The effect of co-occurrence and relational information on speeded evaluation

ABSTRACT

After co-occurrence of a neutral conditioned stimulus (CS) with an affective unconditioned stimulus (US), the evaluation of the CS acquires the US valence. This effect disappears when information about the CS-US relation indicates that they are opposite in valence. In that case, people often show a contrastive effect, evaluating the CS with valence opposite of the US. We investigated whether the assimilative effect of co-occurrence persists and is only obscured by a stronger counteracting contrast effect of the inference from the CS-US opposition relation. Participants evaluated CSs that had opposite relations with the USs under time pressure, a condition that restricts inference processes more than it restricts the associative processes that might underlie the assimilative effect of co-occurrence. Evidence supporting the persistence of the assimilative effect emerged only in Experiment 2 (N = 79): Under time pressure, people evaluated creatures that ended positive sounds more favourably than creatures that ended negative sounds. However, no difference between the creatures’ evaluations occurred under time pressure in Experiments 1 and 3 (Ns = 78, 460). These results are inconclusive because they might reflect equal contrastive and assimilative effects or no effects at all. We discuss further research directions to test our question.

KEYWORDS:

• Evaluative conditioning
• co-occurrence
• relational information
• speeded evaluation

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1 We report all data exclusions, manipulations, and measures, and how we determined sample size. Decisions to stop collecting data did not depend on the obtained results. Materials and data are in osf.io/37myb/.

2 A shorter duration in a pilot study resulted with low success-rate of rating the creatures on time.

3 In Experiment 1 and 2, analyzing only accurate participants found the same results as the analyses that included all the participants (see osf.io/n7wy6/ for full details).

4 Computed using the Metafor package in R (Viechtbauer, 2010). Following the recommendations by Lakens (2013), Cohen’s d was calculated using SD pooled.

Additional information

Funding

This work was supported by grants from the Israel Science Foundation [779/16] and the United States Israel Binational Science Foundation [2013214] to Y. B.A.