Testing for Potential Contextual Bias Effects During the Verification Stage of the ACE-V Methodology When Conducting Fingerprint Comparisons

• NCJ Number: 226893
• Journal: Journal of Forensic Sciences Volume: 54 Issue: 3 Dated: May 2009 Pages: 571-582
• Author(s): Glenn Langenburg M.Sc.; Christophe Champod Ph.D.; Pat Wertheim B.A.
• Date Published: May 2009
• Length: 12 pages
• Grant Number(s): 428-25-03
• Sponsoring Agency: Midwest Forensics Resource Ctr
• Publication Format: Article
• Publication Type: Report (Study/Research)
• Language: English
• Country: United States of America

Annotation

Given concern about contextual bias and confirmation bias that has come to the fore in recent cases of fingerprint comparisons, the current study tested whether fingerprint specialists could be influenced by extraneous contextual information during a verification process.

Abstract

The study found a contextual bias effect for “novice” participants (laypersons with no training or experience in comparing friction ridge impressions) and “expert” participants (specialists in fingerprint identification) when comparing and assessing fingerprints, with the effect being stronger for novice participants. Experts were biased toward resisting a conclusion of “individualization” (the images are from the same source to the exclusion of all others). Under this cautious posture, experts exposed to the contextual bias made significantly fewer errors (4 errors) than novices (46 errors). The “experts” were divided into three groups that differed in the level of contextual bias. One group (n=16) was exposed to “high bias,” which consisted of a worksheet they were told was produced by a prominent internationally recognized expert in friction ridge comparison, and that the worksheet reported his opinions from an actual case. Also included was a copy of an official agency report stating his conclusions. The “low bias” “expert” group (n=12) was given a work sheet that provided conclusions for each of the trials and was told that these conclusions were opinions provided by a latent print examiner trained to competency. The third “expert” group (n=15), the control group, received the same images as the other 2 groups, but with no context information. All groups were asked to render their opinions on the same set of six side-by-side comparisons of a fingermark (unknown impression) and a fingerprint exemplar. The novices (n=86), medical students, were divided into 3 groups under the same conditions as the experts. 5 tables, 13 figures, and 26 references