Comparing Categorical and Probabilistic Fingerprint Evidence

Abstract

Fingerprint examiners traditionally express conclusions in categorical terms, opining that impressions do or do not originate from the same source. Recently, probabilistic conclusions have been proposed, with examiners estimating the probability of a match between recovered and known prints. This study presented a nationally representative sample of jury-eligible adults with a hypothetical robbery case in which an examiner opined on the likelihood that a defendant's fingerprints matched latent fingerprints in categorical or probabilistic terms. We studied model language developed by the U.S. Defense Forensic Science Center to summarize results of statistical analysis of the similarity between prints. Participant ratings of the likelihood the defendant left prints at the crime scene and committed the crime were similar when exposed to categorical and strong probabilistic match evidence. Participants reduced these likelihoods when exposed to the weaker probabilistic evidence, but did not otherwise discriminate among the prints assigned different match probabilities.

Error Rates, Likelihood Ratios, and Jury Evaluation of Forensic Evidence

“…The guilty vote rate among people who stated that freeing a guilty person was the worse type of error was 58%, compared with a guilty vote rate of 25% among people who stated that convicting the innocent is the worse type of error. These findings are similar to those observed in prior studies regarding lay assessment of forensic evidence, including Mitchell, Scurich, and Garrett (12), which focused on the role of risk aversions when laypersons evaluated proficiency information concerning forensic experts. The findings raise troubling questions regarding how effective the beyond‐a‐reasonable‐doubt instruction, which was provided verbatim in this mock trial scenario, is in informing jurors of their obligations under the Constitution.…”

“…We also asked participants a question about risk aversion in the context of criminal cases generally, a question that relates to people’s preferences regarding the standard of proof in a criminal case. This question has been posed in several prior studies concerning lay evaluation of forensic evidence (5,12). We asked, “ Which type of error is worse ?” and 46.9% (421 of 897) answered “erroneously convicting an innocent person”; 6.9% (62 of 897) chose “failing to convict a guilty person”; and 46.2% (414 of 897) responded “the errors are equally bad.” We ran a series of one‐way ANOVAs with their response to this as the independent variable and a variety of views on evidence and justice system as outcomes to see how this belief affected other relevant views about convictions.…”

“…Practitioners in many forensic science disciplines increasingly acknowledge that errors are possible in their standardized reporting language and that error rates are calculable and observable (11). One laboratory, the Defense Forensic Science Center, reports fingerprint results in reports and in trial testimony using a probabilistic method (12). Most recently, the Department of Justice, as noted, made a series of recommendations concerning language to be used in forensic techniques, such as latent fingerprint comparisons and firearms comparisons.…”

The Utility of Expanded Conclusion Scales During Latent Print Examinations

“…Consumers would have to learn to accept the strength‐of‐support statements into the overall evidence flow, rather than relying on examiners to make a decision for them. In addition, jurors have difficulty interpreting statements that include large probabilities , and accommodations involving verbal statements would have to be calibrated . Perhaps because of these challenges, the PCAST report argued for a continuation of the categorical conclusions, supported by more extensive error rate (black box) studies, rather than suggesting a move to pure strength‐of‐support statements.…”

“…However, the translation of evidence from the examiner’s comparison all the way to the consumer has multiple places where information can be lost, and inaccuracies can occur. This mapping of evidence to decisions and the subsequent understanding by the consumer has been addressed using a variety of approaches, including measuring the utility of different outcomes using proposed gambling paradigms and direct comparison between different statements .…”