Children’s Testimony

Children may experience or witness crime and may need to provide reports to authorities. Children’s eyewitness accounts can contain critical information about serious acts such as murder, domestic violence, kidnapping, and assault. Child sexual abuse is particularly likely to bring children into contact with the criminal justice system because the case may boil down to the child’s word against that of the accused. Although even young children can provide accurate accounts of their experiences, including highly traumatic incidents, such children on average are both less complete in their memory reports and more suggestible than older children and adults.

Like adults’ accounts, children’s accounts are influenced by numerous factors, including cognitive, social, and individual ones. Developmentally appropriate interview protocols may contribute to obtaining complete and accurate accounts while reducing inaccuracies in a child’s testimony. As part of a forensic interview, children may have to identify culprits from photo lineups. Children 5 years and older can perform quite well if the culprit is pictured in the lineup; however, in “target-absent” lineups, even older children have a strong tendency to guess. Children’s emotional and attitudinal reactions to providing eyewitness testimony in criminal cases can be long lasting. For example, testifying multiple times, especially in severe intrafamilial child sexual abuse cases, is associated with adverse emotional and attitudinal reactions into adulthood. Children in such cases may need additional legal protections.

Memory and Suggestibility in the Child Witness

During the past several decades, there has been an exponential increase in the number of children who provide statements in legal cases, thus magnifying the need to determine the credibility of their testimony. In general, older children are more accurate in eyewitness reports than are younger children, although even preschool-age children can provide accurate accounts of salient or personally meaningful events, including their own victimization. When asked free recall and open-ended questions, preschoolers can recall relevant and accurate information, but on average they are less responsive and provide fewer spontaneous statements than older children and adults. Because young children’s free reports are generally relatively brief and incomplete, they are often exposed to specific and leading questions in forensic situations, which are indeed more likely to elicit the child’s memory of an event. On the negative side, however, children are less accurate than adults in response to specific questions and more vulnerable to interviewers’ implied suggestions. Particularly, closed questions, such as yes/no and forced-choice questions, can be problematic for young children, because they may guess instead of providing “I don’t know” responses. Children also often have considerable difficulty in using standardized units of measurement, such as minutes and months, and in indicating the number of times highly repeated acts have occurred, even though such information can be vital in a legal case.

Usually, children’s testimony is required for crimes or experiences that are negative, if not traumatic. Although this is a subject of debate, considerable research with adults suggests that for stressful compared with nonstressful events, central features (e.g., the main stressors) are retained particularly well, whereas peripheral details are less well remembered. Several studies confirm such findings for children; however, the results of developmental studies are mixed.

Child sexual abuse often involves trauma to child victims, leading to feelings of self-blame and helplessness. These characteristics have contributed to make child sexual abuse situations of special interest in debates about trauma and memory. Research suggests that memory of traumatic events, in many ways, follows the same cognitive principles as memory of distinctive nontraumatic events. However, there is debate as to whether “special memory mechanisms” (e.g., repression) are also involved.

Some of the main theoretical accounts of trauma and memory suggest that traumatized individuals remember trauma-related information particularly well. Empirical evidence confirms that traumatized individuals, especially those who have developed posttraumatic stress disorder, overfocus on trauma cues, have difficulty ignoring trauma stimuli, and remember their trauma experiences. In contrast, other theories indicate that trauma victims, such as incest survivors, may experience amnesia for the trauma and that children who have suffered a larger number of traumatic events tend to forget or remember more poorly those experiences compared with children who have been exposed to a single traumatic event.

Children’s memory and testimony about negative emotional experiences also depend on individuals’ coping strategies. Avoidant coping strategies lead children to evade thoughts, conversations, or reminders about the traumatic experiences. Parents’ attempts to minimize or ignore their own or their children’s distress facilitate avoidant coping. These postevent avoidance processes may prevent the creation of a complete, detailed, and verbally accessible account of the traumatic experience and the integration of these memories with the individual’s other autobiographical memories. In contrast, positive parent-child interactions provide an opportunity for rehearsal and reactivation of event details, which may help maintain and strengthen memory traces, thus reducing the effects of decay while enhancing long-term retention. For example, children who received maternal support after disclosure of child sexual abuse and who discussed the event with their mothers provided more accurate reports, with fewer omission errors, of their maltreatment experiences years after the abuse reportedly ended compared with those who did not.

Children’s suggestibility in the forensic context has been a flash point in the debate over children’s testimonial competence. Suggestibility concerns the degree to which the encoding, storage, retrieval, and reporting of events can be influenced by a range of internal (e.g., developmental, cognitive, and personality) and external (e.g., social and contextual) factors. False information given before, during, and after an event can lead to difficulty in retrieving the original (true) information, alteration of true memory representations, and/or conscious acquiescence to social demands. Young children, specifically preschoolers, are disproportionately susceptible to the effects of leading questions and suggestions. However, of importance in the legal context, children are often less suggestible about negative than positive or neutral events.

Both cognitive and social factors can underlie developmental differences in eyewitness memory and suggestibility. Due to a less complete knowledge base and more limited capabilities of using memory strategies, young children have greater difficulty recalling events on their own. Also, compared with adults’ memories, children’s memories of the original event may be weaker and thus more vulnerable to being altered or overwritten by the suggestions of others. “False memory” may occur when the erroneous suggestion is particularly strong, such as in multiple suggestive contexts where not only misleading questions but also an accusatory context is involved. In addition, preschoolers are less able to distinguish between different sources of memories and thus misattribute an interviewer’s suggestions to actual experiences. Moreover, without understanding the ramifications of their statements, children may adopt suggestions to gain the adult investigator’s approval and avoid negative reactions, perceiving pressure to conform to the suggestions of the authority figure.

Although there is consensus that misleading questions and highly suggestive contexts are to be avoided when interviewing children, such questioning does not necessarily lead to false reports. For example, if the child’s memory is strong, blatantly misleading questioning in a highly misleading context can actually bolster resistance to misinformation, at least compared with the effects of such questioning after a long delay, when the child’s memory traces have weakened. However, with such questioning, the risks of memory contamination are potentially great, and the child’s credibility may be destroyed in the process.

Individual Difference Factors

Although chronological age is almost always the strongest predictor of suggestibility, with preschool children being the most suggestible, even adults are suggestible. Moreover, there is much variability within age groups depending on the characteristics of the individual. However, findings concerning individual differences tend to be somewhat inconsistent, and the predictive power of individual difference factors tends not to be strong. That said, global, comprehensive measures of language ability are sometimes associated with preschool-age children’s suggestibility. Children with mental retardation are more suggestible than typically developing children with normal intelligence, although intelligence is not significantly related to suggestibility within the normal population. Young children with poor self-concepts or poor supportive relationships with their parents are at risk of being more suggestible. Children raised by secure and supportive parents may develop positive self-concepts, which in turn may make them resistant to suggestions that are inconsistent with their own experiences. Cultural factors may also play a role; in cultures where children are trained to be especially polite or obedient to adult authority, they may have more difficulty disagreeing with adult interviewers who falsely suggest information.

Interview Techniques and Protocols

How likely children are to disclose crimes such as child sexual abuse when simply asked free-recall and open-ended questions is the subject of debate. Researchers have developed child interview techniques and standardized child interview protocols intended to increase the likelihood of disclosure as well as the amount and accuracy of the information obtained, while reducing inaccuracies. These protocols (e.g., cognitive interview, narrative elaboration) derive from the application of mnemonic, communication, and social facilitative techniques to forensic practice and can in principle be used to interview child witnesses about a wide variety of events; however, some protocols are specifically designed for interviewing alleged child victims of child sexual abuse (e.g., the National Institute of Child Health and Human Development [NICHD] structured interview protocol). Overall, interview protocols and interview guidelines (e.g., the guidelines developed by the American Professional Society on the Abuse of Children) recommend that forensic interviewers rely as much as possible on free-recall/open-ended prompts. However, the use of some specific questioning is typically also allowed. We review a subset of the protocols/techniques next.

The cognitive interview (CI) relies on well-established principles of encoding specificity (i.e., how the items to be retrieved were encoded and stored determines the effectiveness of a particular retrieval cue) and varied retrieval. According to these principles, the original CI (developed for adults) included four mnemonic techniques: (1) “mental reinstatement” of the external and internal contexts of the experienced event; (2) the “report everything” instruction; (3) the “reverse-order-recall” instruction, which refers to recalling the event in an alternative temporal order; and (4) the “change perspective” instruction, which refers to recalling the event from an alternative perspective. Also, to avoid the common problems observed during the administration of the CI by professionals, the revised CI includes several social techniques intended to facilitate communication (e.g., rapport building, no interruptions). Compared with control interviews, the developmentally adapted CI for children ranging in age from 4 to 12 years, tends to elicit more correct information, although the reverse-order-recall and change perspective instructions may increase the reporting of incorrect details by young children. Moreover, the mental reinstatement and report everything mnemonics appear to be useful in reducing the negative effects of misinformation even in preschool-age children (i.e., 4-5 years).

Rather than supplying specific cues derived from the event itself, narrative elaboration (NE) provides child witnesses with pre-interview training, instructions, and techniques that could be applied to any event of interest. NE’s main objective is to help overcome potential developmental limitations in communication and memory, such as lack of knowledge about the expectations of the listener and ineffective use of memory search strategies, by training children about the level of detail required and by providing picture cards as external cues to report forensically important categories of information. Overall, NE is helpful in enhancing children’s eyewitness recall without increasing the amount of inaccuracies provided by 3- to 11-year-old children.

Similarly, after an initial rapport-building phase, the NICHD interview protocol incorporates training of children to respond to open-ended prompts during the presubstantive phase of the investigative interview. Next, the interviewer attempts to shift the child’s focus to the substantive issue in a nonsuggestive manner (e.g., “Tell me why you came to talk to me today”), so that the recollection process can begin. During this substantive phase, interviewers maximize the use of open-ended questions and probes, introducing focused questions only after exhausting the open-ended-question modes. At the end of the session, interviewers may use option-posing questions to obtain essential information. This protocol is flexibly structured and aimed to translate research-based recommendations into operational guidelines to enhance children’s retrieval using recall-memory prompts. It has been extensively investigated with real alleged child victims of sexual offenses, and it appears to be useful with children 4 years and older.

Basic and applied research underlies the development of interview techniques and protocols. However, further research on the accuracy of children’s eyewitness memory—for example, concerning highly emotional and embarrassing information—is necessary to elucidate how extensive these benefits are. And, of special relevance, improved strategies and tools that can be effectively used with young children (e.g., 3-year-olds) to obtain evidence about specific details of an event without compromising the accuracy of their reports are still needed.

Props and Cues

Children typically have more information in memory than they report in response to free-recall or open-ended questions. Props such as real objects, scale models, dolls, toys, photographs, and drawings can provide concrete external retrieval cues for young children. They also can potentially extend memory retrieval by engaging children in the forensic interview for a longer period than do mere verbal prompts. According to the principle of encoding specificity, the effectiveness of a particular retrieval prop or cue depends on its match with the items to be retrieved with regard to how they were encoded and stored. Especially for younger children, an optimal match should include the original sensory/perceptual features as well as a clear symbolic correspondence.

Overall, props can facilitate children’s reports but also increase the number of errors children make (e.g., if they are too young to understand dual representations). The extent to which props facilitate or compromise children’s testimony depends on factors such as the nature of the event and of the prop, the mode of presentation, and the time that has elapsed between the event and the interview. And the age of the child may be critical in determining the influence of these factors.

Real props have maximal overlap with event information and can effectively aid retrieval for 3- to 10-year-olds. Real props and scale models increase the correct information that children report, but they also introduce additional errors, especially for younger children. In contrast to real props, toys and dolls, including anatomically detailed dolls, can increase commission errors and decrease accuracy, especially when preschool-age children are interviewed with misleading questions or when “distractor” or play-evoking props are involved. Under certain circumstances (i.e., in combination with specific but nonleading prompts), drawings can facilitate the completeness and accuracy of 5-years-olds’ and older children’s accounts, although there are mixed findings in relation to the effectiveness of drawings with preschool children. Finally, human figure drawings can produce a considerable amount of new details during the interview, especially for children aged from 4 to 7 years, but at the same time, these drawings may also increase inaccuracies in children’s testimony.

In summary, research has shown that props and drawings can, under certain circumstances, facilitate memory accuracy in children older than 5 years, whereas they may add error to the reports of younger children. Although there is currently no “gold standard” method of interviewing children, different combinations of free-recall, specific, and prop-assisted questions are being researched to determine which of them facilitates the most accurate and complete memory reports from children.

Photo Lineups

When interviewed in forensic situations, children may be presented with photographic lineups to identify culprits. A lineup may include a criminal (target-present lineup) or only innocent individuals (target-absent lineup). When they are shown a target-present lineup, preschool-age children are less likely than adults to make correct identifications, although children around age 5 and above are typically comparable with adults in making correct identifications. Shown a target-absent lineup, however, even early adolescents are inferior to adults, making fewer correct rejections and more false identifications. As with leading questions, the photo lineup may entice children to guess.

Witnesses are usually shown a simultaneous lineup, in which all lineup members are presented at once and only one decision is made. This method has been criticized for encouraging a relative judgment, whereby witnesses compare all lineup members and choose the member who looks most like the criminal relative to other members. Although this strategy is successful in target-present lineups, it may lead to errors in target-absent lineups. Fortunately, fairly simple training techniques can reduce guessing on target-absent lineups in older children.

An alternative procedure is the sequential lineup, in which witnesses are shown photographs one at a time and make a decision for each photograph. Compared with simultaneous lineups, sequential lineups reduce adults’ false identifications by increasing correct rejections of target-absent lineups while having a minimal effect on correct identifications from target-present lineups. When they are shown a sequential lineup, witnesses might make an absolute judgment for each photograph by comparing the photograph with their recollection of the criminal. However, target-absent errors by children are not reduced in sequential compared with simultaneous procedures.

Children in the Courtroom

As a result of involvement with legal authorities, children may experience social and emotional distress. Although repeated interviewing of children can keep accurate memories alive, child victims report that being interviewed multiple times by legal authorities is stressful for them. Speaking about traumatic experiences (particularly in open court), lack of parental support, harsh cross-examination, facing the defendant, and not being believed add to children’s distress and may reduce significantly the amount of information provided by child witnesses. Moreover, in child sexual abuse cases, a child’s initial disclosure of the abuse to a parent, teacher, or other trusted adult may include a more detailed account than the testimony that the child gives in court months or even years later. Although testifying may be helpful for some children, it causes others to recover from the criminal and legal experience more slowly than their nontestifying counterparts. Child sexual abuse victims who had to testify multiple times in severe intrafamilial cases tend to have the most negative long-term emotional effects and are thus most in need of protection during criminal prosecutions. To remedy these negative consequences, procedural modifications (e.g., testifying via closed-circuit television) and multidisciplinary investigations, conducted at child advocacy centers and involving teams of legal professionals (e.g., the police, prosecuting attorneys, and child protective services workers), who coordinate their efforts into a single interview of the child victim/witness, are being developed and tested in the United States and abroad.

Having an adult (e.g., a mother, social worker, or police officer) recount children’s out-of-court statements (e.g., hearsay) at trial has recently attracted research and legal interest. In criminal trials regarding child sexual abuse, hearsay is often introduced in addition to the child’s live testimony. Although hearsay is normally discouraged in the American legal system, there are special hearsay exceptions, some of which apply specifically to children’s statements. However, a recent U.S. Supreme Court ruling suggests that if the out-of-court statement was made to an authority (e.g., a forensic interviewer) and is thus “testimonial,” the authority cannot testify in place of the child.

Mock jurors find children’s statements more credible when the child testifies live in court than if the child is replaced by a hearsay witness. Mock jurors also find children less credible if they testify via closed-circuit television instead of face-to-face at trial. Both hearsay and closed-circuit television are potential ways to protect children from the stress of testifying live in court and are used in many European countries.


  1. Bruck, M., & Melnyk, L. (2004). Individual differences in children’s suggestibility: A review and synthesis. Applied Cognitive Psychology, 18, 947-996.
  2. Castelli, P., Goodman, G. S., Edelstein, R. S., Mitchell, E. B., Paz-Alonso, P. M., Lyons, K. E., et al. (2006). Evaluating eyewitness testimony in adults and children. In I. B. Weiner & A. K. Hess (Eds.), The handbook of forensic psychology (3rd ed., pp. 243-302). Hoboken, NJ: Wiley.
  3. Ceci, S. J., & Bruck, M. (1995). Jeopardy in the courtroom: A scientific analysis of children’s testimony. Washington, DC: American Psychological Association.
  4. Eisen, M. L., Quas, J. A., & Goodman, G. S. (2002). Memory and suggestibility in the forensic interview. Mahwah, NJ: Lawrence Erlbaum.
  5. Pipe, M.E., Lamb, M., Orbach, Y., & Cederborg, A.C. (2006). Child sexual abuse: Disclosure, delay, and denial. Mahwah, NJ: Lawrence Erlbaum.
  6. Poole, D. A., & Lamb, M. E. (1998). Investigative interviews of children: A guide for helping professionals. Washington, DC: American Psychological Association.
  7. Pozzulo, J. D., & Lindsay, R. C. L. (1998). Identification accuracy of children versus adults: A meta-analysis. Law and Human Behavior, 22, 549-570.

Return to the overview of Trial Consulting in Forensic Psychology.