The hippocampus is a brain structure that is important for forming new memories of autobiographical events or episodes. An important part of such memories is context, which includes the spatial location in which an event occurred. Neurons in the hippocampus appear to play a specialized role in encoding spatial location: when researchers record from individual hippocampal neurons while a rat explores a maze, they find some cells that respond only when the rat enters one area of the maze, and others that respond only when the rat is in a different area. Just by knowing which of these neurons (called “place cells”) are active, researchers can deduce where in the maze the rat is. Our brains may do the same: our understanding of where we are in space may reflect which hippocampal neurons are currently firing.
Several studies have shown that hippocampal place cells encode not only where we are, but where we’re going. For example, rats can be trained in a maze where they start at one end, run down an alley, and are presented with a fork in the road: they can go left or right. While the rat is still running down the alley, some hippocampal neurons begin to fire if the rat is going to turn left, and others fire if the rat is going to turn right. Thus, just by observing which group of neurons is firing, researchers can predict what the rat is about to do. Presumably, such “prospective encoding” by neurons helps rats (and us) plan a sequence of actions that will help us get where we want to be in the future.
A recent study by Pamela Kennedy and Matthew Shapiro shows that this prospective encoding is not limited to purely spatial information. In their study, the rat is always placed at one end of a long alley and, at the far end, has a choice of entering a white goal box or a black goal box. The location of these boxes can vary. On some days, the rat is deprived of water, making it thirsty; if the rat runs into the white goal box, it will find water there. On other days, the rat is deprived of food, making it hungry; if the rat runs into the black goal box, it will find food there. (After the testing session, the rats are generally allowed free access to food and water – it’s only right before the testing that they are made hungry or thirsty.)
Kennedy and Shapiro found that some hippocampal place cells fired when the rat was running down the long alley – but only if the rat was thirsty and intending to head for the water-associated goal box. Other neurons also fired when the rat was running down the alley – but only if the rat was hungry and intending to head for the food-associated goal box. Because the location of the goal boxes varied, this wasn’t purely spatial information (“when you reach the end of the alley, turn left”); rather, these neurons seemed to encode what goal the rat was seeking (“when you reach the end of the alley, turn towards the white to obtain water”).
Assuming that similar encoding happens in the hippocampus of other mammals, including humans, such neurons could be an important component helping our brains to retrieve particular memories that are especially relevant to our current goals, allowing us to select among possible actions based on our memories of how well those actions helped us achieve particular goals in the past.
Further reading:
Pamela J. Kennedy & Matthew L. Shapiro (2009). Motivational states activate distinct hippocampal representations to guide goal-directed behaviors. Proceedings of the National Academy of Sciences, vol. 106, no. 26, pp. 10805-10810.
