By James M. Peters1

Attorneys representing men accused of child sexual abuse or exploitation sometimes try to introduce evidence in criminal trials showing that their clients do not respond on psychological tests the same as men known to have molested children.2 Most courts that have considered such testimony have excluded it,3 but the issue has recently reappeared with a device called the Abel Assessment for Sexual Interest™ (AASI).

What is the AASI?

Psychiatrist Gene Abel, who developed and markets the AASI, holds it out as an objective test that can reliably differentiate between child molesters and non-molesters.4 A man taking the AASI answers a 389-item questionnaire about his sexual behavior, beliefs and willingness to admit common social problems.5 He then views 160 suggestive slides and reports his arousal to each. The time he spends viewing each slide is recorded.6 The results from these procedures are used to derive three scores that allegedly reveal the probability of the subject being a sexual molester of pre-adolescent girls or boys, or lying when denying an act of molestation.7

Recent Criminal Cases on the AASI

Judges assessing the admissibility of the AASI in criminal trials have reached disparate results. A Louisiana man charged with molesting two twelve-year-old girls successfully introduced expert testimony that he did not have a sexual preference for underage females, based on his performance on the AASI.8 The man was acquitted. In deciding what weight to give this case, it is worth noting that the prosecutor was unable to find an expert to rebut Dr. Abel’s testimony, so the judge’s decision was based on a tendentious record.9

Recently, in South Dakota, a man charged with assaulting his six-year-old son sought to introduce testimony regarding his performance on the AASI to show that he did not have a sexual interest in underage boys.10 After hearing experts from both sides, the judge concluded that the AASI does not satisfy the Daubert11 standard and further stated “. . . Part I of the Abel Assessment is an impermissible comment on (the defendant’s) credibility which would not be admissible even if the Abel Assessment did meet the Daubert standard.”12 A three-judge panel in a Texas criminal appeals case reached a similar conclusion, finding that the results of the AASI “. . . are much like the results of a polygraph test, which are per se inadmissible.”13

Challenging the AASI in Criminal Trials

While the AASI has certain advantages as a tool for evaluating admitted sex offenders,14 it has no identified value as evidence in the guilt phase of criminal trials. First, the procedure has been tested on limited classes of admitting offenders and specifically excludes “incest-only” cases.15 It has also not been tested using significant samples of minority groups such as Native Americans.16

Second, the developer withholds as “trade secrets” data that might be analyzed and replicated, a factor that confounds rigorous scientific review.17 While that does not prove it is inferior or biased, conflict of interest can occur when the same company tests, provides expert testimony, and markets a product.18

The government’s expert in the South Dakota case, psychologist Susan Sachsenmaier, analyzed available data on the AASI and identified relevant details. First, studies upon which the first phase of the AASI is based have used penile plethysmography (PPG) as the standard against which to validate the hypothesis that viewing time positively correlates with sexual interest, and sexual interest positively correlates with sexual arousal.19 Using PPG results to validate an experimental device offered as evidence in the fact-finding portion of a criminal trial is problematic because PPG results are, themselves, not reliable enough to be admissible evidence.20 Second, one of Dr. Abel’s own studies on viewing time inaccurately classified as non-pedophiles almost one quarter of admitted pedophiles.21 This information was provided to the judge in the South Dakota case, who held that “. . . a 24 percent rate of false negative results does not assist the jury in understanding the evidence or determining a fact in issue.”22 Third, the few independent researchers who have published their data have argued that the reliability and validity of the first phase of the AASI is not yet established.23 Fourth, Dr. Abel’s recently published study on the second phase of the AASI,24 claiming to discriminate between non-child molesters and molesters who deny molesting children, has notable drawbacks. The “non-child molesters” were sex offenders whose formal conviction history did not show child sex assault, but whose actual offense history was unknown.25 Subjective determinations were used to decide which “alleged” offenders were truly guilty but “lying-denying.” The probability score is generated by comparing the current subject to a pre-determined “sex-offender profile,” in direct opposition to ethical and practice standards that say there is no identifiable sex offender profile.26 The “probability/profile” phase of the AASI has not yet been peer reviewed in the published literature.27

Acceptance in the relevant scientific community is always important when determining the admissibility of expert testimony. It can be argued that the AASI has gained acceptance because many treatment professionals have purchased it—more than 1000 according to the Abel web site,28 300 according to Dr. Abel’s testimony in South Dakota.29 Conversely, it can be argued that the scientific community has not accepted it because scientists have not been allowed to access and review the data, much of which is closely-held, and therefore remain skeptical. The Judge in the South Dakota case concluded that the AASI has not achieved widespread acceptance within the scientific community.30

Finally, two leading professional organizations have established ethical provisions that call into question the legitimacy of presenting AASI results in criminal trials. The Association for the Treatment of Sexual Abusers has professional Practice Standards and Guidelines,31 including two sections that are directly on point. Section 9.02 provides that: “Members shall not knowingly provide court testimony during the guilt phase of a criminal trial from which a reasonable person would draw inferences about whether the client did or did not commit a specific (sexual) crime.” Standard 9.03 states that “Members should recognize that there is no known psychological or physiological test, profile, evaluation procedure, or combination of such tools that can be used to prove or disprove whether the client has committed a specific sexual crime.” (emphasis in original).

The American Psychological Association’s Ethical Standards are more general.32 Among several pertinent sections is Ethical Standard 7.04, which states that: “(a) In forensic testimony and reports, psychologists testify truthfully, honestly, and candidly and, consistent with applicable legal procedures, describe fairly the bases for their testimony and conclusions. (b) Whenever necessary to avoid misleading, psychologists acknowledge the limits of their data or conclusions.” These guidelines make it clear that members of either ATSA or the APA are required to acknowledge the limits of their ability to distinguish child molesters from non-child molesters. While not every sex offender therapist is necessarily a member of these organizations, the reasoning underlying these ethical codes should carry significant weight.

Conclusion

Given the lack of adequate, independent studies to support the accuracy of the AASI, prosecutors should argue that the research falls short of proving the relevance, reliability and acceptance necessary for its use in criminal trials. Prosecutors confronted with a proffer of testimony about the results of the AASI should ask for a pre-trial hearing to challenge the device under standards for the admission of scientific evidence in their jurisdiction.

1 The author is an Assistant United States Attorney in Boise, Idaho. Previously, he worked in the U.S. Attorney’s office in the Virgin Islands, was a senior attorney at the National Center for Prosecution of Child Abuse, and was a deputy prosecuting attorney in Vancouver, Washington. The opinions stated here are those of the author and not the United States Department of Justice. Special thanks to Charles Phipps at the University of South Carolina for his help with this article
2 See, e.g., Dennis Saccuzzo, Still Crazy After All These Years: California’s Persistent Use of the MMPI As Character Evidence in Criminal Cases, 33 U.S.F. L. Rev. 379 (1999); William Murphy & James Peters, Profiling Child Sexual Abusers: Psychological Considerations,” 19 Crim Just. & Behav. 24 (1992); and James Peters & William Murphy, Profiling Child Sexual Abusers: Legal Considerations, 19 Crim Just. & Behav. 38 (1992).
3 See Brian Holmgren, Expert Testimony Profiling Sex Offenders - A Dead Issue Rears Its Head Again, UPDATE, National Center For Prosecution of Child Abuse, Vol. 12, No. 4, (1999); J. Myers, Evidence in Child Abuse and Neglect Cases, John Wiley & Sons, Inc. §§ 5.53 & 5.54 (3d ed. 1997); American Prosecutor’s Research Institute, Investigation and Prosecution of Child Abuse, p. 378 (2d ed. 1993).
4 Abel, Gene G., Jordan, Alan, Hand, Cynthia G. Holland, Laura A., & Phipps, Alexandra, Classification models of child molesters utilizing the Abel Assessment for Sexual Interest, 25 Child Abuse and Neglect, 703, 715 (2001).
5 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001). (Transcript of testimony of Gene G. Abel, page 9.)
6 Id. (Transcript of testimony of Dr. Abel at page 9).
7 Id. (Transcript of testimony of Dr. Abel, at pp. 8-11 and 60-61.)
8 United States v. Robinson, 94 F. Supp. 2d. 751 (W.D. La. 2000).
9 Telephone communication with AUSA John “Luke” Walker, May 24, 2000.
10 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001).
11 Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993); see also Federal Rules of Evidence 702 (December 1, 2000).
12 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001).
13 In the Matter of J.G., 1998 WL 271053 (Tex. App.-Austin, May 29, 1998) (No. 03-97-00217-CV) (unpublished) (text in Westlaw).
14 Susan Sachsenmaier and James M. Peters, Sexual Offender Risk Assessment Methods and Admissibility as Expert Witness Evidence in Prosecuting Online Child Exploitation Cases, James M. Peters, ed., U.S. Department of Justice (in press).
15 See, United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001), (citing Abel, Gene G., Jordan, Alan, Hand, Cynthia G, Holland, Laura A., & Phipps, Alexandra, Classification models of child molesters utilizing the Abel Assessment for Sexual Interest, 25 Child Abuse and Neglect, 703, 705 (2001).
16 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001).
17 See, endnote 14.
18 Marcia Angell, Is Academic Medicine for Sale?, New Eng.J.Med., Vol. 342, No. 20 (May 18, 2000). See also www.Abelscreen.com (last visited 12-27-01) advertising the device and noting that it recently passed Daubert scrutiny in United States v. Robinson and in a New Mexico family court case, but omitting the fact that it failed to pass the Daubert hearing in United States v. White Horse.
19 See, endnote 14.
20 See, e.g., Doe v. Glanzer, 232 F.3d 1258, 1266 (9th Cir. 2000); Leftwich v. The State, 538 S.E.2d 779 (Ga. Ct. App. 2000); United States v. Powers, 59 F.3d 1460 (4th Cir. 1995); State v. Spencer, 459 S.E.2d 812 (N.C. App. 1995); Gentry v. State, 443 S.E.2d 667 (1994); State v. Emery, 156 Vt. 364, 593 A.2d 77, 78-79 (1991).
21 See, Id. and United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001), (citing Gene Abel et al., Screening Tests for Pedophilia, 21 Criminal Justice and Behavior, 115-131 (1994).
22 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001).
23 See, Id. See also Lane Fischer, The Abel Screen: A Nonintrusive Alternative?, in Remaking Relapse Prevention with Sex Offenders: A Sourcebook, (D. Richard Laws, et. al. eds, (2000); Lane Fischer, and Gillian Smith, Statistical Adequacy of the Abel Assessment for Interest in Paraphilias,11 Sexual Abuse: A Journal of Research and Treatment,195-205, 1999.
24 Abel, Gene G., Jordan, Alan, Hand, Cynthia G, Holland, Laura A., & Phipps, Alexandra, Classification models of child molesters utilizing the Abel Assessment for Sexual Interest, 25 Child Abuse and Neglect, 703-18 (2001).
25 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001). (Testimony of Susan Sachsenmaier).
26 Practice Standards and Guidelines for Members of the Association for the Treatment of Sexual Abusers, 2001, Standard 9.03. See http://www.atsa.com (last visited 12-13-01). ATSA is located at 4900 S.W. Griffith Drive, Suite 274, Beaverton, OR 97005, phone (503) 643-1023.
27 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001).
28 See www.abelscreen.com (last visited 12-27-01).
29 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001). (Transcript of testimony of Dr. Abel, at p. 35.)
30 United States v. White Horse, — F. Supp. —, 2001 WL 1464093 (D.S.D. 2001).
31 Practice Standards and Guidelines for Members of the Association for the Treatment of Sexual Abusers, 2001. ATSA is located at 4900 S.W. Griffith Drive, Suite 274, Beaverton, OR 97005, phone (503) 643-1023. See http://www.atsa.com.
32 American Psychological Association, Ethical Principles of Psychologists and Code of Conduct (1992) (available at www.apa.org).