Mugshot Exposure Effects Retroactive Interference, Mugshot Commitment, Source Confusion, and Unconscious Transference

• Published in: Law and human behavior Vol. 30; no. 3; pp. 287 - 307
• Main Authors: Deffenbacher, Kenneth A, Bornstein, Brian H, Penrod, Steven D
• Format: Journal Article
• Language: English
• Published: United States Springer 01.06.2006; Kluwer Academic/Plenum Publishers; American Psychological Law Society
• Subjects: Behavior; Bias; Commitment; Confidence interval; Confusion; Criminal investigations; Culture; Effects; Experiments; Eyewitness identification; Eyewitness memory; Failure; False alarms; False identification; Human; Humans; Hypotheses; Hypothesis testing; Memory; Memory interference; Mental Confusion; Meta-analysis; Police lineups; Portraits as Topic; Proportions; Psychology; Retention; Retroactive Inhibition; Sample size; Studies; Time Factors; Transference; Transference (Psychology); Unconscious (Personality Factor); Unconscious (Psychology); Visual Perception; Witnesses; mugshots; photobiased lineups; source confusion; unconscious transference; retroactive interference
• ISSN: 0147-7307; 1573-661X
• DOI: 10.1007/s10979-006-9008-1

More than 25 years of research has accumulated concerning the possible biasing effects of mugshot exposure to eyewitnesses. Two separate metaanalyses were conducted on 32 independent tests of the hypothesis that prior mugshot exposure decreases witness accuracy at a subsequent lineup. Mugshot exposure both significantly decreased proportion correct and increased the false alarm rate, the effect being greater on false alarms. A mugshot commitment effect, arising from the identification of someone in a mugshot, was a substantial moderator of both these effects. Simple retroactive interference, where the target person is not included among mugshots and no one in a mugshot is present in the subsequent lineup, did not significantly impair target identification. A third metaanalysis was conducted on 19 independent tests of the hypothesis that failure of memory for facial source or context results in transference errors. The effect size was more than twice as large for "transference" studies involving mugshot exposure in proximate temporal context with the target than for "bystander" studies with no subsequent mugshot exposure.