A biased opinion: Demonstration of cognitive bias on a fingerprint matching task through knowledge of DNA test results
A biased opinion: Demonstration of cognitive bias on a fingerprint matching task through knowledge of DNA test results
One study is presented which explores the biasing effects of irrelevant contextual information on a fingerprint matching task. Bias was introduced by providing the outcomes of a DNA test relating to each fictitious case under consideration. This was engineered to suggest either a match, no match, or an inconclusive outcome, and was thus either consistent, misleading or unbiased depending on the ground truth of each fingerprint pair. The results suggested that, when the difficulty of the fingerprint matching task was measurably increased, participants became more vulnerable to the biasing information. Under such conditions, when performance was good, misleading evidence lowered accuracy, and when performance was weaker, consistent evidence improved accuracy. As such, the results confirmed existing demonstrations of cognitive bias from contextual information in the fingerprint task. Moreover, by taking a process-based approach, it became possible to articulate the concerns, and the potential solutions, at each stage of the workflow. The results offer value for the forensic science community in extending the evidence-base regarding cognitive bias, and in articulating routes to improve the credibility of fingerprint decisions.
93-106
Stevenage, Sarah V.
493f8c57-9af9-4783-b189-e06b8e958460
Bennett, Alice
2350affd-26bf-45e7-848e-d01377fbd788
1 July 2017
Stevenage, Sarah V.
493f8c57-9af9-4783-b189-e06b8e958460
Bennett, Alice
2350affd-26bf-45e7-848e-d01377fbd788
Stevenage, Sarah V. and Bennett, Alice
(2017)
A biased opinion: Demonstration of cognitive bias on a fingerprint matching task through knowledge of DNA test results.
Forensic Science International, 276, .
(doi:10.1016/j.forsciint.2017.04.009).
Abstract
One study is presented which explores the biasing effects of irrelevant contextual information on a fingerprint matching task. Bias was introduced by providing the outcomes of a DNA test relating to each fictitious case under consideration. This was engineered to suggest either a match, no match, or an inconclusive outcome, and was thus either consistent, misleading or unbiased depending on the ground truth of each fingerprint pair. The results suggested that, when the difficulty of the fingerprint matching task was measurably increased, participants became more vulnerable to the biasing information. Under such conditions, when performance was good, misleading evidence lowered accuracy, and when performance was weaker, consistent evidence improved accuracy. As such, the results confirmed existing demonstrations of cognitive bias from contextual information in the fingerprint task. Moreover, by taking a process-based approach, it became possible to articulate the concerns, and the potential solutions, at each stage of the workflow. The results offer value for the forensic science community in extending the evidence-base regarding cognitive bias, and in articulating routes to improve the credibility of fingerprint decisions.
Text
AUTHOR ACCEPTED A Biased Opinion
- Accepted Manuscript
More information
Accepted/In Press date: 10 April 2017
e-pub ahead of print date: 1 May 2017
Published date: 1 July 2017
Organisations:
Cognition
Identifiers
Local EPrints ID: 411123
URI: http://eprints.soton.ac.uk/id/eprint/411123
ISSN: 0379-0738
PURE UUID: 7ab9f7da-25e0-40be-9958-f3e17e75daba
Catalogue record
Date deposited: 14 Jun 2017 16:31
Last modified: 16 Mar 2024 05:25
Export record
Altmetrics
Contributors
Author:
Alice Bennett
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics