PERSPECTIVE Navigating from hippocampus to parietal cortex Jonathan R. Whitlock*, Robert J. Sutherland*†, Menno P. Witter*, May-Britt Moser*, and Edvard I. Moser*‡ *Kavli Institute for Systems Neuroscience, Norwegian University of Science and Technology, 7489 Trondheim, Norway; and †Canadian Centre for Behavioral Neuroscience, University of Lethbridge, AB, Canada T1K 3M4 Edited by Jeremy Nathans, Johns Hopkins University School of Medicine, Baltimore, MD, and approved July 17, 2008 (received for review May 29, 2008). The navigational system of the mammalian cortex comprises a number of interacting brain regions. Grid cells in the medial entorhi- nal cortex and place cells in the hippocampus are thought to participate in the formation of a dynamic representation of the ani- mal’s current location, and these cells are presumably critical for storing the representation in memory. To traverse the environment, animals must be able to translate coordinate information from spatial maps in the entorhinal cortex and hippocampus into body- centered representations that can be used to direct locomotion. How this is done remains an enigma. We propose that the posterior parietal cortex is critical for this transformation. entorhinal cortex ͉ grid cell ͉ place cell ͉ Path integration ͉ spatial memory nimals have a number of strat- the cells were mostly silent. Subsequent The multiple firing fields of cells in this egies for optimizing movement work showed that neighboring hip- region of entorhinal cortex formed a toward goal locations. The rep- pocampal cells had nontopographically strikingly regular grid-like pattern com- ertoire of navigation strategies organized place fields such that the en- posed of equilateral triangles tessellating A the entire environment covered by the ranges from simple approach and avoid- tire surface of the environment could be ance, such as following odor trails and represented by a group of neurons in a animal (16; Fig. 1B). Based on the firing chemical gradients or moving toward local circuit (5). Cells with firing fields patterns of small numbers of grid cells, prominent visual beacons, to the use of in two environments fired at unrelated the position of a moving animal could complex representations, such as geo- locations in those environments (6). Fol- be reconstructed on a second-by-second metric maps based on perceived and lowing these observations, it was pro- basis (15), suggesting that self-position remembered spatial relationships be- posed that the concerted activity of was represented already in inputs to the tween distributed landmarks, and path place cells provides the physiological hippocampus. integration based on the continuous substrate for a ‘‘cognitive map,’’ where Grid cells in the entorhinal cortex are flow of motion-generated speed and di- cell populations throughout the hip- organized in a map-like manner accord- rection signals (1–3). In most species, pocampus maintain a coherent, up-to- ing to the basic parameters of the grid the mechanisms are complementary and date representation of allocentric space (15, 16). All grid fields display the same used in combination. Landmarks and and the animal’s location in that space iterative triangular geometry, but they geometrical relationships enable the ani- at any point in time (1). may differ in spacing (distance between mal to store maps and routes for indi- fields), orientation (the angle to which vidual environments, but on their own Grid Cells and the Spatial Map in the the maps are tilted), and phase (xy off- these cues provide limited information Medial Entorhinal Cortex set of the fields relative to an external about the direction and distance that Despite major advances in understand- reference). Grids of neighboring cells the animal has moved from a given ref- ing hippocampal spatial computation, have similar orientation and spacing, but erence position. Conversely, path inte- the neural mechanisms for computing a randomly shifted vertices, such that the gration can be used to build a metric dynamic representation of the animal’s fields of adjacent cells do not overlap representation of the animal’s position own location have remained elusive. The more than expected by chance. Collec- in the environment, but without regular large number of nonoverlapping spatial tively, grid cells with different spatial calibration against perceived or recalled representations stored in the hippocam- phase, orientation, and spacing provide unambiguous information about the ani- landmarks and geometric boundaries, pus (7, 8) pointed to an extrahippocam- mal’s current position (17). Grid cells errors will accumulate and the represen- pal location for general navigational are the predominant cell type of the me- tation will drift. The composite nature computations (5, 9, 10). This possibility dial entorhinal cortex (MEC), but in of navigation suggests that multiple was supported by the persistence of layers III–VI of this area they intermin- brain regions and mechanisms may place fields in CA1 in rats where intra- gle with head direction-responsive cells be involved. hippocampal connections were disrupted and cells with conjunctive grid and head but direct projections from the entorhi- direction properties (18). Place Cells in the Hippocampus nal cortex were spared (11, 12). For a Our understanding of the neural repre- long time, however, the possibility of an Path Integration and Entorhinal sentation of navigational space began extrahippocampal origin was neglected Grid Cells with the discovery of ‘‘place cells’’ by because neurons in the entorhinal cor- The expression of a strongly periodic O’Keefe and Dostrovsky in 1971 (4). By tex, from which the hippocampus re- spatial firing pattern in the presence of using microelectrodes chronically im- ceives most of its cortical inputs, showed planted in freely behaving rats, these only weak spatial modulation (13, 14). authors found that the firing of putative A closer look at the firing properties of Author contributions: J.R.W., R.J.S., M.P.W., M.-B.M., and pyramidal cells in the dorsal hippocam- entorhinal neurons finally revealed that E.I.M. wrote the paper. pus exhibited a striking behavioral mod- neurons in more dorsomedial parts of The authors declare no conflict of interest. ulation, where cells fired only when the structure, far more dorsal than This article is a PNAS Direct Submission. animals occupied particular locations, or where activity was recorded in the ear- ‡To whom correspondence should be addressed. E-mail: ‘‘place fields,’’ in the recording environ- lier studies, exhibit fine spatial tuning [email protected]. ment (Fig. 1A). Outside of these fields, similar to that observed in CA1 (15). © 2008 by The National Academy of Sciences of the USA www.pnas.org͞cgi͞doi͞10.1073͞pnas.0804216105 PNAS ͉ September 30, 2008 ͉ vol. 105 ͉ no. 39 ͉ 14755–14762 Downloaded by guest on September 27, 2021 40Hz B 17Hz C 8Hz A max 0 1.0m 1.5m 0.6m Fig. 1. Spatial firing properties of neurons in hippocampus (A; place cell), medial entorhinal cortex (MEC) (B; grid cell), and posterior parietal cortex (PPC) (C). In A and B, the rat runs freely in an open-field environment; in C, the rat traverses a complex maze with multiple segments and turns (Left, path of the animal; Center, rate map with lights on; Right, rate map in darkness). Firing rates are color-coded, with red showing maximal rate and blue minimum (scale bar to the right). Note that the parietal neuron fires at specific epochs along the trajectory and that firing is independent of visual inputs. [C is adapted with permission from ref. 66 (Copyright 2006, Neuron).] constantly changing running speed and MEC and hippocampus should display In this task, the way back could only be running direction suggests that the grid navigational impairments. Several de- found by integrating self-movement on must rely on path integration computa- cades of research have indeed shown the outward journey. Whereas control tions (16, 19, 20), where changes in ve- that lesions of the hippocampus disrupt rats returned to the location that the locity and direction are integrated over the ability to navigate efficiently to the refuge had held at the beginning of the time to allow a constant representation goal location in various kinds of mazes trial, rats with lesions in the hippocam- of space (3, 19, 21). The relative invari- (see ref. 26) for a review). In the major- pus had long return paths with no ap- ance of the grid representation is consis- ity of these studies, the impairments in parent preference for the original start tent with this idea (22). Unlike place navigation can unfortunately not be dis- location. A similar impairment was ob- cells in the hippocampus, grid cells are tinguished from effects on memory. served after lesions of the entorhinal activated in a stereotypic manner across However, in one informative series of cortex (29; Fig. 2B). The conclusions environments, irrespective of the partic- studies, path integration was assessed from these studies are somewhat contra- ular landmarks of the environment. Two cells in different parts of MEC whose more directly by analyzing the trajecto- dicted by a study in which rats with par- grid fields are shifted and rotated 30° ries of animals that returned to a start- tial hippocampal lesions did return to relative to each other in one environ- ing refuge after searching for food on a the start position (30). The authors sug- ment will show the same relative shift slowly rotating arena (27, 28; Fig. 2A). gested that the hippocampal part of the and rotation in a different environment. The proposed dependence on path inte- gration is further supported by the fact that grid-like spacing is expressed imme- diately as an animal starts to explore an environment and that, like place fields, the grids persist after removal of exter- nal sensory cues (16). The continued firing of place cells and grid cells in the absence of environ- mental cues does not mean they are not normally anchored to extrinsic inputs.