Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published June 2023 | v2
Journal Article Open

Extracting probability in the absence of visual awareness


Extracting statistical regularities from the environment is crucial for survival. It allows us to learn cues for where and when future events will occur. Can we learn these associations even when the cues are not consciously perceived? Can these unconscious processes integrate information over long periods of time? We show that human visual system can track the probability of location contingency between an unconscious prime and a conscious target over a period of time of minutes. In a series of psychophysical experiments, we adopted an exogenous priming paradigm and manipulated the location contingency between a masked prime and a visible target (i.e., how likely the prime location predicted the target location). The prime's invisibility was verified both subjectively and objectively. Although the participants were unaware of both the existence of the prime and the prime-target contingency, our results showed that the probability of location contingency was tracked and manifested in the subsequent priming effect. When participants were first entrained into the fully predictive prime-target probability, they exhibited faster responses to the more predictive location. On the contrary, when no contingency existed between the prime and target initially, participants later showed faster responses to the less predictive location. These results were replicated in two more experiments with increased statistical power and a fine-grained delineation of prime awareness. Together, we report that the human visual system is capable of tracking unconscious probability over a period of time, demonstrating how implicit and uncertain regularity guides behavior.

Attached Files

Supplemental Material - 13415_2022_1057_MOESM1_ESM.docx

In Press - s13415-022-01057-1.pdf


Files (1.1 MB)
Name Size Download all
26.1 kB Download
1.1 MB Preview Download

Additional details

November 20, 2023
January 9, 2024