Abstract Each day brings new experiences and the opportunity to form new episodic memories. However, our everyday experiences are not isolated episodes; rather, there is significant spatial and temporal structure that holds across experiences, allowing us to build up structured knowledge about the world. We can then leverage this known structure to make predictions about how new experiences will unfold. In a recent study in Open Mind, Poskanzer, Tarder-Stoll, et al. found that such predictions benefitted episodic memory. Specifically, on trials in which participants were successfully able to make predictions about an upcoming experience, participants were more likely to encode that predictive information into memory. These findings seem to stand in stark contrast to other recent work, which found the opposite of worse episodic memory for predictive cues (Sherman et al., 2022; Sherman & Turk-Browne, 2020). How can these discrepant findings be reconciled? Here, I discuss several key task differences that might explain the discrepancy and highlight avenues for future research which might help to theoretically disentangle the contexts under which prediction may impede vs. facilitate episodic encoding.
Brynn E. Sherman (Thu,) studied this question.