Author(s) Moritz Ingendahl (Ruhr-Universität Bochum) - moritz.ingendahl@rub.de Franziska Schäfer (University of Mannheim) - franziska.schaefer@uni-mannheim.de
Pre-registered on 02/25/2023 02:16 AM (PT)
1) Have any data been collected for this study already? No, no data have been collected for this study yet.
2) What's the main question being asked or hypothesis being tested in this study? We investigate the role of metamemory in evaluative conditioning (EC).
We expect that pairings with positive/neutral/negative USs lead to corresponding evaluations of CSs, such that positive > neutral > negative.
We expect that US valence also impacts JOLs, such that positive and negative USs lead to higher JOLs than neutral USs.
We expect that EC effects are stronger for pairings where participants can identify the correct US.
We expect that JOLs predict correct US memory, and therefore, JOLs also moderate the EC effect.
3) Describe the key dependent variable(s) specifying how they will be measured. We assess JOLs with the question: "How likely is it that you will remember this picture if your are presented with this brand name?", for each of the 24 CS.
We assess CS Evaluations and CS-US memory with the same instructions and tasks as Ingendahl and Vogel (2023).
4) How many and which conditions will participants be assigned to? US valence (positive/neutral/negative, within-subjects). There are 8 CS-US pairs per valence level.
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. We will analyze each dependent variable with a multilevel regression in lme4, using the highest converging random effect structure. For CS-US memory, we will use a binomial multilevel regression. In each model, we will use two dummy variables (positive + negative valence, baseline = neutral) as predictors.
For CS evaluations, we expect a positive effect of positive valence and a negative effect of negative valence.
For JOLs, we expect two positive effects.
For CS-US memory, we do not expect any effects.
Next, we will compute within-subjects gamma correlations between JOLs and CS-US memory.
Next, we will do one multilevel model where we add the z-standardized (within-person-centered) JOLs as a moderator to the model predicting CS evaluations. The model will have a main effect of within-person JOL, the two dummies, and two interaction terms. We expect two significant interaction terms (positive for positive, negative for negative interaction term).
Next, we will do one multilevel model where we add CS-US memory (0 vs. 1) as a moderator to the model predicting CS evaluations. The model will have a main effect of memory, the two dummies, and two interaction terms. We expect two significant interaction terms (positive for positive, negative for negative interaction term).
Finally, we will add standardized memory (grand-mean centered) into the model with JOL as the moderator, again with the two interaction terms. We leave open whether the JOL*positive and JOL*negative interactions are still significant.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. We do not plan any exclusions.
7) How many observations will be collected or what will determine sample size? No need to justify decision, but be precise about exactly how the number will be determined. We will collect 50 finished interviews.
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) --
