How do we remember places?
How do we recognize places and why do we sometimes get lost? In everyday life we use many strategies to help us keep track of where we are and get quickly and directly to an intended destination. For example, we might look out for distinctive landmarks and signs, we might consult maps. People also talk about having a sense of direction, for example, you might be able to point accurately toward some distant location which you cannot see (think of a local shop for example), and you may be able to keep track of these locations as you move about. This is the first of three posts that look at the brain basis of such skills. Based on this understanding my colleagues and I developed a test to assess them. It is called the Four Mountains test and is featured in Channel 4′s Hidden Talent show (to be shown on Tuesday 15th May at 9pm). These posts give a flavour of the scientific background to the test.
Imagine you are trying to find a friend’s house that you’ve visited just once before. In principle you could recognize it in several ways. One way would be to match the visual characteristics of the scene with a mental image stored from the previous visit. Perhaps you have a distinct memory that there was a flowerpot to the left of a blue door. This purely visual type of memory might be useful if you could find our way near to the place where we stored the image, but it would not be much good for getting there. It would be vulnerable to any change which disrupted the visual features we are relying on (for example, if someone moved the flowerpot or painted the door). In addition, the visual features of a place change a great deal as we move around, perhaps encountering it from different directions on different occasions. As time passes its appearance will also be affected by weather and lighting conditions and seasonal changes. For these reason purely visual memories are unlikely to be very useful over long periods of time.
We already know quite a lot about how places are represented in the brain. Brain cells communicate by sending electrical signals to one another, and by analysing signals from cells in different parts of the brain, neuroscientists can see how different forms of information are represented in different parts of the brain. In the hippocampus, neurons called ‘place cells’ signal where an animal is as it moves around. Each place cell signals as it passes through a different location, so that the current position is indicated by the pattern of cells which are signalling at any time.
Place cells are now known to be just one of several types of spatial cell forming a circuit connecting to the hippocampus – other neurons represent the current heading (something like a compass) and distances and directions to nearby boundaries (such as walls). Cells immediately downstream of the hippocampus fire in a grid-like pattern, with each neuron signalling a set of evenly spaced locations. A characteristic of all these cells is that they signal spatial properties (such as location and heading) which don’t depend on specific sensory cues. One idea is that they provide a kind of internal map (a ‘cognitive map’). Like a real map, and unlike a photograph, the information in the cognitive map doesn’t depend on your viewpoint or on individual visual details. Instead it maintains a sense of location relative to the environment around you, updated as you move about. It allows you to find new direct routes between locations and to form new memories of places that don’t depend on changeable visual features of the scene, so that locations can be recognized from whichever direction they are seen.
The video below is a TED talk by my colleague Neil Burgess, explaining how the different cell types might contribute to our memory for places, and to memory and spatial imagery more generally. I was involved in much of the experimental work he talks about as a postdoc in his research group and subsequently as a collaborator, and you can read some of the original papers on this site.
The hippocampus is one of the first brain regions to be affected in Alzheimer’s Disease, and so testing abilities that depend on hippocampal function could one day be useful in identifying more serious memory problems that might require further investigation. In the next post, I describe a simple test we have developed, based on what we know about the function of the hippocampus, which provides a measure of the ability to recognize places as distinct from their visual properties.