Introduction

In many child sexual abuse cases the report of a child is the only evidence to implicate a suspect and is usually the only indication that abuse occurred. Also, the ability of adults involved in the case to accurately assess the children’s testimony is equally as important.

Children’s memory over long delays: Highly biased questioning procedures have shown that false information may be elicited by children; therefore, statements made by children after specific questioning techniques should be looked at closely. Children’s reports of abuse may be influenced by leading questions rather than actual victimization. Another aspect to look at when basing an assessment off of a child’s memory is the possibility of reconstructive errors. As time progresses children use "scripts" (general semantic knowledge) to account for information that is hard to remember.

Children’s suggestibility: Children may be resistant to suggestibility concerning abuse versus other types of information because it is generally a significant experience, as well as personally threatening. On the other hand, the opposite may occur if children are prone to sexual fantasy or are as easily misled in responding to abuse-related questions as they are to questions about other information. Phase 1 involves a biased interview and accusatory environment to measure the likelihood that children produce erroneous reports or even false accusations of abuse.

Evaluations of children’s statements: Phase 2 investigates the ability of professionals and nonprofessionals to evaluate children’s accuracy when children were interviewed about abuse following a four-year delay.

Professionals’ perceptions and assessments of child abuse and child accuracy: Based on previous studies, professional’s attitudes toward child sexual abuse cases seem to be correlated with: being a professional in specific disciplines, like social work; length of experience in certain professions; and past personal experience of child sexual victimization.

Laypersons’ perceptions of children’s credibility and accuracy: Even though older children are perceived to be more credible than younger children, cases where honesty, innocence, and naivety are of concern rather than cognitive competence give younger children more credibility than older children and/or adults. Also, children’s confidence and presentation style make a bearing on laypersons’ judgments; confident children are viewed as more accurate regardless of their actual accuracy.

Predictions: Phase 1: 1. Older children would be more accurate and less suggestible than younger children, 2. Children would be less susceptible to abuse-related than nonabuse-related false suggestions. Phase 2: 1. There would be as significant correlation between professionals’ estimates of children’s accuracy and children’s actual accuracy, but that the same correlation for laypersons would be nonsignificant, 2. Compared to untrained nonprofessionals, social workers would be less likely to rate the children as abused and less likely to rate the case as "founded", 3. A significant correlation would exist between professionals’ and undergraduates’ ratings of older children’s accuracy and their actual accuracy, but that this same correlation would be nonsignificant when participants’ rated the accuracy of younger children, 4. Adults with abuse experience would be more likely than other adults to rate the children as having suffered abuse.

Phase 1

Participants included 15 children, one group of seven 7-year-olds and another group of eight 10-year-olds, who had experienced a mundane social interaction 4 years earlier. In the original study (4 years earlier), 3- and 6-year-old children individually conversed and played an arm-coordination game for 5 minutes with an unfamiliar male confederate. After 4 or 5 days, each child was asked about the event by specific, correctly leading, and misleading questions. In Phase 1 of the present study, children were questioned about 4 years after their interaction with the confederate. They were asked to recall as much as possible about what took place. Children were then shown a picture of the confederate to cue their memory. Again, they were asked to recount what happened. After that, 17 "nonabuse" and 11 "abuse" questions were asked, followed by another recall attempt. Next, the children were given a 15-minute "cookie and juice break." After the break, the children were asked a few times to recall what happened, twice with the aid of puppets. Children were finally asked four more nonabuse and four more abuse questions.

Nonabuse questions concerned actions related to the game, if the children had watched cartoons or received a toy, etc. Abuse questions concerned actions associated with child abuse. Also, statements and questions intended to create an atmosphere of accusation were included.

Results: Children recalled as much incorrect as correct information, but the amount recalled was low in both cases, which was expected from a group of children who mostly failed to remember the target event. None of the children recalled sexualized information and little of the incorrect information presented in the questions intruded into children’s later recall. Children did answer abuse questions more accurately than nonabuse questions. However, a few children falsely agreed with suggestions of intimate contact that could cause suspicion in a real abuse investigation.

Phase 2

Investigates professionals’ and nonprofessionals’ a) perceptions of children’s eyewitness memory and b) abilities to assess the accuracy of children’s statements.

Participants included 24 social workers (16 females and 8 males), 32 nonprofessionals (16 males and 16 females). Based on the children’s accuracy scores and videotape sound quality, 8 out of 15 children were chosen for inclusion in Phase 2. Participants watched, either alone or in small groups, one of the eight children, so that each child was viewed by the same amount of participants. They were led to believe that this was a real child sexual abuse case and the results were pending. Participants read a written case scenario and were given a written transcript of the interview to review as needed, yet they were told to pay close attention to the videotape. The participants filled out a questionnaire after viewing a videotape of one child to make judgments concerning the child’s accuracy, believability, confidence, and similarity of the child’s responses to those of sexually abused children.

Results: Although the sample size was smaller, the children’s memory reports conformed to patterns reported for Phase 1. Adults had difficulty assessing the accuracy of children’s eyewitness reports. The expectation that they would be able to predict older children’s accuracy but not younger children’s accuracy was not met. However, participants were able to predict the accuracy of children’s answers to abuse questions. Child gender did not significantly affect social workers’ judgments but did significantly affect nonprofessionals’ judgments, indicating that boys were more accurate than girls; however, boys and girls’ accuracy were about equal. This shows that professionals are generally more reliable in making evaluative assessments than nonprofessionals. Both groups rated the younger children as having been abused more than the older children, yet they were unsure that the children had been victimized. Even though professionals were less likely to rate the children as abused and did not judge the case as "founded", those who reported personal experience with abuse were more likely to do so. The opposite was found for nonprofessionals; those who had personal experience with abuse were less likely to think abuse occurred and less likely to judge the case as founded.

General Discussion

After a 4-year delay, children were not able to remember much about the mundane interaction and were not easily led to suggestible false reports of abuse. Although the adults were not impressive in assessing the statements of older and younger children, professionals and nonprofessionals were able to reliably evaluate children’s answers to abuse questions. Also, nonprofessionals were more likely to attribute younger children’s failures to report abuse to fear than professionals, and they were more biased by gender in their assessments.