In the 1980s there was a substantial increase in reporting of child sexual abuse allegations in cases all across the United States. Media attention to highly charged cases, such as the Jordan, Minnesota incident and the McMartin preschool trials, rode the wave of popular and media hysteria. The result led to intense criticism of the methods used by those conducting the interviews. Articles by attorneys criticized the heavy-handed and leading questions used by the interviewers. The McMartin case was notable in the way the defense was able to demonstrate that the procedures used by Kee MacFarlane of Children’s Institute International were faulty. The videos of the child interviews in these cases provided ammunition for defense attorneys who criticized these techniques. The problems outlined in this article aren’t confined to the United States, but appear to be common throughout much of the world.
The number of reported incidents of child abuse in the United States is now roughly about 3.6 million assaults per year (National Center on Child Abuse and Neglect, (2007).
Training of police and child protective investigators since the 1990s has improved with respect to the heavy-handed leading questioning and badgering of children in which interviewers, consciously or unconsciously, prompted children to give the answers the interviewer was looking for. However, “confirmatory bias,” while less obvious today, continues to dominate prosecution-oriented interviews by police and social services agencies.
Today, the interviewer is less likely to confront or bribe the child to give the desired answers; instead interviewers talk over the child or change the subject when the child begins to give answers not favorable to the interviewer’s objective. Interviewers may ignore contradictory responses that don’t fit their view of the case. In this process, interviewers may see their mission as confirming or validating the key allegations needed to prosecute the case. Information that contradicts the allegations is ignored or discarded.
The process begins to go wrong when forensic examiners are in contact with investigating detectives, or other officers, agency workers, and/or family members with definite agendas. Police and social services agencies ignore the fact these contacts tend to contaminate, or “taint,” the interviewer’s attitude toward of the case. Close personal working alliances tend to confound the process and taint the information obtained. To date, however, research on the impact of this problem is lacking.
Most agencies that conduct forensic Interviews tend to “over interview” children. Mental health professions now recommend interviewing children a minimal number of times. Most experts suggest no more than three interviews; a small number accept a maximum of six. The consensus is that more than three interviews of a child who was allegedly victimized, including informal interviews by a parent, risk modifying the child’s statements, thus making them unreliable.4
It takes great effort to compensate for problems associated with multiple interviews of children. However, they can be ameliorated by limiting the number of interviews to three. Law enforcement contact with the child should be confined to one well-trained specialist. Parents should be cautioned against satisfying their curiosity by repeatedly questioning the child.
The problems of the police-child contacts go beyond the number of interviews; a police station isn’t an appropriate place to interview a child. The setting can be intimidating and send misleading signals to the child, especially if the officer is in uniform, or wearing a badge, and carrying handcuffs and a gun. These factors are problematic and should be avoided. Ideally, there should be as few distractions as possible.
Some interviewers have video cameras in the room. They seem to believe that they should tell the child they’re “making a movie,” which is a very bad idea. This introduces an element of play acting to the interview. Does the child say what really happened, or has the child adopted make-believe, as if it’s a movie? Children don’t have an intrinsic need to know why the session is being taped. Young children lack the skills to make that kind of cognitive distinction between reality and play. By the same token, it’s important not to clutter up, and further contaminate, the interview by concluding that it’s necessary to do this so that “child testimony can be avoided.” That’s not why we’re here today! Most experts agree that ideally children should not be made aware that a video is being made.
Confirmatory BiasWe tend to see what we expect to see. This is especially true of forensic interviews when the interviewer has a collegial relationship with other members of the agencies who have had contact with the child, as well as family members. A work dynamic forms in a police or social services agency in which other individuals involved in the process feel compelled to “fill in” the interviewer on the “background” of the case. This only serves to enhance confirmatory bias. A clear agency or inter-agency policy against this should be posted for all forensic interviewers to see. The forensic interviewer should have only basic information as to what is alleged and some demographic information on the child. The person conducting the forensic interview should be as unbiased and uncontaminated as possible.
Nothing contaminates a forensic interview of a child as much as a bias that prompts someone to become a “validator” on a mission to extract information necessary to successfully prosecute the accused. They ignore or reject all information that doesn’t fit in with their preconception of the case and focus the interview toward getting information to support charges against the accused. They ignore children’s exculpatory statements because their sole mission is confirm the allegations, not to elicit the child’s own statements.
A common and severe problem for forensic interviewers is the danger of becoming jaded. Interviewers who become involved in these cases will, over time, build a set of prejudices based on their experiences. This causes them to project their past experiences into the contemporary interview. As well-intentioned as the individual interviewers may be, by filtering how they conduct a forensic interview based on their extensive past experience, they are likely to contaminate the present interview by confirmatory bias.
While total objectivity is impossible, we need to find some mechanism to identify this condition before it becomes irredeemable. New testing techniques are needed to determine the level of the contamination of interviewers’ objectivity, and some manner to tell us if and when they can be rehabilitated to resume the work. When interviewers detect in themselves that they are “anticipating” the child’s responses, and arrive at a point of, “I know what you mean” in their own minds (because they have done this so many times before) the risk of contamination eventually becomes inescapable. The problem is that you can arrive at a point where you are seeing things that aren’t really there.
Flawed Environment (Law Enforcement )
The process within most agencies is problematic because the premise they start out with doesn’t help find out the truth. It’s aimed at securing a conviction, whether justified or not. The process is designed as a tool for prosecuting the accused, not for finding the truth. It is outcome oriented, a process in which truth becomes secondary to getting information to support the prosecution’s goal.
Most agencies reason: “It’s not our job to do the work of the defense.” So if exculpatory information comes up in an interrogation, the interviewer will ignore it and get away from it as quickly as possible. If the child starts to make exculpatory statements, the interviewer then talks over the child and reverts back to a point where the child gave “helpful” responses.
When we examine police interviews today, we commonly see the total absence of any effort to ascertain whether the child might have been coached by someone prior to the interview, or the extent to which such contacts may render the child’s statements unreliable. There are a number of ways to detect and overcome such external influences. Dr. William O’Donahue from the University of Nevada at Reno outlined these techniques in a training program he developed under a grant from the United States Department of Human Services.5 But forensic interviewers all too often ignore these problems because they lack the training and skills to use them. And they go against the mission they have undertaken.
One suggestion is to ask the child if there is anything important that they need to tell. In many cases a child who has been coached will blurt out details they were told, because which they don’t want to forget what they were told. Sometimes the coaching is well intentioned, sometimes it isn’t. The forensic interviewer should separate one from the other.
The interviewer should pay close attention to things the child may say, such as comments about what is going to happen to the accused. “Will he be arrested?” “Will he go to jail?” The interviewer should explore these areas to determine how much influence has been exerted on the child. The goal is to ascertain how much information might have been transferred to the child. Children may also give clues by saying, “(naming a person) said that …” This should be, but often isn’t, a red flag that should be pursued.
A good forensic interviewer needs to begin with this information in order to have quality control over what is to come later — the meat of the allegations. A good forensic interview should have a solid foundation; the job should be seen not as only to get statements to aid the prosecution, but to get statements that are accurate. Forensic interviewers need be aware of any external influences on the child, such as a divorce, child custody proceedings, or an overly involved grandparent or other family member.
In a recent example, a Midwestern city police detective who does the forensic interviews was dealing with a case in which a specific charge was indecent exposure. The child gave contradictory statements about the events as to whether the alleged perpetrator knew she was in the room or not. The interviewer ignored the child’s description of the man’s penis being flaccid. The arrest documents, however, claimed he had an erection. In that state an essential element of the crime is that the exposure was deliberate, the purpose being for his sexual gratification or to induce a sexual interest in the child.
Not only did the interviewer ignore the child’s statement and talk over it and change the subject, she missed completely the child’s flat affect. There were many clues that screamed for further follow-up. But these were ignored, most likely because the interviewer knew it was a black hole. It was a situation that in all probability would have made the child’s statements unreliable and sunk the case.
Ideally, the forensic interviewer is neutral. This is sometimes virtually impossible, as people who work in the field of child protection often become jaded and bring their expectations into the room with them as they conduct the interviews. It is a form of wish fulfillment or a self-fulfilling prophecy. They find what they expect or want to find. They lose professional detachment and assume an advocacy role. The only way to guard against this is to rotate people in the job frequently, and not assign them to do field investigations of child abuse if there is any hope to bring them back after a break.
The main issue, however, is the training of the forensic interviewers who work with the children. It is important they have interviewing skills and recognize their own biases. They must zealously guard against being contaminated with too much biased external information, such as through fellow police officers or case workers who want to help by giving them information they feel is important. This “team” concept is flawed when it comes to the child interview. The interviewer needs to know minimum facts as to what is alleged and minor demographics on the child. Too much additional information may result in confirmatory bias unless the interviewer is aware of this danger and consciously explores all possible hypotheses. But few real world interviewers do this. Perfect neutrality is probably the impossible dream, but perfect bias is the perfect nightmare.
Additional Resources and Credit for this article:
Footnotes 1 Schreiber, N., Bellah, L. D., Martinez, Y., McLaurin, K. A., Strok, R., Garven, S., & Wood, J. M. (2006). Suggestive interviewing in the McMartin Preschool and Kelly Michaels daycare abuse cases: A case study. Social Influence, 1(1), 16-47. [Back]
Ceci, S. J., & Bruck, M. (1995). Jeopardy in the courtroom: A scientific analysis of children’s testimony. Washington, DC: American Psychological Association, Chapter 8.
Protecting Children Undergoing Abuse Investigations and Testimony Approved by Council, February 1986, American Academy of Child and Adolescent Psychiatry.
Dr. O’Donahue’s training program is the best one I have seen. There are other good ones, but the O’Donahue model is the best in dealing with difficult cases.
Elliott, A. N., O’Donohue, W. T., & Nickerson, M. A. (1993). The use of sexually anatomically detailed dolls in the assessment of sexual abuse. Clinical Psychology Review, 13(3), 207-221.
Wolfner, G., Faust, D., & Dawes, R. M. (1993). The use of anatomically detailed dolls in sexual abuse evaluations: The state of the science. Applied & Preventive Psychology, 2, 1-11.
Bourg, W., Brokerick, R., Flagor, R., Kelly, D. M., Ervin, D. L., & Butler, J. (1999). A child interviewer’s guidebook. Thousand Oaks, CA: Sage Publications.
Ceci, S. J., & Bruck, M. (1995). Jeopardy in the courtroom: A scientific analysis of children’s testimony. Washington, DC: American Psychological Association.
Doris, J. (Ed.). (1991). The suggestibility of children’s recollections: Implications for eyewitness testimony. Washington, DC: American Psychological Association.
Eisen, M. L., Quas, J. A., & Goodman, G. S. (Eds). (2002). Memory and Suggestibility in the Forensic Interview. Lawrence Erlbaum Associates, Publishers, Mahwah, NJ.
Kuehnle, K. (1996). Assessing allegations of child sexual abuse. Sarasota, FL: Professional Resource Press.
O’Donahue, W., & Geer, J. (Eds.) (1992). The sexual abuse of children, Volumes 1 & 2. Hillsdale, NJ: Lawrence Erlbaum.
Poole, D. A., & Lamb, M. E. (1998). Investigative interviews of children: A guide for helping professionals. Washington, DC: American Psychological Association.
Robin, M. (1991). Assessing child maltreatment reports. New York: Haworth Press.
Wilson, C., & Powell, M. (2001). A guide to interviewing children: Essential skills for counselors, police, lawyers and social workers. London: Taylor and Frances.
Zaragoza, M., Graham, J. R., Hall, G. C. N., Hirschman, R. & Ben-Porath, Y. S. (Eds.) (1995). Memory and testimony in the child witness. Newbury Park, CA: Sage.
Credit for this article is given to Kenneth R. Pangborn, MS Palm Harbor, FL