At the heart of consciousness research lies the contrastive method, in which a conscious condition is compared to an unconscious condition to identify the neural basis of consciousness and/or the function of consciousness. Vast literature attempts to identify the extent and complexity of unconscious processing and unconscious cognition. However, there are well-known flaws in using statistical procedures to establish such unconscious processes, most notably the practice of using frequentist statistics to show the absence of a direct effect on consciousness. This may overestimate the degree to which certain processes can take place unconsciously, and thereby decrease the validity of the contrastive method. A new project led by Johannes Fahrenfort at Vrije Universiteit Amsterdam aims to introduce a valid novel solid statistical framework for establishing unconscious cognition in the brain and/or in behavior.

This project will implement and utilize a new technique to mitigate these issues through a statistical framework that uses:

• Pre-registration
• Bayesian ordinal model selection
• Optional stopping

The ability to estimate the evidence (Bayes-factor) for the absence of a direct effect on consciousness while concurrently establishing the presence of an indirect effect on unconscious processing or unconscious cognition within the same statistical model is central to this method.

The research team will use this framework to evaluate several currently accepted claims in the literature about the extent and complexity of unconscious cognition and will examine two distinct questions:

• Is feedforward processing conscious or unconscious?
• To what extent is unconscious priming really excluded from conscious perception? 

The project is aimed at consciousness scientists, and the framework and empirical data will be published in several scientific papers. If successful, the findings could have a significant impact on the scientific community and may change the way consciousness scientists do consciousness science. All experiments will be pre-registered on the Open Science Framework, and all data will be openly available after completion of the project.