Abstract
The present research aimed to assess the effect of recognition decision on subsequent affective evaluations of recognised and non-recognised objects. Consistent with the proposed account of post-decisional preferences, results showed that the effect of recognition on preferences depends upon objective familiarity. If stimuli are recognised, liking ratings are positively associated with exposure frequency; if stimuli are not recognised, this link is either absent (Experiment 1) or negative (Experiments 2 and 3). This interaction between familiarity and recognition exists even when recognition accuracy is at chance level and the “mere exposure” effect is absent. Finally, data obtained from repeated measurements of preferences and using manipulations of task order confirm that recognition decisions have a causal influence on preferences. The findings suggest that affective evaluation can provide fine-grained access to the efficacy of cognitive processing even in simple cognitive tasks.
This study has been supported by Russian Foundation for Humanities [project #12-36-01298a2].
This study has been supported by Russian Foundation for Humanities [project #12-36-01298a2].
Notes
1 These predictions, of course, can contradict real data, as when we have an elaborated hallucination, but we will not know it.
2 The author uses the operational definition of “objective familiarity” as a total time of previous processing of an object starting with zero. This definition includes both novel and old stimulus in a single continuum.
3 Dr Lee also generously provided additional data, which was necessary for the proper analysis of the results of the experiments reported in her paper.
4 Portions of the research in this paper use the FERET database of facial images collected under the FERET program, sponsored by the DOD Counterdrug Technology Development Program Office.
5 Could it be that the obtained differences are due to the amplification of recognition decision? The decisional amplification suggests that repeating the same decision is self-supporting, that is, the decision becomes faster and more confident. It is possible that there is an amplification of preferences with repeated evaluation: positive items become more positive and negative items become more negative. As recognition correlates with preferences this effect can be used to explain the observed pattern of preferences for targets in Block 2. Still, it does not explain why amplification is observed only for targets. The inclusion of random stimuli effects in LMER also decreases such a possibility. Nevertheless, an analysis of liking differences equivalent to the one described above was conducted, using liking valence (positive vs. negative) in Block 1 as a predictor. The results were quite clear: Preferences increased for initially negative stimuli, B = 0.50, SE = 0.03, t = 14.96, p < .001, and decreased for initially positive stimuli, B = −0.53, SE = 0.04, t = −15.00, p < .001. This finding corresponds to a regression to the mean effect and not to exaggerated ratings. Consequently, the amplification of preferences cannot explain the previously described results.