Back

Regularity extraction under stress: boosted statistical learning but unaffected sequence learning

Toth-Faber, E.; Janacsek, K.; Szollosi, A.; Keri, S.; Nemeth, D.

2020-09-18 neuroscience
10.1101/2020.05.13.092726 bioRxiv
Show abstract

Acute stress can crucially influence learning and memory processes. One of the key processes underlying human learning and memory is the ability of our brain to rapidly detect and extract regularities from sensory input across time and space leading to effective predictive processing. Here, we aimed to get an in-depth look into the effect of stress on the acquisition of two aspects of regularity extraction. We examined whether and how stress affects (1) processes and learning of probability-based, statistical regularities and (2) serial order-based, sequential regularities in the same experimental design, and (3) explicit access to the acquired information. Considering that the acquisition of statistical regularities is a rapid process, we primarily focused on the early phases of the task. We induced stress with the Socially Evaluated Cold Pressor Test in 27 young adults, while 26 participants were enrolled in the control group. Salivary cortisol levels and subjective ratings of affective states showed successful stress induction. After the stress induction, we measured regularity extraction with the cued Alternating Serial Reaction Time task. We found that stress promoted the acquisition of statistical regularities and did not alter the learning of sequential regularities. Post-block reports showed weaker explicit access to the sequential regularities in the stress group. Our results give a process-level understanding on how stress alters learning and memory functions related to predictive processes.

Matching journals

The top 11 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.