Global Similarity and Pattern Separation in the Human Medial Temporal Lobe Predict Subsequent Memory

Recent advances in neuroscience have given us unprecedented insight into the neural mechanisms of false memory, showing that artificial memories can be inserted into the memory cells of the hippocampus in a way that is indistinguishable from true memories. However, this alone is not enough to explain how false memories can arise naturally in the course of our daily lives. Cognitive psychology has demonstrated that many instances of false memory, both in the laboratory and the real world, can be attributed to semantic interference. Whereas previous studies have found that a diverse set of regions show some involvement in semantic false memory, none have revealed the nature of the semantic representations underpinning the phenomenon. Here we use fMRI with representational similarity analysis to search for a neural code consistent with semantic false memory. We find clear evidence that false memories emerge from a similarity-based neural code in the temporal pole, a region that has been called the "semantic hub" of the brain. We further show that each individual has a partially unique semantic code within the temporal pole, and this unique code can predict idiosyncratic patterns of memory errors. Finally, we show that the same neural code can also predict variation in true-memory performance, consistent with an adaptive perspective on false memory. Taken together, our findings reveal the underlying structure of neural representations of semantic knowledge, and how this semantic structure can both enhance and distort our memories.false memory | semantic | temporal pole | fMRI | pattern similarity E ach of us has a vast store of semantic knowledge that we apply to incoming sensory data to extract meaning from the world around us. Semantic representations are capable of capturing important structural features of the world at many different levels of abstraction, which allows for rapid and flexible responses to a diverse array of environmental challenges. This preexisting knowledge structure guides ongoing cognition, which usually aids performance, but under some circumstances can lead us into error (1-3). A striking example is the widely studied DRM (Deese, Roediger, and McDermott) false-memory illusion (4, 5). In a typical DRM task, subjects are asked to memorize a set of words such as "snow," "winter," "ice," and "warm." After a delay, subjects will typically falsely remember having seen the semantically related word "cold." It is widely agreed that this memory illusion is driven by the semantic relatedness between words contained in the encoding list (e.g., "snow") and falsely remembered words that were not actually presented (e.g., "cold"). As such, it is thought that each list item automatically, but weakly, activates the semantically related concept (Fig. 1A). This activation leads to memory confusion, either through a cumulative priming of the related lure (5, 6) or the encoding of the semantic overlap as a "gist" memory (3), resulting in a false memory unless the error is detected by some internal monitoring p...

Section: Discussionmentioning

confidence: 99%

Semantic representations in the temporal pole predict false memories

Chadwick

Anjum

Kumaran

et al. 2016

“…Is it necessary to consider attentional states to predict memory from the hippocampus during encoding? To assess the selectivity of our findings to the state-template match variable, we examined two additional variables that have been linked to encoding in prior subsequent memory studies: (i) patterns of activity without respect to attentional state (37)(38)(39), and (ii) the overall level of univariate activity (5,23,40).…”

Section: Predicting Memory From the Attentional State Of The Hippocampusmentioning

confidence: 99%

Attention promotes episodic encoding by stabilizing hippocampal representations

Aly

Turk‐Browne

2016

176

157

Attention influences what is later remembered, but little is known about how this occurs in the brain. We hypothesized that behavioral goals modulate the attentional state of the hippocampus to prioritize goal-relevant aspects of experience for encoding. Participants viewed rooms with paintings, attending to room layouts or painting styles on different trials during high-resolution functional MRI. We identified template activity patterns in each hippocampal subfield that corresponded to the attentional state induced by each task. Participants then incidentally encoded new rooms with art while attending to the layout or painting style, and memory was subsequently tested. We found that when task-relevant information was better remembered, the hippocampus was more likely to have been in the correct attentional state during encoding. This effect was specific to the hippocampus, and not found in medial temporal lobe cortex, category-selective areas of the visual system, or elsewhere in the brain. These findings provide mechanistic insight into how attention transforms percepts into memories.long-term memory | selective attention | hippocampal subfields | medial temporal lobe | representational stability W hy do we remember some things and not others? Consider a recent experience, such as the last movie complex you visited, flight you took, or restaurant at which you ate. More information was available to your senses than was stored in memory, such as the theater number of the movie, the faces of other passengers, and the color of the napkins. The selective nature of memory is adaptive, because encoding carries a cost: newly stored memories can interfere with existing ones and with our ability to learn new information in the future. What is the mechanism by which information gets selected for encoding?Attention offers a means of prioritizing information in the environment that is most relevant to behavioral goals. Attended information, in turn, has stronger control over behavior and is represented more robustly in the brain (1, 2). If attention gates which information we perceive and act upon, then it may also determine what information we remember. Indeed, attention during encoding affects both subsequent behavioral expressions of memory (3) and the extent to which activity levels in the brain predict memory formation (4-7). Although these findings suggest that attention modulates processes related to memory, how it does so is unclear.According to biased competition and other theories of attention (1, 8), task-relevant stimuli are more robustly represented in sensory systems, and thus fare better in competition with taskirrelevant stimuli for downstream processing. Indeed, there is extensive evidence that attention enhances overall activity in visual areas that represent attended vs. unattended features and locations (2, 9). Moreover, attention modulates cortical areas of the medial temporal lobe that provide input to the hippocampus (10-12).Attention can also modulate the hippocampus itself. Specifically, there is g...

“…This overlapping representation should interfere with the original representation in memory, in some cases leading to difficulty in correctly rejecting the lure image. Based on prior work in humans (13)(14)(15), it is thought that correct rejection of these similar lures depends at least in part on successful resolution of this interference in the hippocampus.…”

Section: Significancementioning

confidence: 99%

“…This computation is widely thought to occur as a function of sparse firing patterns of granule cells of the hippocampal dentate gyrus (DG), which form powerful mossy fiber synapses with large numbers of pyramidal cells in subregion CA3 (3)(4)(5)(6). A critical role for the DG in orthogonalizing interfering inputs has been demonstrated in animal studies by using lesions (7)(8)(9), NMDA receptor knockouts (10), electrophysiological recordings (11,12), and human functional MRI (13)(14)(15).…”

mentioning

confidence: 99%

Object and spatial mnemonic interference differentially engage lateral and medial entorhinal cortex in humans

Reagh

Yassa

2014

211

207

Recent models of episodic memory propose a division of labor among medial temporal lobe cortices comprising the parahippocampal gyrus. Specifically, perirhinal and lateral entorhinal cortices are thought to comprise an object/item information pathway, whereas parahippocampal and medial entorhinal cortices are thought to comprise a spatial/contextual information pathway. Although several studies in human subjects have demonstrated a perirhinal/parahippocampal division, such a division among subregions of the human entorhinal cortex has been elusive. Other recent work has implicated pattern separation computations in the dentate gyrus and CA3 subregions of the hippocampus as a mechanism supporting the resolution of mnemonic interference. However, the nature of contributions of medial temporal lobe cortices to downstream hippocampal computations is largely unknown. We used high-resolution fMRI during a task selectively taxing mnemonic discrimination of object identity or spatial location, designed to differentially engage the two information pathways in the medial temporal lobes. Consistent with animal models, we demonstrate novel evidence for a domain-selective dissociation between lateral and medial entorhinal cortex in humans, and between perirhinal and parahippocampal cortex as a function of information content. Conversely, hippocampal dentate gyrus/CA3 demonstrated signals consistent with resolution of mnemonic interference across domains. These results provide insight into the information processing capacities and hierarchical interference resolution throughout the human medial temporal lobe.T he encoding of episodic memories is known to rely on a network of brain regions within the medial temporal lobes (MTL) (1). Past models of episodic memory have largely focused on the functional role of the hippocampus and its subregions. A wealth of recent evidence has implicated pattern separation, a computation by which overlapping inputs are orthogonalized into nonoverlapping outputs, as a key process in resolving interference among similar memories (2). This computation is widely thought to occur as a function of sparse firing patterns of granule cells of the hippocampal dentate gyrus (DG), which form powerful mossy fiber synapses with large numbers of pyramidal cells in subregion CA3 (3-6). A critical role for the DG in orthogonalizing interfering inputs has been demonstrated in animal studies by using lesions (7-9), NMDA receptor knockouts (10), electrophysiological recordings (11, 12), and human functional MRI (13-15).Numerous studies have provided evidence for selective information processing in cortical areas that project to the hippocampus. Much of this research has focused on perirhinal cortex (PRC) and parahippocampal cortex (PHC, postrhinal in the rat). PRC has been shown to be critical for memory of item or object information, whereas PHC has been shown to play a role in memory of contextual or spatial information both in rats (16-18) and in humans (19)(20)(21). These dissociations are consistent with diff...