Moderators of post-identification feedback effects on eyewitnesses' memory reports

Purpose. Information provided to eyewitnesses suggesting that their identification was correct (i.e., post-identification feedback) distorts witnesses' memory reports. More pronounced effects of feedback have been detected for subjective than for objective judgments. We investigated two variabl...

Ausführliche Beschreibung

Gespeichert in:  
Bibliographische Detailangaben
1. VerfasserIn: Douglass, Amy Bradfield (VerfasserIn)
Beteiligte: Brewer, Neil ; Semmler, Carolyn
Medienart: Elektronisch Aufsatz
Sprache:Englisch
Veröffentlicht: 2010
In: Legal and criminological psychology
Jahr: 2010, Band: 15, Heft: 2, Seiten: 279-292
Online Zugang: Volltext (lizenzpflichtig)
Volltext (lizenzpflichtig)
Journals Online & Print:
Lade...
Verfügbarkeit prüfen: HBZ Gateway
Beschreibung
Zusammenfassung:Purpose. Information provided to eyewitnesses suggesting that their identification was correct (i.e., post-identification feedback) distorts witnesses' memory reports. More pronounced effects of feedback have been detected for subjective than for objective judgments. We investigated two variables that might explain this finding: response format and question verifiability. Methods. Response format was manipulated by varying whether participants (N = 160) answered objective questions using fill-in-the-blank options or Likert scales. Question verifiability was manipulated within-subjects by including objective questions for which the answers either could or could not be known by the experimenter. Results. Objective measures produced non-significant feedback effects, whereas most subjective measures produced large feedback effects. Neither response format nor verifiability moderated the feedback effect on any of the measures. Conclusions. Researchers should continue to search for moderators of the post-identification feedback effect, guided by theoretical memory research. In addition, they should publish descriptive statistics for individual dependent measures, even if the ideal statistical techniques require collapsing across measures. This approach will allow future researchers to evaluate the difference between objective and subjective measures. Finally, researchers should avoid sweeping statements about the breadth of susceptibility to feedback - objective measures, for reasons unknown, are generally unaffected by post-identification feedback.
ISSN:2044-8333
DOI:10.1348/135532509X446337