The Role of Flexible Retrieval Mechanisms in False Memories for Contextual and Value-Based Details in Young and Older Adults: Behavioral and Neuroimaging Evidence

Abstract

Research suggests that certain errors in episodic memory can be viewed as byproducts of adaptive constructive processes that play a functional role in memory but produce errors or distortions as an unintended consequence. An important, adaptive feature of episodic memory is the capacity to flexibly retrieve and recombine information from past experiences into novel representations. For example, recent evidence suggests that flexible recombination plays a key role in our capacity to make inferences based on distinct past events that share a common feature. The current dissertation aims to examine whether the domain of adaptive memory distortions, where a memory error (source memory errors for contextual details/reward) results from carrying out a cognitive operation that has demonstrably beneficial consequences on another aspect of performance, extends to associative inference by investigating a) the cognitive byproducts of flexible retrieval mechanisms that support successful associative inference (Paper 1), b) how flexible retrieval mechanisms and the consequences of such mechanisms may change as a result of normal aging (Paper 2), and c) how such mechanisms affect the neural representations of overlapping, yet distinct events in support of successful inference (Paper 3). Paper 1 examines the impact of flexible retrieval on false memories by extending our associative inference paradigm to the domain of value-based decision-making by targeting false memories for value information. We demonstrate that after successful inference, participants were more likely to misattribute specific value information across event boundaries to a previously unrewarded item compared to after unsuccessful inference. Paper 2 investigates how age-related changes in flexible retrieval mechanisms affect participants’ susceptibility to false memories resulting from successful inference. In contrast to younger adults, results reveal that older adults were not more likely to misattribute contextual details across event boundaries after successful relative to unsuccessful inference. Paper 3 aims to determine how flexible retrieval mechanisms affect the neural reinstatement of overlapping, yet incorrect contextual details and false memories during subsequent retrieval attempts. After successful inference, neural patterns in the anterior hippocampus, posterior medial prefrontal cortex (i.e., regions associated with the flexible retrieval of past events), and content-reinstatement regions, were more similar to the overlapping, yet incorrect context compared to pattern similarity after unsuccessful inference. Further, the degree of univariate activity in the hippocampus during the directly learned/associative inference test correlated with the reinstatement of the overlapping, yet incorrect context in our content-reinstatement region, which correlated with the strength of participants’ behavioral false memory effects. The results of these papers suggest that flexible retrieval mechanisms that support successful inference also play a role in the misattribution of contextual details across event boundaries in younger adults but not older adults, and that for younger adults these misattributed details are reinstated during subsequent retrieval attempts, resulting in false memories.