As I was preparing my presentations for today’s seminars, I recalled a comment that was made by a college chemistry professor during one of his lectures. He was discussing some of the problems in doing research, and he said that the terms “hard science” and “soft science” were really misused. It was his firm belief that research such as psychology, or sociology, or political science were really the “hard sciences;” that it was much more difficult to do good research in these areas than in disciplines like chemistry or physics, simply because you couldn’t do the research in a laboratory and it was much more difficult to isolate the variables one is researching from the other variables in the environment.
Besides, I hope, giving you some useful information on the topics that we will be discussing, I hope that you will also leave today realizing that, for all of these topics, the subtleties and nuances associated with them are very important; that much of what we need to discern are varying shades of gray, and that little is black and white.
CSA and Interviewing Methods for Children
Questions concerning the reliability and credibility of children's accounts are often raised in adversarial legal proceedings, and these have fostered a highly contentious debate concerning the value and limitations of children's testimony (Lamb, 1994). Much of the research dealing with interviewing children and the reliability of children's testimony is the result of the increased attention and increased reports of child sexual abuse over the last 30 years. Most of the examples I give on improper interviewing techniques will center on child sexual abuse. However, the interviewing techniques that have been developed to properly interview children regarding possible child sexual abuse are valid for any circumstance.
Bruck and Ceci (1995) note that the current research indicates that certain interview conditions have a high risk of contaminating young children's reports. Contrary to previously-held beliefs, errors made by children are not limited to errors of omission (i.e., failure of the child to report important events).
Several factors affect the likelihood of children giving misinformation about personal events. Foremost among these are the interviewing methods which are used in obtaining the testimony. Repeated questioning, delayed questioning, suggestive and misleading interviewing, the emotional tone of the interview, and the status of the person conducting the interview can all have a significant negative impact on the testimony which is obtained from the child. Furthermore, if the interviewer has a bias, the interviewer's methods of questioning may be adversely affected and the child's response or testimony might be consistent with the interviewer's bias.
Improper Interviewing Methods and Their Negative Effect on Children's Testimony
Interviews with children, whether in suspected sexual abuse cases or for other reasons, far too often are conducted using improper interviewing techniques. Mapes (1995) has identified a continuum of questioning methods, ranging from spontaneous disclosure by the child with non-leading techniques, through minimally leading, moderately leading, to maximally leading techniques.
Mapes notes that if the child is presented with inaccurate post-event information, interpretations, or explanations, the misleading information will modify the child's memory of the event being recalled. Generally speaking, the younger the child, the more likely the child will accept someone else's interpretation of an event if it bears a resemblance to the original event. Numerous studies demonstrate that improper interviewing methods may negatively affect the reliability of children's testimony.
The New Jersey Supreme Court, in State v. Michaels, relied on the social science literature on child witnesses and "accepted as its central inquiry whether the interview techniques used by the State in this case were so coercive or suggestive that they had a capacity to distort substantially the children's recollections of actual events and thus compromise the reliability of the children's statements and testimony based on their recollections (McGough, p. 371).
After the use of improper interviewing methods, a determination of what, in fact, actually happened to the child becomes virtually impossible.
Unfortunately, the legal system, according to McGough (1995), does little to insulate a child from this susceptibility; on the contrary, it compounds and undermines a child's resistance. Substantial intervals often occur between the observed events and testimony at trial. Multiple pretrial interviews are conducted. Interviewers who are authority figures (police officer, prosecutor, parent, or counsel) may consciously or unconsciously influence the child's accounts. In light of these factors, substantial impairment and distortion of a child's memories seem inevitable.
Repeated Questioning
Several studies show that asking children the same question repeatedly within an interview and across interviews often results in the child changing his or her answer (Ceci & Bruck, 1995; Poole & White, 1991). Preschool age children are particularly vulnerable to the contaminating effects of repeated questioning. Bruck and Ceci (1995) note that children seem to reason, "The first answer I gave must be wrong, that is why they are asking me again. Therefore, I should change my answer" (p. 279). They cite studies in which children who were questioned in a contradictory manner quickly changed their stories to conform to the suggestions and beliefs of the interviewer. Suggestions planted in the first interview session were quickly taken up and mentioned in the second interview session. This means that if a child is originally interviewed using improper methods, the "testimony" of that child can be contaminated through all subsequent interviews.
In the Michaels case, the defendant, Kelly Michaels was accused of hundreds of counts of sexual abuse and later convicted of 115 counts of sexual abuse against 20 three- to five-year old children. She was said to have licked peanut butter off children's genitals, played the piano while nude, made children drink her urine and eat her feces, and raped and assaulted these children with knives, forks, spoons, and Lego blocks. She was accused of performing these acts during regular school hours over a period of 7 months. During this time, none of the alleged acts were noticed by staff or reported by children to their parents. Ms. Michaels was later sentenced to 47 years in prison, but was released on bail after 5 years (Ceci & Bruck, 1995).
In addition to conversations likely held between the children and their parents, the children were repeatedly interviewed by a child therapist, by an investigator for the Division of Youth and Family Services, and by an expert for the prosecution. Bruck and Ceci (1995) analyzed available audiotapes of the interviews and describe the highly suggestive, coercive, and inappropriate interview techniques. They note that the children who ultimately became witnesses in court were interviewed dozens of times before they "testified." The risks of contamination of the children's testimony were mammoth. The interviewers' misconduct in conducting the interviews furnished the pivot of the New Jersey appellate's decision to reverse Kelly Michaels' conviction (McGough, 1995). (For a more detailed description of that case and other famous day-care cases, see Bruck and Ceci [1995] and Ceci and Bruck [1995]).
Unfortunately, even today, in many parts of the country, repeated interviewing is far more likely to be the rule rather than the exception in suspected sexual abuse investigations.
Suggestive or Misleading Questioning
A number of studies show that repeatedly giving children misleading information in a series of interviews can have serious effects on the accuracy of their later reports. Not only can misinformation become directly incorporated into the children's subsequent reports, it can also lead to fabrications or inaccuracies which do not directly mirror the content of the misleading information or questions. It is now known that misinformation presented in initial or early interviews can taint the testimony that the child gives later. In other words, as Bruck and Ceci (1995) note, the "primary evidence has been destroyed" (p. 272).
Investigators approaching a crime scene are taught not to touch objects, weapons, or features of the surroundings because of the imminent risk of destroying the evidence. Yet investigators who engage in repetitive child sexual abuse cases appear to have no knowledge whatsoever of the potential of their own interviewing methods in destroying the evidence, i.e., the child's testimony.
Emotional Tone of the Interview
Children may give incorrect information to misleading questions when the interviewer creates an emotional tone of accusation (Bruck & Ceci, 1995). In some studies, when an accusatory tone is set by the examiner, children may fabricate reports of past events even in cases when they have no memory of any event occurring. Goodman, Wilson, Hazan, and Reed (1989) demonstrated that "an atmosphere of accusation" can be created, in which children will produce reports of abuse when none has occurred.
Effects of Peer Pressure
In some cases, interviewers have told children that their friends have "already told" about being abused, even when this is not true. Bruck and Ceci (1995) note numerous examples of the use of peer pressure in the Michales case, both by telling the child what other children had allegedly reported, and by threatening to tell other children that the child being interviewed had been uncooperative. In one example, the interviewer said, "Oh, come on, we talked to a few more of your buddies. We talked to everyone now. And everyone told me about the nap room, and the bathroom stuff and the music room stuff and the choir room stuff and the peanut butter stuff and everything. Nothing surprises me anymore."
Effects of Being Interviewed by Adults with High Status
Young children are sensitive to the status and power of their interviewers (Bruck & Ceci, 1995). As a result, they are especially likely to comply with the implicit and explicit agenda of such interviewers. Children are more likely to believe adults than other children, they are more willing to go along with the wishes of adults, and are more apt to incorporate adults' beliefs into their reports.
Children's testimony is susceptible to contamination from several factors.
Effects of Interviewer Bias on Children's Reports
Several researchers note a distinct "confirmatory bias" on the part of interviewers, i.e., that abuse has occurred and the purpose of the interview is to get the child to admit the abuse. Bruck and Ceci (1995) observe that some interviewers blindly pursue a single hypothesis that sexual abuse has occurred and fail to explore rival hypotheses that might explain the behavior of the child. When Bruck and Ceci reviewed tapes in the Wee Care case, they found them replete with interviewer bias. They report that when children said something that was not part of the interviewer's hypothesis, the interviewer ignored it.
When interviewers of the children in the Wee Care case were asked how they conducted their interviews, one admitted that "her goal was to induce the children to discuss sexual abuse" (Bruck & Ceci, 1995, p. 278), while another primary interviewer stated, "The interview process is in essence the beginning of the healing process" (p. 279). A third of the primary interviewers said, "So you open the interview in an effort to disempower (the accused person) of these super powers that she allegedly has . . . and also to let the children know that telling about these things was okay and they would be safe" (p. 279). Such statements reveal a clear assumption that abuse has occurred and demonstrate that these interviewers held preconceived biases that pervaded their interviewing methods and subsequent conclusions.
In a study conducted by Pettit, Fegan, and Howie (1990), children who were interviewed by biased interviewers gave the most inaccurate information. If an interviewer's belief was contrary to what the child had experienced, interviews were characterized by an overabundance of misleading questions which resulted in children providing highly inaccurate information.
It is generally recognized that the most reliable and accurate information is obtained from children who are responding to open-ended questions designed to elicit free narrative accounts of events that they have experienced. When direct questions are asked, they should be formulated as non-suggestively as possible, using developmentally and individually appropriate vocabulary and sentence construction. Repeated, highly-leading, or suggestive questions asked in an accusatory manner are most likely to promote distortion on the part of the child and may introduce details that are incorporated into and contaminate subsequent accounts (Lamb, 1994, p. 155).
Risks of Inaccuracies in Hearsay Testimony in Child Sexual Abuse Cases
A great deal of literature reveals that the credibility and reliability of testimony obtained from children may be compromised when improper interviewing methods are used. Several major concerns emanate from this discussion. As noted above, repeated interviews or interviews employing improper questioning methods can irreversibly taint the evidence, i.e., destroy the original testimony of the child, as early as the second interview. Improper questioning, even if the interviewer is unaware of using improper methods, can also destroy the primary evidence. Any attempt that is made to re-evaluate, re-examine or cross-examine the testimony of the child for presentation to a jury after several interviews may fail, since accurate information may no longer be obtained. The more interviews to which the child has been subjected, the greater the likelihood that the child's testimony has already been contaminated.
Hearsay testimony from mental health and other professionals plays a key role in many cases. Compounding the risks of contamination from improper methods of securing testimony from the child, the resultant hearsay testimony is fraught with dangers of inaccuracy. Mental health practitioners or others who testify on behalf of the child are not necessarily accurate in their statements about what the child testified and may misinterpret what the child said.
McGough (1995), addressing the issue of hearsay in cases of child abuse, notes that most child abuse trials do not depend solely on in-court testimony of the child witness. Although rules of evidence prefer in-court testimony, many types of "hearsay" — out of court statements — are admissible. In a great number of cases, the child does not testify directly, but a medical or mental health practitioner testifies regarding alleged reports obtained from the child, which is hearsay testimony. In such cases, the child is considered, by legal criteria, to be "psychologically unavailable" to testify directly. According to McGough (1995), there are three primary dangers when hearsay is received as trial evidence:
1. Faulty memory — the danger that the declarant will forget key material;
2. Ambiguity — the danger that the meaning intended by the declarant will be misinterpreted by the witness, and hence, the fact finder;
3. Misperception — the danger that the declarant misjudged, misinterpreted, or misunderstood what she heard or saw;
Faulty Memory
The first potential danger to the accuracy of hearsay is the faulty memory of the declarant. A number of researchers (Bruck & Ceci, 1995; Ceci & Bruck, 1993; Loftus & Davies, 1984; Loftus & Ketcham, 1991) have provided scientific evidence which bears on the potential danger of faulty memory on the part of the witness. This is especially true for children's memory but is also true for health care professionals. Difficulties with memory could fill and has already begun to fill volumes. Summarizing the memory literature in a few sentences is difficult, but generally, memory can be seen as constructive, often inaccurate, and susceptible to numerous influences which reduce accuracy. The reader is directed to the published works of these authors for more complete analysis. Such studies are intriguing from the scientific standpoint, but appalling from the legal standpoint, with heavy implications for the judicial system.
Here the possibility of faulty memory on the part of the hearsay declarant, i.e., the mental health professional, must also be considered. Bruck and Ceci (1995) have pointed out that hearsay declarants' statements are likely to be affected by faulty memory. According to these authors, adult interviewers are often inaccurate in recalling what was said or what took place in an interview. Neither the interviewer's remembrance of the interview nor the written documentation of the interview may be accurate. In general, adult remembrance of interviews is inaccurate. When asked to recall conversations, most adults may recall the gist, but they cannot recall the exact words used, nor the sequences of interactions between speakers. Bruck and Ceci (1995) observe that written summaries of unrecorded interviews may be subject to a number of distortions, and conclude that summaries of missing interviews do not substitute for an audio or video recording.
Ambiguity
The second danger affecting hearsay testimony, ambiguity (McGough, 1995), is also prominent in child sexual abuse investigations. For example, for mental health professionals to ask a four-year-old child if someone "touched" him or her invites inaccuracy. The child who is four years old often cannot distinguish a general, affectionate touch from a sexualized touch. The child may honestly relate, on the basis of a concrete interpretation of the word "touch," that he or she was "touched." The investigator may interpret this as meaning that the child was touched in a sexual way, whereas the child may have been referring to a non-sexual touching. Immediately, an allegation of sexual abuse arises, with tremendous personal and social costs to all involved.