Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 11 (1), 789

The Place-Cell Representation of Volumetric Space in Rats

Affiliations

The Place-Cell Representation of Volumetric Space in Rats

Roddy M Grieves et al. Nat Commun.

Abstract

Place cells are spatially modulated neurons found in the hippocampus that underlie spatial memory and navigation: how these neurons represent 3D space is crucial for a full understanding of spatial cognition. We wirelessly recorded place cells in rats as they explored a cubic lattice climbing frame which could be aligned or tilted with respect to gravity. Place cells represented the entire volume of the mazes: their activity tended to be aligned with the maze axes, and when it was more difficult for the animals to move vertically the cells represented space less accurately and less stably. These results demonstrate that even surface-dwelling animals represent 3D space and suggests there is a fundamental relationship between environment structure, gravity, movement and spatial memory.

Conflict of interest statement

The authors declare the following competing interests: K.J. is a non-shareholding director of Axona Ltd.

Figures

Fig. 1
Fig. 1. The recording room and apparatus.
a Room and maze schematic, shown in aligned configuration. b Photographs showing the aligned lattice maze in position for recording (left) the tilted lattice maze in position for recording (middle) and a rat implanted with an Axona microdrive exploring the aligned lattice whilst connected to the wireless headstage (right).
Fig. 2
Fig. 2. Animals moved parallel to the maze axes.
Statistical test results can be seen in Table 1. Source data are provided as a Source Data file. a For each maze: a schematic of the maze configuration color-coded to show height (top right inset); three-dimensional heat plot of heading direction distribution for all sessions combined (middle); same three-dimensional heat plot for a single session (bottom right inset). Note concentration around horizontal trajectories for the arena and aligned maze, and along the three axes for the tilted maze. b Each marker represents an animal: graphs show proportion of total time spent moving roughly parallel to each possible maze axis. Red lines show the 1st, 50th, and 99th percentile of a shuffle distribution. Inset plots show the result of a shuffle testing the probability of observing this ratio of total XYZ time to total ABC time by chance. Red lines denote the 1st and 99th percentile rank positions in the shuffled distribution of ratio values (gray area), blue line denotes the overall ratio value averaged across rats, and blue dotted line denotes the ratio value observed in the single session shown in b.
Fig. 3
Fig. 3. Representative example place cells and their activity in the lattice mazes.
For additional examples see Supplementary Figs. 3 and 4 and Supplementary Movies 1 and 2. Five cells are shown, one per row. First column shows the path of the animal and spikes plotted as red markers. Second column shows the three-dimensional firing rate map. Colors denote firing rate and areas of low or no firing are transparent. Third column shows the convex hull of the dwell time map as a gray outline and the convex hull of any detected place field(s) as separate (color-coded) polygons. Last column shows the spike and firing rate maps when the data are projected onto the three possible cardinal planes.
Fig. 4
Fig. 4. Fields were less numerous in the lattice mazes but larger in volume.
Source data are provided as a Source Data file. Markers in boxplots represent fields. Omnibus test results can be seen in Table 1; post hoc test results are displayed here: *significant at the 0.05 level, **significant at the 0.01 level, ***significant at the 0.001 level. a Schematic demonstrating the process of place field detection and analysis. An example firing rate map (left) is thresholded at 20% of the peak firing rate and regions which passed our criteria were considered place fields (second plot). We can visualize these regions as convex hulls (third plot) and extract features such as their centroid (right plot). b Aligned lattice maze schematic and place field convex hulls of four example place cells exhibiting 1–4 place fields. c Number of place fields exhibited by place cells in each maze. Inset: same data in boxplot representation. d Number of fields per cubic meter exhibited by place cells in each maze. e Distribution of place field volumes observed in each maze. f Enclosing diameter of place fields in each maze.
Fig. 5
Fig. 5. Place fields were distributed throughout the three mazes.
Source data are provided as a Source Data file. a Representative arena, aligned and tilted lattice recording sessions demonstrating homogenous distribution of fields. To allow clearer distinction of separate (color-coded) fields, only fields with a volume less than 300 voxels (one voxel = 50 mm cube) are shown (~2/3 total). Inset schematics show the maze orientation and axes. b Location of all recorded place field centroids. Colors denote vertical position. c Columns follow the mazes as above; the kernel smoothed distribution of place fields along the X, Y and Z dimensions of each maze (Z is not shown for the arena) relative to the maze center. Vertical lines represent the median value of these distributions. d Columns follow the mazes as above; from top to bottom graphs show the results of a shuffle analysis on the distribution of fields along the X, Y and Z axes (Z is not shown for the arena). Blue lines denote the median position of real place fields along these axes relative to the maze center. Shaded areas represent the distribution of median values obtained from 1000 shuffles. Red lines show the 2.5 and 97.5 percentile rank positions in the shuffled distributions, red dotted lines denote the 50th percentile rank or median of the shuffle distributions.
Fig. 6
Fig. 6. The majority of place fields were elongated instead of spherical.
Source data are provided as a Source Data file. Markers represent place fields. Omnibus test results can be seen in Table 1, post hoc test results are displayed here: *significant at the 0.05 level, **significant at the 0.01 level, ***significant at the 0.001 level. a Cartesian axis and principal axis lengths of place fields in all three mazes (P1-3 are the principal, semi-major and semi-minor axes respectively). b Elongation index of all place fields in each maze. An index of 1 (gray line) would indicate a spherical field, higher values indicate elongation. The percentage of fields that are equally or more spherical than would be expected by chance is given by the text below each boxplot. c Same as b but for sphericity. Values <1 indicate deviation away from a sphere. d Probability density functions of the XYZ and ABC field length data in a. The distribution of Z lengths differs significantly from X and Y in the aligned lattice and the distribution of Z lengths appears to be bimodal (Supplementary Data: Field elongation). In the tilted lattice there are no differences between X, Y & Z or A, B & C. See Supplementary Fig. 7 for further field elongation analyses.
Fig. 7
Fig. 7. Place fields were oriented parallel to the maze axes.
Source data are provided as a Source Data file. a Schematic showing how the orientation of place fields was extracted and visualized. A place field is detected in an example firing rate map through thresholding, the principal or longest axis of this field can then be extracted (red arrows, third plot). To visualize the orientation of multiple fields we project these axes onto a unit sphere and generate a spherical Von-Mises kernel smoothed density map, where hot colors denote that many fields ‘pointed’ their principal axis in this orientation. b Three-dimensional heat plots of place field orientation for the three maze configurations; inset heat plots (top right) show data for a single session, large plots (bottom) show data for all place fields. Note concentration around the three axes of the aligned and tilted mazes. Flat cylindrical projections can be seen in Supplementary Fig. 8. c Graphs show proportion of total fields oriented roughly parallel to each possible maze axis. Circles give the observed proportion per axis, error bars represent 95% confidence intervals calculated through a bootstrapping procedure. Red lines show the 50th percentile of a shuffle distribution while shaded red areas denote the interval between the 2.5th and 97.5th percentiles. Inset plots show the result of a shuffle testing the probability of observing this ratio of total XYZ fields to total ABC fields by chance. Red lines denote the 1st and 99th percentile rank positions in the shuffled distribution of ratio values (gray area), blue line denotes the overall ratio value averaged across rats, and blue dotted line denotes the ratio value observed in the single session shown in c.
Fig. 8
Fig. 8. Spatial information was lowest along the Z dimension of the aligned lattice.
Source data are provided as a Source Data file. Omnibus test results can be seen in Table 1; post hoc test results are displayed here: *significant at the 0.05 level, **significant at the 0.01 level, ***significant at the 0.001 level. a Mean and SEM autocorrelation found for all place cells in the aligned lattice (left) and tilted lattice (right) at increasing distances or spatial lag. b Proportion of total spatial information found per projection for the aligned lattice (left) and tilted lattice (right). c Area under the firing rate curve (AUC) produced when place fields are projected onto each axis for both lattice mazes. d Span and normalized activity of every recorded place field in the lattice mazes. Vertical lines represent place fields, ordered along the x-axis by their position relative to the lattice central node. Line color represents normalized firing rate.

Similar articles

See all similar articles

References

    1. Tolman EC. Cognitive maps in rats and men. Psychol. Rev. 1948;55:189–208. doi: 10.1037/h0061626. - DOI - PubMed
    1. O’Keefe, J. & Nadel, L. The Hippocampus as a Cognitive Map (Clarendon Press, 1978).
    1. Jeffery KJ, Wilson JJ, Casali G, Hayman RM. Neural encoding of large-scale three-dimensional space—properties and constraints. Front. Psychol. 2015;6:927. doi: 10.3389/fpsyg.2015.00927. - DOI - PMC - PubMed
    1. Jeffery KJKJKJ, Jovalekic A, Verriotis M, Hayman R. Navigating in a three-dimensional world. Behav. Brain Sci. 2013;36:523–543. doi: 10.1017/S0140525X12002476. - DOI - PubMed
    1. Finkelstein A, Las L, Ulanovsky N. 3-D Maps and compasses in the brain. Annu. Rev. Neurosci. 2016;39:171–196. doi: 10.1146/annurev-neuro-070815-013831. - DOI - PubMed
Feedback