Search

The hippocampal formation encodes maps of the physical environment [1-5]. A key question in neuroscience is whether its spatial coding principles also provide a universal metric for the organization of non-spatial information. Initial evidence comes from studies revealing directional modulation of fMRI responses in humans [6, 7] during navigation through abstract spaces and the involvement of place and grid cells in encoding of non-spatial feature dimensions [8]. However, a critical feature of a map-like representation is information about distances between locations, which has yet only been demonstrated for physical space [4, 9]. Here, we probe whether the hippocampus similarly encodes distances between points in an abstract space spanned by continuous stimulus-feature dimensions that were relevant to the acquisition of a novel concept. We find that, after learning, two-dimensional distances between individual positions in the abstract space were represented in the hippocampal multivoxel pattern as well as in the univariate hippocampal signal as indexed by fMRI adaptation. These results support the notion that the hippocampus computes domain-general, multidimensional cognitive maps along continuous dimensions., (Copyright © 2019 Elsevier Ltd. All rights reserved.)

Grid cells are one of the core building blocks of spatial navigation [1]. Single-cell recordings of grid cells in the rodent entorhinal cortex revealed hexagonal coding of the local environment during spatial navigation [1]. Grid-like activity has also been identified in human single-cell recordings during virtual navigation [2]. Human fMRI studies further provide evidence that grid-like signals are also accessible on a macroscopic level [3-7]. Studies in both non-human primates [8] and humans [9, 10] suggest that grid-like coding in the entorhinal cortex generalizes beyond spatial navigation during locomotion, providing evidence for grid-like mapping of visual space during visual exploration-akin to the grid cell positional code in rodents during spatial navigation. However, electrophysiological correlates of the grid code in humans remain unknown. Here, we provide evidence for grid-like, hexadirectional coding of visual space by human high-frequency activity, based on two independent datasets: non-invasive magnetoencephalography (MEG) in healthy subjects and entorhinal intracranial electroencephalography (EEG) recordings in an epileptic patient. Both datasets consistently show a hexadirectional modulation of broadband high-frequency activity (60-120 Hz). Our findings provide first evidence for a grid-like MEG signal, indicating that the human entorhinal cortex codes visual space in a grid-like manner [8-10], and support the view that grid coding generalizes beyond environmental mapping during locomotion [4-6, 11]. Due to their millisecond accuracy, MEG recordings allow linking of grid-like activity to epochs during relevant behavior, thereby opening up the possibility for new MEG-based investigations of grid coding at high temporal resolution., (Copyright © 2018 Elsevier Ltd. All rights reserved.)

The German psychiatrist and neuropathologist Alois Alzheimer was fascinated by the symptoms of Auguste D., a 50-year-old woman admitted to the Frankfurt Psychiatric Hospital in 1901 who suffered from memory disturbances, paranoia and progressive confusion. After her death and autopsy, Alzheimer described histological alterations in her brain that later came to be known as amyloid plaques and neurofibrillary tangles (Figure 1). The case report was published in a psychiatric textbook some years later, and this peculiar and (at the time) seemingly rare illness was later named Alzheimer's disease., (Copyright © 2018.)

Memories are thought to be retrieved by attractor dynamics if a given input is sufficiently similar to a stored attractor state [1-5]. The hippocampus, a region crucial for spatial navigation [6-12] and episodic memory [13-18], has been associated with attractor-based computations [5, 9], receiving support from the way rodent place cells "remap" nonlinearly between spatial representations [19-22]. In humans, nonlinear response patterns have been reported in perceptual categorization tasks [23-25]; however, it remains elusive whether human memory retrieval is driven by attractor dynamics and what neural mechanisms might underpin them. To test this, we used a virtual reality [7, 11, 26-28] task where participants learned object-location associations within two distinct virtual reality environments. Participants were subsequently exposed to four novel intermediate environments, generated by linearly morphing the background landscapes of the familiar environments, while tracking fMRI activity. We show that linear changes in environmental context cause linear changes in activity patterns in sensory cortex but cause dynamic, nonlinear changes in both hippocampal activity pattern and remembered locations. Furthermore, the sigmoidal response in the hippocampus scaled with the strength of the sigmoidal pattern in spatial memory. These results indicate that mnemonic decisions in an ambiguous novel context relate to putative attractor dynamics in the hippocampus, which support the dynamic remapping of memories., (Copyright © 2016 Elsevier Ltd. All rights reserved.)

Integration of separate memories forms the basis of inferential reasoning--an essential cognitive process that enables complex behavior. Considerable evidence suggests that both hippocampus and medial prefrontal cortex (mPFC) play a crucial role in memory integration. Although previous studies indicate that theta oscillations facilitate memory processes, the electrophysiological mechanisms underlying memory integration remain elusive. To bridge this gap, we recorded magnetoencephalography data while participants performed an inference task and employed novel source reconstruction techniques to estimate oscillatory signals from the hippocampus. We found that hippocampal theta power during encoding predicts subsequent memory integration. Moreover, we observed increased theta coherence between hippocampus and mPFC. Our results suggest that integrated memory representations arise through hippocampal theta oscillations, possibly reflecting dynamic switching between encoding and retrieval states, and facilitating communication with mPFC. These findings have important implications for our understanding of memory-based decision making and knowledge acquisition., (Copyright © 2016 Elsevier Ltd. All rights reserved.)

Our memories are remarkably dynamic and allow us to reinterpret the past once new information comes to light. Gaining novel insights can lead to mental reorganization of previously unrelated events, thus linking them into narratives. The hippocampus and medial prefrontal cortex (mPFC) support integration of partially overlapping events, but the neural mechanisms underlying the reorganization of memories for the formation of coherent narratives remain elusive. Here, we combine fMRI with The Sims 3 videos of life-like animated events, which could either be integrated into narratives or not. We show that insight triggers the emergence of de novo mnemonic representations of the narratives and is associated with increased neural similarity between linked event representations in the posterior hippocampus, mPFC, and autobiographical-memory network. Simultaneously, events irrelevant to the newly established memory of the narrative were pruned out. This process was accompanied by increased neural dissimilarity between non-linked event representations in the posterior hippocampus and mPFC and was additionally signaled by a mismatch response in the anterior hippocampus. Our results demonstrate that insight leads to neural reconfiguration of representational networks within a memory space and have implications for knowledge acquisition in educational settings., (Copyright © 2015 Elsevier Ltd. All rights reserved.)

Two recent imaging studies have shed new light on information representation in human parahippocampal cortex. Despite their different approaches, the two studies both support the view that this brain region represents space at an elementary level., (Copyright © 2011 Elsevier Ltd. All rights reserved.)

The neural responses that distinguish novel from familiar items in recognition memory tasks are remarkably fast in both humans and nonhuman primates. In humans, the earliest onsets of neural novelty effects emerge at about approximately 150-200 ms after stimulus onset. However, in recognition memory studies with nonhuman primates, novelty effects can arise at as early as 70-80 ms. Here, we address the possibility that this large species difference in onset latencies is caused experimentally by the necessity of using reward reinforcement to motivate the detection of novel or familiar items in nonhuman primates but not in humans. Via magnetoencephalography in humans, we show in two experiments that the onset of neural novelty signals is accelerated from approximately 200 ms to approximately 85 ms if correct recognition memory for either novel or familiar items is rewarded. Importantly, this acceleration is independent of whether the detection of the novel or the familiar scenes is rewarded. Furthermore, this early novelty effect contributed to memory retrieval because neural reward responses, which were contingent upon novelty detection, followed approximately 100 ms later. Thus, under the contextual influence of reward motivation, behaviorally relevant novelty signals emerge much faster than previously held possible in humans.