Author(s) Sascha Meyen (University of Tübingen) - sascha.meyen@uni-tuebingen.de Miguel A. Vadillo (Universidad Autónoma de Madrid) - miguel.vadillo@uam.es Ulrike von Luxburg (University of Tübingen, MPI IS, Tübingen) - ulrike.luxburg@uni-tuebingen.de Franz H. Volker (University of Tübingen) - volker.franz@uni-tuebingen.de
Pre-registered on 2021/10/26 - 07:22 AM (PT)
1) Have any data been collected for this study already? It's complicated. We have already collected some data but explain in Question 8 why readers may consider this a valid pre-registration nevertheless.
2) What's the main question being asked or hypothesis being tested in this study? In the contextual cueing paradigm, many studies infer from their results that participants recognized repeated configurations throughout the experiment implicitly. Typically, studies infer implicit recognition from the following pattern of results: (a) low sensitivity in the awareness task (interpreted as poor explicit recognition) and (b) a clear reaction time effect in the search task with faster reaction times for repeated configurations as compared to new ones (interpreted as some recognition that then must have been implicit). However, this inference is only valid if the reaction times provide evidence for a higher sensitivity performance than in the awareness task-we call this situation an indirect task advantage (ITA). Otherwise, the search task effect can be fully explained by residual explicit recognition. Thus, we want to test whether sensitivity from responses in the search task is higher than sensitivity from the awareness task.
We have investigated an ITA already in masked priming studies (Meyen et al., 2021, JEP:G; Zerweck et al., 2021, AP&P). This is a straightforward application of the method we used there to contextual cueing.
Some studies base their argument for implicit recognition on the correlation between the search task effect and the awareness task performance (e.g., Colagiuri & Livesey, 2016, PB&R). We will give a theoretical reason why our proposed sensitivity comparison is relevant also for their argument.
Meyen, S., Zerweck, I. A., Amado, C., von Luxburg, U., & Franz, V. H. (2021). Advancing research on unconscious priming: When can scientists claim an indirect task advantage? Journal of Experimental Psychology: General. Advance online publication. https://doi.org/10.1037/xge0001065
Zerweck, I.A., Kao, CS., Meyen, S. et al. Number processing outside awareness? Systematically testing sensitivities of direct and indirect measures of consciousness. Atten Percept Psychophys 83, 2510–2529 (2021). https://doi.org/10.3758/s13414-021-02312-2
3) Describe the key dependent variable(s) specifying how they will be measured. We will reanalyze studies that measure reaction time effects in the search task and performance (accuracy or sensitivity) in the awareness task.
4) How many and which conditions will participants be assigned to? The typical conditions in these studies are repeated vs. new stimulus configurations.
5) Specify exactly which analyses you will conduct to examine the main question/hypothesis. We will reanalyze the sensitivity for repeated vs. new conditions of the reaction times in the search task. For this, we will median split reaction times and "classify" (in the sense of predict) trials with fast reaction times to come from repeated configurations; trials with slow reaction times will be classified as coming from new configurations. We will then compare these classifications to the true labels and compute signal detection theory's sensitivity. We will compare these sensitivities to those from the awareness task. For justification and details, see Meyen et al. (2021).
The median split will be computed for each participant individually. If a study presents multiple experiments and conditions (such as high/low contrast conditions as in Zhao & Ren, 2020, AP&P), we will conduct the analysis separately for each condition.
If the study's trial-by-trial data is not available, we will conduct a reanalysis based on the study's reported results (see methods in Meyen et al., 2021). Because the cueing effect typically increases over time, we will focus on reanalyzing results from the end of the search task depending on which summary statistics are available in this case. We will conduct a t test for a difference in sensitivities as well as Bayes Factors.
6) Describe exactly how outliers will be defined and handled, and your precise rule(s) for excluding observations. We will exclude studies from our reanalysis that do not provide the trial-by-trial data or report statistics needed for the reanalysis. Within each study, we will adopt the exclusion criteria they used.
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. Based on an initial literature search, we attempt to reanalyze the following 21 studies.
8) Anything else you would like to pre-register? (e.g., secondary analyses, variables collected for exploratory purposes, unusual analyses planned?) We have already started reanalyzing some studies (cf. https://doi.org/10.1167/jov.21.9.1831):
Chun & Jiang (1998)
Chun & Phelps (1999)
Zhao & Ren (2020)
Vadillo, Malejka, Lee, Dienes, Shanks (2021)
Zellin, Conci, von Mühlenen, & Müller (2011)
We will mark these as "analyzed before preregistration" when preparing a manuscript. We may add additional studies to our reanalysis list. We will also mark these as "not listed in the preregistration".
