Guilty by association: an investigation of bias in fingerprint identification

Searston, Rachel (2012). Guilty by association: an investigation of bias in fingerprint identification Honours Thesis, School of Psychology, The University of Queensland.

       
Attached Files (Some files may be inaccessible until you login with your UQ eSpace credentials)
Name Description MIMEType Size Downloads
SEARSTONRachel4071thesis2012.pdf Thesis full text application/pdf 11.55MB 12
Author Searston, Rachel
Thesis Title Guilty by association: an investigation of bias in fingerprint identification
School, Centre or Institute School of Psychology
Institution The University of Queensland
Publication date 2012-10-09
Thesis type Honours Thesis
Supervisor Jason Tangen
Total pages 81
Language eng
Subjects 1701 Psychology
Abstract/Summary There has been significant resistance from within the forensic science community to acknowledge the integral role of human judgment and perception in the process of forensic fingerprint identification. The dehumanization of the field has led to claims of infallibility and zero error rates on the basis of longstanding use, and has led to individuals who have made errors being targeted as “bad apples” rather than considering the root cause of the problem. There has been minimal research investigating potential sources of bias in fingerprint identification and the research that is currently available is limited methodologically. We conducted three experiments investigating the biasing effects of contextual factors including crime severity, extrinsic case familiarity and intrinsic case familiarity. We exposed subjects to biasing information in each of the three experiments and tested their ability to discriminate between matching and non-matching prints. In line with predictions, the results of Experiment 1 revealed that subjects tended to adopt a more liberal response bias (tended to say “match” more) overall, and more so in cases involving crimes perceived as more severe than in cases perceived as less severe. Results of Experiment 2 and 3 were also consistent with predictions, demonstrating that novices’ were biased to say “match” more and were less accurate when presented with familiar case information, and familiar fingerprints similar to previously encountered cases. Overall, our project highlights that both extrinsic information (i.e., crime severity, case familiarity) and intrinsic information (i.e., familiarity of print features) can bias a person’s judgment when comparing fingerprints. Future studies may look to replicating this research with expert populations.
Keyword Bias
Fingerprint identification
Human judgement and perception

 
Citation counts: Google Scholar Search Google Scholar
Created: Fri, 08 Mar 2013, 10:07:03 EST by Mrs Ann Lee on behalf of School of Psychology