Reliability of fingerprint experts in extracting and evaluating minutiae in individualization tests of fingerprint traces

This study examines the reliability of fingerprint experts in assessing the individualization value of minutiae during the analysis of latent fingerprint traces. Despite the widespread use of fingerprint evidence in criminal investigations, growing concerns about examiner variability and the lack of verification protocols have prompted critical scrutiny of forensic practices. In this study, 30 Polish fingerprint experts were asked to identify and evaluate seven minutiae in two fingerprint traces of differing quality. Experts classified each minutia and rated its individualization value on a five-point scale. Inter-rater reliability was assessed using Krippendorff’s alpha, and the relationship between selection frequency and individualization value was analyzed via linear regression and correlation.

Results revealed low inter-expert agreement in both minutiae selection and value assessment, with alpha coefficients of 0.39 for the trace of high-quality and 0.42 for the traces of a lower-quality. A strong negative correlation was found between how often a minutia was selected and its perceived individualization value (r = –0.84 for the trace of a high-quality and r = –0.89 for the trace of a lower-quality), suggesting that commonly selected features were viewed as less discriminative. These findings indicate that while experts may converge on general patterns at the group level, their individual assessments are often inconsistent.

The study underscores the need for systematic verification of expert conclusions. Given the variability observed, relying on a single unverified fingerprint opinion poses a substantial risk of error. Requiring that initial reports be prepared by multiple independent experts could enhance the reliability of conclusions and reduce vulnerability to evidentiary challenges.