Pre-recorded evidence and juror decision-making: evidence review

Paper on the impact on juror decision-making of pre-recorded evidence and live-link testimony by child and vulnerable adult witnesses in criminal trials.

4. Jurors’ Perceptions of Pre-Recorded Evidence Given by Vulnerable Adults

This chapter outlines the findings of existing research in relation to the impact upon jurors of the use by adults of pre-recorded evidence or live-links within criminal trials. Far fewer studies have been conducted in relation to adult witnesses vis-à-vis children, but the sophistication of trial simulations within these studies – which have typically focused on rape complaints – offer a high degree of confidence in respect of their findings. Research suggests that, in respect of adult rape complainers, the influence upon mock jurors of the use of pre-recorded evidence is unpredictable and unlikely to have a significant impact upon verdict outcomes. The use of special measures may increase one person’s empathy for an adult rape complainer at the same time that it raises another’s suspicion about her credibility in equal measure.

Beyond sexual offences trials, existing research is extremely limited. Some studies suggest that mode of testimony may have a stronger influence here, which can work to the detriment of assessments of the witness’ credibility and likeability. It is difficult to place great confidence in these findings, however, on account of key weaknesses in these studies’ underlying methods: in particular, their failure to incorporate a group deliberation stage, which has been shown elsewhere to neutralise individual influences and counteract any significant impact of mode of testimony on verdicts.

4.1 Introduction

In contrast to the relatively large number of studies which have explored the extent and ways in which mode of evidence effects observers’ perceptions of children’s testimony, there is a surprisingly sparse evidence base in relation to adults. It is likely of course that some of the considerations raised in relation to child witnesses in Chapter 3 will also be relevant in respect of adults. However, there are a number of distinguishing features – bound up, amongst other things, with jurors’ particular preconceptions about children’s maturity and memory, their perceptions and understanding of different types of adult vulnerability, and their views as to what it is reasonable to expect an adult accusing someone of a crime to withstand as part of the trial process, as well as, in some cases, distinctive procedural and evidential rules applying to adult versus child witnesses. This makes it important for jury researchers to also dedicate specific attention to adults (in all their diversity).

Of the work which has been undertaken to date on the impact upon jurors of adult witnesses’ use of out-of-court live-links and pre-recorded evidence, the vast majority has focused upon sexual offence complainers. These studies have been marked by a more robust methodology than many previous simulation studies, which bolsters the confidence that can be attached to their findings, at least in respect of this type of trial.

4.2 Jurors’ Assessments of Adult Testimony in Rape Trials

Research has identified a generally high level of appreciation of the protection afforded by special measures amongst adult rape complainers. Hamlyn et al (2004) found that vast proportions of adult witnesses (including sexual offence complainers) who had used them found them helpful, with one-third indicating that they had enabled them to give evidence that they would not otherwise have been willing or able to give. At the same time, as with child witnesses, substantial concerns have been raised regarding the potential undue influence of the use of special measures on the proof process, and on juror decision-making in particular. In the specific context of adult sexual offences trials, critics have argued, for example, that the use of screens, live-links and pre-recorded interviews may unfairly prejudice the defence or imbue the complainer’s testimony with an undeserved level of credibility (Burton et al, 2006; Payne, 2009a). Meanwhile, others have feared that the absence of the complainer in the courtroom and the mediating effect of the live-link may create a distance between the complainer and the jury which will make it less likely that his or her account will incite sympathy or be believed (Council of HM Circuit Judges, 2006; Hamlyn et al, 2004; Payne, 2009b).

Concerns have also been expressed regarding the extent to which the use of video-recorded police statements as evidence-in-chief places complainers at a disadvantage, since officers are engaged at that stage in an investigatory process, receiving and pursuing fresh information as the account unfolds. This means that these accounts can lack the kind of logical, sequential narrative that can be imposed by advocates who take the complainer through their testimony at trial, whether in open court or via live-link (Wade et al, 1998; Davies et al, 1999; Welbourne, 2002; Tinsley & McDonald, 2011).

In an effort to begin to test some of these concerns, Taylor & Joudo conducted a simulation study in Australia involving 18 different trial conditions, across which an adult female rape complainer gave testimony variously via CCTV, pre-recorded videotape or face-to-face in the courtroom, and with different levels of observable emotionality. The transcript was based upon an actual sexual assault case, and across the study 210 volunteer participants from the local community observed a 75 minute live trial re-enactment. Having done so, they were asked to complete questionnaires measuring both their pre- and post-deliberation preferences, as well as their broader attitudes in relation to rape. These responses were analysed alongside discussions undertaken in groups of 12 members (on average) within which participants were given up to one hour to deliberate and agree upon a collective, unanimous verdict.

Overall, the researchers were unable to identify any significant relationship between verdict preferences at the individual or collective level as a result of different modes of the witness’ evidence delivery. This is not to say that, for some individual jurors, the use of live-link or pre-recorded evidence did not provoke specific reactions. Taylor and Joudo note, for example, that “although no systematic differences were found in pre-deliberation perceptions or propensity for jury verdict between face-to-face, CCTV and video conditions, some jurors in the CCTV condition expressed the view that they would have preferred the complainant to be physically present in the courtroom. These jurors stated that they felt unable to make an accurate assessment of her character, demeanour and truthfulness on the screen and that they would have liked her to be in the courtroom so that they could watch her non-verbal actions and whether she looked at the accused in giving her testimony” (2005: 62). The impulse to prefer live evidence and face-to-face confrontation may be particularly prevalent in rape trials where the physical ability of the complainer to resist her attacker is often paramount in jurors’ minds (Ellison & Munro, 2010b) and where a lack of corroborating evidence often means the trial outcome depends upon the perceived credibility of the allegation (and of the complainer herself). But while some jurors expressed this preference during the deliberations, it was not consistently shared and, in line with Davies (1999) in respect of child witnesses, Taylor & Joudo concluded that it did not, in any significant or reliable way, influence jurors’ decision-making either pre- or post-deliberation.

Similarly, in respect of the complainer-witness’ emotionality, while Taylor and Joudo found that individual jurors might seek to draw inferences from this in the course of deliberations, a calm or emotional demeanour were not interpreted and understood consistently. This meant, in the end, that any associated influence tended to be neutralised. As the researchers put it, “emotionality of testimony did not impact in any consistent or systematic way on jurors’ perceptions of the complainant or the accused (other than perceiving her to be more stressed when emotional than neutral). This is likely to be explained by individual differences in preferences for victim testimony. Some jurors indicated that they found her testimony more plausible because she was not emotional; other jurors indicated they would have found her more plausible if she had been emotional” (2005: 66). Ultimately, therefore, the researchers concluded that “what may be much more important than the manner in which testimony is presented are the pre-existing attitudes, biases and expectations that jurors bring with them into the courtroom” (2005: 66) which, in their view, were the key determinants of verdicts.

The Taylor and Joudo study makes an important contribution to our understanding of juror assessments of vulnerable adult, as opposed to child, witnesses; and in a context in which much previous research has relied on a 1-way CCTV model in which counsel (and possibly the judge) relocate to a witness room in order to hear testimony in person, the use of a 2-way live-link is particularly valuable. But in the pre-recorded evidence condition within this study, jurors observed a videotape of the evidence that the witness gave live, rather than viewing a videotape of her forensic police interview as evidence-in-chief. As discussed in Chapter Three, while this allowed the researchers to fully isolate the mode of delivery by keeping content and form constant, it sits at odds from courtroom practice in a number of jurisdictions and as such reduces the transferability of findings arising in the pre-recorded condition to the ‘real’ world.

In this respect, the recent research by Ellison and Munro, in the context of England and Wales, is particularly instructive. Here, a mini-trial transcript was created with input from legal professionals, which was then re-enacted live by actors and barristers in front of an audience of 160 jury service eligible community volunteers across the study. After observing the 75 minute trial simulation, jurors were split into deliberating groups of 8 members and given 90 minutes to reach a verdict. Trial conditions varied only in respect of mode of evidence delivery, with the witness giving evidence either by means of a live TV link, from behind a screen in the courtroom, in open court, or by replacing her examination-in-chief with a pre-trial police forensic interview and undergoing cross-examination via a live TV link. Where the police interview was used, it covered the same substantive points as the examination-in-chief but repetition was increased and some narration was adjusted to render it less logically sequential in an effort to replicate the less structured way in which the account would unfold at this early stage.

Akin to Taylor & Joudo’s research, Ellison & Munro (2014) explored data arising from jury verdicts, individual jurors’ pre- and post-deliberation questionnaires, and the substantive content of group deliberations, which were recorded and coded for analysis. The researchers similarly failed to establish any consistent pattern in terms of final verdict outcomes and jurors’ preferences for in-court versus out-of-court video-mediated evidence. They specifically addressed three themes relevant to the question of the influence of mode of delivery upon jurors – (1) reducing the emotional impact of testimony; (2) influencing credibility assessments; and (3) fairness to the accused.

4.2.1 Reducing the Emotional Impact of Testimony

Ellison and Munro’s study provides little support for the suggestion that the emotional impact of testimony will be reduced when a witness appears on a screen, translating into a loss of juror empathy. Across the deliberations, there were only a handful of exchanges which raised concerns of this sort and it was not clear that these had any impact upon jurors’ approach to other aspects of the deliberation, or to their verdict. The majority of participants made no reference to the live-link or to the use of video-recorded evidence during deliberations, and when asked in post-verdict questionnaires to reflect on specific questions about the complainers’ emotional state, there was no evidence to suggest that the mode of delivery had an impact. Indeed, 76% (n=63/83) of jurors in the conditions in which evidence was given either in open court or from behind a screen agreed or strongly agreed that the complainer was distressed whilst giving testimony, compared with 74% (n=57/77) of jurors in the trials where the complainer testified ‘out of court’ via either a live-link or through pre-recorded testimony. Similarly, 66% (n=55/83) agreed or strongly agreed that the complainer was nervous whilst giving her testimony in the ‘in-court’ conditions, compared with 55% (n=42/77) in the out of court conditions. The reduced response rate in the latter case reflects, no doubt, the perception that giving testimony inside the courtroom will be more stressful for witnesses than making use of live-links or pre-recorded evidence. Overall, however, the researchers concluded that these findings indicate that the use of special measures did not significantly preclude jurors from appreciating the emotional difficulties complainers experience when giving testimony.

4.2.2 Influencing Credibility Assessments

Ellison and Munro also found little evidence in support of concerns either that the use of special measures will imbue witness testimony with undeserved credibility or, conversely, that the witness’ absence from the courtroom will undermine her perceived reliability or trustworthiness. This is not to say that there were not occasional jurors who expressed such views, but the direction of those comments – which were themselves much in the minority – was variable, suggesting that special measures could work in favour of the complainer as much as they could work against her. This stance was further reflected, moreover, in pre-deliberation questionnaires where 35% (n=29/83) of jurors in the open court and screen conditions reported that they believed the complainer’s account and 39% (n=30/77) felt similarly in respect of the ‘out of court’ conditions. And while, as a consequence of deliberations, views shifted in this study somewhat away from a belief in the complainer’s credibility, the broad parameters of that shift were comparable across in-court and out-of-court conditions. Thus, 28% (n=23/83) of jurors in the in-court conditions and 29% (n=22/77) in the live-link and video+link conditions combined continued to believe the complainer’s account.

4.2.3 Fairness to the Accused

Ellison and Munro’s study also found little evidence to suggest that the use of special measures prejudices the defence, either by suggesting to the jury that the witness cannot face being in the same room as, and needs to be protected from, the accused, or more generally by creating an imbalance in the procedures by which competing accounts are provided. The risk that the use of special measures may impact adversely upon the accused was explicitly raised in the deliberations by jurors in three trials – all of which involved use of a CCTV live-link – but in each instance the suggestion was quickly challenged by other jurors. When asked specifically in post-deliberation questionnaires if they felt that the trial they had observed was fair, the majority of participants across all trial conditions answered affirmatively; and where jurors did register concerns, there was no suggestion of a clear association between this and the mode of the witness’ testimony. Indeed, participants in the open court condition were most likely to give a negative answer to this question – 20% did so, as compared to 16% for the video+link and 15% in the live-link condition – and their supplementary comments established that their main concern was the perceived insufficiency of evidence provided in the trial rather than the use of special measures.

4.3 Jurors’ Assessments of Adult Testimony Beyond Rape

Rape trials – though an obvious and important terrain upon which to conduct experiments testing for the impact of the use of special measures upon jurors – provide a very particular context. Sexual offences by their nature require recounting an intimate form of violation, and rape trials are often characterised, or presumed to be characterised, by heightened levels of emotionality that may influence the ways in which jurors perceive alternative modes of evidence delivery, and the credibility of the adult witness who relies upon them. Existing research in relation to adult witnesses in other types of trial is extremely limited, but where it has been conducted, there is some suggestion of a stronger influence upon mock jurors as a result of mode of testimony.

In a study by Landström et al (2005), 122 students were recruited as mock jurors in a trial regarding a staged car accident. 12 adult witnesses observed the accident and were divided into groups of ‘truth-tellers’ and ‘liars’ to then be interviewed. That interview was either pre-recorded and played to jurors or re-enacted live in front of them. In line with previous studies, the researchers found that while jurors who watched truthful witnesses often rated them as having to think less hard than did jurors who watched their lying counterparts, this did not translate into an effective cue for deception. Jurors were no better than chance in assessing the veracity of the adult witness testimony, regardless of mode of presentation. But on the question of jurors’ credibility assessments, there was some evidence of an impact. The researchers reported that “the live observers rated the witnesses more positively than did the video observers. Specifically, the results showed that live (versus video) observers rated the witnesses as being more eloquent and more pleasant” (2005: 928).

Similar results also emerged in a study by Fullwood et al (2008), in which 60 undergraduate students were recruited and grouped across three conditions to observe an adult female give testimony concerning a ‘fairly innocuous’ domestic incident, either face-to-face, via a video, or via a video preceded by a brief face-to-face introduction. Here, the authors concluded that jurors’ perceptions were influenced by mode of delivery to the extent that they “perceived themselves to be less able to emotionally engage with the witness and felt that the testimony was less believable when presented via video compared to the face-to-face presentation” (2008: 7).

Further, in a recent study by Landström et al (2015a), it was suggested that mode of evidence delivery had an influence upon a group of 81 law student volunteers who observed an adult male witness give a statement in relation to a physical assault either live or via a videotaped interview, and were then asked to rate their confidence in the truth of the allegation and their perceptions of the witness’ emotional demeanour. The researchers concluded that “male assault victims are considered less credible if their statement is presented on video, as opposed to live” and suggest this is “troublesome” since it poses “real-life implications in light of the increased use of visual technology in the legal arena” (2015: 103). These latter claims in particular over-reach, however.

The findings emerging in these studies merit further exploration, but in evaluating their contribution to the existing field of research, it is crucial to bear in mind their methodological limitations. In marked contrast to the studies on adult rape complainers, where considerable effort has been taken to ensure realism to trial proceedings, with recruitment from the local community and extended scope for collective deliberations, these non-rape studies of adult witnesses rely on student volunteers, do not contextualise the witness testimony in the broader setting of a trial re-enactment, and do not include any group deliberative process, despite the importance that has been attributed to this in mediating and moderating jurors’ potential preferences for live testimony in previous research involving child witnesses.

In a context in which Fullwood et al explicitly acknowledge that “there is no reason necessarily to believe that such perceptions would affect the decision-making process of the jury” (2008: 3), the omission of this deliberative component is particularly disappointing, and marks as all the more premature the sort of claims made by Landström et al (2015) in relation to any purported “real life implications” of this work. These studies raise the prospect that – notwithstanding findings of a minimal impact in the rape trial context – adult witness testimony in relation to other offences may be perceived differently by jurors depending upon the mode of its delivery. Without further, and more careful, investigation, however, it is not possible to draw this conclusion with any confidence, and it is certainly not possible to predict what impact such difference in perception might have in the overall context of verdict deliberations.

4.4 Future Research

It is clear from the above discussion that there are several dimensions to the impact on jurors of adult witnesses’ use of special measures that are yet to be properly investigated, let alone understood. A substantial research project is currently underway, funded by the Nuffield Foundation, which may address a number of gaps in our existing understanding ( As of yet, however, no findings are available, and to the extent that it relies on videotaped trial simulations as a background context into which video-mediated forms of testimony are inserted, its conclusions may be limited. From the evidence that is currently available, there is little support for concerns regarding the impact of mode of evidence delivery upon verdict outcomes, notwithstanding the influence it may have on some individual jurors in a range of unpredictable, and often contradictory, ways. The research base in respect of non-sexual offences and male adult complainers is less established, and while there is some suggestion that pre-recorded evidence may work to the disadvantage of witnesses in these contexts, the absence of a deliberative component within existing studies – particularly given the significance attached to them in studies on child and other adult witnesses, and their practical import in real cases – is a major omission.

Further work is clearly needed in order to provide a richer and more diverse spectrum of studies in relation to adult witnesses and mode of evidence delivery. In addition, future research should seek to supplement the findings of existing studies on rape trials by including an evaluation of the impact on jurors of the use of pre-recorded cross-examination (as currently being rolled out in England and Wales). While Taylor and Joudo’s study included a pre-recorded cross-examination, and found that this, as part of the overall pre-recorded evidence condition, had no significant effect on jurors, this cross-examination followed on from a pre-recorded examination-in-chief that was very different in style and format from the forensic police interviews routinely utilised in many jurisdictions. As such, the question of the impact upon jurors of this format of evidence in adult cases remains unsatisfactorily addressed by existing research.

More broadly, as will be discussed briefly in Chapter Five, there may be a need to conduct further research to explore the impact of differential modes of introducing pre-recorded and live-link evidence within the criminal courtroom. In both the Taylor & Joudo and Ellison & Munro studies, best practice was followed. Images of the witness were presented on a very large plasma screen, which was clearly visible to all jurors. This is not always the case in real courtrooms where smaller screens may be used or screens may be located at a distance that impairs jurors’ view. In addition, Ellison & Munro’s pre-recorded interview was scripted on the basis of examples of best practice identified within police training videos and in accordance with the ‘Achieving Best Evidence’ guidelines (Ministry of Justice, 2011). As a result, the videotape did not replicate certain interviewing practices which have been identified as a cause for concern ( HMCPSI/ HMIC, 2007), and a case could legitimately be made for research that better reflects the variable quality of interviews and screen technology in courts.


Back to top