Influence of credibility, warnings, and source-monitoring questions on the misinformation effect.

Wyler, Helen; Oswald, Margit (27 June 2014). Influence of credibility, warnings, and source-monitoring questions on the misinformation effect. (Unpublished). In: Annual Conference of the European Association of Psychology and Law. St. Petersburg. 24.-27.06.2014.

[img] Text (Poster)
POSTER_EAPL_Wyler.pdf - Presentation
Restricted to registered users only
Available under License BORIS Standard License.

Download (2MB)

Introduction. Erroneous answers in studies on the misinformation effect (ME) can be reduced in different ways. In some studies, ME was reduced by SM questions, warnings, or a low credibility of the source of post-event information (PEI). Results are inconsistent, however. Of course, a participant can deliberately decide to refrain from reporting a critical item only when the difference between the original event and the PEI is distinguishable in principle. We were interested in the question to what extent the influence of erroneous information on a central aspect of the original event can be reduced by different means applied singly or in combination. Method. With a 2 (credibility; high vs. low) x 2 (warning; present vs. absent) between subjects design and an additional control group that received neither misinformation nor a warning (N = 116), we examined the above-mentioned factors’ influence on the ME. Participants viewed a short video of a robbery. The critical item suggested in the PEI was that the victim was given a kick by the perpetrator (which he was actually not). The memory test consisted of a two-forced-choice recognition test followed by a SM test. Results. To our surprise, neither a main effect of erroneous PEI nor a main effect of credibility was found. The error rates for the critical item in the control group (50%) as well as in the high (65%) and low (52%) credibility condition without warning did not significantly differ. A warning about possible misleading information in the PEI significantly reduced the influence of misinformation in both credibility conditions by 32-37%. Using a SM question significantly reduced the error rate too, but only in the high credibility no warning condition. Conclusion and Future Research. Our results show that, contrary to a warning or the use of a SM question, low source credibility did not reduce the ME. The most striking finding was, however, the absence of a main effect of erroneous PEI. Due to the high error rate in the control group, we suspect that the wrong answers might have been caused either by the response format (recognition test) or by autosuggestion possibly promoted by the high schema-consistency of the critical item. First results of a post-study in which we used open-ended questions before the recognition test support the former assumption. Results of a replication of this study using open-ended questions prior to the recognition test will be available by June.

Item Type:

Conference or Workshop Item (Poster)

Division/Institute:

07 Faculty of Human Sciences > Institute of Psychology > Social Psychology and Social Neuroscience

UniBE Contributor:

Wyler, Helen and Oswald, Margit

Subjects:

100 Philosophy > 150 Psychology
300 Social sciences, sociology & anthropology

Language:

English

Submitter:

Helen Wyler

Date Deposited:

06 Mar 2015 14:07

Last Modified:

10 Mar 2015 21:08

BORIS DOI:

10.7892/boris.64117

URI:

https://boris.unibe.ch/id/eprint/64117

Actions (login required)

Edit item Edit item
Provide Feedback