Rebel Without a (Conscious) Cause: Priming a Nonconscious Goal for Psychological Reactance

Abstract

Psychological Reactance is typically studied using conscious manipulations and individual-difference measures. We hypothesized that, similar to other goals, a reactance motive can be primed nonconsciously. In this experiment, participants were given an explicit expectation that a pill would improve their performance on an accuracy task, or they were not given this expectation. Participants also received a reactance or neutral prime. On a subsequent accuracy task, participants given both the reactance prime and the explicit accuracy expectation committed the most errors. The findings suggest that reactance can be nonconsciously primed, generating behavioral effects. Evidence supporting the goal-priming interpretation is also presented.

ACKNOWLEDGEMENTS

This article is based on a Master's thesis submitted by the first author to the Department of Psychology, University of Toledo. The authors wish to thank Alexander Czopp, Trina Floyd, Stephanie Fowler, Lisa Neff, and Patricia Smithmyer for their helpful comments and their assistance with various aspects of this research. This research was supported by National Institutes of Health grant R03 NS051687.

Notes

¹Of these participants, 2 were dropped from analyses because they refused to take the caffeine placebo. An additional 5 participants were dropped from analyses because they failed to follow the experimental procedures.

Note. Standard deviations are in parentheses.

²Given their large number and their exploratory nature we applied a more conservative alpha of .01 to all individual comparisons and correlations involving sex to prevent inflated Type I error rates.

Note. Shared subscripts indicate correlations do not significantly differ (p > .05).

∗p < .01 (significance value of r).

³This ANOVA failed Levene's test of equality of error variances, F(7, 85) = 2.43, p = .026. Importantly, the strong significance level of the expectation by prime interaction exceeded the most stringent adjustment criteria suggested for factorial ANOVAs that violate this assumption (Tabachnick & Fidell, 2007).