Latest reports about 3D Spatial Representation by Gily Ginosar at Weizmann Institute of Science and Misun Kim at UCL in the Grid Cell Meeting on May 21-22, 2018. (http://www.cognitive-map.com/img/GCMposters.pdf )

Gily Ginosar, Weizmann Institute of Science

Grid cells recorded from animals exploring 2D planes, fire in a hexagonal pattern across the environment. However, many animals navigate through 3D space, but no studies have characterized the 3D volumetric firing of grid cells.

Here, they trained Egyptian fruit bats (Rousettus aegyptiacus) to fly in a large room, while they wirelessly recorded single-neuron activity in medial entorhinal cortex (MEC).

Their results revealed structured firing in the 3D firing-rate maps, with multiple firing-fields. The spacing between firing-fields was more variable than in perfect synthetic 3D lattices, but was less variable than shuffled data. Thus, some neurons exhibited a fixed distance scale, without forming a global lattice – supporting a distance-coding function for grid cells.

They also found a number of other spatial cell types in the MEC, including (i) 3D border cells, (ii) 3D head-direction cells, and (iii) a new subset of MEC neurons that fired near landing-balls-oftentimes only at very specific balls.

Taken together, these data suggest a rich 3D spatial representation in the MEC of flying bats – including coding of 3D space by grid cells, coding of 3D geometry by border cells, as well as object-related coding in the bat MEC.

References:

Finkelstein A, Las L & Ulanovsky N, 3-D maps and compasses in the brain, Annu. Rev. Neurosci. 39, 171-196 (2016)

Finkelstein A, Derdikman D, Rubin A, Foerster JN, Las L & Ulanovsky N, Three-dimensional head-direction coding in the bat brain, Nature 517, 159-164 (2015).

Yartsev MM & Ulanovsky N, Representation of three-dimensional space in the hippocampus of flying bats. Science340, 367-372 (2013).

Ulanovsky N, Neuroscience: How is three-dimensional space encoded in the brain? Curr. Biol. 21, R886-888 (2011).

Misun Kim, UCL

Recent studies suggest that grid cells are an efficient neural mechanism for encoding knowledge about spatial location and also abstract cognitive information. The world, be it physical or abstract, is high-dimensional. However, grid cells have been mainly studied on a two-dimensional plane and little is known about how grid cells encode three-dimensional (3D) space.

Here, they consider how to study 3D grid cells in humans using a novel 3D virtual reality paradigm and fMRI analysis method. This analysis relies on the known property of grid cells where their activity is modulated by the movement direction of animals relative to the grid axis.

They found that signals in left entorhinal cortex were best explained by one particular 3D grid cell model – a face-centred cubic lattice model. This is the first empirical evidence for 3D grid cells in the brain.

In conducting this study, they also developed interactive software to help researchers visualize 3D grid cells and predict the activity of these cells for varying movement directions and grid axis orientations.

Their findings and software serve as an initial stepping-stone for studying grid cells in realistic 3D worlds and also potentially for interrogating abstracting high-dimensional cognitive processes.

References:

Kim, Misun, and Eleanor A. Maguire. “3D grid cells in human entorhinal cortex: Theoretical and methodological considerations and fMRI findings.” bioRxiv (2018): 282327.

Kim, Misun, Kate J. Jeffery, and Eleanor A. Maguire. “Multivoxel pattern analysis reveals 3D place information in the human hippocampus.” Journal of Neuroscience 37, no. 16 (2017): 4270-4279.

Kim, Misun, and Eleanor A. Maguire. “Hippocampus, Retrosplenial and Parahippocampal Cortices Encode Multicompartment 3D Space in a Hierarchical Manner.” Cerebral Cortex 28, no. 5 (2018): 1898-1909.