Abstract
Developmental factors affect the way that children and young people behave in legal contexts. We first discuss developmental factors such as cognitive and emotional development, social expectations and suggestibility that affect young victims and suspects. We then describe some implications of these developmental factors for police interviewers and for the youth justice system more generally and call for the more differentiated treatment of young people according to their age and development.
Young people or children – terms used here interchangeably to refer to all individuals under the age of 18 – interact with the justice system daily as witnesses, victims, and suspects. While the incidence of substantiated child abuse is not published in the United Kingdom, in 2011 approximately 50,500 children were registered on child protection plans and considered to be at risk for physical, emotional or sexual abuse and neglect (NSPCC, n.d.). In England and Wales, 241,737 juveniles were arrested in the fiscal year 2009/10, accounting for 17 per cent of all arrests during that same period (Ministry of Justice, 2012). With such numbers of young people within the welfare and justice systems, it is crucial for researchers and practitioners to recognize that various developmental factors – cognitive, emotional, and social – can compromise the effective participation of young people in legal contexts (Bruck et al., 2006). In this article, we discuss in turn the developmental factors that affect youthful victims and suspects, while recognizing that these factors overlap and are not unique to either group. We then comment on the implications of these developmental factors and conclude by highlighting a proposal for an overhauled youth justice system.
Young People as Victims
It is helpful to begin with a historical account that contextualizes the key changes in our understanding within the last 20 years of the 20th century. Much of this program of research can be attributed to the aftermath of the infamous McMartin case in Southern California. In 1983, the mother of a toddler attending the family-run McMartin daycare centre in Manhattan Beach, California, became concerned that her child had been sexually molested while in the centre’s care. Her complaints to the police triggered an investigation that, by the time of its conclusion, had elicited reports that more than 300 ‘pre-schoolers’ enrolled in the centre had been abused by seven members of staff, including the owner and her son, who were charged with multiple instances of child sexual abuse. After a series of false starts and a very expensive trial, the prosecutors eventually dropped all charges against the defendants.
For our present purposes, the McMartin trial was important because it provided a forum for competing accounts of children’s capacities as witnesses in legal contexts. In brief, one group of experts argued that young children were incapable of lying about the sorts of abusive experiences that lay at the heart of the trial. By contrast, another group of experts argued forcefully that young children possessed limited memorial skills and linguistic capacities, making them highly vulnerable to confusion when questioned about their experiences. Strikingly, both sets of experts could make reference to only a very small body of supportive research. This prompted a generation of researchers, many of them students of cognitive development, to initiate studies designed to elucidate children’s developing capacities. They produced hundreds of reports, presentations and articles that greatly enriched our understanding of the factors that influence the capacity of children to be competent informants about their own experiences (for reviews, see for example, Lamb et al., 2008; Lamb et al., 2011).
Indeed, the accumulating evidence was sufficiently rich and clear that by the beginning of the 1990s, professionals and investigative agencies published summaries of the available evidence and formulated guidelines outlining how forensic interviews with children should be conducted to maximize the quality and quantity of information that children were able to provide. Perhaps the most impressive of these documents was the ‘Memorandum of Good Practice on Video-Recorded Interviews with Child Witnesses for Criminal Proceedings’ (Home Office, 1992) jointly published by the Home Office and the Department of Health in the UK. Similar, though typically less detailed and less thorough, guidelines were created and published by agencies in many other countries during the 1990s (see for example, American Professional Society on the Abuse of Children, 1991, 1997). We thus owe a considerable debt of gratitude to those researchers and agencies because they helped bring about a dramatic change in how young alleged victims were questioned and understood in legal contexts.
Four Developmental Factors Affecting Child Victims
Memory
Researchers confirmed that young children remember fewer details about their experiences than older children and adults (Eisen et al., 2002; Flin et al., 1992; Lamb et al., 2003; Leippe et al., 1991; Ornstein et al., 1992; Poole and Lindsay, 1995; Thierry et al., 2001) while making clear that the information that they retain is just as likely to be accurate as the information provided by older children and adults (Flin et al., 1992; Goodman and Reed, 1986; Johnson and Foley, 1984; Marin et al., 1979; Oates and Shrimpton, 1991). Researchers further demonstrated that the accuracy of the information recalled by young people – as well as by adults – was profoundly affected by the nature of the questions used to elicit that information from memory (Bjorklund et al., 1998; Ceci and Bruck, 1995; Waterman et al., 2001). Specifically, researchers made clear that information retrieved using recall memory processes was much more likely to be accurate than information retrieved using recognition memory probes (Dent, 1982, 1986; Dent and Stephenson, 1979; Hutcheson et al., 1995; Lamb and Fauchier, 2001; Lamb et al., 2007a; Orbach and Lamb, 2001).
Communicative skills
As far as language was concerned, it was clear that young children often had rich and extensive vocabularies, even though communication was sometimes undermined by the use of unusual and idiosyncratic words and labels, especially for body parts (Thackeray and Readdick, 2003). Communication was further compromised by the failure of interviewers to use age-appropriate vocabulary (Saywitz et al., 1990; Saywitz et al., 1993; Walker, 1994), and even more importantly by their failure to frame their questions simply and unambiguously (Carter et al., 1996; Imhoff and Baker-Ward, 1999; Perry et al., 1995). Because interviewers instead often attempted to ask complicated questions, frequently including forms of language like the passive voice with which young children were unfamiliar, their questioning was often less informative and more confusing than it might have seemed (Dent, 1982; Perry et al., 1995).
Social orientation
Researchers also drew attention to the highly unusual circumstances in which young children found themselves when being questioned by forensic investigators. Young children are, of course, very familiar with situations in which they are questioned by adults, but those adults typically ask questions designed to test the children’s knowledge (Lamb et al., 2007b). In such circumstances, the questioners typically know the correct answers to their questions; by contrast, in the forensic context, the interviewer typically does not know the ‘right’ answer. Instead, the child is the expert. It is thus extremely important for forensic interviewers to communicate to children the uniqueness of the forensic interview, and the children’s special and unique roles as expert informants (see for example, Sternberg et al., 2002). Similarly, children may be shy or reticent toward unfamiliar adults, especially adults with the status of police officers, and this makes it important for interviewers to establish rapport before they begin questioning the children (Carter et al., 1996; Wood et al., 1996).
Suggestibility
Notwithstanding their shyness, of course, young children are typically eager to impress adults and to gain their approval and this eagerness is both a blessing and a potential problem. It can be a special problem when children, in their eagerness to gain the approval of adult questioners, focus on real or perceived cues of approval and disapproval, and adjust their answers in order to curry favour (Ceci et al., 1987; Gudjonsson and Singh, 1984; Richardson et al., 1995). The problem is exacerbated when children are being asked about incidents that they may not remember very well and when the questions themselves are hard for the children to understand. In these circumstances, researchers showed, children’s suggestibility can become a considerable problem, with children providing information that does not accurately describe their experiences and is misinterpreted by the interviewers (Warren et al., 1996). Indeed, analyses of the interviews conducted in the McMartin case clearly showed that the interviewers elicited considerable amounts of false information from the alleged victims because they failed to recognize just how suggestible children could be when questioned inappropriately and in inappropriate contexts (Garven et al., 2000; Garven et al., 1998; Schreiber et al., 2006).
Taken together, the research on young witnesses provides compelling evidence that children can be either competent or incompetent witnesses, with the quality and quantity of the information they provide determined not simply by developmental differences in the children’s capacities but also, and perhaps even more importantly, by differences in the quality of the interviewers’ questioning. Professional guidelines, like the Memorandum of Good Practice (Home Office, 1992), provide an important and authoritative compendium of practices to utilize or avoid in order to maximize the quality of the information elicited from young victim-witnesses. It is important to underline that the research conducted in the 20 years since the Memorandum was published has elaborated on, or enriched, the advice and guidance provided for interviewers without significantly altering any of the conclusions supported by this early wave of research (for reviews, see for example, Lamb et al., 2008; Lamb et al., 2011).
Children as Suspects
As we have indicated, the initial wave of research focused on children who were forensically interviewed because they were suspected victims of crime, typically child abuse, and especially sexual abuse. However, the strengths and limitations identified in the research emphasized characteristics of children and their behaviour when being questioned about their experiences.
As a result, this research is also helpful for understanding the capacities of children who are not alleged victims of crime. Whether questioning children as alleged victims, as witnesses, or as suspects, it is equally important to recognize the strengths and limitations of their memory, their communicative capacities, their social styles and orientation to adult questioners, and their susceptibility to suggestion. Clearly, children can be competent witnesses if they are well interviewed by experts who understand the developmental factors influencing their abilities and limitations. Unfortunately, this rather obvious conclusion has not been widely honoured in the breach. In 1993, the year after the ‘Memorandum of Good Practice’ (Home Office, 1992) was published, Jon Venables and Robert Thompson were arrested by the police for their role in the murder of James Bulger. These 10 year-old children would have benefited from the involvement of skilled and specially trained interviewers had they been suspected victims of abuse; instead, they were subjected to some 20 hours of questioning by police officers who were guided more by their experiences as interviewers of adult criminals than by developmentally sensitive evidence documenting the special needs and circumstances of young children in forensic contexts. The Police and Criminal Evidence Act of 1984 (PACE) recognizes that suspects under 17 should have access to a responsible ‘appropriate adult’, whose role is to give advice, facilitate communication, and ensure that the interview is conducted properly and fairly (Home Office, 2008). Unfortunately, further guidance is not given. Compared to professional guidelines for interviewing child victims, the rubric pertaining to young suspects is considerably less detailed, less informed by the developmental research and outdated in light of the considerable amount of research conducted since 1984.
Additional Developmental Factors Affecting Youthful Suspects
Cognitive development
In addition to the developmental changes in memory, communicative capacity, social orientation and disposition and suggestibility that we have outlined above, there are a number of other factors that are particularly important when considering the role and capacities of children who are being questioned as suspects. First of all, developmental researchers have long been aware that there are significant increases in the ability to engage in logical thinking and problem-solving between the ages of approximately 11 and 15 years (see for example, Neimark and Lewis, 1967; Saarni, 1973). In Piaget’s (1952) terminology, the beginning of this phase is marked by the transition from ‘concrete operational’ to ‘formal operational’ thinking, and by the end of this period most young people have an understanding of risk and probability that is roughly comparable to that of adults (Acredolo et al., 1989; Schlottmann, 2001; Schlottmann and Wilkening, 2011). However, this ability to understand risk and probability is not accompanied by an equivalently sophisticated ability to apply that information (Gardner and Steinberg, 2005; Van Leijenhorst et al., 2008)! Most importantly, older teenagers appear less capable than adults at applying information about risk (including estimates of the probability of being caught following transgression) because they tend to overvalue the possible benefits (for instance, of shoplifting) while simultaneously underestimating the costs (for example, of being caught) until they are approximately 20 years of age (Gardner and Herman, 1990; Halpern-Felsher and Cauffman, 2001; Scott and Grisso, 1997).
Peer influence
Perhaps even more obvious than these changes in young peoples’ logical reasoning and the application of their knowledge is a significant social shift from parents to peers as the primary reference sources in early adolescence. This shift begins in middle childhood, as children find increasing pleasure in interactions with peers rather than adults, but it accelerates during the period between 10 and 14 years of age. This period is marked by a shift from a focus on behaviour that will elicit approval from parents to focus on behaviour that would elicit approval from peers (Nickerson and Nagle, 2005; Steinberg and Silverberg, 1986). The transition is often accompanied by considerable anxiety and concern on the part of adolescents about their own identity and social standing (Erikson, 1968). As at other points in the life cycle, such uncertainty increases the extent to which individuals look to others for guidance and approval. The increased desire to gain social approval parallels the focal shift from parents to peers, a shift that leads young people to be particularly focused on obtaining approval from peers who, like the individuals themselves, have much less mature and appropriate means of judging the appropriateness and/or suitability of behaviour.
Puberty
Another important developmental shift that characterizes the second decade of life is associated with puberty. Puberty is, of course, a biological transitional event but apart from the reproductive maturation of boys and girls, puberty is also associated with increases in emotionality (Lewin, 1939), emotional lability (Larson and Lampman-Petraitis, 1989), sensation-seeking (Steinberg et al., 2008) and inadequate inhibitory self-control (Greenberger, 1982). These characteristics appear to be associated with the immature status of the limbic system in adolescence; a number of studies have documented that the structural and functional maturation of the limbic system is typically incomplete until mid-adolescence (see for example, Giedd, 2008; Schneider, and Vergesslich, 2007).
These developments in emotional maturity are paralleled by changes in the ability to appraise the future consequences of behaviour – a skill that does not fully develop until around 20 years of age (see for example, Crone and van der Molen, 2004; Spear, 2000). Such reasoning is located, neuro-anatomically speaking, in the prefrontal cortex, which is widely viewed as the seat of executive function. Recent research assessing cerebral functioning using sophisticated techniques shows that the structural and functional maturation of this part of the brain is not complete until the early 20s (see for example, Blakemore and Choudhury, 2006; Casey et al., 2005; Sowell et al., 1999), a finding that complements behavioural research showing that the ability to employ executive functions reaches maturity at approximately the same age (see for example, Leon-Carrion et al., 2004; Luciana et al., 2005).
Developmental Factors: Applications and Implications for Youth Justice
We have for many years recognized high levels of exploratory and sometimes impulsive misbehaviour as normative characteristics of teenagers, behaviour likely to be attributable to emotional lability, poor self-control, and eagerness to impress. Importantly, however, behavioural research on juvenile misbehaviour and delinquency shows that this period of nearly normative misbehaviour peaks around 15 to 18 years of age and that, in the majority of cases, involvement in delinquent behaviour completely ceases at around this point (Moffitt, 1993). Only a small minority continue to engage in inappropriate behaviour at later ages. These more persistent offenders are often characterized by continuing susceptibility to peer pressure, extremely poor evaluation of the risks associated with inappropriate or criminal behaviour and continuing poor future planning, all of which are associated with the continued involvement in criminal activity that persists from adolescence into adulthood and beyond.
Regardless of whether young people are destined to be on this ‘life-course persistent’ criminal trajectory or else to be characterized by time-limited involvement in misbehaviour, those who are apprehended and questioned by the police are characterized by a number of deficits in their cognitive capacity, poor communicative skills, and elevated suggestibility that have profound implications. For example, they might diminish the ability of these young people to avoid self-incrimination or affect their ability to understand the caution, a standard legal requirement of any suspect interview, which is even difficult for many adults to understand (Clare et al., 1998; Fenner et al., 2002). Our own ongoing research underscores the importance of having legal representatives present when young people are formally interviewed by the police. In particular, it appears that appropriate persons such as parents and other relatives infrequently intervene in ways that prevent children from incriminating themselves, replicating previous research that the contribution of ‘appropriate adults’ is limited (Drizin and Colgan, 2004; Medford et al., 2003). The presence of solicitors is systematically associated with the reduced likelihood that young people will provide information or make an admission during questioning, which is consistent with previous findings (Clarke et al., 2011; Medford et al., 2003). Presumably, solicitors are more aware than other ‘appropriate’ persons of the fact that arrest and prosecution are typically impossible in the absence of corroborated evidence concerning the suspects’ misbehaviour.
Such findings are important because of their relevance to false confessions, and their far more serious and common counterpart, coerced confessions (Drizin and Colgan, 2004). Juveniles are more likely than older suspects to confess (Redlich et al., 2004) and to confess falsely (Drizin and Leo, 2004). These differences are especially significant when we consider the crucial role that confessions play in the criminal justice system. In the past, confessions were often sufficient to ensure conviction but growing awareness of the uncomfortably large number of false confessions (which may in turn lead to false convictions: Kassin and Gudjonsson, 2004) has prompted an increased emphasis on corroborative evidence, even when confessions have been obtained. Nevertheless, confessions still play a crucial role. Confessions may establish a confirmatory bias on the part of investigators, a bias that leads them to discount possibly exculpatory evidence while interpreting the available evidence or interrogating suspects in a way that makes it appear to confirm the confessions (see for example, Hill et al., 2008; Kassin et al., 2003; Narchet et al., 2011; Snook et al., 2012). Suspects who have provided confessions are treated differently at every subsequent stage of the criminal process (Leo, 1996).
As Malloy and Lamb (2010) have observed, furthermore, it is revealing to compare the contradictory way in which fact finders and investigators view inconsistencies in the testimonies of alleged offenders and victims. When victims change their testimony, either by adding additional embellishments to their accounts or by recanting allegations they have made, considerable scepticism ensues, and there is substantial evidence that courts often fail to convict when a principal victim-witness significantly alters his or her account. By contrast, when suspects change their accounts, especially when they claim to have confessed falsely, this tends to have very little effect on fact finders, who instead tend to regard such recantations as tactical changes that do not undermine the probative value of the initial confessions. Malloy and Lamb (2010) propose that fact finders and investigators must be similarly cautious when considering changes in any forensic statements, particularly when the individuals involved are children or adolescents whose developmental characteristics may affect the quality, reliability, and trustworthiness of the statements they are believed to have provided.
The developmental differences that we have summarized in this article – notably differences in the ability of young people to fully understand their circumstances and to anticipate the future consequences of their actions and statements – not only affects their behaviour (that is, the likelihood that they may become involved in delinquent behaviour) but also their ability to appropriately instruct counsel, make important decisions (for example, whether or not to confess), and to accept pleas. The latter capacity is especially important in jurisdictions, such as those in the United States, where the majority of convictions are obtained following pleas negotiated by prosecutors (Bureau of Justice Statistics, n.d.).
Looking beyond the immediate implications, developmental differences also need to be considered when we focus on the possibility for rehabilitation. As we have already noted, the majority of children and young people who engage in delinquent behaviour in adolescence are most likely not to engage in further misbehaviour as they grow older. Because of this, it is extremely important not to punish young offenders in ways that will increase the likelihood that they will commit to lives of crime. It is particularly undesirable to punish young offenders by placing them in custodial circumstances where they may encounter and learn from others who are more criminally inclined than themselves. Similarly, it is important not to stigmatize or label them in ways that will limit their ability to resume a more conventional law-abiding behavioural trajectory as they mature (for a comparative analysis of the cases of James Bulger in the UK and Silje Redergard in Norway, see Green, 2008). One of the best predictors of further criminality on the part of young offenders is educational attainment (Blomberg et al., 2012; Katsiyannis et al., 2008), and thus any forms of punishment that restrict the ability of young people to complete their education and training can significantly and destructively impede their chances to become productive members of society in the future. This is a matter of particular concern in those jurisdictions, especially but not only in the United States, where young offenders are increasingly treated as adults and may find themselves spending part or all of their incarceration in adult prisons where there are fewer opportunities for education and training in preparation for the return to the civilian world.
For many of the reasons that we have summarized in this article, we are impressed by the proposals offered by Scott and Steinberg (2008) in their important recent book. Scott and Steinberg argue forcefully that misbehaving children under the age of 10 should never be treated within the criminal justice system, and that any state intervention should be aimed at promoting children’s welfare. Young people who offend between the ages of 10 and 15 should be treated as individuals whose developmental capacities and tendencies usually require them to be treated within an educational/welfare system that emphasizes the opportunity for maturation and rehabilitation, rather than within a criminal justice system that places a heavy emphasis on punishment. For those between the ages of 15 and 18, Scott and Steinberg suggest a case-by-case determination of whether the seriousness of the offences and the individuals’ history of misbehaviour require disposition within the adult criminal justice system with some attention to developmental factors that may affect both culpability and amenability to intervention or whether, like younger offenders, they should be processed within a juvenile justice system with more lenient rules of evidence and with a greater focus on rehabilitation and the potential for constructive change.
Scott and Steinberg’s recommendations focused on the American system; nevertheless, their suggestions have considerable relevance elsewhere, not least in the UK, which lays claim to the youngest age of criminal responsibility in the European Union. The National Association for Youth Justice in the UK recently urged the government to raise the age of criminal responsibility to 16, and to treat younger children who offend through a welfare- rather than punishment-oriented system (Bateman, 2012). This request was rejected by Jeremy Wright, the Minister for Prisons and Rehabilitation, who reiterated the government’s position that 10 year-old children are able to differentiate between right and wrong, and that ‘for young people who do offend there must be an effective response that is targeted at reducing reoffending’ (P Hibbert, personal communication, 4 January 2013). As we have shown above, the growing empirical evidence on development and rehabilitation contrasts with the government’s point of view.