The revelation effect: A test of theories and the impact on witness and juror judgments
Final Report Abstract
Judgments are an important and almost ubiquitous part of life. Will it rain today? Is it safe to cross the road? Is this news story believable? It is a little known fact that the activity directly preceding a judgment influences its outcome—in the literature this is known as the revelation effect. In research studies, participants are more likely to consider that something is familiar after a brief problem-solving task (e.g., solving a math problem) compared to a condition without task. Problem-solving tasks also affect whether a statement (e.g., “the leopard is the fastest animal”) is judged to be true, how preferable a product appears, and even whether participants believe that they have experienced a specific event in their childhood (e.g., I” broke a window playing ball”). In the present project, I tested the discrepancy-attribution hypothesis. According to this hypothesis, some types of judgments, such as recognition or aesthetical pleasure, rely on the fluency—that is, the ease and speed—of information processing: More fluently processed stimuli appear more familiar and more appealing than less fluently processed stimuli. Consider, for example, a revelation-effect experiment in which participants solve anagrams in the problem-solving task. Participants process the anagram (e.g., nushines) less fluently than the anagram solution (sunshine). The relatively high processing fluency for the anagram solution compared to the anagram serves as a cue that the anagram solution is familiar—the revelation effect. However, in a series of experiments the results contradicted the discrepancyattribution hypothesis, when the problem-solving task manipulated perceptual fluency (e.g., stimuli appeared blurred vs. clear). For example, faces did appear as more attractive after some problem-solving tasks but not after others. Even more traditional problem-solving tasks such as anagrams failed to produce a revelation effect for emotional-valence judgments. Further, one variant of the discrepancy-attribution hypothesis also predicts that research participants who were warned about the revelation effect should be able to prevent the effect. However, in a now published study, the revelation effect appeared even when participants were confident that they had successfully suppressed it. In another set of experiments, my collaborators and I manipulated fluency through the degree of cognitive effort involved in the intervening task. We found a revelation effect only for problem-solving tasks that included unpracticed and, arguably, more effortful elements. For example, a revelation effect appeared when the problem-solving task involved different anagrams in each trial of the test procedure: After each anagram, participants judged stimuli as more familiar compared to a condition without anagram. However, the revelation effect was absent when participants received the same anagram repeatedly. Similarly, participants judged stimuli as more familiar after an unpracticed task—that is, typing words in reverse letter order (e.g., sunshine - enihsnus). Conversely, the revelation effect was absent with a well-practiced task—that is, when participants simply typed words as they appeared (e.g., sunshine - sunshine). Overall, the results are inconsistent with the discrepancy-attribution hypothesis. Contrary to the hypothesis’ predictions, the revelation effect did not appear when experiments included manipulations of perceptual fluency and judgments of aesthetical pleasure. Conversely, the revelation effect should have been diminished when participants were warned about the effect, which was not the case. Instead, cognitive effort seems to play a more central role in the revelation effect than previously assumed.
Publications
- (2015). A word of warning: instructions and feedback cannot prevent the revelation effect. Consciousness and Cognition, 34, 75-86
Aßfalg, A., & Nadarevic. L.
(See online at https://doi.org/10.1016/j.concog.2015.03.016) - Looking backward and forward on hindsight bias. In: John Dunlosky & Sarah Tauber (Hrsg.), Oxford Handbook of Metamemory. Oxford University Press, (2016): Kap. 15 S. 289
Bernstein, D. M., Aßfalg, A., Kumar, R., & Ackerman, R.
- Task difficulty moderates the revelation effect. Memory & Cognition May 2017, Volume 45, Issue 4, pp 664–676
Aßfalg, A., Currie, D., & Bernstein, D. M.
(See online at https://doi.org/10.3758/s13421-016-0685-9) - The revelation effect: A meta-analytic test of hypotheses. Psychological Bulletin & Review December 2017, Volume 24, Issue 6, pp 1718–1741
Aßfalg, A., Hockley, W., & Bernstein, D. M.
(See online at https://doi.org/10.3758/s13423-017-1227-6)
