Forensic Psychology

by Graham M Davies

  Anchored narratives are potential biasing factors in judicial decision-making. Consequentially, in court, if decision-makers have to choose between two narratives they are likely to choose the most intuitively plausible one (Baudet, Jhean-Larose, & Legros, 1994). Moreover, should a witness fail to articulate a coherent series of events that is both plausible and anchored in the appropriate narrative for the listener, then that witness’ story may carry little weight, irrespective of its truth-value. However, for such critical reasoning to occur, judges’ awareness need to be raised (Perkins, 1989) and training provided of the need for multiple anchor points (Plous, 1993) and source biasing. But do such biases apply to judicial decision-making when groups of magistrates or judges are involved?

  13.4.2 Decision-making by Groups of Magistrates or Judges: Group Dynamics

  Research has demonstrated that magistrates’ decision-making is inconsistent and sometimes uninformed. For example, in bail use, limited information may be provided by the police that does not provide the full picture to the courts (Dhami, 2004; Doherty & East, 1985). One implication is that defendants may be unnecessarily remanded in custody and may be subsequently acquitted or given a non-custodial sentence. For the penal system, unnecessary detentions before trial add pressure on precious prison accommodation and forces penal authorities to release other prisoners prematurely.

  The gravest penalty available is the decision to impose an immediate custodial sentence, the average length of which has risen in the UK in recent years. In 1995 the average sentence for a man over 21 was 2.8 months compared to 2.6 months in 1990, while the corresponding figures for women were 2.3 and 2.4 months respectively (Flood-Page & Mackie, 1998). It is unclear whether increases reflect judgements derived from the facts of the cases, and that such increased punishment was warranted (Rumgay, 1995). More recent figures from England and Wales, October–December 2011, show the average sentence (less or equal to 6 months) for men is 3.3 months and 3.1 months for women (see Yarwood, 2013). Regardless, relatively little formal guidance exists to help judges and magistrates assess whether an offence is “so serious that only a custodial sentence can be justified”. In Cox (1993) however, Lord Justice Taylor affirmed that a “right-thinking person’s test” was the correct approach. However, the seriousness of an offence is a subjective judgement, which can differ between individuals and/or over time (Ashworth & Hough, 1996). Worryingly, past research has repeatedly revealed inconsistency in sentencing decisions concerning similar cases both between different courts (Parker, Casburn, & Turnbull, 1981) and within the same court (Parker, Sumner, & Jarvis, 1989), for the same individuals (Ashworth, Genders, Mansfield, Peay, & Player, 1984). Indeed, Flood-Page & Mackie (1998) examined current UK sentencing practice in magistrates’ courts and concluded that, whilst past research had shown considerable differences in sentences applied, “since 1993 it is sentencing practice, not the legal framework that has changed” (p. 139). Magistrates have also been found to impose more severe sentences on the basis of video evidence than equivalent text accounts (Chenery, Henshaw, Parton, & Pease, 2001), potentially communicating incidents graphically and directly in ways that produce greater impact than a written statement (Davies, 2003).

  The evolution and perpetuation of “court-cultures” that routinise interactions and decision-making could be invoked as an explanation for differences in magistrates’ sentencing patterns. Some influences on group decision-making can be drawn from the literature on conformity to group pressure, where conformity increases significantly up to group sizes of three to four, but then plateaus and may even decline (Stang, 1976). Furthermore, social impact theory specifies that we conform when the group is one we care about, when members are unanimous in thoughts or behaviour, and when the group size is three or more (Latané, 1981). Other significant psychological research into group processes suggests that groupthink (i.e. maintenance of group cohesiveness; Janis, 1982); polarization (i.e. making extreme decisions; Isenberg, 1986) and social loafing (Latané, Williams, & Harkins, 1979) can all influence decision-making. A bench of magistrates or judges is a small group consisting commonly of three people who interact with each other and are interdependent, in the sense that to fulfil their needs and goals, they must rely on each other (see Aronson, Wilson, & Akert, 2005 for details). Thus groups of judges or magistrates are not immune to bias; indeed they may be subject to additional pressures not experienced by judges sitting on their own.

  PHOTO 13.3 There are numerous aspects that pertain to both process and the individual that might impact upon the outcome of jury decision-making.

  Source: © PNC/Getty Images

  13.5 JURIES AS DECISION-MAKERS

  13.5.1 The Impact of Selection and Profiling on Outcome

  There are numerous aspects that pertain to both process and the individual that might impact upon the outcome of jury decision-making. However, with the jury room closed to scrutiny in the UK (due to the Contempt of Court Act, 1981), the extent of impact remains the subject of constant attention by academics, psychologists and practitioners. In the absence of direct observation of juries, a rigorous debate rages in respect of the validity of jury research. Bornstein (1999), who conducted an analysis of jury simulation studies published in the first 20 years of Law and Human Behavior, identified a number of perceived weaknesses in simulation studies and asserted (p. 2):

  “these concerns are justified not only by fundamental principles governing the sound conduct of scientific research, but also by the desire to apply findings from simulation studies to understanding, and ultimately improving, the legal system.”

  The principal concerns include the samples used in mock jury studies (i.e. undergraduates vs. community adults), the research setting (i.e. laboratory vs. courtroom), the trial medium (i.e. written summaries vs. listening to an actual trial), the trial elements included (e.g. presence or absence of deliberation), the dependent variables used (e.g. dichotomous verdicts vs. probability-of-guilt judgments), and the consequentiality of the task (i.e. hypothetical vs. real decision) (Diamond, 1997; Konecni & Ebbesen, 1979). One obvious drawback with the use of mock juries, for example, is that the defendant’s future is not at risk and so gravity and importance may well be lost (Darbyshire, 2001). The call for allowable juror research is subject to ongoing debate in the UK, although Zander (2005) has cautioned that such research might show “an intolerably high degree of irrationality, prejudice, stupidity and other forms of undesirable conduct in the jury retiring room” (p. 2) that might lead to calls for the abolition of juries.

  Regardless, studies of selection and composition have shown that males are over-represented and non-whites underrepresented (Zander & Henderson, 1993) and as a consequence the perspectives of women and ethnic minorities are still minimised on juries (Lloyd-Bostock & Thomas, 1999). The system by which potential jurors are summoned and excused can impact upon representativeness. For example, Airs and Shaw (1999) found that, in England and Wales, 38% of potential jurors were excused for a variety of reasons and that only 34% were available for service. Additionally, non-registration was found to be highest for ethnic minorities, people aged between 20–24, and those living in rented accommodation. Thus, random electoral roll samples do not necessarily lead to representative population samples.

  One of the main factors expected to influence a jury is the evidence (Kalven & Zeisel, 1966) and to a large extent this is probably still the case. However, Ellsworth (1993) comments that individual jurors draw different conclusions about the right verdict on the basis of exactly the same evidence; thus evidence alone is likely to be insufficient to produce a uniform verdict.

  Studies examining the sex and age of jurors have also been found to influence decision-making. For example, women are significantly more likely to convict on circumstantial evidence (Sealy & Cornish, 1973). In this study, females also gave 78% initial guilty verdicts in cases of rape and 71% for murder cases, whilst the corresponding rates for men were 53% and 50%, respectively. More recently, Thomas
(2010) found that first and final guilty verdicts remained stable for males whilst female guilty verdicts dropped from 41% (first) to 33% (final), suggesting that females might be more open to persuasion or reconsideration of facts. The same report argues no apparent gender bias in cases of rape. However, 52% of verdicts found in the “female 16 or over” category were classified not guilty, whilst 77% of verdicts shown in the “male 16 or over” category were classified as guilty, which perhaps suggest otherwise (see Wheatcroft et al., 2009 for discussion). Upon further examination of the Sealy & Cornish data it was found that the difference in verdicts lay between black males and black females rather than between white males and white females, where no significant difference was observed. Indeed, between 2006 and 2008, white jurors have been found to convict a larger percentage of black defendants (67%) than white or Asian defendants (63% and 63%, respectively); though it is not clear whether these figures differed significantly. Kemmelmeier (2005) maintains race is a critical factor in white jurors’ decision-making.

  On the other hand, in a mock trial, male jurors alone were more likely to impose harsher punishments, such as longer sentences on criminal defendants when committing rape, particularly when the female victim was attractive (Villemar & Hyde, 1983). A more recent examination of crime in England and Wales (Hough, Radford, Jackson, & Roberts, 2013) showed longer sentences were associated with; gender (men more punitive); socio-economic class (non-professionals more punitive); education (those without degrees more punitive); newspaper readership (broadsheet readers less punitive), together with expressed concern about disorder and a perception of rising crime in the jurisdiction concerned. Other studies have also found that levels of education lead to a higher rate of acquittals by males (Mills & Bohannon, 1980). Such impact factors could be mediated by cognitive resource. For example, Hastie, Penrod and Pennington (1983) found that only 48% of case facts presented in testimonies was recalled by jurors with low education levels, compared with 70% for those jurors with high levels of educational attainment. Case Study 13.3 highlights some jury selection difficulties (the interested reader is directed to Darbyshire (2001) for a review of jury research and Hale-Starr & McCormick (2001) for jury selection discussions).

  CASE STUDY 13.3 THE TRIALS OF SCOTT DYLESKI AND O. J. SIMPSON

  Never forget, almost every case has been won or lost when the jury is sworn.

  Clarence Darrow, defence lawyer, 1936

  Juries decide thousands of cases each year. It is not surprising therefore that the jury system, where it operates throughout the world, is of central importance. Opportunities for jury selection are severely limited in the UK, but it is widespread in the United States, where jury selection services have become available to those willing to pay the associated fees. These services tend to be based on social-psychological and behavioural principles in order to apply the most sympathetic jurors to particular cases.

  One such case was that of Scott Dyleski, who at 17 was sentenced to life in prison in August 2006 without parole for the murder of Pamela Vitale, the wife of prominent Californian defence lawyer Daniel Horowitz, on 15 October 2005. Dyleski was convicted after the victim was found bludgeoned to death with a piece of crown moulding on the site of the couple’s planned dream home. In such high profile cases finding jurors with little knowledge of or opinion on the case can be more difficult than usual and the courts may call hundreds of potential jurors in an attempt to find those who have not yet formed opinions. In Dyleski’s case, Judge Barbara Zuniga called 200 jurors to the Contra Costa County Superior Court, but many more may have been called if so required. In similar fashion, more than 1,000 jurors were summoned to Redwood City in the Scott Peterson trial during the same year.

  In theory, large jury pools attempt to find people who are representative of the general population and exclude people with extreme views. However, “in practice what you end up with is a jury that looks like the middle of a bell curve – it’s not jury selection, it’s a de-selection” (M. Rice, quoted in Krupnick, 2006). The 1995 O. J. Simpson trial famously publicised jury selection techniques and, whether classified as selection or de-selection, such methods were demonstrable through the voir dire process. Oral voir dire consists of follow-up questions to questionnaires completed by prospective jurors and questions the lawyer specifically wants to ask orally in order to observe juror responses. In the O. J. case, an advisor to the prosecution, Don Vinson, compiled the jury profile to include an African American male, a Hispanic male, eight African American females and two Caucasian females. The emphasis on African American females came from voir dire questioning which showed that these women were more tolerant of physical force within marriage, the belief that every relationship has its troubles, and people do get slapped around – “it just happens”. Nevertheless, contrary to this view, Marcia Clark, the lead prosecutor in the case, believed that these women would be more likely to have experienced domestic violence in the past and would judge O.J. more harshly than males would.

  What one can conclude is that jury selection is far from an exact science. The voir dire process of juror questioning might eliminate jurors whose opinions differ with either side but, in addition, and in accordance with much academic opinion outlined in this chapter, juries can be influenced by a range of factors. It has been reported, however, that judges agree with 80% of case verdicts (Eisenberg et al., 2005), confirming perhaps, that juries can arrive at just conclusions.

  Visher (1987) has argued that personal characteristics are insignificant in predicting trial outcome compared to the wider factors of witness credibility and other evidence. Yet, there is a body of research that counters such an assertion and suggests that it is inappropriate to expect jurors to negotiate the justice process in an unbiased manner. This includes less obvious aspects of judicial processes, such as the use of specialist language and complexity, which also impact in particular ways.

  13.5.2 The Comprehension of Jurors of Complex Material and Legal Terminology

  Law would not exist without its specialist use of language and were not for the close relationship between legal language and the legal system (Danet, 1984). As such, terminology embraced by the law has two main characteristics: first, it is technical because it uses a specialised vocabulary and second, it is culturally bound because law is above all a social science (Terral, 2004). It is not surprising therefore, that legal terminology can lead to collective misunderstandings by juries. For example, a study conducted by Bornstein, Miller, Nemeth, Page and Musil (2005) demonstrated that the two most important negative juror perceptions of the court system were trial complexity and making the decision. According to Horowitz, ForsterLee and Brolly (1996) jurors’ perceptions of complexity fail to recognise that several evidential factors such as volume to be processed, clarity, and comprehensibility, are relevant to the term and may further add to misconceptions.

  Research suggests legal definitions or how they should apply or operate are actually poorly understood by jurors. For example, according to a survey of 1,225 jurors in New South Wales, Australia, jurors seem to have great difficulty in understanding concepts such as “circumstantial evidence”, “reasonable doubt”, “presumption of innocence” and “intent” (Trimboli, 2008; see also Tinsley, 2001). Further, Wheatcroft et al. (2004) noted that jurors may well tend to focus on witness responses at trial rather than legal terminology.

  Trials often involve complex material presented to juries in a haphazard and disorganised manner, which hampers juries in reaching an appropriate verdict (Bennett & Feldman, 1981). Instead, juries synthesise the information into an organised narrative that makes sense to them but this may be at variance with the perceptions of the court (Wiener, Richmond, Seib, Rauch, & Hackney, 2002). Indeed, cognitive research would suggest that complex cases require greater amounts of cognitive work by jurors without previous experience of such tasks that could lead to overload (see Wagstaff, Wheatcroft, Cole, Brunas-Wagstaff, Blackmore, & Pilkington, 2007). For example, Heuer and Penrod (
1995) questioned real jurors who reported that they became progressively less confident in their verdict as the quantity of information increased, without additional instruction or other support. Thus, a positive correlation might be expected between cognitive overload and the employment of inappropriate heuristics (Gilovich, Griffin, & Kahneman, 2001); excessive information might lead jurors to become confused, disengaged and seek the most available common-sense outcome. Conversely, some lawyers have argued it is precisely because juries are not allowed access to important evidence that they reach perverse decisions (see Taylor, 2004). The testimony of expert witnesses might assist or conversely, skew comprehension. Cooper, Bennett and Sukel (1996) found that mock jurors in the US were more persuaded by high-level expert witnesses than less expert witnesses, but only when the testimony was highly complex. These effects, observed in mock juror studies, may be even greater in real trials, where the pressures on jurors are that much greater (Honess, Levi, & Charman, 1998; Jackson, 1996).

  To assist, jurors in England and Wales may be provided with a route to verdict, which sets out important questions for each jury member to consider in respect of the case under consideration. Whilst this approach has an intuitive appeal, the protocol has not yet undergone any form of empirical assessment. The first trial of Vicky Pryce in 2013 illustrates the misunderstanding that can arise. The jury of eight women and four men were dismissed after handing a list of 10 questions to the judge, sparking concerns and his comments of their “absolutely fundamental deficits in understanding” of their role (see Huffington Post, 2013).