> Having the goal in the centre of the room, so that approach paths covered 360 degrees, turned out to be the essential element in pinpointing the vector cells. It meant that the team could verify that the same neurons were firing whatever the direction of approach.<p>This reminds me of hacking on old Mac OS games with a program called Pandora's Box.<p><a href="http://www.macgamefiles.com/item/11768/Pandoras-Box/" rel="nofollow">http://www.macgamefiles.com/item/11768/Pandoras-Box/</a><p>The basic idea is to dump the game's memory, then do something like lose a life, then scan the memory dump for values that changed from the old value to the new value. It almost always worked!
"That’s a surprise, considering how well this area has been studied by researchers"<p>Is it? We don't know much about the individual types of neurons at all - it's a sisyphean task, there are just too many.<p>We don't even know how a brain really functions, because it looks like it's not just neurons and there has to be much more to it. Considering the complexity, I'm almost certain there are many more mechanisms involved.<p>Remember: the brain doesn't come into existence with a fixed structure, it is built up from an over-connected state, synapses are pruned during development and after that it continues to change its structure over time depending on the input from the environment. It's not that simple and I don't think that we'll ever be able to understand it completely.