CONFIDENTIAL

Rachel Wolkenstein, Esq.

67 WaIl Street,

Suite 2411

New York, NY 11217

RE: Polygraph Examination of Arnold R. Beverly

Ms. Wolkenstein:

At your request, I conducted a psychophysiological detection of deception examination on Arnold R. Beverly. The purpose of this examination was to assess Mr. Beverly's credibility concerning his knowledge of and participation in the murder of Officer Daniel Faulkner. The examination was conducted in Oakland, California, on 24 April 1999. At the beginning of the pretest interview, Mr. Beverly signed a consent form giving his permission for the examination. At that time, Mr. Beverly also gave his permission to have the examination audio tape-recorded. He refused to allow video recording. During the pretest interview I discussed the murder of Officer Daniel Faulkner with Mr. Beverly. Mr. Beverly stated that he was present at the time that Officer Faulkner was shot and killed. Mr. Beverly stated that Mumia Abu-Jamal was not present at the time that Officer Faulkner was shot the first time, but that he was present when Officer Faulkner was shot the second time. Mr. Beverly stated that Mumia Abu-Jamal did not shoot Officer Faulkner. When asked if he, Arnold Beverly, shot Officer Faulkner, Mr. Beverly, said "No" but subsequently he asked that the tape be turned off. The tape was turned off and Mr. Beverly stated that he did in fact shoot Officer Faulkner, but that he would not say so on the tape and that he wanted the polygraph to show that he did the shooting. I interrupted the examination so that I could discuss this turn of events with Ms. Wolkenstein, and it was agreed that the examination would continue as Mr. Beverly wanted. Later in the interview Mr. Beverly stated that he had been paid to shoot Officer Faulkner. At the conclusion of the pretest interview, an acquaintance test was conducted. That procedure was designed to demonstrate to the subject and to me that he was a suitable subject for a psychophysiological detection of deception examination. Adequate recordings of Mr. Beverly's physiology were obtained during this procedure, and the examination was continued. A comparison question test was then conducted using the methods developed and validated at the University of Utah. The test induded the following relevant questions that were reviewed with Mr. Beverly prior to the test: Ri. Did you shoot Officer Daniel Faulkner? Answered "No" o Page 2 May 21, 1999 R2. Were you present when Officer Daniel Faulkner was shot? Answered "Yes" R3. Was Mumia Abu-Jamal present when Officer Faulkner was shot the first time? Answered "No" R4. Were you paid to shoot Officer Daniel Faulkner? Mr. Beverly's physiological responses were monitored during the presentation of the questions by means of a Stoelting digital polygraph instrument running the Scientific Assessment Technologies Computerized Polygraph System software, Version 2.20. The following physiological measures were recorded in digital form on the computer's hard disk: cardiovascular activity (Erlanger Method), skin conductance, vasomotor activity, abdominal and thoracic respiration. In addition, data were obtained from a movement sensor placed under the subject's chair. Data from three presentations of the questions were obtained. The digitized physiological data were displayed on the computer's monitor as analog waveforms and were then subjected to a numerical scoring analysis using the criteria developed and validated at the University of Utah. The digitized physiological data were also subjected to computer-based analyses using the algorithms developed at the University of Utah. The numerical scores after three charts on the above relevant questions were -3, +1, -1, and +1, respectively, for a total examination score of -2. In the Utah Scoring System, when evaluating a single relevant question, a score of +3 or greater is considered to be indicative of truthfulness, and a score of -3 or less is considered to be indicative of deception. Single question scores between +3 and -3 are considered inconclusive. A total numerical score of -6 or less is considered as indicative of deception. A total numerical score of +6 or greater is considered as indicative of truthfulness. Total numerical scores between -6 and +6 are considered inconclusive. In the present case, interpreting the total numerical score does not make sense since the relevant questions cover different issues. Mr. Beverly's score of -3 to Relevant question RI. is indicative of deception to that question. The numerical scores to the other three questions are inconclusive. The conference room where the initial test was conducted had become very warm by the time the first series had been numerically evaluated. It was decided that a second series should be run and that it would be better if we moved the testing to a suite at the Airport Hilton Hotel. The equipment was relocated to the Hilton and testing was continued. Mr. Beverly was confronted with his deceptive result to Relevant question Ri. His comment was "Good, maybe now they will believe me." A second series was then conducted with the following relevant questions. Ria. Did Mumia Abu-Jamal shoot Officer Daniel Faulkner? Answered "No" R2. Were you present when Officer Daniel Faulkner was shot? Answered "Yes" R3. Was Mumia Abu-Jamal present when Officer Faulkner was shot the first time? Answered "No" R4. Were you paid to shoot Officer Daniel Faulkner? Data from three presentations of the questions were obtained. The data were then subjected to numerical scoring and computer-based analyses. The numerical scores after three charts on the above relevant questions were +1, o, -i, and o, respectively, for a total examination score of o. All of these numerical scores are inconclusive. The entire data set of six question repetitions was then subjected to computer based statistical analysis. The first analysis was the discriminant classification analysis developed at the University of Utah. That analysis gives an a posteriori probability of truthfulness based on a statistical analysis of the physiological data. The discriminant classification analysis of Mr. Beverly's physiological data resulted in the following aposteriori probabilities of truthfulness: RI. = 0.001; R2 = 0.381; R3 = 0.340; R4 = 0.604; and Ria = 0.811. Scientific research has shown that probabilities of truthfulness that exceed 0.70 can be used to validly infer truthfulness while probabilities of truthfulness of less than 0.30 can be used to validly infer deception. This analysis procedure has been cross-validated in a number of scientific studies using data from a variety of sources, including actual case data obtained from the United States Secret Service. In that scientific research, the discriminant classification analysis has consistently been found to be a highly valid method for classifying persons as truthful or deceptive. This statistical analysis has also consistently been found to perform as well as, or better than, the best human evaluators. The results of the discriminant analysis suggest that Mr. Beverly was being deceptive when he said that he did not shoot Officer Faulkner (Ri), and was being truthful when he stated that Mumia Abu-Jamal did not shoot Officer Faulkner. The computer was then used to perform a rank order analysis of the strength of the combined physiological response across all of the repetitions. This analysis produced the following rank ordering of relevant and comparison questions (shown from strongest response to weakest response, comparison questions begin with the letter "C"): R1>CQ1>CQ3>R3>R2>R4>CQ2> Cal>Ca3>R1a. The theory of the comparison question test predicts that deceptive subjects will produce larger responses to relevant than to comparison questions, while truthful subjects will produce larger responses to control than to relevant questions. The pattern resulting from the rank order analysis is ambiguous for all but two questions. Relevant question Ri produced a larger response and any of the comparison questions; the pattern expected for deception. Relevant question, Ria, produced a response that was smaller than all of the comparison questions; the pattern expected for truth telling. On the basis of the results of the numerical scoring and the computer-based analyses, it is my opinion that Mr. Beverly was not being truthful when he said "No" to the question: Did you shoot Officer Daniel Faulkner? it is also my opinion that Mr. Beverly was being truthful when he answered "No" to the question: Did Mumia Abu-Jamal shoot Officer Daniel Faulkner? The results of the analyses of the other relevant questions are inconclusive.

Charles Robert Honts, Ph. D. Detection of Deception Examiner Professor of Psychology

There have been

visitors to my site
since 05/21/99