**2.1. Memory development – Witnessing an event**

Most critical for child witnesses is memory development. Although the ability to create memories seems to be present at least from birth, lasting autobiographical memories that a child can report of an experienced event are rare until approximately 3 years of age (Fivush, 1994, 1997; Jack, Simcock, & Hayne, 2012), likely facilitated and organized by language development (Simcock & Hayne, 2002). Nevertheless, once developed, children's memories can be highly accurate. In terms of eyewitness identifications, child witnesses can be as accurate in identifying a culprit as adult witnesses (Humphries, Holliday, & Flowe, 2012; Pozzulo & Lindsay, 1998).

Generally speaking, to be effective witnesses, children must be able to communicate facts of a forensic nature about which they are knowledgeable because they directly perceived and remember them. Thus, a child's ability to properly encode an event in memory is paramount. Effective encoding can be influenced by attention and prior knowledge, while memory retention might be affected by memory strength, script knowledge, and memory storage capacity.

**Attention.** One does not necessarily encode into one's memory information that did not receive attention. This is relevant to child witnesses, as children often do not attend to the same aspects of an event as adults (Johnson & Foley, 1984; Mandler, 1990; Yarmey, 1979). This could be protective for children, as recent findings using the Deese-Roediger-McDermott paradigm (DRM) suggest. The DRM paradigm asks participants to recall lists of semantically related words (e.g., slumber, dream, tired, snooze) from which a critical lure word is missing (e.g., sleep). The procedure tests for false memory of the lure words. Otgaar, Peters and Howe (2012) found that children presented with a divided attention task were *less* likely to have false memories in the DRM paradigm, while adults with divided attention were *more* likely to have false memories. In both groups, however, true memories were not impacted by the attention manipulation. Also, while adults and children might focus on different aspects of a perpetrator's appearance, they nevertheless both seem to be vulnerable to weapon focus effects (Davies, Smith, & Blincoe, 2008). That is, adults and children may focus on a weapon (or other object inconsistent with their expectations) to the detriment of their recall of other aspects of an event that captured less attention (Davies et al., 2008; Pickel, Narter, Jameson, & Lenhardt, 2008).

84 Current Topics in Children's Learning and Cognition

detailed below.

Pozzulo & Lindsay, 1998).

storage capacity.

accuracy and credibility. Yet, this impact likely varies as a function of case content. That is, cognitive maturity might be an important consideration in weighing a child's custody preferences, but be considered less relevant for an alleged sexual abuse victim, for whom perceived trustworthiness is more critical (Crossman, Powell, Principe, & Ceci, 2002; Duggan, Aubrey, Doherty, Isquith, Levine & Scheiner, 1989). To understand the role and impact of child witnesses in the legal system, it is thus important to consider what aspects of

In the current chapter, cognitive development is explored as it pertains to children in different forensic contexts. First, aspects of memory development are considered in terms of how they likely impact child witness perception and reporting of events, in addition to factors that might influence witness memory, such as trauma. Then, additional factors related to child witness strengths and vulnerabilities are considered, including intelligence, source monitoring abilities, and suggestibility. Finally, the interplay between case content and cognitive development is explored for its impact on child witness effectiveness. While the central focus of this chapter is on children as victims/witnesses, children also interact in the justice system as suspects and perpetrators. One of the central concerns in those cases is with lie-telling, which is addressed elsewhere in this volume, but here we consider issues

There is no one "child witness" and the variety of forensic contexts in which children participate makes it difficult to define uniform aspects of cognitive development that are relevant to all child witnesses. Nevertheless, various aspects of cognitive development have been heavily examined for their importance to child witness accuracy and credibility, as

Most critical for child witnesses is memory development. Although the ability to create memories seems to be present at least from birth, lasting autobiographical memories that a child can report of an experienced event are rare until approximately 3 years of age (Fivush, 1994, 1997; Jack, Simcock, & Hayne, 2012), likely facilitated and organized by language development (Simcock & Hayne, 2002). Nevertheless, once developed, children's memories can be highly accurate. In terms of eyewitness identifications, child witnesses can be as accurate in identifying a culprit as adult witnesses (Humphries, Holliday, & Flowe, 2012;

Generally speaking, to be effective witnesses, children must be able to communicate facts of a forensic nature about which they are knowledgeable because they directly perceived and remember them. Thus, a child's ability to properly encode an event in memory is paramount. Effective encoding can be influenced by attention and prior knowledge, while memory retention might be affected by memory strength, script knowledge, and memory

cognitive development, if any, are related to child witness accuracy and credibility.

such as juvenile confessions in relation to their cognitive development.

**2. The developing cognitive abilities of the child witness** 

**2.1. Memory development – Witnessing an event** 

**Prior knowledge.** Children's knowledge, which grows with age, impacts the accuracy of their event memory as well. It influences what they attend to, how they understand and interpret events and what they encode into memory (Principe, Ornstein, Baker-Ward, & Gordon, 2000). For example, a child expert in a domain (e.g., chess) can outperform an adult in recalling locations of chess pieces on a chess board (Chi, 1978), and prior expert knowledge can overwhelm the effect of general intellectual aptitude on story recall (Schneider, Bjorklund, & Maier-Brückner, 1996).

On the other hand, children begin to develop awareness of culturally-based stereotypes during the preschool years (e.g., obesity, mental illness, sex, race; Corrigan & Watson, 2007), and these stereotypes can lead children to selectively encode or inaccurately reconstruct their memories to be in line with their stereotypes (Bigler & Liben, 1993; Klaczynski, 2008; Martin & Halverson, 1983). This suggests that child witness memory accuracy could be undermined by children's stereotypes (Leichtman & Ceci, 1995a, b). Overall, then, knowledge can enrich and deepen a child's event recall and understanding, but also undermine their accuracy.

**Memory strength and scripts.** Once encoded, strength of a memory likely influences a child's ability to recall that memory. That is, events that are highly salient or meaningful to a child, or events that are repeated, might be associated with a stronger memory trace. This can protect the memory over time from degradation or other influences, such as suggestion (Christianson, 1992; Hamann, 2001; Holliday, Douglas, & Hayes, 1999; Howe, 1997; Pezdek & Roe, 1995). With sufficient replication, children can develop scripts, or generalized event representations abstracted from repeated instances of an event. For example, a child might develop a script for 'going out to eat' that involves multiple actors (e.g., host/hostess; waiter/waitress) and a particular set of ordered events (e.g., order beverages; read menu; order; eat; pay bill). The same abstraction can occur for abuse events (Roberts & Powell, 2001).

Children become more efficient in their ability to abstract event scripts with age. Farrar and Goodman (1992) showed that 4-year-olds did not seem to develop an event script following three repetitions of an event, while 7-year-olds did. More generally, children increasingly connect meaning across different exemplars of a target with age (e.g., apple juice, lemonade, and orange juice are all sweet drinks; Brainerd, Reyna, & Ceci, 2008). Thus, in some cases, younger children might not yet have the elaborate scripts that older children have developed. This can insulate them from making erroneous assumptions or interpretations about events (Brainerd et al., 2008; Lindberg, 1991; Pickel et al., 2008).

Cognition and the Child Witness 87

assistance, in the form of questions, to provide a narrative account of an event. Older children, on the other hand, are increasingly able to provide a structured, coherent narrative description of their experiences in response to simpler prompts, such as "What happened?". Thus, interviewers (and parents) frequently ask questions that are more supportive, or "leading", with younger children, in part to provide the narrative structure children need to produce a coherent event account (e.g., Dent & Stephenson, 1979; Hudson, 2006; Poole &

However, researchers consistently find that the least directive questions (e.g., free recall) tend to elicit the most accurate information and have the lowest risk of eliciting false information from children. They also elicit less information overall from children (Bjorklund et al., 2000; Poole & Lindsay, 1995). Interview strategies have thus been designed that attempt to provide additional structure and cues to children to maximize the amount of accurate event recall, while minimizing false recall. Techniques include contextreinstatement (La Rooy et al., 2007) and full interviewing protocols such as the Cognitive Interview (Fisher & Schreiber, 2007) and the NICHD Protocol (Lamb, Orbach, Hershkowitz, Esplin, & Horowitz, 2007). These practices seem to facilitate child witnesses' attempts to

**Communication and linguistic ability**. A child who cannot communicate effectively in the justice system may lack credibility, leading perhaps to future victimization or to conviction, in the case of a juvenile defendant. Most child victims, for instance, must be able to disclose abuse in order for the abuser to be stopped or prosecuted. However, it seems clear from the number of adults who never reported childhood abuse that this remains a problem for many children (Bottoms, Rudnicki, & Epstein, 2007; London, Bruck, Wright, & Ceci, 2008). Yet, when abuse is suspected, the interview strategies noted above can help children to disclose

Some child witnesses must testify in court, subsequent to a disclosure. This can be traumatic, particularly if it must be repeated (Quas et al., 2005). In many cases, children must first undergo a competency evaluation. These typically attempt to assess their understanding of truths and lies, which emerges in the preschool years and becomes more nuanced with age (Talwar & Crossman, 2012). However, it is likely that competency evaluations underestimate children's understanding of truth/lie concepts and are not likely to be related to a child witness' actual truth-telling (Talwar & Crossman, 2012), raising

When they testify, children are immersed in the linguistically foreign legal system, for which they are often not developmentally prepared (Walker, 1993). They may face the challenge of deciphering legal jargon, as well as linguistically complex questions that they are ill equipped to answer (as are many adults; Carter, Bottoms, & Levine, 1996; Perry et al., 1995). However, this might not undermine children's credibility, as one study of trial transcripts found that complex questions asked by defense attorneys predicted convictions, rather than acquittals in those cases (Evans, Lee, & Lyon, 2009). Nevertheless, children also may be required to offer testimony on concepts that are developmentally beyond their

Lindsay, 1998).

(Pipe et al., 2007).

reconstruct events from memory.

questions about their usefulness.

However, despite their efficiency, event scripts can interfere with recall of specific, *individual* instances of an event, such as abuse (Powell & Thomson, 1996). This could be problematic for a testifying child witness asked to describe and differentiate between multiple abuse incidents – perhaps because the defense is seeking to present alibi evidence. If the incidents were so similar that they became script-like for the child, it could be extremely difficult for the child to differentiate the events or to identify a unique incident at a specific date and time (Roberts & Powell, 2001), potentially undermining their credibility.

**Memory storage.** Children's memory capacity grows with age (Gathercole, Pickering, Ambridge, & Wearing 2004), such that they become increasingly able to recall larger amounts of information over time. For child witnesses, this can become visible when they are asked to provide free recall reports in response to open-ended questions, such as "What happened?". Younger children tend to provide very little event detail when asked such questions. Nevertheless, even young children are capable of accurately recalling and reporting their experiences from memory (Goodman, 2006).

An important moderator of children's ability to store memories is the amount of time between encoding and retrieval. With longer delays, child witnesses have greater difficulty providing complete and accurate memory reports, particularly younger children, as their memory traces seem to decay more rapidly than older children or adults (Brainerd & Rayna, 1995). However, offering children an opportunity to recall an event after a short delay might help to inoculate their reports and maintain them over time (Peterson, Pardy, Tizzard-Drover, & Warren, 2005), particularly when the event context is reinstated during the interview (La Rooy, Pipe, & Murray, 2007). This suggests that the development of children's memory storage facilitates better witness reports with age. It also suggests that there are means of helping even young children to overcome some of their memory shortcomings.
