Dr. Diana McCoy
  Curriculum Vitae
  Contact Dr. McCoy
  Forensic Psych Pages


Suppression Hearings and Investigator Bias

Diana McCoy, Ph.D.

A police officer in a suppression hearing in a child sex abuse case testified that “the defendant appeared nervous,” with his presumed nervousness at the outset having been preliminary to the investigator’s interrogation of the defendant getting under way.  By so testifying, he insinuated that the defendant’s nervousness, in and of itself, was indicative of guilt.  Since there is no electronic recording of the interrogation, the court must take the officer’s word on the defendant’s demeanor despite the fact that “appearing nervous” takes many forms, may range from mild to severe, and may signify many different things, including being in a state of innocence.

The officer’s subsequent testimony was that since he believed the child’s allegations of abuse as well as believed the defendant’s demeanor signaled guilt, by his own admission he approached the defendant with a presumption of guilt.  This same presumption of guilt, incidentally, was readily communicated to other witnesses this same officer  interviewed, who complained of feeling intimidated by his overbearing manner when they did not support his insistence on the defendant’s guilt. 

An allegedly well-trained investigator whose credentials included years of seminars on how to conduct investigations of this nature, I was stunned to hear that he did not bother to clearly determine the defendant’s meaning of terms used in reference to relevant  human anatomy. It can be very important if “butt,” for example, means buttocks to the defendant whereas it means anus to an investigator. The officer could not at first recall what words the defendant even used. He seemed unfazed when the terms he eventually came up with, that he ascribed as having come from the defendant’s own mouth, turned out to be different from those actually used in his “summary” of the defendant’s statement, despite this reportedly having been written contemporaneously, i.e., coinciding in real time with the defendant’s utterances.

The officer appeared equally unperturbed when it became apparent that he had “blended” the statement of the child with that of the defendant such that he could not clearly recall the anatomical terminology used in either statement by either person, a point he conceded was probably the case.  He asserted that he saw his job as that of putting people in jail, so at the very least there was no hypocritical pretense of fairness, impartiality, or neutrality on his part. 

Neither a transcript of the child’s actual statement nor an electronic recording of this statement was provided the defendant, nor, as already noted, was there an electronic recording of the defendant’s statement.  Whatever transpired in the interviews between investigator and alleged child victim and between investigator and defendant was communicated to the court by a law enforcement officer who seemed well pleased with his supposed ability to discern the truth, based on his interview with the child and his perception of the defendant’s “nervousness,” and thus single-handedly bring the guilty to justice.

What is wrong with this picture?  Let us first consider the issue of the defendant’s alleged nervousness, which, as any well trained investigator knows, signals guilt and hence the likelihood of deception.  Despite the confidence of the police that they are able to use verbal and nonverbal behavioral cues to make accurate judgments, police investigators perform only slightly better than chance, if at all, in detecting deception (Bull, 1989; DePaulo, 1994; DePaul & Pfeiffer, 1986; Ekman & O’Sullivan, 1991; Ekman, O’Sullivan, & Frank, 1999; Koehnken, 1987; Porter et al., 2000), and “training” does not help on a consistent basis (Bull, 1989; Kassin & Fong, 1999; Porter, Woodworth & Birt, 2000; Vrij, 1994; Zuckerman, Koestner, & Alton, 1984; Zuckerman, Koestner, & Colella, 1985).  In a study aimed at assessing the ability of police officers and college students to determine whether videotaped guilty or innocent suspects were lying versus being truthful, law enforcement personnel did not outperform the students.  However, police officers were significantly more confident they were right despite their often erroneous judgments (Kassin & Fong, 1999).

Consider the findings of a recent study (Kassin, Goldstein, & Savitsky, 2003) that found that when investigators presume guilt, they put into motion a process known as “behavioral confirmation” wherein they select more guilt-presumptive questions, use more interrogation techniques such as the promise of leniency and presentation of false evidence, and exert more pressure to get a confession, particularly with subjects actually innocent.  This process, more familiarly known as the self-fulfilling prophecy, tends to lead to aggressive, guilt-provoking interrogations, with the behavior of the suspects in response to such methods that of feeling constrained, thereby confirming the investigators’ assumption of their guilt.  The authors of the study note that erroneous prejudgment of guilt colors the information-gathering process such that plausible denials on the part of the defendant are discounted or misinterpreted.

As Kassin, et al .further conclude, it is hardly surprising that in the face of coercive interrogation techniques defendants become defensive such that they sigh in despair, slouch in their seats, and look away, supposedly classical signs of deception as noted in Inbau, Reid, Buckley, and Jaynes’ Criminal Interrogation and Confession (2001), the bible of law enforcement investigators, now in its 4th edition.  One of the authors of this book, in marked contrast to what is known in the social science literature about the inability of trained investigators to discern truth from lying any better than chance, nonetheless reports that investigators trained in their methods are able to distinguish truth from deception at an 85% accuracy level (http://www.reid.com/service-bai-interview.html).

That confirmation bias is alive and well is indicated in the case above, where the police officer, believing the alleged child victim was telling the truth and confident the suspect was guilty, testified that the defendant admitted to sexually molesting the child.  Another individual present at this very same interrogation, caught off guard at finding himself unexpectedly called as a defense witness, testified to the contrary.  Right to the very end of the interrogation, said this state witness, the defendant consistently maintained, in response to whether particular scenarios presented to him were “possible,” that although anything was possible, nothing had happened but that if it did, it was by accident.  Here we have two people present at the same interrogation, both of them state witnesses, who arrived at entirely different conclusions: one said the defendant admitted guilt and the other said he did not.

It is a common practice, even in First Degree murder cases, for investigators to “summarize” their interrogation findings.  They then present this summary to the often by now cowed defendant to sign, who is typically in such a state of fright, confusion, and agitation that he has little comprehension of what he is actually signing, to his later chagrin.  It then becomes the defendant’s word versus that of law enforcement as to what actually may have transpired in the all too frequent absence of electronic recording.  Of course, it cuts both ways.  An electronic record of a defendant’s statement, properly taken, can be very hard to challenge. 

Despite the best of intentions, human memory is simply not accurate in recording every detail of what takes place in an interrogation, and details count when the stakes, notably life and liberty, are high.  Memory may be inaccurate when there has been a delay of only a few seconds.  This translates to memory being inaccurate when the writer is contemporaneously writing what the defendant is saying while the defendant is in the process of speaking, a practice, at least according to their testimony, favored by investigators.  To the contrary, Ceci and Bruck (1995) found that when adults are asked to recall conversation or passages they have just heard that their memory fades within seconds.  Rayner and Pollatsek (1989) determined that subjects tend to extract the meaning or gist of what they hear but forget the form or the exact wording.  Warren and Woodall (1999) found in their study of experienced interviewers that their notes reflected only 20% of the questions they actually asked, leaving the misleading and potentially damaging impression that the information had been spontaneously provided.

Lamb, Orbach, Sternberg, Hershkowitz, and Horowitz (2000) compared the verbatim contemporaneous accounts of 20 investigative interviews with audiotaped recordings of these same interviews and found that the investigators’ notes misrepresented both the information elicited and the way the information was elicited.  They found that 57% of the interviewers’ utterances as well as 25% of the incident-relevant details provided by those being interviewed were omitted from the so-called verbatim notes.  In addition, the structure of the interviews was inaccurately represented in these accounts.  Investigators systematically misidentified details as resulting from open rather than focused prompts.  What is particularly disturbing about this study is that the investigators were among the best in the field, were well aware of their legal and moral responsibility to fully and completely record the interview structure and content as completely as possible and not to simply summarize, and knew the interviews were being recorded and that their verbatim notes would be compared to these recordings.  The authors conclude, “These results underscore the superiority of electronic recording when the content and structure of investigative interviews must be preserved…interviewers cannot be expected to provide complete and accurate accounting of their interviews without electronic assistance.” 

In addition to the known research findings regarding the virtual impossibility of 100% accurate recall, a summary of an interrogation, aside from what we know about the often confirmatory bias of investigators, simply cannot capture the all-important pauses, nuances of speech, gestures, body language, number of times a question may have been asked, indications of whether the response was given spontaneously or in response to a prompt, and so forth.  The officer above, for example, never mentioned in his “summary” or his testimony that “scenarios” were presented wherein the defendant was asked to respond to the “possibility” of something having happened, a far cry from spontaneously admitting guilt, as was this officer’s testimony.  He never described the young man’s confusion in the face of being bombarded by questions from three interrogators, with this being a severely learning disabled individual with a lengthy, well-documented history of problems processing and understanding verbal material.  Instead, the so-called summary statement presented an image of a much more verbally astute, responsive, coherent, and competent individual than my interviews with him suggested, with the trier of fact now placed in the unenviable position of somehow reconciling grossly contradictory and inconsistent testimony.  Listening to just a few minutes of an audiotape of the defendant’s presentation during the interrogation might have clarified this. 

The implications are all too obvious.  The testimony of law enforcement can have a  devastating effect on the credibility of a defendant to the trier of fact, who needs a true and accurate picture of the totality of the circumstances of the interrogation in order to provide this individual his constitutional right to due process.  Although the family of the defendant who I was evaluating luckily had the financial resources to hire topnotch legal representation, who in turn hired an expert witness in an effort to bring these important issues to the attention of the court, one wonders how many defendants are not afforded this same opportunity and instead silently shuffle off to the penitentiary. 

When police, trained in powerful, coercive interrogation techniques, armed with confirmatory bias as well as unbounded (and unfounded) confidence in their truth-detecting abilities, and motivated by career aspirations, conclude guilt and then misinterpret and/or misrepresent statements made to them by the defendant, as in the case above, they may set into motion a tragic chain of events ranging from increasing the likelihood of a false confession to facilitating an unjust conviction.  All too common scenarios of this nature underscore the need for electronic recording of investigative interviews, preferably videotaped, rather than relying on the uncertain memory and fairness of law enforcement.


Bull, R. (1989).  Can training enhance the detection of deception? In J.C. Yuille (Ed.), Credibility assessment (pp. 83-99). London:  Kluwer Academic Publishers.

Ceci, S., & Bruck, M. (1995).  Jeopardy in the courtroom.  American Psychological Association, Washington, D.C.

DePaulo, B.M. (1994). Spotting lies:  Can human learn to do better?  Current Directions in Psychological Science, 3, 83-86.

DePaulo, B.M., & & Pfeifer, R.L. (1986) On-the-job experience and skill at detecting deception.  Journal of Applied Social Psychology, 16, 249-267.

Ekman, P., & O’Sullivan, M. (1991).  Who can catch a liar?  American Psychologist, 46, 913-920.   

Ekman, P., & O’Sullivan, M. & Frank, M.G. (1999). A few can catch a liar.Psychological Science, 10, 263-266.

Inbau, F.E., Reid, J.E., Buckley, J.P., & Jayne, B.C. (2001).  Criminal interrogation and confessions (4th ed.). Gaithersburg, MD: Aspen Publishers.

Kassin, S.M., & Fong, C.T. (1999).  “I’m innocent!”:  Effects of training on judgments of truth and deception in the interrogation room. Law and Human Behavior, 23, 499-516. 

Kassin, S.M.,Goldstein, C.J., & Savitsky, K. (2003). Behavioral confirmation in the interrogation room:  On the dangers of presuming guilt.  Unpublished manuscript.

Koehnken, G. (1987).  Training police officers to detect deceptive eyewitness statements: Does it work?  Social Behavior, 2, 1-17.

Lamb, M.E., Orbach, Y., Sternberg, K.J., Hershkowitz, I., & Horowitz,, D. (2000). Accuracy of investigators’ verbatim notes of their forensic interviews with alleged child abuse victims.  Law and Human Behavior, 24, 699-708.

Meissner, Christian A., & Kassin, Saul M. (2002).  “He’s guilty!”: Investigator bias in judgments of truth and deception.  Law and Human Behavior, 26, 469-480.

Porter, S., Woodworth, M., & Birt, A.R. (2000).  Truth, lies, and videotape:  An investigation of the ability of federal parole officers to detect deception.  Lawand Human Behavior, 24, 643-658.

Rayner, K. & Pollatsek, A. (1989).  The psychology of reading. Englewood Cliffs, NJ: Prentice-Hall.

Vrij, A. (1994).  The impact of information and setting on detection of deception by police detectives.  Journal of Nonverbal Behavior, 18, 117-132.

Warren, A., & Woodall, C.E. (1999).  The reliability of hearsay testimony:  How well do interviewers recall their interviews with children?  Psychology, PublicPolicy, and the Law, 5, 355-371.

Zuckerman, M., Koestner, R. & Alton, A.O. (1984).  Learning to detect deception. Journal of Personality and Social Psychology, 46, 519-528.

Zuckerman, M. Koestner, R., & Colella, M.J. (1985).  Learning to detect deception from three communication channels.  Journal of Nonverbal Behavior, 9, 188-194. 

Dr. McCoy specializes in death penalty mitigation. For further information, including a bibliography on mitigation and other articles written by Dr. McCoy.

About Dr. McCoy | Curriculum Vitae | Articles | Contact Dr. Diana McCoy

5019 Jacksboro Pike - Knoxville, TN . 37918
phone:  (865) 687-0340 - fax: (865) 689-1611

© 2011 Forensic Psych Pages