Broadly speaking, I am interested in the fine-grained architecture of cognition - in other words, how is information represented in the brain? A particularly interesting forum for exploring this question is spatial cognition, because work has revealed the existence of a map-like representation of space in the brain that can be used for self-localisation and navigation. This cognitive map seems to reside in a network of neural structures including the hippocampus and entorhinal cortex. The big questions concerning the cognitive map are (1) How is it constructed, and (2) What is it for?
My work involves recording single neurons from these structures in freely moving and exploring rats, to determine how the cells respond to spatial information. The hippocampal neurons (place cells) encode location in a complex, multidimensional space, and some entorhinal neurons (grid cells) have the recently discovered property that they mark out distances across the environment, forming a grid-like array of activity that can presumably be used by other brain structures in spatial computations. I am currently pursuing a number of questions related to these neurons: (1) How does a place cell determine where it is? (2) How do place cells integrate spatial and non-spatial information? (3) What determines the spacing between the activity peaks of grid cells? (In other words, how does a grid cell calculate how far the animal has gone, and in what direction?) (4) How are spatial and non-spatial inputs integrated (by both place cells and grid cells)?
As well as single neuron studies, I have been exploring behavioural tasks that will enable us to determine how an animal determines where it is, or where it is going. These studies will help uncover what the cognitive map is used for.