David J. Foster

Hippocampal place-cell sequences depict future paths to remembered goals

Effective navigation requires planning extended routes to remembered goal locations. Hippocampal place cells have been proposed to have a role in navigational planning, but direct evidence has been lacking. Here we show that before goal-directed navigation in an open arena, the rat hippocampus generates brief sequences encoding spatial trajectories strongly biased to progress from the subject’s current location to a known goal location. These sequences predict immediate future behaviour, even in cases in which the specific combination of start and goal locations is novel. These results indicate that hippocampal sequence events characterized previously in linearly constrained environments as ‘replay’ are also capable of supporting a goal-directed, trajectory-finding mechanism, which identifies important places and relevant behavioural paths, at specific times when memory retrieval is required, and in a manner that could be used to control subsequent navigational behaviour.

Memory and Space: Towards an Understanding of the Cognitive Map

More than 50 years of research have led to the general agreement that the hippocampus contributes to memory, but there has been a major schism among theories of hippocampal function over this time. Some researchers argue that the hippocampus plays a broad role in episodic and declarative memory, whereas others argue for a specific role in the creation of spatial cognitive maps and navigation. Although both views have merit, neither provides a complete account of hippocampal function. Guided by recent reviews that attempt to bridge between these views, here we suggest that reconciliation can be accomplished by exploring hippocampal function from the perspective of Tolmans (1948) original conception of a cognitive map as organizing experience and guiding behavior across all domains of cognition. We emphasize recent studies in animals and humans showing that hippocampal networks support a broad range of domains of cognitive maps, that these networks organize specific experiences within the contextually relevant map, and that network activity patterns reflect behavior guided through cognitive maps. These results are consistent with a framework that bridges theories of hippocampal function by conceptualizing the hippocampus as organizing incoming information within the context of a multidimensional cognitive map of spatial, temporal, and associational context. SIGNIFICANCE STATEMENT Research of hippocampal function is dominated by two major views. The spatial view argues that the hippocampus tracks routes through space, whereas the memory view suggests a broad role in declarative memory. Both views rely on considerable evidence, but neither provides a complete account of hippocampal function. Here we review evidence that, in addition to spatial context, the hippocampus encodes a wide variety of information about temporal and situational context, about the systematic organization of events in abstract space, and about routes through maps of cognition and space. We argue that these findings cross the boundaries of the memory and spatial views and offer new insights into hippocampal function as a system supporting a broad range of cognitive maps.

Hippocampal Replay Captures the Unique Topological Structure of a Novel Environment

Hippocampal place-cell replay has been proposed as a fundamental mechanism of learning and memory, which might support navigational learning and planning. An important hypothesis of relevance to these proposed functions is that the information encoded in replay should reflect the topological structure of experienced environments; that is, which places in the environment are connected with which others. Here we report several attributes of replay observed in rats exploring a novel forked environment that support the hypothesis. First, we observed that overlapping replays depicting divergent trajectories through the fork recruited the same population of cells with the same firing rates to represent the common portion of the trajectories. Second, replay tended to be directional and to flip the represented direction at the fork. Third, replay-associated sharp-wave–ripple events in the local field potential exhibited substructure that mapped onto the maze topology. Thus, the spatial complexity of our recording environment was accurately captured by replay: the underlying neuronal activities reflected the bifurcating shape, and both directionality and associated ripple structure reflected the segmentation of the maze. Finally, we observed that replays occurred rapidly after small numbers of experiences. Our results suggest that hippocampal replay captures learned information about environmental topology to support a role in navigation.

Autoassociative dynamics in the generation of sequences of hippocampal place cells

Memory storage in neural networks Neuronal networks can store and retrieve discrete memories, but often fail to retrieve stored sequences. This is because error decreases over time for a static attractor, but builds up drastically over time if patterns are not trained to retrieve themselves but to retrieve the next item in a sequence. Pfeiffer and Foster studied brain activity in awake but immobile rats. Recording simultaneously from a large number of place cells in the hippocampal formation, they found that internally generated sequences alternated between periods of hovering in place while being strengthened, and periods of abrupt transition to a new place. Science, this issue p. 180 Place cell representations of spatial trajectories in awake but immobile animals jump from one location to another. Neuronal circuits produce self-sustaining sequences of activity patterns, but the precise mechanisms remain unknown. Here we provide evidence for autoassociative dynamics in sequence generation. During sharp-wave ripple (SWR) events, hippocampal neurons express sequenced reactivations, which we show are composed of discrete attractors. Each attractor corresponds to a single location, the representation of which sharpens over the course of several milliseconds, as the reactivation focuses at that location. Subsequently, the reactivation transitions rapidly to a spatially discontiguous location. This alternation between sharpening and transition occurs repeatedly within individual SWRs and is locked to the slow-gamma (25 to 50 hertz) rhythm. These findings support theoretical notions of neural network function and reveal a fundamental discretization in the retrieval of memory in the hippocampus, together with a function for gamma oscillations in the control of attractor dynamics.

Sequence Learning and the Role of the Hippocampus in Rodent Navigation

The hippocampus has long been associated with navigation and spatial representations, but it has been difficult to link directly the neurophysiological correlates of hippocampal place cells with navigational planning and action. In recent years, large-scale population recordings of place cells have revealed that spatial sequences are stored and activated in ways that may support navigational strategies. Plasticity mechanisms allow the hippocampus to store learned sequences of locations that may allow predictions of future locations based on past experience. These sequences can also be activated during navigational behavior in ways that may allow the animal to learn trajectories toward goals. Task-dependent alterations in place cell firing patterns may reflect the operation of the hippocampus in associating locations with navigationally relevant decision variables.

Reverse Replay of Hippocampal Place Cells Is Uniquely Modulated by Changing Reward

Hippocampal replays are episodes of sequential place cell activity during sharp-wave ripple oscillations (SWRs). Conflicting hypotheses implicate awake replay in learning from reward and in memory retrieval for decision making. Further, awake replays can be forward, in the same order as experienced, or reverse, in the opposite order. However, while the presence or absence of reward has been reported to modulate SWR rate, the effect of reward changes on replay, and on replay direction in particular, has not been examined. Here we report divergence in the response of forward and reverse replays to changing reward. While both classes of replays were observed at reward locations, only reverse replays increased their rate at increased reward or decreased their rate at decreased reward, while forward replays were unchanged. These data demonstrate a unique relationship between reverse replay and reward processing and point to a functional distinction between different directions of replay. VIDEO ABSTRACT.

Dissociation between the Experience-Dependent Development of Hippocampal Theta Sequences and Single-Trial Phase Precession

Theta sequences are circuit-level activity patterns produced when groups of hippocampal place cells fire in sequences that reflect a compressed behavioral order of place fields within each theta cycle. The high temporal coordination between place cells exhibited in theta sequences is compatible with the induction of synaptic plasticity and has been proposed as one of the mechanisms underlying the encoding of episodic memory of recently acquired experience. Yet how theta sequences develop with experience has not been directly addressed. Here we simultaneously recorded large numbers of cells in the dorsal hippocampal CA1 area from rats exploring on a novel linear track. Although place cell firing activities accurately represented the animals current location, distinct theta sequences were absent on the first lap but emerged immediately thereafter and remained stable once established. The absence of theta sequences on the first lap was not due to place field instability, decreased overall excitability of place cells, behavior variables, or the absence of individual neuronal phase precession. We observed strong single-lap phase precession in a significant percentage of place fields on the first lap and throughout the recording. Individual neuronal phase precession was stable from the first lap to subsequent laps but, across neurons, phase precession became more synchronized after experience, suggesting a novel mechanism for the generation of theta sequences. These results suggest that experience-independent temporal coding in individual neurons is combined with rapid plasticity of hippocampal neural networks during experience to acquire predictive representations of the immediate future.

Disordered Ripples Are a Common Feature of Genetically Distinct Mouse Models Relevant to Schizophrenia

We present results from a novel comparative approach to the study of the mechanisms of psychiatric disease. Previous work has examined neural activity patterns in the hippocampus of a freely behaving mouse model associated with schizophrenia, the calcineurin knockout mouse. Here, we examined a genetically distinct mouse model that exhibits a similar set of behavioral phenotypes associated with schizophrenia, a transgenic model expressing a putative dominant-negative DISC1 (DN-DISC1). Strikingly, the principal finding of the earlier work is replicated in the DN-DISC1 mice, that is, a selective increase in the numbers of sharp-wave ripple (SWR) events in the hippocampal local field potential (LFP), while at the same time other LFP patterns such as theta and gamma are unaffected. SWRs are thought to arise from hippocampal circuits and reflect the coordinated activity of the principal excitatory cells of the hippocampus in specific patterns that represent reactivated memories of previous experiences and imagined future experiences that predict behavior. These findings suggest that multiple genetic alterations could converge on distinct patterns of aberrant neurophysiological function to give rise to common behavioral phenotypes in psychiatric disease.

Discovering the Brain’s Cognitive Map

Identifying our position in space is critical for navigation, and also for our ability to form memories of behavioral episodes, because these occur in a specific time and place. Understanding the neuronal mechanisms underlying these abilities has been an enduring question in neuroscience: how do our brains combine external sensory information with internal self-motion cues to produce a circuit-level representation of spatial location, and how are these representations used to help us understand where we are and to allow us to successfully navigate through our environment? Three neuroscientists, John O’Keefe, May-Britt Moser, and Edvard Moser, were recently awarded the 2014 Nobel Prize in Physiology or Medicine for their seminal research in understanding how 2 neighboring brain regions, the hippocampus and the entorhinal cortex, represent spatial information. Their groundbreaking work not only established these areas as primary sources of spatial processing in the brain but created and sustained a research field within neuroscience dedicated to unveiling the processes by which positional and episodic information is encoded and extracted at the cellular level. Prior to the work of O’Keefe, Moser, and Moser, there was considerable debate and speculation regarding the nature of spatial representation in the brain. In 1948, Edward Tolman coined the term cognitive map to describe the apparent ability of animals to create an internal representation of the external environment.1 This concept was primarily driven by a series of experiments in which rats demonstrated spatial learning behavior that could not be fully explained by simple stimulus-response behavior. Rather, rats appeared to acquire, through experience, a more complex understanding of how their environment was structured. It was hypothesized that access to this mental map allowed animals to create shortcuts or establish novel trajectories to obtain a reward more quickly or effectively. In 1971, O’Keefe and Dostrovsky2 provided the first evidence that neurons within the hippocampus specifically contribute to a cognitive map by directly representing an animal’s physical location. In a series of studies using the then-relatively-new technique of in vivo electrophysiology, O’Keefe demonstrated that single neurons in the dorsal hippocampus of the rat (homologue of the human posterior hippocampus) consistently fired action potentials only when the animal was in a specific, restricted location of an environment. Because the activity of these neurons consistently reflected the rat’s position in space, they were given the moniker place cells. Critically, it was established that although place-cell representations could be influenced by sensory cues in the environment, their activity patterns were not a trivial reflection of primary sensory input but, rather, were more consistent with a representation of spatial location. The discovery of place cells laid the foundation for a neuronal basis for the cognitive map.3 Decades of continued research, led by O’Keefe and many others, established that the preferred firing location, or “place field,” of a place cell in one environment did not predict the field in another, suggesting that, unlike topographic maps in the brain, the hippocampus “remaps” in new environments, enabling a combinatorial code that represents without interference the thousands of environments an individual might be expected to encounter in life. For several decades following the discovery of place cells, it was postulated that hippocampal neurons obtained their position-specific firing patterns as a result of the intrinsic circuitry within the hippocampus. The work of Jim Ranck and colleagues established that cells in many areas closely connected to the hippocampus represent head direction, providing a compass-like representation.4 However, early results had suggested that spatial location itself was coded more weakly in upstream regions. May-Britt Moser and Edvard Moser decided to record from the specific region that projected to hippocampal place cells, despite the fact that these neurons in layers II and III of the dorsal medial entorhinal cortex could only be targeted with deeply implanted and awkwardly angled electrodes, compared with the more accessible upstream regions that had been studied before. This strategy was transformative; the Mosers discovered “grid cells” in the medial entorhinal cortex that fire in multiple locations within the same environment in a precise grid of equilateral triangles.5 The spatial frequency of grid nodes increases systematically along the dorsoventral axis of the medial entorhinal cortex, while, at any given point along the axis, different grid cells fire at different spatial phases. Thus, while each grid cell fires in multiple locations within an environment, each location in that environment is represented by a unique set of active grid cells. An explosion of interest followed, with observations of grid cells in various species, as well as reports of conjunctive cells that simultaneously code for both grid and head direction and of border cells that code for walls within an environment. Place cells and grid cells reside in the medial temporal lobe, a region best known for its role in memory. Fascinatingly, the spatial mapping firing properties of neurons in these areas appear to provide a scaffold for memory construction. First, hippocampal place cells do not only report current location, but their activity is sensitive to task demands and modulated by the past and future journeys taken. Second, as the technology has advanced to allow recording from large numbers of hippocampal place cells simultaneously, it has become clear that place cells are activated in temporal sequences. During exploration, place cells are modulated by a strong VIEWPOINT

Abnormal Sleep Architecture and Hippocampal Circuit Dysfunction in a Mouse Model of Fragile X Syndrome

Fragile X syndrome (FXS) is the most common heritable cause of intellectual disability and single-gene cause of autism spectrum disorder. The Fmr1 null mouse models much of the human disease including hyperarousal, sensory hypersensitivity, seizure activity, and hippocampus-dependent cognitive impairment. Sleep architecture is disorganized in FXS patients, but has not been examined in Fmr1 knockout (Fmr1-KO) mice. Hippocampal neural activity during sleep, which is implicated in memory processing, also remains uninvestigated in Fmr1-KO mice. We performed in vivo electrophysiological studies of freely behaving Fmr1-KO mice to assess neural activity, in the form of single-unit spiking and local field potential (LFP), within the hippocampal CA1 region during multiple differentiated sleep and wake states. Here, we demonstrate that Fmr1-KO mice exhibited a deficit in rapid eye movement sleep (REM) due to a reduction in the frequency of bouts of REM, consistent with sleep architecture abnormalities of FXS patients. Fmr1-KO CA1 pyramidal cells (CA1-PCs) were hyperactive in all sleep and wake states. Increased low gamma power in CA1 suggests that this hyperactivity was related to increased input to CA1 from CA3. By contrast, slower sharp-wave ripple events (SWRs) in Fmr1-KO mice exhibited longer event duration, slower oscillation frequency, with reduced CA1-PC firing rates during SWRs, yet the incidence rate of SWRs remained intact. These results suggest abnormal neuronal activity in the Fmr1-KO mouse during SWRs, and hyperactivity during other wake and sleep states, with likely adverse consequences for memory processes.