Models of the disclosure process
Recently, two models have been advanced that describe the often difficult route of disclosure for sexual abuse victims. These descriptive models highlight the difficult process of disclosure. The first model, the child abuse accommodation syndrome, was proposed by Summit.30 It consists of five categories: (1) secrecy, (2) helplessness, (3) entrapment and accommodation, (4) delayed, conflicted and unconvincing disclosure, and (5) retraction. These categories were drawn from clinical accounts of the secondary trauma associated with reporting abuse, where children's allegations were typically disbelieved. It highlights both the secretive nature of abuse (because so many children are sworn to secrecy), and the fact that disclosure is rarely a one off event. As well, many children take a long time to disclose sexual abuse and often retract their initial allegations.
More recently, Sorensen and Snow proposed a typical disclosure pattern for children who had been sexually abused.31 They retrospectively analysed 116 cases, drawn from 630 cases of confirmed sexual abuse, and indeed found support for some of the stages in Summit's model. (The cases had been confirmed either by a confession or guilty plea from the offender, a conviction in a criminal court, or medical evidence consistent with sexual abuse.) As a result of qualitative analysis of the clinical reports associated with these cases a four stage progressive disclosure process was identified: denial, disclosure (first tentative and then active disclosure), recant, and reaffirm. In Sorensen and Snow's sample almost 75 per cent of the children denied being sexually abused when questioned about it after the initial allegation.32 Most of their sample (78 per cent) then moved to the middle ground of first tentatively then actively, disclosing the abuse. Fully 22 per cent of children then recanted their initial allegations of sexual abuse. Finally, 92 per cent of these children reaffirmed their initial allegations.
These two models of the disclosure process are extremely valuable in drawing attention to the difficulty that children experience in disclosing sexual abuse and in highlighting the frequency of retractions. Disclosure of sexual abuse is a difficult process. Although these models are important for understanding the difficulties involved in such disclosure, they do little to show how to increase the disclosure of sexual abuse without simultaneously increasing potential false allegations.
More recently, an alternative model of the disclosure process, based on social cognitive theory, has been proposed by Bussey and Grimbeek .33 In that model a distinction is made between children's ability to remember the events that took place and their willingness to report those events. Factors that are postulated to influence accurate and truthful reporting of experienced sexual abuse include four main processes. First, attentional processes refer to the attention paid to the original event (both central and peripheral aspects). Second, retention processes refer to the mental representation of the event (as children increase in age, visual encoding will be replaced by verbal/conceptual encoding) and the amount of rehearsal of the experienced event. Third, production processes refer to the assessment techniques used to establish children's memory for the events (for example, children are asked to demonstrate what happened, asked to describe what happened, and asked specific questions). Finally, motivational processes involve three sub-processes: outcome expectations which are anticipated punishments or rewards for accurate and truthful disclosures, false allegations or false denials: internal evaluative reactions which include anticipated embarrassment, self blame, and pride for accurate and truthful disclosures, false allegations or false denials: and self efficacy expectations which are beliefs about one's ability to disclose sexual abuse in the face of disbelief or negative reactions from others to the disclosure, and to resist leading questions, et cetera. It is proposed that there is a developmental progression in the extent to which these motivational factors influence children's reporting of events. Initially, only highly salient factors would serve as a motivational basis for inhibiting children's disclosure of witnessed events, for example, the presence of the alleged perpetrator .34 Between four and five years of age, the mere anticipation of punishment or other negative reaction by others could inhibit children's disclosure and conversely, strong social pressures could promote false allegations. With increasing age, all three motivational factors would be expected to influence children's reporting of abuse that happened, as well as their reporting of an event that did not happen.
From this viewpoint, reporting is dependent on both children's memory for the event and motivational factors. It is also clear from the model proposed by Bussey and Grimbeek that children's reporting of the event is dependent on the way memory is assessed. Whether it is assessed through direct or open-ended questions, for example, will influence the amount and accuracy of information obtained. Also, the manner in which the interview is conducted will influence motivational factors that could affect the truthfulness of the information reported. This model departs from the other two models in stressing the importance of interview procedures to obtain the most accurate and truthful information about children's experiences of past events, and to reduce suggestive questioning procedures that may lead to false allegations. Issues of suggestibility are accorded great importance, although they were not elaborated in the initial model proposed by Bussey and Grimbeek. 35 The conceptualisation of suggestibility advanced here departs considerably from other researchers' conceptualisations of it.
Most research and theoretical accounts of suggestibility make children the central focus. Can they resist suggestive questioning and provide accurate and truthful evidence? In this paper, drawing on the social cognitive model, 36 a different approach is proposed. Rather than making children the focus of attention, interviewing strategies and other factors that can influence the accuracy and truthfulness of the information that children report are accorded equal importance in this model. From this theoretical perspective it is argued that children's suggestibility is related not only to the cognitive competence of the child, but more importantly, to the social factors and interviewing strategies used in these contexts.
The traditional definition of suggestibility has been that information provided after the event influences the recollections of the actual event. 37 This definition is memory-based and implies "[that suggestibility can only be unconscious (i.e., interfering information is unwittingly incorporated into memory); suggestibility results from the provision of information following an event as opposed to preceding it; and suggestibility is a memory-based, as opposed to a social, phenomenon". 38 In contrast, a broader definition of suggestibility has been proposed by Ceci and Bruck in which, "suggestibility concerns the degree to which children's encoding, storage, retrieval, and reporting of events can be influenced by a range of social and psychological factors." 39 This definition differs from earlier definitions in three important ways. First, it implies that "it is possible to accept information and yet be fully aware of its divergence from some originally perceived event, as in the case of 'confabulation' ... acquiescence to social demands, or lying ... thus, these forms of suggestibility do not involve alteration of memory". Second, "suggestibility can result from the provision of information preceding or following an event." And finally, it "can result from social as well as cognitive factors". 40
From the Ceci and Bruck position, suggestibility occurs when information is wittingly accepted even though it is known to be false. 41 This contrasts with earlier definitions that regarded suggestibility as occurring unwittingly. The position adopted here lies between these two positions. The distinction here is not whether the introduced information is accepted either wittingly or unwittingly, rather, the issue is whether the suggested information is believed to be true and accurate or not. It is possible to believe that false information is true and accurate, either wittingly or unwittingly, because of mnemonic factors, or a combination of mnemonic and motivational factors. For example, the suggested information can be accepted as a result of mnemonic processes when the original information is unavailable, not accessible, or the source is confused Further, the memory for the event may be faded inaccessible, confused, or the suggested information may be believed, even though the memory for the event can be accessed because of the statue and perceived knowledge of the person providing it Under such conditions, one's own memory for the event is wittingly discounted in favour of the suggested information. Thus mnemonic and/or motivational influences can lead to belief in the accuracy and truth of the suggested information and this can occur either wittingly or unwittingly. Furthermore, suggestible influences can be either external (for example, suggestive interviewing practices) or internal (for example, an imagined event). However. in this article the focus is on external influences because only these can be regulated and monitored in a forensic setting.
Regardless of the processes involved, suggestibility as defined here occurs when the person believes the suggestions as being an accurate and truthful account of their own recollections, despite their inaccuracy. Because such recollections are inaccurate, yet truthful, they are unintentional falsehoods and not lies. In other instances, however, where the information can be remembered but the interviewee acquiesces to the suggestions of the interviewer, lying has occurred. That is, when an interviewer reports the interviewer's suggestions as being a true and accurate account of their experience despite no believing the account, this constitutes an intentional falsehood or lie. Over time, individuals may forge the true and accurate information so that technically lies are transformed into unintentional falsehoods or truthful but inaccurate recollections of events.
Therefore, this perspective differs from that of Ceci and Bruck in that a distinction is made between the reporting of false information as a result of suggestive interviewing based on whether the suggestive information is believed or not. 42 If it is believed, suggestibility has occurred, if it is not believed, lying has occurred. Suggestibility can occur wittingly or unwittingly, and can involve either or both mnemonic and motivational processes in accepting the false information. In contrast, lying is always witting and involves only motivational, not mnemonic processes, since children can still access their recollections of the events accurately. Both types of false statements, that is, suggestible falsehoods (truthful but inaccurate reports as a result of suggestibility) and lies can occur, for both false denials and false allegations.
The important question then is how to reduce the possibility of both false allegations and false denials, either of which can be made intentionally or unintentionally. There is no foolproof way to determine if a child is falsely alleging or denying abuse or accurately and truthfully reporting it. It is possible, however, to ensure that both suggestibility and lying are minimised through the use of interviewing procedures that do not either wittingly or unwittingly encourage children to report false information and by providing a context that allows and requires children to report truthful information. Only from laboratory-based studies in which the event that the child needs to disclose, is it known (for example, where they were touched in a medical examination) that the efficacy of non-suggestive interviewing strategies can be assessed. Only in such studies is it possible to assess those interview techniques that promote truth-telling and accurate disclosures and reduce false allegations and false denials. Once their efficacy has been established in such controlled settings, they can then be used in forensic settings.
As noted above, in most of the recent accounts of children's suggestibility, it is conceptualised as intrinsic to the child. The child's lack of cognitive competence and social dependence in relation to adults renders them more vulnerable to adults' suggestive questioning. However, from the social cognitive theory model, human behaviour is explained in terms of a model of triadic reciprocality in which behaviour, personal factors, and environmental events all operate as interacting determinants of each other (see Figure 1). From this perspective human behaviour is neither totally shaped by environmental forces, nor totally guided by internal dispositions.
Although the child's memory for a witnessed event is cognitive representation internal to the child, for the interviewer to become aware of that memory, communication must occur. There are different factors that influence children's memory for an event versus their reporting of it. The personal contribution includes cognitive skills, appraisal of the event, stress level, attention to the event, subsequent thought about the event, ability to communicate about the event, and the ability to understand the interviewer's questions.
The behavioural contribution refers to actions, verbal and nonverbal processes (for example, gestures). The environmental contribution includes both events occurring at the time of the event (for example, threatened by the perpetrator, relationship to the perpetrator, and the nature of the abuse) and events occurring at the time of reporting of the event (for example, quality of the interview, appraisal of the purpose of the interview, and the number of interviews). Therefore, the accuracy and truthfulness of children's reports of experienced events will depend not only on internal factors (their memory for the event) but also on events that are external to the child (the type of question asked by the interviewer).
From this theoretical perspective, suggestibility is not totally an internal factor that resides in the child, but rather, suggestibility only comes into play when the memory is to be elicited from the child by an interviewer. That is, suggestibility is an interaction between the child's memory of the event, their appraisal of the event, and how they interpret the interview situation. When children tell something spontaneously it is unlikely that they will succumb to suggestion. However, the more input there is from the interviewer, particularly erroneous material the greater the possibility that children will comply with that suggestion and incorporate it into their reporting of the event on later occasions either intentionally or unintentionally. A major issue then, is not the suggestibility that resides in the child, but rather the misleading interviewing practices of interviewers that reduces the reliability of children's evidence.
From a forensic viewpoint, information can be forgotten and inaccessible, so that sometimes it may be impossible for children to retrieve their recollections of abusive events. There are a number of methods that can be used to facilitate the retrieval of information, but it is essential that these procedures do not include the provision of suggestive information. Geiselman's cognitive interview and various contextual reinstatement methods can be used successfully. 43 The crucial point is that remembrance should not be hurried by using props or verbal suggestions. The other issue that is particularly important from a forensic viewpoint, is that apart from not using props or words that will interfere with the memorial representation of the abuse, it is essential that there is no motivational incentive for children to lie about abuse by either falsely alleging or denying it. Interviewers need to make it clear to the children that they do not know what happened to them. and make sure that their biases are not apparent to the children who will be placed in an invidious situation if they need to contradict the adult interviewer.
A recent study that represents the new wave of research on child witnesses is provided as an example of how interview practices can promote false allegations rather than accurate and truthful disclosures by children. 44 Preschool children were instructed to think about four events that actually happened to them and four events that did not happen to them, although in the latter instances they were led to believe that the events had happened to them. They were read the following instructions :
I am going to read some things that happened to you when you were little, and I want you to think real hard about each one of them. Try to make a picture of it in your head. What do you think you would have been wearing when it happened? Who would have been with you? How do you think you would have felt? We made this list up by talking to your mother to get her to tell us about some things that happened to you when you were younger. So, after you make a picture of it in your head, and think real hard about each thing for a minute. I want you to tell me if you can remember it or not. OK? Don't worry if you cannot remember it though. 45
Children were required to visualise the eight events and then try to recollect them, on 12 separate occasions, spaced approximately one week apart. Although initially children indicated that they could recall the real events but not the fictional ones, after 12 weeks of visualizing and talking about the events, children began to believe that the fictional events did happen to them and provided increasing details across this time period about these events to the interviewer. Whether children were lying about these events or were suggestible (truthful but inaccurate reporting) is not possible to determine. However, children were interviewed by a different interviewer at the end of all the sessions and when she stated that the first interviewer had made mistakes about things that happened to the child, the number of children who alleged that the fictional events occurred decreased. Not surprisingly, over the time period some children had started to believe that the events were true. Initially children may have lied about an event and generated information about it to please the interviewer. Through repeated interviewing, however, it is apparent that many of the children had come to believe the veracity of their own reports.The important message from this study is that providing false information to children is likely to lead to them providing false information to the interviewer. This is particularly likely when interviewers insist that fictional events" did occur and this is confirmed by the child's mother. In such circumstances, it is not surprising that children are more likely to trust their mother's memory than their own, particularly if they have no recollection of an event which did not occur. This study is silent about what would happen if a child believes an event has occurred and the interviewer states that the mother said it did not occur. Would the child be so readily swayed in this situation? If interviewers lie to children, it is more difficult for them to provide truthful and accurate information. Interviewers undermine the reliability of children's evidence by falsely asserting they have knowledge about events children have experienced.
Rather than conducting more studies to show the myriad of ways that false allegations can occur, more studies are needed to understand why children withhold information from interviewers. Earlier in this article, it was shown that children do not readily disclose their abusive experience. However, there is little information, from the child's perspective, to explain this lack of disclosure. An understanding of why children withhold information is necessary to facilitate accurate and truthful reporting of their experiences. The section below discusses recent studies that have examined children's secret keeping.