Effects Of Domestic Violence And Evaluation Of Children'

Child Health And Human Development

Investigators

Linked publications & trials

Abstract

This program involves the development and assessment of techniques for enhancing the informativeness of child witnesses and for evaluating the credibility of their accounts. Most of our studies focus on the relationship between interviewer practices and the quality of information provided by young children, confirming that open-ended questions elicit longer and more detailed responses than more focused questions. Information elicited in response to open-ended prompts of recall memory is also more likely to be accurate in both analog and forensic contexts. Such findings have strengthened the generalizability of the results obtained in many laboratory studies. In research conducted in collaboration with investigative agencies in the United States, United Kingdom, and Israel, we have shown that, regardless of children's ages, interviewers can increase the length and richness of children's accounts by following protocols we designed to probe recall memory and reduce the reliance on more focused questions, which are more likely to elicit erroneous information. Use of the NICHD Investigative Interview Protocol dramatically increases the amount of information retrieved from four- to 13-year-old alleged victims responding to open-ended, free-recall, prompts. Although preschoolers are often deemed incapable of responding to open-ended prompts, our research shows that a similar proportion of details can be elicited by using open-ended prompts from children as young as four years and as old as 13 years and that action-based cues are consistently more effective than all other types of cues, regardless of age. There are, however, important age differences in the types of information that children provide. For example, ongoing analyses of interviews show steady increases with age in the amount and quality of information children provide about the timing of events. This research has provided unique insight into children's developing appreciation of temporal information about experienced events, whereas most research on the understanding of time has involved laboratory research. A study exploring references to temporal attributes made by alleged victims and interviewers in the context of court interviews is currently conducted in collaboration with Friedman and Lyon. All data has been coded and is being entered into SPSS analyses files. Adaptations of the NICHD Investigative Interview Protocol for use with young witnesses and alleged perpetrators are currently being used in the field as well. Our research has shown that young witnesses recall as much information in total, as well as in response to open-ended prompts, as do alleged victims. Suspects tend to be more reluctant, but those who agree to talk provide as much information about their experiences as age mates who are alleged victims and were very responsive to free-recall prompts, although interviewers used more risky focused prompts when interviewing suspects rather than alleged victims. Comparisons of victims? and offenders? accounts of the same incidents of abuse are currently being analyzed. Effects of different types of interviewers? suggestive prompts, in terms of the inclusion of undisclosed input, and of repeated interviews on children?s responses are being explored in other currently ongoing field research. The first study involves a systematic assessment of children?s responses to different types of interviewers? suggestive prompts in the course of non-protocol forensic interviews, exploring the effects of age and prompt type on such responses. We developed two detailed coding schemes for identifying interviewers? suggestive prompts and categorizing them based on the suggestive input introduced by the interviewer (e.g., undisclosed information, implicit or explicit assumptions, confrontation using external information source) and the extent of pressure exerted on the interviewees, and for differentiating children?s responses to suggestive prompts, based on their relationship to the eliciting suggestive input introduced by the interviewer (e.g., confirming, contradicting). Analyses are being conducted. The second study investigates the advantages and disadvantages of repeatedly interviewing 5- and 6-year-old children about past experiences. We found that children recalled new (previously undisclosed) information in repeated interviews. The timing and context of the interviews was found to influence the accuracy of the new information however. When repeated interviews occurred soon after the event in question, when little forgetting had taken place, new information was highly accurate (92%). Under the same conditions, following a 6 month delay, new information was highly inaccurate (54%). Further investigation of recall after a 6-month delay showed that by using context reinstatement the accuracy of new information increased (72%), relative to when no context reinstatement was used. In other ongoing research, we are exploring the extent to which use of the protocol facilitates decisions and interventions designed to prosecute offenders and protect children. Given, however, that many children do not disclose suspected abuse when interviewed, we are also exploring case characteristics and the dynamics of interviews with children who do and do not make allegations or make allegations only reluctantly. We hope to understand the factors that lead children not to report abuse that they actually experienced. These studies, too, should help us develop procedures that can be implemented nonsuggestively in forensic settings in order to enhance the sensitivity and specificity of conclusions drawn from investigative interviews. Child Maltreatment Another program of research is concerned with the effects of child and spouse abuse on the development of children and adolescents. Research has focused on clarifying the short and long-term effects of being a victim of parental physical abuse, a witnesses of inter-parental abuse, and an abused witness (victims and witnesses), in comparison to no-violence controls. Our previous studies in the domain of family violence show that the effects of abuse on children?s behavior problems differ depending on the form of the abuse and children?s age and gender, but results across the literature have been mixed, partially due to small sample sizes. Thus, a mega-analytic study was undertaken that combined the raw data from multiple studies that used the Child Behavior Checklist (CBCL) as an outcome measure. Our combined, cross-sectional data set of 1870 4- to 17-year old boys and girls provided adequate power for testing the effects of different forms of abuse and multiple moderating variables on children?s behavior problems. Results confirmed that abused witnesses are at greater risk for problems than children who experience only one form of abuse or no abuse, and that children?s age, though not gender, moderate the effects. These results, together with our previous findings, underscore the need for longitudinal research on the effects of family violence, and for information about abuse history and children?s adjustment from multiple informants with differing perspectives. We have also learned, however, that, regardless of the type of abuse, age, or gender, the majority of children do not show clinically severe behavior problems. Even in the abused-witness group from the mega-analysis study, for which odds of severe problems were consistently highest, only 28% to 50% of the children in any age/gender group had clinically significant behavior problems. By the same token, clinically significant behavior problems were evident in 14% to 35% of the children not exposed to family violence, underscoring the fact that family violence is only one of many factors that adversely affects children?s adjustment.