Motivation modulates the effect of approach on implicit preferences

Abstract

With three studies, we investigated whether motivational states can modulate the formation of implicit preferences. In Study 1, participants played a video game in which they repeatedly approached one of two similar beverages, while disregarding the other. A subsequent implicit preference for the target beverage emerged, which increased with participants’ thirst. In Study 2, participants approached one brand of potato chips while avoiding the other: Conceptually replicating the moderation observed in Study 1, the implicit preference for the approached brand increased with the number of hours from last food intake. In Study 3, we experimentally manipulated hunger, and the moderation effect emerged again, with hungry participants displaying a higher implicit preference for the approached brand, as compared to satiated participants. In the three studies, the moderation effect was not paralleled in explicit preferences although the latter were affected by the preference inducing manipulation. Theoretical implications and open questions are discussed.

Keywords:

• Approach
• Implicit attitudes
• Explicit attitudes
• Hunger
• Thirst
• Implicit attitude formation

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplementary material is available via the ‘Supplementary’ tab on the article's online page (http://dx.doi.org/10.1080/02699931.2015.1032892).

Notes

¹ Affective valence and motivational direction are distinct concepts. Positive affect is typically associated with approach motivations and negative affect with avoidance motivations, but there are exceptions, most notably anger. Anger is indeed generally experienced as negative affect, but it often evokes approach orientations (Carver & Harmon-Jones, 2009; Harmon-Jones, Harmon-Jones, Abramson, & Peterson, 2009).

² We expected and found no effect of order of administration of the explicit measures. In a preliminary analysis in which explicit preference was regressed on manipulation, thirst, and order of blocks, only an effect of the predictor emerged while no main or interaction effect involving the order of the blocks approached significance, all ps> .35. Therefore, order of administration was not considered in the main analyses.

³ Also after controlling for IAT order, the effect of approach on implicit attitudes was not significant, as indicated by the regression analysis (see Table 2, lower part, Step 2).

⁴ Hours from food intake and subjective rating of hunger correlated, r = .37, p < .001. We based our moderator analyses on number of hours from food intake, because this variable is less influenced by subjective feelings. However, similar results emerged with an aggregated index of motivation (see Table 1a, Supplementary Online materials).

⁵ The same pattern of results emerged on the three explicit measures separately.

⁶ Study 1 presented a pattern of results that was dissimilar for Studies 2 and 3, with a moderation of motivation explicit attitude change in a direction opposite to its effect on implicit attitude change. Thus, we performed a GLM with the same predictors and interactions, but with only two levels of Study (Study 1 vs. Other studies). Again, no main or interaction effect involving the variable Study approached significance and the significance level for the four-way interaction was p = .19, = .01.