关键词: 错误记忆, 决策偏好, 奖赏学习, DRM范式, 加工分离程序

Abstract: The constructive nature of memory makes it susceptible to retrieval-related distortions that give rise to false memories. However, whether—and how—false memories directly influence value-based decisions remains unclear. We integrated the Deese–Roediger–McDermott (DRM) paradigm with a reward-learning task to test whether false memories shape decision preferences via conscious or unconscious processing. Experiment 1 combined the DRM paradigm with a reward-learning task to examine the effect of false reward memories on decision preferences. Building on Experiment 1, Experiment 2 introduced an attention-distraction task to reduce available conscious resources; we hypothesized that if false reward memories impact decision-making via conscious processing, attention distraction should attenuate false memory-based decision preferences. Experiment 3 employed the process-dissociation procedure (PDP) to dissociate conscious from unconscious components of false memory. This further tested the predictive role of both components in decision preferences, clarifying mechanisms by which false memory shapes value-based choices. We conducted three experiments to validate these hypotheses. Experiment 1 used a within-subjects two-level Reward Condition (Reward vs. Non-Reward) and included 37 participants. Building upon Experiment 1, Experiment 2 introduced an attention-distraction task, yielding a 2 (Reward: Reward vs. Non-Reward) × 2 (Attention: Full vs. Divided) mixed design with 73 participants. Experiment 3 used PDP to dissociate conscious and unconscious contributions of true and false memory in guiding decision-making and included 50 participants. Across experiments, during reward learning, items from some DRM lists were paired with reward whereas others were not. In the memory phase, participants made old/new judgments and, for ‘old’ responses, indicated whether the picture had been rewarded or not. In the decision-making phase, participants maximized earnings by choosing, between two pictures, the one most likely to yield reward. Across three experiments, the paradigm reliably elicited false reward memories for lures, which in turn consistently biased decision preferences. In Experiment 2, regarding memory for studied items and decision preferences, the attention-distraction manipulation during the encoding phase significantly reduced participants' true reward memory, and further significantly diminished their preference for selecting rewarded items. Next, we conducted trial-level mixed-effects logistic regressions (GLMER) to examine whether true reward memory (yes or no) and false reward memory (yes or no) could predict one’s decision-making preference (yes or no). Results showed that whether under the full-attention or divided-attention condition, false reward memory significantly predicted decision preferences. Simultaneously, the full attention group exhibited significantly higher preference for selecting the lure pictures compared to the distracted group, indicating that attentional distraction markedly reduced participants' decision preferences toward the lure pictures. By introducing PDP in Experiment 3, we found that false memory consisted of a greater unconscious component than conscious memory. Analyses incorporating both components accounted for decision preferences of lure pictures. Results showed that conscious and unconscious components of false memory significantly and positively predicted decision preferences of lure pictures, suggesting that both conscious and unconscious memory components of lures substantially contribute to decision preferences. These findings support a spreading-activation account whereby reward associations propagate within semantic networks, fostering false memories that subsequently bias value-based choice. Together, the results show that the influence of false memory on decision preferences cannot be fully explained by limited attentional resources and instead reflects complementary contributions of both conscious and unconscious memory processes. This work clarifies how constructive memory processes shape economically relevant decisions and motivates future neuroimaging research to delineate the neural pathways through which false-memory biases choice.

Key words: False memory, Decision making, Reward learning, DRM paradigm, Process-dissociation procedures