Testing the judgment-related account for the extinction of evaluative conditioning

ABSTRACT

Evaluative conditioning (EC) effects refer to changes in the liking of a neutral (conditioned) stimulus (CS) due to pairing with an affective (unconditioned) stimulus (US). Some research found that EC effects are resistant to presentations of the CS without the US, whereas other studies found evidence for extinction effects. A recent study found extinction of EC only when participants rated the CS before and after the CS-only presentations, but not when CS evaluation was measured once or indirectly with the evaluative priming task. In two experiments (total N = 2,181), we found no evidence that indirectly measured evaluation is sensitive to extinction, using an indirect evaluation measure with high sensitivity – the Implicit Association Test. However, unlike previous research, we found that evaluation of any stimuli (and not only the CS) before the CS-only presentations decreases self-reported EC effects. Our results are compatible with the conclusion that the extinction of EC is limited to evaluation measured directly. We discuss the theoretical implications of these results, and conclude that the specific conditions (and mechanisms) that change the direct evaluative response are yet to be clarified.

KEYWORDS:

• Evaluative conditioning
• extinction
• evaluative judgement

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Previous articles have used the term “acquisition” for the procedure of displaying CS-US pairs, and “extinction” for the procedure of showing CS alone. However, because acquisition and extinction are also the common effects of these procedures, we chose to name those procedures after the stimulus presentation, and, when abbreviated, use “pairing” and “CS-only”.

2 To save resources, we planned to stop at 300, 600, and 900 participants and test the Bayes factor for both critical comparisons. If BF₀₁ or BF₁₀ in both t-tests is above 10, we planned to stop collecting the data.

3 Although these exclusions were not preregistered, we included them in all experiments, because they are the common practice when using data from Project Implicit and when using the IAT in general (e.g., Bar-Anan & Nosek, 2014; Greenwald et al., 2003). Including the excluded participants does not change the results.

4 Measure order did not moderate any of the effects.

5 Measurement time factor is relevant only for the group that rated the CSs post-pairing.

6 We excluded from this analysis one participant who did not rate all CSs post-pairing.

7 To save resources, we planned to stop at 500 and add 500 if the results are not significant for the two confirmatory tests. Therefore, we used p = .027, instead of p = .05, as our critical p-value (Sagarin, Ambler, & Lee, 2014). We changed the sequential testing strategy because we were not sure that the strategy used in Experiment 1 is recommended when the main hypothesis testing follows NHST.

8 Exploratory comparisons raveled that the post-pairing EC score (M = 1.37, SD = 1.84) of the participants who rated the CSs post-pairing was stronger than the post-CS-only EC score of those participants (M = 0.74, SD = 1.55), t(225) = 6.19, p < .001, d = 0.41, BF₁₀ = 2,949,833, and, stronger than the post-CS-only EC score of the participants who rated novel stimuli post-pairing (M = 0.77, SD = 1.63), t(451.15) = 3.71, p < .001, d = 0.34, BF₁₀ = 83.57, but not stronger than the post-CS-only EC score of the participants who did not evaluate any stimuli post-pairing (M = 1.33, SD = 2.09), t(469.28) = 0.214, p = 0.830, d = 0.01, BF₁₀ = 0.104. Bayes-factors in this section were computed with the default prior.

9 Measure order did not influence the results of these comparisons.

10 Gawronski et al. (2015) did not report a comparison of the post-CS-only EC effects in both conditions with the post-pairing EC effect. However, the corresponding author confirmed in a personal communication that both conditions in their Experiment 3 also led to a significant extinction effect, as in the present research.

Additional information

Funding

This research was supported by a grant from the Israel Science Foundation [779/16] to Y. B.A.