processes collaborate to form a representation of the outside world,
the product is colloquially known as knowledge. My research interest
has long been in how the brain “makes knowledge” from the raw materials
of neural tissue plus sensory information, and my focus of study is how
the brain constructs a map of its environment – the cognitive map – for
use in navigation and memory. Central to this map is the hippocampus,
and we investigate the functioning of neurons in the hippocampus and
also in the brain areas that send information to it - particularly
entorhinal cortex, subiculum and retrosplenial cortex.
In humans, the hippocampus seems to have an important role not only in spatial behaviour and navigation, but also in memory for life events. It is thought that perhaps our memory for events (episodic memory) is built upon a spatial framework. If this is true, then understanding the "map" in the hippocampus may help us understand episodic memory, and the things that go wrong with it in conditions like Alzheimer's disease.
My lab studies the activity of single neurons in the hippocampus and in those regions that project to it, in order to understand what environmental information the cells use to form their map of space. We study rats and mice, and our collaborators extend those findings into studies in humans. Our experimental setup is shown below, in which a rat explores (searching for food) while the activity of hippocampal neurons is monitored.
Experimental setup for recording spatially responsive neurons. The plot at the top right shows the final data form, in which the activity of a neuron (its action potentials) are shown plotted at the place where the animal was when the cell fired. This place cell has a "place field" in the North-East corner of the apparatus.
are shown typical recordings from each of the four main spatial cell
types; a place cell, a head direction cell, a grid cell and a border
One of the questions my lab is very interested in is whether the cognitive map extends into all three dimensions, or whether it is basically flat. To this end we have developed apparatus (see pictures below) that enables rats to climb up into the vertical dimension.
On the top panel
in the picture is a climbing wall that we call the "pegboard", with a
rat climbing across the pegs.
On the right are two plots showing
recordings from a grid cell, made after a rat had explored either a
horizontal arena (left) or the pegboard (right). Note that the spikes
from the cell cluster into blobs on a horizontal surface (left plot) but
vertical stripes on the pegboard (right plot), suggesting the cells are
not responding to changes in height in the same way that they respond
to horizontal travel.
On the bottom is the helical maze, a spiral track in which rats ascend and descend, collecting food at the top and the bottom.
To the right is shown the spikes from a grid cell, viewed both from above and from the side (with the coils unwound into a straight line). Note that again, the grid cell seems to react only to changes in horizontal, but not vertical position. Our conclusion is that grid cells - the "odometers" in the brain - do not track vertical distance, but only horizontal distance travelled.
Based on these findings, and numerous others, we have proposed that the cognitive map in mammals is essentially "flat".
Current work aims to test this hypothesis to see if it is true and if so, whether this is just in lab rats or whether it is something more general, and maybe true in humans too.