Reward Prediction Error and Declarative Memory

Abstract

Learning based on reward prediction error (RPE) was originally proposed in the context of non-declarative memory. We postulate that RPE may support declarative memory as well. Indeed, recent years have witnessed a number of independent empirical studies reporting effects of RPE on declarative memory. In this paper, we provide a brief overview of these studies, point out emerging patterns, and identify open issues such as the role of signed versus unsigned RPEs in declarative learning.

States of curiosity and interest enhance memory differently in adolescents and in children

“…In one study, adolescents showed better memory for pictures associated with positive reward prediction errors (i.e., receipt of higher monetary reward than initially expected) suggesting that increased reward sensitivity benefits learning during adolescence (Davidow, Foerde, Galván, & Shohamy, 2016; see also, Hallquist, Geier, & Luna, 2018;van den Bos, Cohen, Kahnt, & Crone, 2012). These findings corroborate similar findings in adults demonstrating that greater and more positive prediction errors enhance episodic memory (Ergo et al, 2020). As the effects of prediction error associated with intrinsically valuable information on memory have only recently been demonstrated in adults (Fastrich et al, 2018;Ligneul et al, 2018;Marvin & Shohamy, 2016), it is an open question to what extent they are also present in childhood and adolescence.…”

“…Reward prediction errors have been shown to be a key driver of learning in various contexts (Rescorla & Wagner, 1972;Schultz, 2013). Consistent with the recent findings of positive IPEs on memory, positive reward prediction errors also have beneficial effects on episodic memory in adults (e.g., Jang, Nassar, Dillon, & Frank, 2019;Pine, Sadeh, Ben-Yakov, Dudai, & Mendelsohn, 2018; for a review, see Ergo, De Loof, & Verguts, 2020). Taken together, both pre-information curiosity and post-information interest may contribute to learning success, and IPEs might be a fruitful approach to investigate how higher than initially expected situational interest might affect memory (Fastrich et al, 2018;Ligneul et al, 2018;Marvin & Shohamy, 2016).…”

Failure to modulate reward prediction errors in declarative learning with theta (6 Hz) frequency transcranial alternating current stimulation

“…Larger and more positive RPEs during study improved subsequent declarative memory during testing. The effect of RPEs in this experimental paradigm was further substantiated in a follow-up EEG study, where oscillatory signatures at reward feedback were detected in the theta (4)(5)(6)(7)(8), high-beta (20)(21)(22)(23)(24)(25)(26)(27)(28)(29)(30) and high-alpha (10-15 Hz) frequency ranges, suggesting the experience of RPEs by the participants [11]. Further validation came from an fMRI study using a similar paradigm in which famous faces were associated with Swahili village names [12].…”

“…Although this RPE effect on declarative memory could not be replicated [59,60], several research labs have since then used a range of experimental paradigms to investigate the role of RPEs in declarative learning. Most of these studies revealed positive effects of RPEs on declarative memory [8,9,61], but one study also reported negative effects [62] (for review see [7]). Results.…”

“…RPEs were primarily studied within procedural learning (e.g., [6]). However, recent evidence suggests that RPEs are crucial for declarative learning as well [7][8][9].…”