Continuous Bump Attractor Networks Require Explicit Error Coding for Gain Recalibration.

Representations of continuous variables are crucial to create internal models of the external world. A prevailing model of how the brain maintains these representations is given by continuous bump attractor networks (CBANs) in a broad range of brain functions across different areas, such as spatial navigation in hippocampal/entorhinal circuits and working memory in prefrontal cortex. Through recurrent connections, a CBAN maintains a persistent activity bump, whose peak location can vary along a neural space, corresponding to different values of a continuous variable. To track the value of a continuous variable changing over time, a CBAN updates the location of its activity bump based on inputs that encode the changes in the continuous variable (e.g., movement velocity in the case of spatial navigation)-a process akin to mathematical integration. This integration process is not perfect and accumulates error over time. For error correction, CBANs can use additional inputs providing ground-truth information about the continuous variable's correct value (e.g., visual landmarks for spatial navigation). These inputs enable the network dynamics to automatically correct any representation error. Recent experimental work on hippocampal place cells has shown that, beyond correcting errors, ground-truth inputs also fine-tune the gain of the integration process, a crucial factor that links the change in the continuous variable to the updating of the activity bump's location. However, existing CBAN models lack this plasticity, offering no insights into the neural mechanisms and representations involved in the recalibration of the integration gain. In this paper, we explore this gap by using a ring attractor network, a specific type of CBAN, to model the experimental conditions that demonstrated gain recalibration in hippocampal place cells. Our analysis reveals the necessary conditions for neural mechanisms behind gain recalibration within a CBAN. Unlike error correction, which occurs through network dynamics based on ground-truth inputs, gain recalibration requires an additional neural signal that explicitly encodes the error in the network's representation via a rate code. Finally, we propose a modified ring attractor network as an example CBAN model that verifies our theoretical findings. Combining an error-rate code with Hebbian synaptic plasticity, this model achieves recalibration of integration gain in a CBAN, ensuring accurate representation for continuous variables.

Continuous Bump Attractor Networks Require Explicit Error Coding for Gain Recalibration.

Representations of continuous variables are crucial to create internal models of the external world. A prevailing model of how the brain maintains these representations is given by continuous bump attractor networks (CBANs) in a broad range of brain functions across different areas, such as spatial navigation in hippocampal/entorhinal circuits and working memory in prefrontal cortex. Through recurrent connections, a CBAN maintains a persistent activity bump, whose peak location can vary along a neural space, corresponding to different values of a continuous variable. To track the value of a continuous variable changing over time, a CBAN updates the location of its activity bump based on inputs that encode the changes in the continuous variable (e.g., movement velocity in the case of spatial navigation)-a process akin to mathematical integration. This integration process is not perfect and accumulates error over time. For error correction, CBANs can use additional inputs providing ground-truth information about the continuous variable's correct value (e.g., visual landmarks for spatial navigation). These inputs enable the network dynamics to automatically correct any representation error. Recent experimental work on hippocampal place cells has shown that, beyond correcting errors, ground-truth inputs also fine-tune the gain of the integration process, a crucial factor that links the change in the continuous variable to the updating of the activity bump's location. However, existing CBAN models lack this plasticity, offering no insights into the neural mechanisms and representations involved in the recalibration of the integration gain. In this paper, we explore this gap by using a ring attractor network, a specific type of CBAN, to model the experimental conditions that demonstrated gain recalibration in hippocampal place cells. Our analysis reveals the necessary conditions for neural mechanisms behind gain recalibration within a CBAN. Unlike error correction, which occurs through network dynamics based on ground-truth inputs, gain recalibration requires an additional neural signal that explicitly encodes the error in the network's representation via a rate code. Finally, we propose a modified ring attractor network as an example CBAN model that verifies our theoretical findings. Combining an error-rate code with Hebbian synaptic plasticity, this model achieves recalibration of integration gain in a CBAN, ensuring accurate representation for continuous variables.

Multiplexing of temporal and spatial information in the lateral entorhinal cortex.

Episodic memory involves the processing of spatial and temporal aspects of personal experiences. The lateral entorhinal cortex (LEC) plays an essential role in subserving memory. However, the specific mechanism by which LEC integrates spatial and temporal information remains elusive. Here, we recorded LEC neurons while rats performed foraging and shuttling behaviors on one-dimensional, linear or circular tracks. Unlike open-field foraging tasks, many LEC cells displayed spatial firing fields in these tasks and demonstrated selectivity for traveling directions. Furthermore, some LEC neurons displayed changes in the firing rates of their spatial rate maps during a session, a phenomenon referred to as rate remapping. Importantly, this temporal modulation was consistent across sessions, even when the spatial environment was altered. Notably, the strength of temporal modulation was found to be greater in LEC compared to other brain regions, such as the medial entorhinal cortex (MEC), CA1, and CA3. Thus, the spatial rate mapping observed in LEC neurons may serve as a coding mechanism for temporal context, allowing for flexible multiplexing of spatial and temporal information.

Object-place-context learning impairment correlates with spatial learning impairment in aged Long-Evans rats.

The hippocampal formation is vulnerable to the process of normal aging. In humans, the extent of this age-related deterioration varies among individuals. Long-Evans rats replicate these individual differences as they age, and therefore they serve as a valuable model system to study aging in the absence of neurodegenerative diseases. In the Morris water maze, aged memory-unimpaired (AU) rats navigate to remembered goal locations as effectively as young rats and demonstrate minimal alterations in physiological markers of synaptic plasticity, whereas aged memory-impaired (AI) rats show impairments in both spatial navigation skills and cellular and molecular markers of plasticity. The present study investigates whether another cognitive domain is affected similarly to navigation in aged Long-Evans rats. We tested the ability of young, AU, and AI animals to recognize novel object-place-context (OPC) configurations and found that performance on the novel OPC recognition paradigm was significantly correlated with performance on the Morris water maze. In the first OPC test, young and AU rats, but not AI rats, successfully recognized and preferentially explored objects in novel OPC configurations. In a second test with new OPC configurations, all age groups showed similar OPC associative recognition memory. The results demonstrated similarities in the behavioral expression of associative, episodic-like memory between young and AU rats and revealed age-related, individual differences in functional decline in both navigation and episodic-like memory abilities.

Assessments of dentate gyrus function: discoveries and debates.

There has been considerable speculation regarding the function of the dentate gyrus (DG) - a subregion of the mammalian hippocampus - in learning and memory. In this Perspective article, we compare leading theories of DG function. We note that these theories all critically rely on the generation of distinct patterns of activity in the region to signal differences between experiences and to reduce interference between memories. However, these theories are divided by the roles they attribute to the DG during learning and recall and by the contributions they ascribe to specific inputs or cell types within the DG. These differences influence the information that the DG is thought to impart to downstream structures. We work towards a holistic view of the role of DG in learning and memory by first developing three critical questions to foster a dialogue between the leading theories. We then evaluate the extent to which previous studies address our questions, highlight remaining areas of conflict, and suggest future experiments to bridge these theories.

Global remapping in granule cells and mossy cells of the mouse dentate gyrus.

Hippocampal place cells exhibit spatially modulated firing, or place fields, which can remap to encode changes in the environment or other variables. Unique among hippocampal subregions, the dentate gyrus (DG) has two excitatory populations of place cells, granule cells and mossy cells, which are among the least and most active spatially modulated cells in the hippocampus, respectively. Previous studies of remapping in the DG have drawn different conclusions about whether granule cells exhibit global remapping and contribute to the encoding of context specificity. By recording granule cells and mossy cells as mice foraged in different environments, we found that by most measures, both granule cells and mossy cells remapped robustly but through different mechanisms that are consistent with firing properties of each cell type. Our results resolve the ambiguity surrounding remapping in the DG and suggest that most spatially modulated granule cells contribute to orthogonal representations of distinct spatial contexts.

Superficial-layer versus deep-layer lateral entorhinal cortex: Coding of allocentric space, egocentric space, speed, boundaries, and corners.

Entorhinal cortex is the major gateway between the neocortex and the hippocampus and thus plays an essential role in subserving episodic memory and spatial navigation. It can be divided into the medial entorhinal cortex (MEC) and the lateral entorhinal cortex (LEC), which are commonly theorized to be critical for spatial (context) and non-spatial (content) inputs, respectively. Consistent with this theory, LEC neurons are found to carry little information about allocentric self-location, even in cue-rich environments, but they exhibit egocentric spatial information about external items in the environment. The superficial and deep layers of LEC are believed to mediate the input to and output from the hippocampus, respectively. As earlier studies mainly examined the spatial firing properties of superficial-layer LEC neurons, here we characterized the deep-layer LEC neurons and made direct comparisons with their superficial counterparts in single unit recordings from behaving rats. Because deep-layer LEC cells received inputs from hippocampal regions, which have strong selectivity for self-location, we hypothesized that deep-layer LEC neurons would be more informative about allocentric position than superficial-layer LEC neurons. We found that deep-layer LEC cells showed only slightly more allocentric spatial information and higher spatial consistency than superficial-layer LEC cells. Egocentric coding properties were comparable between these two subregions. In addition, LEC neurons demonstrated preferential firing at lower speeds, as well as at the boundary or corners of the environment. These results suggest that allocentric spatial outputs from the hippocampus are transformed in deep-layer LEC into the egocentric coding dimensions of LEC, rather than maintaining the allocentric spatial tuning of the CA1 place fields.

Loss of functional heterogeneity along the CA3 transverse axis in aging.

Age-related deficits in pattern separation have been postulated to bias the output of hippocampal memory processing toward pattern completion, which can cause deficits in accurate memory retrieval. Although the CA3 region of the hippocampus is often conceptualized as a homogeneous network involved in pattern completion, growing evidence demonstrates a functional gradient in CA3 along the transverse axis, as pattern-separated outputs (dominant in the more proximal CA3) transition to pattern-completed outputs (dominant in the more distal CA3). We examined the neural representations along the CA3 transverse axis in young (Y), aged memory-unimpaired (AU), and aged memory-impaired (AI) rats when different changes were made to the environment. Functional heterogeneity in CA3 was observed in Y and AU rats when the environmental similarity was high (altered cues or altered environment shapes in the same room), with more orthogonalized representations in proximal CA3 than in distal CA3. In contrast, AI rats showed reduced orthogonalization in proximal CA3 but showed normal (i.e., generalized) representations in distal CA3, with little evidence of a functional gradient. Under experimental conditions when the environmental similarity was low (different rooms), representations in proximal and distal CA3 remapped in all rats, showing that CA3 of AI rats is able to encode distinctive representations for inputs with greater dissimilarity. These experiments support the hypotheses that the age-related bias toward hippocampal pattern completion is due to the loss in AI rats of the normal transition from pattern separation to pattern completion along the CA3 transverse axis.

Flexible encoding of objects and space in single cells of the dentate gyrus.

The hippocampus is involved in the formation of memories that require associations among stimuli to construct representations of space and the items and events within that space. Neurons in the dentate gyrus (DG), an initial input region of the hippocampus, have robust spatial tuning, but it is unclear how nonspatial information may be integrated with spatial activity in this region. We recorded from the DG of 21 adult mice as they foraged for food in an environment that contained discrete objects. We found DG cells with multiple firing fields at a fixed distance and direction from objects (landmark vector cells) and cells that exhibited localized changes in spatial firing when objects in the environment were manipulated. By classifying recorded DG cells into putative dentate granule cells and mossy cells, we examined how the addition or displacement of objects affected the spatial firing of these DG cell types. Object-related activity was detected in a significant proportion of mossy cells. Although few granule cells with responses to object manipulations were recorded, likely because of the sparse nature of granule cell firing, there was generally no significant difference in the proportion of granule cells and mossy cells with object responses. When mice explored a second environment with the same objects, DG spatial maps completely reorganized, and a different subset of cells responded to object manipulations. Together, these data reveal the capacity of DG cells to detect small changes in the environment while preserving a stable spatial representation of the overall context.

Wide-angle, monocular head tracking using passive markers.

BACKGROUND: Camera images can encode large amounts of visual information of an animal and its environment, enabling high fidelity 3D reconstruction of the animal and its environment using computer vision methods. Most systems, both markerless (e.g. deep learning based) and marker-based, require multiple cameras to track features across multiple points of view to enable such 3D reconstruction. However, such systems can be expensive and are challenging to set up in small animal research apparatuses. NEW METHODS: We present an open-source, marker-based system for tracking the head of a rodent for behavioral research that requires only a single camera with a potentially wide field of view. The system features a lightweight visual target and computer vision algorithms that together enable high-accuracy tracking of the six-degree-of-freedom position and orientation of the animal's head. The system, which only requires a single camera positioned above the behavioral arena, robustly reconstructs the pose over a wide range of head angles (360° in yaw, and approximately ± 120° in roll and pitch). RESULTS: Experiments with live animals demonstrate that the system can reliably identify rat head position and orientation. Evaluations using a commercial optical tracker device show that the system achieves accuracy that rivals commercial multi-camera systems. COMPARISON WITH EXISTING METHODS: Our solution significantly improves upon existing monocular marker-based tracking methods, both in accuracy and in allowable range of motion. CONCLUSIONS: The proposed system enables the study of complex behaviors by providing robust, fine-scale measurements of rodent head motions in a wide range of orientations.

The Dome: A virtual reality apparatus for freely locomoting rodents.

The cognitive map in the hippocampal formation of rodents and other mammals integrates multiple classes of sensory and motor information into a coherent representation of space. Here, we describe the Dome, a virtual reality apparatus for freely locomoting rats, designed to examine the relative contributions of various spatial inputs to an animal's spatial representation. The Dome was designed to preserve the range of spatial inputs typically available to an animal in free, untethered locomotion while providing the ability to perturb specific sensory cues. We present the design rationale and corresponding specifications of the Dome, along with a variety of engineering and biological analyses to validate the efficacy of the Dome as an experimental tool to examine the interaction between visual information and path integration in place cells in rodents.

Heterogeneity of Age-Related Neural Hyperactivity along the CA3 Transverse Axis.

Age-related memory deficits are correlated with neural hyperactivity in the CA3 region of the hippocampus. Abnormal CA3 hyperactivity in aged rats has been proposed to contribute to an imbalance between pattern separation and pattern completion, resulting in overly rigid representations. Recent evidence of functional heterogeneity along the CA3 transverse axis suggests that proximal CA3 supports pattern separation while distal CA3 supports pattern completion. It is not known whether age-related CA3 hyperactivity is uniformly represented along the CA3 transverse axis. We examined the firing rates of CA3 neurons from young and aged, male, Long-Evans rats along the CA3 transverse axis. Consistent with prior studies, young CA3 cells showed an increasing gradient in mean firing rate from proximal to distal CA3. However, aged CA3 cells showed an opposite, decreasing trend, in that CA3 cells in aged rats were hyperactive in proximal CA3, but possibly hypoactive in distal CA3, compared with young (Y) rats. We suggest that, in combination with altered inputs from the entorhinal cortex and dentate gyrus (DG), the proximal CA3 region of aged rats may switch from its normal function that reflects the pattern separation output of the DG and instead performs a computation that reflects an abnormal bias toward pattern completion. In parallel, distal CA3 of aged rats may create weaker attractor basins that promote abnormal, bistable representations under certain conditions.SIGNIFICANCE STATEMENT Prior work suggested that age-related CA3 hyperactivity enhances pattern completion, resulting in rigid representations. Implicit in prior studies is the notion that hyperactivity is present throughout a functionally homogeneous CA3 network. However, more recent work has demonstrated functional heterogeneity along the CA3 transverse axis, in that proximal CA3 is involved in pattern separation and distal CA3 is involved in pattern completion. Here, we show that age-related hyperactivity is present only in proximal CA3, with potential hypoactivity in distal CA3. This result provides new insight in the role of CA3 in age-related memory impairments, suggesting that the rigid representations in aging result primarily from dysfunction of computational circuits involving the dentate gyrus (DG) and proximal CA3.

Decreased investigatory head scanning during exploration in learning-impaired, aged rats.

"Head scanning" is an investigatory behavior that has been linked to spatial exploration and the one-trial formation or strengthening of place cells in the hippocampus. Previous studies have demonstrated that a subset of aged rats with normal spatial learning performance show head scanning rates during a novel, local-global cue-mismatch manipulation that are similar to those of young rats. However, these aged rats demonstrated different patterns of expression of neural activity markers in brain regions associated with spatial learning, perhaps suggesting neural mechanisms that compensate for age-related brain changes. These prior studies did not investigate the head scanning properties of aged rats that had spatial learning impairments. The present study analyzed head scanning behavior in young, aged-unimpaired, and aged-impaired Long Evans rats. Aged-impaired rats performed the head scan behavior at a lower rate than the young rats. These results suggest that decreased attention to spatial landmarks may be a contributing factor to the spatial learning deficits shown by the aged-impaired rats.

Parallel processing streams in the hippocampus.

The hippocampus performs two complementary processes, pattern separation and pattern completion, to minimize interference and maximize the storage capacity of memories. Classic computational models have suggested that the dentate gyrus (DG) supports pattern separation and the putative attractor circuitry in CA3 supports pattern completion. However, recent evidence of functional heterogeneity along the CA3 transverse axis of the hippocampus suggests that the DG and proximal CA3 work as a functional unit for pattern separation, while distal CA3 forms an autoassociative network for pattern completion. We propose that the outputs of these functional circuits, combined with direct projections from entorhinal cortex to CA1, form interconnected, parallel processing circuits to support accurate memory storage and retrieval.

Hippocampal Place Cells Encode Local Surface-Texture Boundaries.

The cognitive map is often assumed to be a Euclidean map that isometrically represents the real world (i.e., the Euclidean distance between any two locations in the physical world should be preserved on the cognitive map). However, accumulating evidence suggests that environmental boundaries can distort the mental representations of physical space. For example, the distance between two locations can be remembered as longer than the true physical distance if the locations are separated by a boundary. While this overestimation is observed under different experimental conditions, even when the boundary is formed by flat surface cues, its physiological basis is not well understood. We examined the neural representation of flat surface cue boundaries, and of the space segregated by these boundaries, by recording place cell activity from CA1 and CA3 while rats foraged on a circular track or square platforms with inhomogeneous surface textures. About 40% of the place field edges concentrated near the boundaries on the circular track (significantly above the chance level 33%). Similarly, place field edges were more prevalent near boundaries on the platforms than expected by chance. In both one- and two-dimensional environments, the population vectors of place cell activity changed more abruptly with distance between locations that crossed cue boundaries than between locations within a bounded region. These results show that the locations of surface boundaries were evident as enhanced decorrelations of the neural representations of locations to either side of the boundaries. This enhancement might underlie the cognitive phenomenon of overestimation of distances across boundaries.

Egocentric and allocentric representations of space in the rodent brain.

Spatial signals are prevalent within the hippocampus and its neighboring regions. It is generally accepted that these signals are defined with respect to the external world (i.e., a world-centered, or allocentric, frame of reference). Recently, evidence of egocentric processing (i.e., self-centered, defined relative to the subject) in the extended hippocampal system has accumulated. These results support the idea that egocentric sensory information, derived from primary sensory cortical areas, may be transformed to allocentric representations that interact with the allocentric hippocampal system. We propose a framework to explain the implications of the egocentric-allocentric transformations to the functions of the medial temporal lobe memory system.

Dentate Gyrus Mossy Cells Share a Role in Pattern Separation with Dentate Granule Cells and Proximal CA3 Pyramidal Cells.

The complementary processes of pattern completion and pattern separation are thought to be essential for successful memory storage and recall. The dentate gyrus (DG) and proximal CA3 (pCA3) regions have been implicated in pattern separation, in part through extracellular recording studies of these areas. However, the DG contains two types of excitatory cells: granule cells of the granule layer and mossy cells of the hilus. Little is known about the firing properties of mossy cells in freely moving animals, and it is unclear how their activity may contribute to the mnemonic functions of the hippocampus. Furthermore, tetrodes in the dentate granule layer and pCA3 pyramidal layer can also record mossy cells, thus introducing ambiguity into the identification of cell types recorded. Using a random forests classifier, we classified cells recorded in DG (Neunuebel and Knierim, 2014) and pCA3 (Lee et al., 2015) of 16 male rats and separately examined the responses of granule cells, mossy cells, and pCA3 pyramidal cells in a local/global cue mismatch task. All three cell types displayed low correlations between the population representations of the rat's position in the standard and cue-mismatch sessions. These results suggest that all three excitatory cell types within the DG/pCA3 circuit may act as a single functional unit to support pattern separation.SIGNIFICANCE STATEMENT Mossy cells in the dentate gyrus (DG) are an integral component of the DG/pCA3 circuit. While the role of granule cells in the circuitry and computations of the hippocampus has been a focus of study for decades, the contributions of mossy cells have been largely overlooked. Recent studies have revealed the spatial firing properties of mossy cells in awake behaving animals, but how the activity of these highly active cells contributes to the mnemonic functions of the DG is uncertain. We separately analyzed mossy cells, granule cells, and pCA3 cells and found that all three cell types respond similarly to a local/global cue mismatch, suggesting that they form a single functional unit supporting pattern separation.

Origin and role of path integration in the cognitive representations of the hippocampus: computational insights into open questions.

Path integration is a straightforward concept with varied connotations that are important to different disciplines concerned with navigation, such as ethology, cognitive science, robotics and neuroscience. In studying the hippocampal formation, it is fruitful to think of path integration as a computation that transforms a sense of motion into a sense of location, continuously integrated with landmark perception. Here, we review experimental evidence that path integration is intimately involved in fundamental properties of place cells and other spatial cells that are thought to support a cognitive abstraction of space in this brain system. We discuss hypotheses about the anatomical and computational origin of path integration in the well-characterized circuits of the rodent limbic system. We highlight how computational frameworks for map-building in robotics and cognitive science alike suggest an essential role for path integration in the creation of a new map in unfamiliar territory, and how this very role can help us make sense of differences in neurophysiological data from novel versus familiar and small versus large environments. Similar computational principles could be at work when the hippocampus builds certain non-spatial representations, such as time intervals or trajectories defined in a sensory stimulus space.

Aged rats with preserved memory dynamically recruit hippocampal inhibition in a local/global cue mismatch environment.

Similar to elderly humans, aged outbred Long-Evans rats exhibit individual differences in memory abilities, including a subset of aged rats that maintain memory function on par with young adults. Such individuals provide a basis for investigating mechanisms of resilience to age-related decline. The present study examined hippocampal gene expression in young adults and aged rats with preserved memory function under behavioral task conditions well established for assessing information processing central to the formation of episodic memory. Although behavioral measures and hippocampal gene induction associated with neural activity and synaptic plasticity were similar across age groups, a marker for inhibitory interneuron function in the hippocampal formation was distinctively increased only in aged rats but not in young adults. Because heightened hippocampal neural activity is associated with age-related memory impairment across species, including rats, monkeys, and humans, this finding may represent an adaptive homeostatic adjustment necessary to maintain neural plasticity and memory function in aging.

Recalibration of path integration in hippocampal place cells.

Hippocampal place cells are spatially tuned neurons that serve as elements of a 'cognitive map' in the mammalian brain1. To detect the animal's location, place cells are thought to rely upon two interacting mechanisms: sensing the position of the animal relative to familiar landmarks2,3 and measuring the distance and direction that the animal has travelled from previously occupied locations4-7. The latter mechanism-known as path integration-requires a finely tuned gain factor that relates the animal's self-movement to the updating of position on the internal cognitive map, as well as external landmarks to correct the positional error that accumulates8,9. Models of hippocampal place cells and entorhinal grid cells based on path integration treat the path-integration gain as a constant9-14, but behavioural evidence in humans suggests that the gain is modifiable15. Here we show, using physiological evidence from rat hippocampal place cells, that the path-integration gain is a highly plastic variable that can be altered by persistent conflict between self-motion cues and feedback from external landmarks. In an augmented-reality system, visual landmarks were moved in proportion to the movement of a rat on a circular track, creating continuous conflict with path integration. Sustained exposure to this cue conflict resulted in predictable and prolonged recalibration of the path-integration gain, as estimated from the place cells after the landmarks were turned off. We propose that this rapid plasticity keeps the positional update in register with the movement of the rat in the external world over behavioural timescales. These results also demonstrate that visual landmarks not only provide a signal to correct cumulative error in the path-integration system4,8,16-19, but also rapidly fine-tune the integration computation itself.