounced.
Two studies have examined the use of visual images in the presentation of legal directions. Brewer et al tested the impact of computer animations on mock juror comprehension of a self-defence direction. Their 189 mock jurors were a mixture of law students and adults with no legal expertise. Participants were given a case summary (in both written and audio form) which was followed by approximately ten minutes of legal directions on the law of self-defence delivered by a serving trial judge. These were either delivered in audio only or were accompanied by computer animations in colour which highlighted key words and used human figures (in the style of crash test dummies) to illustrate concepts. Mock jurors then completed a questionnaire which measured comprehension via a mixture of multiple choice questions, paraphrasing and applying law to scenarios. The researchers found that while the audio-visual presentation made little difference to the comprehension scores of the law students, it did significantly improve comprehension in the community sample, bringing their comprehension scores up to the same level as the student sample.  The experiment was not a particularly realistic one: in a real trial jurors would be able to see the trial judge delivering directions rather than listening to an audio-tape and would also have heard from witnesses. It does, however, provide some limited basis on which to conclude that audio-visual presentation may improve comprehension for non-legal experts.
Park and Feigenson tested the impact of prosecution and defence lawyers using PowerPoint. They conducted three experiments, all of which used undergraduate psychology students as mock jurors (a total of 567 students) in a civil case for racial discrimination. The mock jurors watched a 54 minute video of actors playing the part of plaintiff and defence lawyers making a statement of their case which involved either none, one or both of the parties using a PowerPoint presentation to accompany their speech. The researchers found that the use of PowerPoint resulted in a significant improvement in memory of the speechs' content.  The effect was strongest when only one party used it.  As the authors themselves recognise, however, the study does have a number of limitations, most notably that the manner in which the information was presented greatly oversimplified the way in which information would be presented in a real trial. 
The use of visual images does have some drawbacks. Brewer et al note that animations, if they are not sufficiently neutral, may prejudice jurors or adversely affect the solemnity of the proceedings  There is some evidence of this from Park and Feigenson's research, where the use of PowerPoint significantly affected liability judgments in favour of the party using it. This might have been linked to improved juror memory for the evidence where PowerPoint was used. However, this may not have been the only factor at work, as lawyers who used PowerPoint were also rated significantly more persuasive, competent and well prepared by the mock jurors compared to those who did not use PowerPoint.  One mock jury study also found that giving jurors graphic photographs of crime scenes significantly increased the likelihood of conviction;  while another found that the use of a 3D computer aided animation of a crime scene in a mock terrorism case had the same effect. 
2.4.2 Digital Presentation Methods
The easiest way in which information might be presented to jurors in digital form is to give jurors a tablet computer, such as an iPad. The information that could be presented to jurors in digital form might include a full or partial case transcript, copies of exhibits or even legal directions. The main reason for doing so is that it might help jurors to remember the evidence. It might also make for shorter deliberation times, especially where the case is a complex one. Information is potentially easier to navigate on a tablet than in hard copy and it has become increasingly commonplace in society for information to be consumed in this form. There is, however, the risk that some jurors - particularly older jurors - may be disadvantaged if they are not experienced in using the technology. 
There have been remarkably few studies that have examined the effect on memory or understanding of providing jurors with information in digital form. There is one ongoing study, funded by the Canadian Social Sciences and Humanities Research Council, that is examining the impact of providing information to jurors via a tablet computer and this has reported some preliminary results.  The study involved 152 community jurors recruited from the jury pool in a Melbourne criminal court, who watched a 60 minute trial video involving a terrorism charge before deliberating in groups of between eight and twelve.  A package of materials was prepared for the jurors, consisting of images of the prosecution's exhibits (such as political leaflets, screenshots of web pages and pictures of chemicals allegedly found at the defendant's home) and a partial transcript of evidence. One group of jurors received the materials in a ring binder, the other on a tablet computer.
The researchers concluded that the use of tablets shortened deliberation time - the groups with iPads moved to discussion more quickly, and were able to locate materials easily, whereas those with the paper copies spent longer flipping through the pages of their ring binders.  They did not find that there were any major differences in the content of the deliberations or the nature of the interactions between jury members in the two conditions.  No results have yet been reported, however, on whether the tablets led to any improvements in memory or comprehension, nor in terms of how easy the jurors found it to use the tablets or whether jurors who were more experienced in technology dominated the discussions in any way. An earlier pilot study conducted by some of the same researchers did find, however, that juries who used tablets were significantly more likely to convict than juries who did not, which the researchers suggest might be because in their experiment it was only the prosecution that presented visual evidence. 
2.5 Juror Questions
There is only a limited evidence base on allowing jurors to ask questions during the trial. The only mock jury study to assess the impact of this on comprehension found that allowing jurors to question expert witnesses about DNA evidence did not lead to any improvement in understanding. There is evidence from this experiment and from field studies (albeit all conducted some time ago in the US) that jurors appreciate being able to ask questions, but that when they are given the opportunity to do so, they do not tend to ask many questions in practice, except in a minority of trials. There is some evidence that jurors, even when they are told that they can ask questions, feel too intimidated to do so.
This section of the review examines the empirical evidence on allowing jurors to ask questions during the trial. The focus here is specifically on allowing/encouraging questions during the course of the trial, as opposed to once the jury has commenced its deliberations. No empirical evidence exists on the latter - and only a limited amount on the former.
It has been suggested that jurors who are encouraged to ask their own questions - either of the witnesses or the trial judge - may be more active, focused and involved in the trial, decreasing the likelihood of them switching off through a lack of interest. It may help to resolve questions jurors have about the facts or the law, or uncover important evidence and issues that were left out by accident or by design. It may also improve juror satisfaction with the experience of jury service if jurors feel they were given the opportunity to participate.  There are a number of obvious risks, however. If jurors were to ask a lot of questions, this could be time consuming and might interrupt the flow of proceedings, making it harder for other jurors to retain important information. Juror questions may also be inappropriate or focus on irrelevant or otherwise inadmissible matters. They might also attempt to seek information that has been deliberately omitted by the Crown or the defence for tactical reasons.  These last two difficulties in particular could be addressed through trial judge intervention, but there is a risk that jurors might draw inappropriate inferences if they are not allowed to ask a particular question. 
There has been some empirical research that has examined the impact of permitting jurors to ask questions during the trial, but with one exception this has taken the form of field studies rather than mock jury experiments.
2.5.1 Mock Jury Studies
The only mock jury study to examine juror questioning is that of Dann et al. In their experiment, 20 juries were permitted to ask questions and chose to do so in 14 of these trials.  Of those jurors who were in the groups permitted to ask questions, 69 per cent expressed support for being permitted to do so, although question asking received the lowest level of support from jurors of the four innovations that were tested.  It is worth bearing in mind, though, that the study was looking specifically at DNA evidence and that all of the questions asked by jurors related to scientific evidence given by two opposing expert witnesses.  The study did not, however, isolate the impact of question asking - the group who were permitted to ask questions were also asked to take notes. That said, this group did not perform any better in terms of its comprehension of the evidence than the group who were permitted neither of these things,  which suggests that the ability to ask questions does not necessarily lead to improved comprehension of scientific evidence. Dann et al's study was a relatively realistic one which included jury deliberation. Comprehension was tested both pre- and post-deliberation and in neither case did the ability to ask questions (combined with note-taking) improve it.
2.5.2 Field Studies
Field experiments have focused on two issues - the experience of juror questioning in practice (especially how often is it used) and views of jurors and other trial participants on the usefulness of being able to ask questions during the trial.
The evidence on the first is that jurors tend not to ask a lot of questions, even when they are told they can do so. Heuer and Penrod found in their two field studies (examining both criminal and civil cases)  that even where jurors are specifically told that they can ask questions and are told of the procedure for doing so (here by submitting questions to the trial judge), the number of questions actually asked is fairly small. In their first study, across 33 trials there were a total of 88 questions (an average of 2.7 per trial).  Of these, 15 were objected to by lawyers for either or both parties to the case. In every instance that either lawyer objected to a juror question the objection was sustained.  In the second of their studies, questions were permitted in 71 trials and there was at least one question submitted in 51 of these. The number of questions submitted by jurors ranged from zero to 65, and the average number of questions asked per hour of trial time ranged from zero to 3.6. The number of questions asked per hour was two or greater in only three of the 71 trials. 
Sand and Reiss found that in 26 trials (evenly divided between civil and criminal cases) where questions were permitted, the number of questions asked was generally very small (in eight cases, it was zero, in 16 cases it was one, in two cases it was two), but that two trials involved a very high number of questions (40 in one case, 56 in the other).  Cohen and Cohen found similarly - in the majority of the 60 trials (both civil and criminal) in their field experiment jurors asked only one or two questions, but in a very small minority of trials a much higher number were asked (the most was 37).  Unlike Heuer and Penrod, Cohen and Cohen found that the vast majority of the questions submitted to the judge were allowed.  Finally, Mott collected data on juror questions asked in a sample of 130 real civil and criminal trials in the US between 1995 and 2002. She found that the average (mean) number of questions asked per case was 16, but that there were a small minority of trials where a lot of questions were asked (in one trial there were 130), thus a better guide may be the median figure of seven.  Jurors in criminal trials asked twice as many questions as jurors in civil trials.  Mott also collected data on the subject matter of the questions. In criminal trials, the most common questions that jurors submitted were questions to eyewitnesses about what they had seen and how well they could see it, questions to witnesses about their motives, and questions about common practices ( e.g. would it be normal for someone to do X?). 
In all of these studies, jurors were not permitted to ask questions freely - they had to be submitted to the judge and questions could be objected to by the lawyers for either side. It may be that this discouraged questions that might otherwise have been asked. There is evidence from other studies that some jurors feel too intimidated to ask questions. Matthews et al, for example, found that a significant percentage of their sample of real jurors (67 per cent) reported that they wanted to ask questions at certain points during the trial in order to clarify the evidence, or to request more information, but that only half of those wishing to ask questions felt that they could.  The authors note that: 
In some courts, jurors felt they were actively discouraged from asking questions during the trial and consequently said nothing in the hope that the particular point would be clarified in the course of proceedings and that other relevant information would emerge before the completion of the trial. A number of jurors felt 'it wasn't their place to ask questions' and it would be embarrassing, particularly since the question might be read out in court.
In terms of how useful jurors found the ability to ask questions of witnesses, views have tended to be reasonably positive about this. In Heuer and Penrod's second study (their first did not ask jurors about this), jurors were moderately positive.  Cohen and Cohen's jurors were a bit more enthusiastic, with 70 per cent of jurors reporting that being able to ask a question was "very useful" (only six per cent stated that it was "not useful").  In terms of the legal professionals' views, Heuer and Penrod found that trial judges were generally supportive of the procedure and did not see it as harmful to the trial process,  although it should be said that participation in the project was optional and those trial judges who did not take part might have held different views. In all three of the field studies where question asking was tested, the lawyers involved held mixed views as to its helpfulness. 
2.6 Plain Language Directions
There is a substantial body of evidence from mock juror studies on the use of plain language directions and it points overwhelmingly in the same direction - that simplified directions are highly effective in improving juror comprehension of the legal tests they are asked to apply. Improvements in comprehension have been shown across different types of direction, with both student and community mock jurors, with deliberating and non-deliberating juries and with civil and criminal trials. Only two linked studies have failed to show any improvement as a result of plain language directions and that is most likely attributed to the fact that the directions were not re-written very well.
Care does need to be taken in re-writing directions to ensure that they retain the correct legal meaning and most of the studies that demonstrated improvements in comprehension worked with legal advisers to ensure that their re-written directions did this (or used officially sanctioned directions that had been re-written as part of a jurisdiction-wide simplification project).
Questions might be raised about how relevant this body of evidence is in the Scottish context - almost all of the studies concerned stem from the US where directions tend to be longer and more complex. That said, jurors clearly benefit from directions written in plain language and some jurisdictions have revised their standard judicial instructions with this in mind. 
This section of the review examines the empirical evidence on the effectiveness of plain language directions - directions that simplify language, grammar and structure so that they are more easily comprehensible to jurors. The obvious advantage of plain language directions is that they might improve juror comprehension of legal concepts, something that a number of studies have shown to be problematic,  which in turn might lead to improved confidence in jury verdicts. There are no obvious disadvantages, but there might be a challenge in ensuring that simplified directions retain the correct legal meaning. It might also be seen as unnecessary if directions are already reasonably straightforward - the majority of research on juror comprehension has been undertaken in the US context where directions tend to be relatively long and complex. There is no empirical evidence on the extent to which jurors understand legal directions in the Scottish context. As noted in section 1.2, jury directions in Scotland are understood to be generally shorter than those in other jurisdictions, which may aid, but cannot be assumed to ensure, comprehension.
There is a considerable body of research - most of it in the US context - that has evaluated the effect of using simplified language in jury directions. Because of the obvious difficulties involved in conducting a field experiment in this area - or at least one where some juries are directed differently to others - all of the studies have involved mock juries.
2.6.1 Mock Jury Studies
Three pioneering studies of plain language directions were undertaken in the US in the late 1970s and early 1980s. Charrow and Charrow re-wrote 14 civil jury directions on subjects including causation, witness credibility, expert evidence and negligence, simplifying them by using techniques such as removing technical words, multiple negatives and embedded phrases.  They tested these on 48 jurors who had been cited but not selected to sit on trials - half of the jurors were tested on their comprehension of the original directions, the other half on the re-written directions. They found that comprehension improved by between 35 and 41 per cent (depending on the measure of comprehension used). 
Although Charrow and Charrow's experiment used real jurors, it tested short single issue directions in isolation, rather than full directions in the context of a real trial. Elwork et al, however, did both of these things and achieved similar results. They re-wrote directions used in Michigan in a civil negligence trial, working on similar principles to Charrow and Charrow.  Their experiment involved 154 community jurors who watched a videotaped trial (based on a real case, but re-created using actors) for civil liability following a car accident. They took various steps to heighten the realism of the experimental setting, including conducting the experiment in a building that resembled a courtroom and showing the trial on a large TV screen in three one hour segments with a break of ten minutes between each. The jurors then heard either the original directions or the re-written directions. The mock jurors who received the rewritten instructions scored significantly higher in a twelve question comprehension test than those who received the original instructions. 
The Elwork et al study, realistic as it was, only tested the impact of re-writing civil directions. A follow up study by Elwork et al, however, involved criminal directions.  This study used videotapes of two mock criminal trials - one involving relatively complex issues of law (attempted murder, several alternative charges and a plea of insanity), the other one more straightforward (burglary). The researchers re-wrote both sets of directions to reduce their complexity (the directions in the complex trial were re-written twice, each version being a simplification of the previous one).  Their study participants were 314 community mock jurors who watched the trial video in a courtroom setting and then watched a video of one of the researchers reading either the original or re-written directions. Each participant was asked a series of short answer questions designed to test comprehension. The researchers found that simplified directions resulted in substantial improvements in juror comprehension. For the instructions in the complex trial, the average (mean) percentage of correct answers per juror was 51 per cent. After the first rewriting, this average increased to 66 per cent and after the second rewriting effort it increased to 80 per cent. Each of these increases was statistically significant.  Similar improvements were found in relation to the instructions in the simple trial. 
None of the studies discussed thus far included deliberation in their research design. Elwork et al did undertake a small scale follow up study, in which they repeated the experiment using only the complex criminal trial (comparing the original directions with the second set of re-written directions) with 45 community jurors who deliberated in groups of six. As before, they found significant improvements in comprehension scores of the individual jurors in the group who heard the re-written instructions. 
The third of the pioneering US studies - that of Severance and Loftus - tested re-written criminal instructions and included deliberation as a component of the experiment.  Severance and Loftus tested the impact of simplified directions in four areas (the meaning of reasonable doubt, intent, use of previous conviction evidence and an introductory instruction about jurors' general duties) on 216 undergraduate psychology students. The re-written directions were tested in the context of a trial simulation - a one hour long videotaped burglary trial filmed in a real court, with a real judge and the remaining roles played by actors. Jurors watched the trial and then heard either the original instructions or the re-written instructions before deliberating in groups of six for 30 minutes. Each juror then individually completed two comprehension tests - one in a multiple choice format and one a test of application where they were asked to apply the directions to a novel factual scenario. The researchers found that the re-written instructions led to significantly better performance on the application test compared to the original instruction.  However, it did not lead to a statistically significant increase in overall comprehension scores on the multiple choice test. 
The main limitation of Severance and Loftus' study was that it used student jurors. They went on, however, to repeat their experiment using the same materials with 306 community mock jurors - 162 who had been cited to court but not selected and 144 ex-jurors randomly selected from old jury lists.  The revised directions improved comprehension scores (both in the application test and the multiple choice test), but the only statistically significant improvement occurred in relation to the previous conviction direction.  Comprehension was also tested using a paraphrase test whereby jurors were asked to explain in their own words what each concept meant and the authors found that correct answers outnumbered incorrect answers when the revised directions were given but not when the original directions were given.  The results, while positive, were not as overwhelmingly so as those of Charrow and Charrow and Elwork et al. This may, however, be because some of their re-written instructions, while improvements on the originals, were still relatively complex.  A further re-write might have resulted in greater improvements in comprehension.
Aside from these pioneering studies, the area where most empirical research has been undertaken is - again in the US context - in relation to the comprehension of death penalty sentencing instructions, in which jurors are asked to weigh aggravating and mitigating factors in order to determine whether the death penalty should be imposed. Four studies in the US context have focussed on juror comprehension of death penalty instructions - and all have shown comprehension improvements from the use of plain language directions.  Improvements in comprehension have also been demonstrated in the context of directions on eyewitness identification evidence  and on circumstantial evidence. 
Although the overwhelming majority of studies point to a positive effect of plain language directions, this has not been a universal finding. Wiener et al, in two studies involving respectively 91 and 92 community mock jurors, tested the impact of a simplified death penalty direction, but found no improvement in comprehension.  This may, however, have been because the authors did not re-write the instruction very effectively. Their re-written instruction was, as English and Sales put it, "a single, lengthy (52 word) compound sentence that uses several words and phrases that jurors may not be familiar with". 
There is a considerable body of evidence, both from mock jury studies and field studies, to suggest that pre-instructing jurors on the substantive issues in the case improves comprehension of legal issues. The only study that did not find any improvement in comprehension as a result of pre-instruction did not use particularly realistic trial simulation methods, so little weight should be attached to it. Pre-instruction has been found to be most beneficial when combined with note-taking.
There is also some evidence that pre-instruction improves juror memory of the evidence that was led during the trial, although here the studies are more equivocal. The beneficial effects have, however, tended to be found in the most realistic experiments, so these should be given more weight. Here too, pre-instruction has been found to be most effective in improving memory when combined with juror note-taking.
There is little evidence for any of the possible downsides of pre-instruction. There is no evidence that it causes jurors to decide cases prematurely and, in field experiments undertaken in the US, judges, lawyers and jurors have expressed satisfaction with the procedure.
This section examines the empirical evidence relating to pre-instruction of jurors. Pre-instruction refers to the practice of giving jurors case specific direction on the relevant principles of substantive law prior to the evidence being led.
It has been suggested that pre-instructing jurors may have a number of advantages. It might improve recall of evidence, by assisting jurors in identifying relevant evidence as it is led and providing a framework for organising this. It might improve recall (and possibly even comprehension) of judicial directions, as jurors will hear them twice. It might also reduce juror bias, by identifying at the outset what the relevant issues in the trial are likely to be and reducing the possibility that jurors rely instead on inaccurate beliefs about what the law is or personal biases about the case or the characteristics of the accused. 
It has also been suggested that there may be disadvantages. There is a fear that pre-instruction may overload jurors with too much information at the beginning of the trial. It might be regarded as unnecessary and a waste of valuable court time to give substantive directions twice. It might be difficult for jurors to make sense of legal directions before the evidence has been led. There may be a danger that pre-instruction encourages jurors to reach a verdict prematurely, and to ignore evidence that does not fit in with this hypothesis (although the opposite has also been proposed - that by giving jurors a framework for identifying the relevant issues, it discourages speculation about irrelevant matters). Finally, it may also be impractical to give pre-directions if the trial judge does not know what issues are likely to arise in the case. 
Pre-instruction has been tested in both mock jury studies and field experiments.
2.7.1 Mock Jury Studies
A number of mock jury studies have tested the impact of pre-instruction. Some of these test pre-instruction and post-instruction only as alternatives, others have tested the condition where jurors are both pre-instructed and post-instructed. The latter is preferable in terms of realism - it is difficult to conceive of a situation in reality where jurors would only be directed on the law prior to hearing the evidence. Aside from anything else, unexpected issues may arise during the course of the trial that require directions to be given after the evidence has been led. The studies have focused on the effect of pre-instruction on three variables: memory (for both the facts and the law), comprehension and the point at which jurors decide on their verdict.
In terms of the impact on memory, the evidence is mixed. The first mock jury study to look specifically at pre-instruction was undertaken by Smith and did not demonstrate any benefit of pre-instruction in terms of memory for the facts.  Smith's experiment involved 125 participants (a mixture of community and undergraduate psychology student jurors) who watched a 2.5 hour videotape of a homicide trial. They heard directions (including on the substantive law specific to the case) before the evidence, after the evidence, not at all or both. The jurors did not deliberate. Recall was tested via 20 multiple choice questions. Smith speculates, however, that the reason why pre-instruction did not lead to any improvement in memory for the evidence might have been because the jurors did not understand the substantive law well enough to use it as a framework for processing incoming evidence (as discussed below, comprehension of the law was generally low).  Cruse and Brown likewise found no relationship between timing of instructions and recall for the evidence,  but their experiment (which relied solely on a trial booklet) was so unrealistic that little weight can be attributed to this.
Other studies have demonstrated a relationship between pre-instruction and recall. ForsterLee and Horowitz found that pre-instructed jurors performed significantly better than post-instructed jurors in terms of their memory for the evidence.  Their experiment was a relatively realistic one, involving 120 community mock jurors, who watched a 70 minute video of a civil trial which was filmed using a mixture of professional actors and legal professionals (although the jurors did not deliberate). The researchers also found that the beneficial effect of pre-instruction on memory was greatest when it was combined with juror note-taking.  These findings replicate the results of an earlier, less realistic study undertaken by the same authors.  Fitzgerald also found that pre-instructed jurors performed significantly better on a number of different tests designed to measure their recollection for the evidence.  His experiment was also relatively realistic, involving 124 mock jurors (a mixture of college students and community jurors), who watched a two hour civil trial video (with a ten minute break in the middle), although his jurors did not deliberate. Fitzgerald found that pre-instruction was especially beneficial for the memory of the older jurors who participated in the experiment (specifically those aged 55 to 75). 
Neither ForsterLee and Horowitz nor Fitzgerald tested the effect of pre- and post-instruction. In both experiments jurors were either instructed before the evidence was led or afterwards, not both (as Smith did). It may be therefore that memory for the facts was better not because of the pre-instruction but because juror recollections were tested immediately after the evidence had been led, rather than after a delay caused by hearing the post-instruction directions. The likelihood of this being significant would have depended on the length of the directions, but neither study provides information about this.
Finally, one study has examined the impact of pre-instruction not on memory for the evidence but on memory for judicial directions. Ramirez et al found that hearing a direction about the evaluation of eyewitness identification evidence twice - before and after the evidence had been led - significantly improved mock juror memory for the content of that direction.  It does have to be said though that the direction was relatively short  and that if memory for directions is a concern then the more obvious way to address this is by providing a written copy. 
Some of the mock jury studies discussed above also tested the effect of pre-instruction on comprehension of the legal issues in the case.  In her study, Smith tested comprehension via multiple methods, including multiple choice questions; ten true/false questions about the issues in the case and ten scenarios where the jurors had to apply the law to novel facts (determining whether the accused in each scenario was guilty of first degree murder, second degree murder, manslaughter or should be found not guilty). Participants' performance in all of the comprehension tests was poor, but those jurors who had been both pre- and post-instructed performed significantly better than jurors who had only been post-instructed. Pre-instruction alone had no effect.  Similarly, ForsterLee and Horowitz found that pre-instructed jurors performed significantly better than post-instructed jurors in distinguishing between plaintiffs with differentially worthy cases,  although the study did not test the effect of both pre- and post-instruction. They also found that note-taking enhanced this ability further - the best performance was in the group who were pre-instructed and took notes.  By contrast, Bourgeois et al found no relationship between pre-instruction and mock jurors' ability to differentiate between plaintiffs,  but their use of an audiotaped trial rather than a video reconstruction means that lesser weight should be attached to this study.
Finally, two mock jury studies have examined the hypothesis that pre-instruction might cause jurors to decide cases prematurely (one of the suggested disadvantages of pre-instruction). Neither found any evidence that this was the case. Smith asked her mock jurors midway through the trial whether or not they had yet formed a view about their verdict and found that jurors who had been pre-instructed were significantly more likely to defer their verdict decisions compared to those who were only instructed after trial.  Bourgeois et al, who asked mock jurors half way through the trial to indicate which side they favoured or whether they were still undecided, found that pre-instructed jurors were no more likely to have decided the case mid-trial than those who were not pre-instructed. 
2.7.2 Field Studies
All three of the major US field studies examined the use of pre-instruction. Heuer and Penrod's study was the only one to objectively measure the impact of pre-instruction on comprehension, with the result that jurors who sat on criminal trials who were pre-instructed on substantive law showed a marginally significant improvement in performance in multiple choice tests of comprehension, compared to those who were not pre-instructed.  Given, however, that the jurors were allowed to take the questionnaire home and mail it back, rather than completing it immediately after the trial had concluded,  this probably under-estimates the beneficial effect of pre-instruction (as some of the jurors may have struggled with memory by the time they completed the questionnaire).
Heuer and Penrod did not test memory for the evidence directly, but did ask jurors how confident they were that they could remember the evidence that had been led. Pre-instruction made no difference to their degree of confidence,  but the use of a self-reporting measure means that this finding is of limited value. Jurors who were pre-instructed indicated that they found this helpful  and the judges who participated in the experiment were also positive about it, and indicated that they did not find pre-instruction to be disruptive of the trial process.  Sand and Reiss found likewise, although the number of criminal trials in their sample where pre-instruction was used was very small (there were only four of them) and three involved the same judge, so little reliance can be placed on this.  Cohen and Cohen found that pre-instructed jurors reported satisfaction with the procedure, and that it helped their understanding of what happened during the trial.  Lawyers' questionnaires were virtually unanimous in indicating support for pre-instruction  and as a result new rules of criminal procedure were implemented in Tennessee that require the court to pre-instruct jurors on a number of matters, including the legal principles involved in the case. 
2.8 Written Directions
There is no real debate over the question of whether written directions improve juror memory - they clearly do, as jurors no longer have to rely on their own recall of what the trial judge has said. There is, however, a substantial body of evidence that written directions also improve juror comprehension of the law. The vast majority of studies have demonstrated improvements in comprehension from written directions and most of the studies that have not done so have suffered from methodological flaws. Studies of juror deliberations demonstrate that jurors who are given written directions frequently refer to them and use them to correct mistakes of law made by other jurors.
That said, there is a limit to the role that written directions can play in improving comprehension - putting instructions in writing cannot compensate for instructions that are inherently unclear. In a study undertaken by Ede and Goodman-Delahunty, written directions improved the ability of jurors to recall the content of the directions but they did not help jurors to gain a deeper understanding of how the law should be applied.
There are some practical issues that need to be addressed if written directions are to be provided. A copy should be given to each individual juror - otherwise there is a danger that the person with the written directions dominates the discussion.  It cannot be assumed that all jurors will have levels of literacy that would enable them to read a written text, so provision may need to be made in this respect by, for example, the use of recorded verbal or video directions.
This section of the review examines the empirical evidence relating to written directions. It has been argued that providing jurors with written directions can lead to a number of benefits, including improvements in memory; improvements in comprehension; better quality deliberations (where more time is spent applying the law); reduced deliberation time (as juries spend less time trying to recall the instructions and any disputes about their content are quickly and easily resolved); and improvements in juror confidence and satisfaction. 
In relation to the first of these, Semmler and Brewer make the point that we are asking an awful lot of jurors to retain the information provided by the trial judge - even the simplest charge is likely to run to several pages of directions - and at least some barriers to comprehension may simply stem from limitations in working memory.  The possible objections to written directions include the fear that they might actually increase deliberation time (because jurors become involved in time consuming arguments over how to interpret them); they might be time consuming and burdensome for trial judges to produce; or they assume a level of juror literacy that might not be borne out in practice. 
Written directions have been tested in both mock jury and field experiments. 
2.8.1 Mock Jury Studies
The number of mock jury studies that have tested the impact of written directions is relatively small. None of the studies have directly tested the impact of written directions on memory, but that is probably because it is self-evident that having a written copy of directions would help jurors to remember the relevant legal tests. The focus of the studies has instead been on comprehension. Given that memory for the directions is a necessary step to understanding them, if written directions improve comprehension then it is probably safe to conclude that they also improve memory.
A number of studies have tested the impact of written directions on comprehension. Three of these found that comprehension improved as a result.  Thomas' study is the most extensive and realistic of these.  It was conducted with 243 jurors who had been cited to court in Winchester but not selected to sit on trials. The jurors watched a 60 minute mock trial video before deliberating in groups of between nine and twelve for up to 20 minutes.  She found that the proportion of jurors who were able to answer correctly two questions aimed at testing understanding of a self-defence direction rose from 31 per cent to 48 per cent when the direction was provided in writing. 
In the second study, Semmler and Brewer found that a written summary of the trial judge's directions significantly improved comprehension of the law of self-defence in a group of 234 community mock jurors,  although it does have to be said that the experiment was not very realistic, involving as it did a 700 word case summary read aloud to the participants, followed by an audio tape of a real trial judge reading a 15 minute self-defence direction.
The third study was undertaken by Ede and Goodman-Delahunty and tested the impact of written directions across a range of legal concepts. It involved 183 community mock jurors who watched an hour long trial video based on a real case and performed by professional actors.  After viewing the trial, the jurors were asked a number of true/false questions about the procedural and substantive law in the case, some of which were relatively simple, while others required a deeper understanding and application of the law. They did not deliberate. Jurors who received written directions performed significantly better on the simple comprehension questions than jurors who were given no written aids. Perhaps more surprisingly, they also performed significantly better than jurors who had been given a route to verdict (the other innovation tested in the experiment).  On the complex comprehension questions, however, the written directions made no significant difference to comprehension.  The researchers suggest that the effect of the written directions on simple comprehension may have been solely due to the multiple exposure to the text of the legal tests - in other words, it improved juror memory, which allowed jurors to easily recognise correct statements of the law. The fact that this improvement did not translate to the complex comprehension questions suggests that written directions, while acting as a successful memory aid, did not help jurors obtain a deeper understanding of the law.  The researchers also note that the measure of comprehension they used - true/false questions - was not ideal and might have led jurors to answer correctly simply by guessing. 
By contrast, one mock jury study found that written directions did not improve comprehension. This was undertaken by Rose and Ogloff,  who compared the performance of 39 undergraduate psychology students who were given written directions on conspiracy with 33 students who watched the same directions delivered by a trial judge on a video. However, the sample is very small and their experiment was so far removed from the reality of a criminal trial - it used a written fact summary rather than a full trial video and the group with the written directions did not also hear them orally - that little weight can be attached to it. The inherent complexity of the conspiracy direction may also have been an issue. It ran to fifteen pages and comprehension of it was generally low, regardless of the medium by which it was delivered. 
2.8.2 Field Studies
The impact of written directions has been tested in a number of field studies. Heuer and Penrod's is the most extensive of these. Here, 29 judges in Wisconsin randomly assigned their trials (34 civil and 33 criminal) so that some juries received written instructions and some did not. After the trial was over the jurors were asked to complete questionnaires aimed at testing their understanding of the instructions they received (specifically those relating to the standard and burden of proof, the presumption of innocence, the evaluation of testimony and exhibits and procedural issues such as the allocation of responsibility for findings of law and fact). The researchers also canvassed the views of the judges involved. They found no evidence of any of the potential drawbacks of written directions. Written directions made no significant difference to deliberation time  and the judges involved reported that providing them was not burdensome or disruptive.  In terms of the possible advantages, jurors reported that the written directions were very helpful in settling any disputes that did arise. The researchers did not, however, find that the written directions led to any improvement in the comprehension of legal concepts.  Despite this, the researchers concluded that their results presented "a compelling case" for written instructions and that while they might not have all the advantages claimed of them they did have some clear benefits and they had no harmful consequences. 
As noted above, the failure of Heuer and Penrod's study to demonstrate any effect of written directions on comprehension may have been due to the fact that many jurors did not complete the comprehension questionnaire until several days after the end of the trial,  a point the researchers themselves acknowledge.  Another field study that found that written directions did not improve comprehension also suffered from the same methodological flaw (as well as failing to separate out the effect of written and taped directions).  Kramer and Koening, however, in a study of 600 jurors who had sat on real criminal cases in Michigan, and who received questionnaires immediately after they finished serving on a trial, found that written directions did significantly improve comprehension. 
These are the only field experiments to have objectively compared the comprehension of jurors who received written directions to those who did not. Trimboli reports on a survey of 1,225 real jurors, conducted in New South Wales, in which jurors who received written directions were significantly more likely to report that they understood everything or nearly everything the trial judge said about the legal issues in the case. 
Other field studies have examined juror attitudes towards written directions and have found that jurors generally found them helpful and judges were also positive about the experience of providing them.  Surveys of real jurors who did not receive written directions have reported that they would have found it helpful to have done so.  In Young et al's study for the New Zealand Law Commission, a number of jurors suggested that if they had had a written summary of the law, deliberation time would have been reduced: in one case jurors reported that they spent time collectively putting together their notes to work out what the key elements of the offence were; and in another they had to ask the judge a question to clarify the law, which they believed would have been unnecessary if they had received a written aid. 
A final study worth noting is that of Diamond et al, in which the researchers were permitted to record real jury deliberations in civil trials in Arizona.  Deliberations were recorded in 50 civil cases between 1998 and 2001 and all juries had a written copy of the judge's directions, as the law in Arizona mandates this.  The researchers found that frequent reference was made to the written directions during the course of deliberations. In 46 of the 50 juries, at least one juror read a direction aloud to the group, and in nearly half of the trials (46 per cent), at least half of the jurors each read at least one direction aloud.  This, the researchers conclude, demonstrates that where written directions are provided, jurors do consult them as a source of reference.
Diamond et al also found that the majority of references (79 per cent) made to the law by the jurors were accurate, although there were still some errors made.  In Ellsworth's study, however, which also recorded jury deliberations (albeit in a mock jury study), the jurors did not have written directions and the number of legal errors made was far greater (at around half of all statements about the law).  It is also notable that in Diamond et al's study, around half of the legally incorrect statements that were made were corrected by other jurors.  Compare this to Ellsworth's study, where the jurors did not have written directions. Here only 12 per cent of incorrect legal statements were corrected and a substantial number of correct statements about the law were supplanted by incorrect ones.  These comparisons do need to be treated with caution, as the subject matter of the trials and the methodology used by the two studies was not the same, but they do provide some limited evidence that written directions may improve comprehension, especially following deliberation.
2.9 Structured Decision Aids (Routes to Verdict)
Routes to verdict ( RTVs) are a relatively new innovation - compared for example to written directions or note taking - and as such the body of evidence that has empirically tested them is still small. There are, however, two relatively well designed mock jury studies (those of Weiner et al and Ede and Goodman-Delahunty) that have found RTVs to be effective in improving jurors' ability to comprehend legal concepts and, in particular, to apply legal tests appropriately. Further support for the effectiveness of RTVs in improving applied comprehension comes from the study conducted by Kelly, although the fact that she did not isolate the impact of the RTV means that this must be regarded with more caution.
These studies must be weighed against other studies that have not found RTVs to improve comprehension. However, all of the studies where a positive effect was not demonstrated do have methodological weaknesses which mean that limited weight can be attached to them. Further evidence of the effectiveness of RTVs may be forthcoming when a major study being undertaken in Australia and New Zealand reports its findings - preliminary results from this study support the conclusion that RTVs are effective. Field studies also indicate that jurors who have been given RTVs find them useful, particularly in longer, complex trials.
The experimental research has pointed to one practical issue, which is that if juries are not specifically directed about the RTV and it does not relate to the oral directions they hear, then they may not use it. This suggests that RTVs will be most effective when accompanied by a tailored jury direction.
This section considers the empirical evidence relating to routes to verdict and other structured decision aids. There is an initial question of terminology, which is used inconsistently by the studies (and by the courts). What is being considered here is a written aid that provides juries with a series of primarily factual questions that gradually lead them to a legally justified verdict.  This might be presented as a series of questions or in flowchart or other diagrammatic form.  It has sometimes been referred to as a flowchart, a stepped verdict, a fact based direction, a decision tree or a question trail, but for consistency the term route to verdict ( RTV) will be used here.  The focus is on a written RTV as an aid that jurors can take into deliberations,  which may be accompanied by the trial judge talking through the RTV orally and giving any other necessary directions that are not captured in the RTV.
The potential advantage of a RTV is that it greatly simplifies the task that a jury is required to perform. Rather than requiring jurors to understand and apply complex directions, they are asked instead a series of factual questions that focus their attention on the relevant issues and lead them to a legally justified verdict. As such, it should improve the chance of a fair verdict being reached in accordance with the law and minimise problems that might result from the failure of juries to comprehend directions. RTVs also have the potential to improve transparency, with accused persons being able to see more clearly why a particular verdict was reached, which may lead them - and the wider public - to have greater confidence in the criminal justice process. RTVs may also lead to cost savings through shorter deliberation times, as jurors spend less time attempting to understand their task. 
One objection that might be made to RTVs is that they leave no role for jury nullification  - juries choosing not to apply laws that they consider unjust - although depending on one's view of jury nullification that may equally be seen as an advantage. In any case, simply giving a jury a RTV without also requiring it to provide a legally reasoned verdict does not necessarily prevent nullification from occurring. RTVs might also be felt by judges to be difficult and time consuming to prepare,  although that can be ameliorated by the provision of examples to judges for reference and is likely to improve over time as judges gain practical experience.
The body of empirical evidence on RTVs comes primarily from mock jury studies, although field studies do provide some limited observations.
2.9.1 Mock Jury Studies
RTVs have been tested in a number of mock jury studies. In evaluating these studies there are a number of specific methodological considerations. First, the most useful measures of comprehension in this context are those that focus on the application of legal tests. It is this that a RTV is designed to help, and thus measures that test recall or abstract comprehension are not especially useful. Secondly, consideration needs to be given to the RTV document's design. A failure to find any useful effect might simply be because of a poorly designed RTV rather than an indication that RTVs do not assist jurors. Not all of the studies discussed here provide a copy of the RTV that was tested, so it is not always possible to assess this. Third, the studies vary in terms of whether the RTV was accompanied by any tailored direction explaining how it should be used or indeed any oral direction at all - in some experiments it replaced the oral directions rather than supplemented them.
Two early studies are somewhat equivocal in terms of their support for RTVs. Semmler and Brewer tested the effect of a RTV flowchart on 234 community mock jurors who had heard a 700 word audio summary of a self-defence case. The flowchart improved the ability of the jurors to apply the legal test to both the case in question (in which a legally correct application of the law would lead to the conclusion that the defence was not made out) and to novel fact scenarios, compared to jurors who only received standard (oral) directions. However, neither relationship was a statistically significant one, so little can be drawn from this.  It is worth noting, however, that in terms of applying the law to novel factual scenarios, the improvement was only present when the participants also heard a recording of a trial judge giving directions - the flowchart on its own did not result in any improvement.  The recorded directions did not refer to the flowchart - they were standardised across experimental conditions - and the improvement may well have been greater if they were integrated. The study is also limited by the fact that it was not particularly realistic, relying as it did on a trial summary and lacking any element of deliberation.
In a more realistic experiment, involving 665 community mock jurors who watched a 2.5 hour video incorporating the trial and sentencing of a capital murder case in the US, and then deliberated in groups of between six and 13 for up to 20 minutes, Wiener et al found that a RTV improved juror comprehension of the legal concepts involved in the case.  The comprehension questionnaire was multiple choice, but it contained both abstract questions and questions that involved application of the law to factual scenarios. The study tested a number of innovations, and the researchers found that while the RTV significantly improved comprehension, it did not perform as well as a set of simplified plain language directions. The reason for this may have been that the RTV used in the experiment was not very clear and some elements were still difficult to understand (unlike the simplified directions, it did not, for example, explain what was meant by a mitigating factor).
Three relatively recent experiments provide more robust evidence that RTVs improve juror comprehension. Ede and Goodman-Delahunty tested the effect of a RTV on 183 community mock jurors in Australia, who watched a 60 minute trial video performed by professional actors. Jurors were provided with either a written summary of the trial judge's directions, a RTV or no aids. The researchers found that while the RTV made little difference to comprehension for the simple comprehension questions (that tested memory rather than understanding of the directions), it did significantly improve comprehension scores for the more complex questions that required a deeper understanding of the law.  This experiment, while relatively realistic, did have some limitations. There was no deliberation and the questionnaire used to test comprehension used yes/no answers, rather than multiple choice or free text. Participants in the RTV group heard the same oral directions as the other jurors - the oral directions were not tailored to the use of the RTV (although if they had been, the improvement in comprehension may have been even greater).
In a series of experiments undertaken by Kelly, mock jurors who were given a RTV performed significantly better in multiple choice comprehension tests (both those measuring abstract and applied knowledge) than those who only heard standard oral trial judge directions.  Her experiments - one of which involved community jurors recruited from the jury pool at Southern Tasmania Supreme Court and three of which involved student jurors - both involved a 55 minute trial video, filmed in a real courtroom with a mixture of legal professionals and actors using a script adapted from a real criminal trial. One of her four experiments involved deliberation in groups of four or five for up to 30 minutes. While the trial materials were relatively realistic, unfortunately, the RTV was only one component of a package of materials given to jurors - the package also included a transcript of evidence, written directions and a chronology of key events - so it is not possible to isolate the effect of the RTV alone. 
Finally, McKay et al found some limited support for a RTV's effectiveness on the application of legal tests in an experiment involving a mixed sample of 92 mock jurors (some of whom were students, some of whom were recruited from the community). It was tested in three factual scenarios (aggravated robbery, kidnapping and indecent assault), but significant improvements in comprehension were found only in the aggravated robbery scenario.  It has to be said, though, that the research methods were not as robust as other studies. The mock jurors were given a trial summary rather than a video (although they did hear recorded oral directions afterwards), there was no deliberation and comprehension was measured by asking the jurors to provide a written justification of the verdict they reached (which was coded by the researchers for "quality"  ).
Taken together, the studies described above provide some limited support for the effectiveness of RTVs. They do have to be weighed, however, against three studies that have failed to find any positive effect. Essex and Goodman-Delahunty did not find any improvement in comprehension scores among jurors who were given a RTV compared to those who were only directed orally.  Their experiment was relatively realistic in most respects - it used 236 real jurors in a Sydney criminal court who had been cited but not selected and they watched a 45 minute professionally acted trial video of a sexual assault trial. Comprehension was measured in a variety of ways, including through application. There were, however, also a number of limitations, which may explain why the RTV did not have any effect. The trial was relatively short, it did not involve any defence evidence and the legal issues were reasonably easy to understand, as evidenced by the fact that comprehension scores were high across all experimental conditions. In addition, the jurors who were given the RTV heard the same directions as those who were not - no reference was made to the RTV in the oral directions.
Ogloff also failed to find any significant improvement in comprehension from the use of a RTV compared to the use of oral directions only.  His study involved 545 community mock jurors, who watched a two and a half hour mock trial video based on an actual criminal case. The jurors deliberated for up to two hours (the size of the groups is not specified in the report cited here) before returning a verdict and individually completing comprehension questionnaires. However, when Ogloff and another researcher watched the recorded deliberations, they found that few of the juries were actually using the RTV which, they state, "highlights the need to clearly instruct - and perhaps to direct - the jury in the use of decision making aids such as [these]". 
Finally, in a slightly different context, Dann et al examined the effect of a decision tree on comprehension of DNA evidence, but found that it had no significant effect.  The study was a large scale one and was relatively realistic, using a trial video and including a long period of deliberation.  However, the context perhaps limits the usefulness of the study - the jurors were working their way through a series of questions designed to evaluate the likelihood that hairs found on a discarded sweatshirt belonged to the accused, rather than to reach a verdict.  Perhaps more importantly, like Ogloff, the researchers found that when they analysed the recorded deliberations, jurors made limited references to the decision tree. Only seven out of the 20 juries made any use of it and only two attempted to work their way through the questions, with both efforts being "abandoned about midway through when discussion of related evidence commenced".  This might suggest that the decision tree was not well designed, but it also provides further support for Ogloff's point about needing to direct juries specifically on the use of a RTV if it is to provide maximum benefit. In Dann et al's experiment, the juries who used the decision tree received the same oral directions as those who did not.
2.9.2 Field Experiments
A much more limited set of field experiments have examined RTVs. Strawn et al report on a judge (Strawn himself) who utilised a RTV in a re-trial of a criminal case he presided over, the first case having resulted in the jury being unable to reach a verdict. In the second trial, the jury was able to reach a verdict and commented favourably on the RTV, whereas in the first case, it was reported that they jury had been confused by the instructions, unable to agree on a verdict, and left the jury room "angry with one another and confused, still debating what 'the law' required".  The weight that can be attached to an anecdotal report of a single case is, however, limited. Young et al, in their study of real jurors in New Zealand, found that those jurors who had been given a RTV all reported that they found it useful in deliberations. Other jurors who did not receive a RTV commented that it might have been useful, because they felt that they had "no framework for their decision-making and did not work through the legal points in the case systematically".  Heuer and Penrod also found that the limited number of jurors in their field study who were given RTVs were significantly more likely to feel well informed, satisfied, more confident their verdict was correct, and more confident their verdict reflected a proper understanding of the law. This was most marked in longer and more complex trials. 
There is a further - and very extensive - project testing the impact of RTVs that is in progress in Australia and New Zealand.  The project involves both a major mock jury study and a field experiment comparing matched pairs of cases in New Zealand (where RTVs are routinely used) and Victoria (where they are not). Preliminary results from the project suggest that in the mock jury study, the RTV improved juror comprehension, but only after deliberation.  As in Ogloff's study, however, the jurors were not specifically directed on the RTV and the researchers found that the deliberating juries did not always use it to help them reach a verdict - instead some groups set it aside as they did not see how it related to the information in the verbal charge.
Preliminary results from the field study suggest that RTVs may improve applied comprehension and that the deliberations of juries who used RTVs were significantly shorter than those that did not. However, until these results are formally published, along with full details of the research methods used, it is not possible to determine the weight to be attached to them. The methods used in the field study in particular will need to be scrutinised to assess whether any differences in deliberation times can be confidently attributed to the RTV rather than to other differences in the way criminal trials are conducted in the two jurisdictions or to differences in the nature of the cases concerned.