Place field repetition and purely local remapping in a multicompartment environment.

Hippocampal place cells support spatial memory using sensory information from the environment and self-motion information to localize their firing fields. Currently, there is disagreement about whether CA1 place cells can use pure self-motion information to disambiguate different compartments in environments containing multiple visually identical compartments. Some studies report that place cells can disambiguate different compartments, while others report that they do not. Furthermore, while numerous studies have examined remapping, there has been little examination of remapping in different subregions of a single environment. Is remapping purely local or do place fields in neighboring, unaffected, regions detect the change? We recorded place cells as rats foraged across a 4-compartment environment and report 3 new findings. First, we find that, unlike studies in which rats foraged in 2 compartments, place fields showed a high degree of spatial repetition with a slight degree of rate-based discrimination. Second, this repetition does not diminish with extended experience. Third, remapping was found to be purely local for both geometric change and contextual change. Our results reveal the limited capacity of the path integrator to drive pattern separation in hippocampal representations, and suggest that doorways may play a privileged role in segmenting the neural representation of space.

Navigating in a three-dimensional world.

The study of spatial cognition has provided considerable insight into how animals (including humans) navigate on the horizontal plane. However, the real world is three-dimensional, having a complex topography including both horizontal and vertical features, which presents additional challenges for representation and navigation. The present article reviews the emerging behavioral and neurobiological literature on spatial cognition in non-horizontal environments. We suggest that three-dimensional spaces are represented in a quasi-planar fashion, with space in the plane of locomotion being computed separately and represented differently from space in the orthogonal axis - a representational structure we have termed "bicoded." We argue that the mammalian spatial representation in surface-travelling animals comprises a mosaic of these locally planar fragments, rather than a fully integrated volumetric map. More generally, this may be true even for species that can move freely in all three dimensions, such as birds and fish. We outline the evidence supporting this view, together with the adaptive advantages of such a scheme.

Geometric cues influence head direction cells only weakly in nondisoriented rats.

The influential hypothesis that environmental geometry is critical for spatial orientation has been extensively tested behaviorally, and yet findings have been conflicting. Head direction (HD) cells, the neural correlate of the sense of direction, offer a window into the processes underlying directional orientation and may help clarify the issue. In the present study, HD cells were recorded as rats foraged in enclosures of varying geometry, with or without simultaneous manipulation of landmarks and self-motion cues (path integration). All geometric enclosures had single-order rotational symmetry and thus completely polarized the environment. They also had unique features, such as corners, which could, in principle, act as landmarks. Despite these strongly polarizing geometric cues, HD cells in nondisoriented rats never rotated with these shapes. In contrast, when a cue card (white or gray) was added to one wall, HD cells readily rotated with the enclosure. When path integration was disrupted by disorienting the rat, HD cells rotated with the enclosure even without the landmark. Collectively, these findings indicate that geometry exerts little or no influence on heading computations in nondisoriented rats, but it can do so in disoriented rats. We suggest that geometric processing is only a weak influence, providing a backup system for heading calculations and recruited only under conditions of disorientation.

Experience-dependent rescaling of entorhinal grids.

The firing pattern of entorhinal 'grid cells' is thought to provide an intrinsic metric for space. We report a strong experience-dependent environmental influence: the spatial scales of the grids (which are aligned and have fixed relative sizes within each animal) vary parametrically with changes to a familiar environment's size and shape. Thus grid scale reflects an interaction between intrinsic, path-integrative calculation of location and learned associations to the external environment.

How heterogeneous place cell responding arises from homogeneous grids--a contextual gating hypothesis.

How entorhinal grids generate hippocampal place fields remains unknown. The simplest hypothesis-that grids of different scales are added together-cannot explain a number of place field phenomena, such as (1) Summed grids form a repeating, dispersed activation pattern whereas place fields are focal and nonrepeating; (2) Grid cells are active in all environments but place cells only in some, and (3) Partial environmental changes cause either heterogeneous ("partial") remapping in place cells whereas they result in all-or-nothing "realignment" remapping in grid cells. We propose that this dissociation between grid cell and place cell behavior arises in the entorhinal-dentate projection. By our view, the grid-cell/place-cell projection is modulated by context, both organizationally and activationally. Organizationally, we propose that when the animal first enters a new environment, the relatively homogeneous input from the grid cells becomes spatially clustered by Hebbian processes in the dendritic tree so that inputs active in the same context and having overlapping fields come to terminate on the same sub-branches of the tree. Activationally, when the animal re-enters the now-familiar environment, active contextual inputs select (by virtue of their clustered terminations) which parts of the dendritic tree, and therefore which grid cells, drive the granule cell. Assuming this pattern of projections, our model successfully produces focal hippocampal place fields that remap appropriately to contextual changes.

Entorhinal Neurons Exhibit Cue Locking in Rodent VR.

The regular firing pattern exhibited by medial entorhinal (mEC) grid cells of locomoting rodents is hypothesized to provide spatial metric information relevant for navigation. The development of virtual reality (VR) for head-fixed mice confers a number of experimental advantages and has become increasingly popular as a method for investigating spatially-selective cells. Recent experiments using 1D VR linear tracks have shown that some mEC cells have multiple fields in virtual space, analogous to grid cells on real linear tracks. We recorded from the mEC as mice traversed virtual tracks featuring regularly spaced repetitive cues and identified a population of cells with multiple firing fields, resembling the regular firing of grid cells. However, further analyses indicated that many of these were not, in fact, grid cells because: (1) when recorded in the open field they did not display discrete firing fields with six-fold symmetry; and (2) in different VR environments their firing fields were found to match the spatial frequency of repetitive environmental cues. In contrast, cells identified as grid cells based on their open field firing patterns did not exhibit cue locking. In light of these results we highlight the importance of controlling the periodicity of the visual cues in VR and the necessity of identifying grid cells from real open field environments in order to correctly characterize spatially modulated neurons in VR experiments.

The boundary vector cell model of place cell firing and spatial memory.

We review evidence for the boundary vector cell model of the environmental determinants of the firing of hippocampal place cells. Preliminary experimental results are presented concerning the effects of addition or removal of environmental boundaries on place cell firing and evidence that boundary vector cells may exist in the subiculum. We review and update computational simulations predicting the location of human search within a virtual environment of variable geometry, assuming that boundary vector cells provide one of the input representations of location used in mammalian spatial memory. Finally, we extend the model to include experience-dependent modification of connection strengths through a BCM-like learning rule - the size and sign of strength change is influenced by historic activity of the postsynaptic cell. Simulations are compared to experimental data on the firing of place cells under geometrical manipulations to their environment. The relationship between neurophysiological results in rats and spatial behaviour in humans is discussed.

Disrupting the Grid Cells' Need for Speed.

Hinman et al. demonstrate the presence of two speed signals in the rodent medial entorhinal cortex that are differentially affected by muscimol inactivation of medial septum. The results reveal important constraints on several computational models of grid cell firing.

How cumulative error in grid cell firing is literally bounded by the environment.

In this issue of Neuron, Hardcastle et al. (2015) show that the spatial firing patterns of grid cells accumulate error, drifting coherently, until reset by encounters with environmental boundaries. These results reveal important aspects of the neural dynamics of self-localization from self-motion and environmental information.

Neural encoding of large-scale three-dimensional space-properties and constraints.

How the brain represents represent large-scale, navigable space has been the topic of intensive investigation for several decades, resulting in the discovery that neurons in a complex network of cortical and subcortical brain regions co-operatively encode distance, direction, place, movement etc. using a variety of different sensory inputs. However, such studies have mainly been conducted in simple laboratory settings in which animals explore small, two-dimensional (i.e., flat) arenas. The real world, by contrast, is complex and three dimensional with hills, valleys, tunnels, branches, and-for species that can swim or fly-large volumetric spaces. Adding an additional dimension to space adds coding challenges, a primary reason for which is that several basic geometric properties are different in three dimensions. This article will explore the consequences of these challenges for the establishment of a functional three-dimensional metric map of space, one of which is that the brains of some species might have evolved to reduce the dimensionality of the representational space and thus sidestep some of these problems.

Grid cells on steeply sloping terrain: evidence for planar rather than volumetric encoding.

Neural encoding of navigable space involves a network of structures centered on the hippocampus, whose neurons -place cells - encode current location. Input to the place cells includes afferents from the entorhinal cortex, which contains grid cells. These are neurons expressing spatially localized activity patches, or firing fields, that are evenly spaced across the floor in a hexagonal close-packed array called a grid. It is thought that grids function to enable the calculation of distances. The question arises as to whether this odometry process operates in three dimensions, and so we queried whether grids permeate three-dimensional (3D) space - that is, form a lattice - or whether they simply follow the environment surface. If grids form a 3D lattice then this lattice would ordinarily be aligned horizontally (to explain the usual hexagonal pattern observed). A tilted floor would transect several layers of this putative lattice, resulting in interruption of the hexagonal pattern. We model this prediction with simulated grid lattices, and show that the firing of a grid cell on a 40°-tilted surface should cover proportionally less of the surface, with smaller field size, fewer fields, and reduced hexagonal symmetry. However, recording of real grid cells as animals foraged on a 40°-tilted surface found that firing of grid cells was almost indistinguishable, in pattern or rate, from that on the horizontal surface, with if anything increased coverage and field number, and preserved field size. It thus appears unlikely that the sloping surface transected a lattice. However, grid cells on the slope displayed slightly degraded firing patterns, with reduced coherence and slightly reduced symmetry. These findings collectively suggest that the grid cell component of the metric representation of space is not fixed in absolute 3D space but is influenced both by the surface the animal is on and by the relationship of this surface to the horizontal, supporting the hypothesis that the neural map of space is "multi-planar" rather than fully volumetric.

A proposed architecture for the neural representation of spatial context.

The role of context in guiding animal behavior has attracted increasing attention in recent years, but little is known about what constitutes a context, nor how and where in the brain it is represented. Contextual stimuli can take many forms, but of particular importance are those that collectively define a particular place or situation. The representation of place has been linked to the hippocampus, because its principal neurons ('place cells') are spatially responsive; behavioral experiments also implicate this structure in the processing of contextual stimuli. Together, these findings suggest a hippocampal role in representing 'spatial context'. The present article outlines a proposed architecture for the encoding of spatial context in which spatial inputs to place cells are modulated (or 'gated') by non-spatial stimuli. We discuss recent experimental evidence that spatial context is population-coded, a property which could allow both discrimination between overlapping contexts and generalization across them, and thus provide a foundation for animals' capacity for flexible context-linked place learning.

Plasticity of the hippocampal place cell representation.

The role of the hippocampus in the representation of 'place' has been attributed to the place cells, whose spatially localised firing suggests their participation in forming a cognitive map of the environment. That this map is necessary for spatial memory formation is indicated by the propensity of almost all navigational tasks to be disrupted by hippocampal damage. The hippocampus has also long been implicated in the formation of episodic memories, and the unusually plastic nature of hippocampal synapses testifies to its probable mnemonic role. Arguably, the place cell representation should, if it is to support spatial learning, be modifiable according to known principles of synaptic reorganization. The present article reviews evidence that the place cell representation is indeed plastic, and that its plasticity depends on the same neurobiological mechanisms known to underlie experimentally induced synaptic plasticity. Inferences are drawn regarding the architecture of the spatial representation and the principles by which it is modified. Spatial learning is promising to be the first kind of memory which is completely understood at all levels, from molecular through circuitry to behaviour and beyond.

Allocentric directional processing in the rodent and human retrosplenial cortex.

Head direction (HD) cells in the rodent brain have been investigated for a number of years, providing us with a detailed understanding of how the rodent brain codes for allocentric direction. Allocentric direction refers to the orientation of the external environment, independent of one's current (egocentric) orientation. The presence of neural activity related to allocentric directional coding in humans has also been noted but only recently directly tested. Given the current status of both fields, it seems beneficial to draw parallels between this rodent and human research. We therefore discuss how findings from the human retrosplenial cortex (RSC), including its "translational function" (converting egocentric to allocentric information) and ability to code for permanent objects, compare to findings from the rodent RSC. We conclude by suggesting critical future experiments that derive from a cross-species approach to understanding the function of the human RSC.

Are new place representations independent of theta and path integration?

Brandon et al. (2014) show that the formation of place cell representations in new environments is preserved under septal inactivation, and is thus likely independent of the hippocampal theta rhythm and, by implication, the firing of entorhinal grid cells and the process of path integration.

Specific evidence of low-dimensional continuous attractor dynamics in grid cells.

We examined simultaneously recorded spikes from multiple rat grid cells, to explain mechanisms underlying their activity. Among grid cells with similar spatial periods, the population activity was confined to lie close to a two-dimensional (2D) manifold: grid cells differed only along two dimensions of their responses and otherwise were nearly identical. Relationships between cell pairs were conserved despite extensive deformations of single-neuron responses. Results from novel environments suggest such structure is not inherited from hippocampal or external sensory inputs. Across conditions, cell-cell relationships are better conserved than responses of single cells. Finally, the system is continually subject to perturbations that, were the 2D manifold not attractive, would drive the system to inhabit a different region of state space than observed. These findings have strong implications for theories of grid-cell activity and substantiate the general hypothesis that the brain computes using low-dimensional continuous attractors.

Anisotropic encoding of three-dimensional space by place cells and grid cells.

The subjective sense of space may result in part from the combined activity of place cells in the hippocampus and grid cells in posterior cortical regions such as the entorhinal cortex and pre- and parasubiculum. In horizontal planar environments, place cells provide focal positional information, whereas grid cells supply odometric (distance measuring) information. How these cells operate in three dimensions is unknown, even though the real world is three-dimensional. We investigated this issue in rats exploring two different kinds of apparatus: a climbing wall (the 'pegboard') and a helix. Place and grid cell firing fields had normal horizontal characteristics but were elongated vertically, with grid fields forming stripes. It seems that grid cell odometry (and by implication path integration) is impaired or absent in the vertical domain, at least when the rat itself remains horizontal. These findings suggest that the mammalian encoding of three-dimensional space is anisotropic.

Horizontal biases in rats' use of three-dimensional space.

Rodent spatial cognition studies allow links to be made between neural and behavioural phenomena, and much is now known about the encoding and use of horizontal space. However, the real world is three dimensional, providing cognitive challenges that have yet to be explored. Motivated by neural findings suggesting weaker encoding of vertical than horizontal space, we examined whether rats show a similar behavioural anisotropy when distributing their time freely between vertical and horizontal movements. We found that in two- or three-dimensional environments with a vertical dimension, rats showed a prioritization of horizontal over vertical movements in both foraging and detour tasks. In the foraging tasks, the animals executed more horizontal than vertical movements and adopted a "layer strategy" in which food was collected from one horizontal level before moving to the next. In the detour tasks, rats preferred the routes that allowed them to execute the horizontal leg first. We suggest three possible reasons for this behavioural bias. First, as suggested by Grobety and Schenk, it allows minimisation of energy expenditure, inasmuch as costly vertical movements are minimised. Second, it may be a manifestation of the temporal discounting of effort, in which animals value delayed effort as less costly than immediate effort. Finally, it may be that at the neural level rats encode the vertical dimension less precisely, and thus prefer to bias their movements in the more accurately encoded horizontal dimension. We suggest that all three factors are related, and all play a part.

Context-specific acquisition of location discrimination by hippocampal place cells.

The spatially localized firing of rodent hippocampal place cells is strongly determined by the local geometry of the environment. Over time, however, the cells can acquire additional inputs, including inputs from more distal cues. This is manifest as a change in firing pattern ('remapping') when the new inputs are manipulated. Place cells also reorganize their firing in response to non-geometric changes in 'context', such as a change in the colour or odour of the environment. The present study investigated whether the new inputs acquired by place cells in one context were still available to the cells when they expressed their altered firing patterns in a new context. We found that the acquired information did not transfer to the new context, suggesting that the context inputs and the acquired inputs must interact somewhere upstream of the place cells themselves. We present a model of one possible such interaction, and of how such an interaction could be modified by experience in a Hebbian manner, thus explaining the context specificity of the new learning.