Fillers can help control for contextual bias in forensic comparison tasks.

Abstract

Forensic examiners are often exposed to contextual information that can bias their conclusions about evidence samples (e.g., fingerprints, fibers, tool marks). We tested the recently proposed filler-control method for moderating the biasing effects of contextual information for forensic comparisons. Borrowing from an analogy to eyewitness lineups versus showups, the filler-control method embeds a suspect's sample among known-innocent samples rather than the standard practice of presenting the analyst with only the suspect's sample for comparison. Our test of the filler-control method used fingerprints. After brief training, 234 participants compared eight sets of fingerprints in which suspect prints either matched the crime print or not, the prints were high or low in ambiguity, there was or was not contextual information suggesting there should be a match, and the suspect print was either embedded among filler prints or presented alone. Although the filler-control procedure reduced both hits and false alarms, the filler-control procedure produced better results overall as measured by d' analyses on suspect samples. These findings suggest the filler-control procedure should be considered for use in everyday forensic examination judgments, particularly when the error rate for a technique is unknown, or the risk of contextual bias is obvious, such as when examiners are called to make verification decisions. (PsycINFO Database Record