Think about what happens when you walk. Really THINK about it. What does it take to walk? Well, your feet and legs have to move (far more complicated than they look), which means your muscles have to move, which means your nerves have to control your muscles, which means your brain has to send the signals in the first place. All of this is based on further information, knowing where you are in space and where you're going, how fast you need to get there. And then there's even more! How do you know where you are? How do you know how fast you're going? How do you know which direction you're headed? And behind all of this are thousands and millions of neurons firing, together and separately. And underlying THAT are thousands of biochemical processes which allow the neurons to fire... take that walking speed, and make it a run.

The sheer number of neurobiological processes and number of things that need to happen to make you walk into your workplace every morning is the kind of thing that makes neuroscientists stop in their tracks with wonder.

And today, we're going to talk about a paper that may have worked out a tiny piece of how the brain might deal with things like increased speed. How does your brain keep up with your feet?

By running a little faster.

Ahmed and Mehta. "Running Speed Alters the Frequency of Hippocampal Gamma Oscillations" Journal of Neuroscience, 2012.

To understand how this works. We need to talk about two major things: place neurons, and oscillatory networks.

Place neurons are just what they sound like. In the hippocampus, certain cells will fire in response to certain places. When you record cell firing from a rat's brain with an electrode array, you can actually see and hear the neurons firing as a rat looks around a place it has been before (and this is definitely one of the awesome and defining moments of a young neuroscientis. Certain neurons will fire for, say, the left corner, and others will fire for the bottom right.

But here's where we get to oscillation. You see, a neuron doesn't fire alone. It fires repetitively with large numbers of nearby neurons, in turn triggering other neurons to fire, resulting in a firing network of neurons. These networks fire at certain frequencies and patterns, which we describe by their frequencies, with names like theta and gamma and beta.

The question is, how do things like place cell firing and neural oscillations vary as you move? The authors of this study wanted to see just how place cell firing varied as a function of an animal's speed. To do this, they hooked electrode arrays into the hippocampi of rats, and watched the oscillations of their place cells as they ran a maze. They set up a nice large Y Maze, with rewards at all the ends, and let the animals run it as fast as they wanted. They looked to see how the activity in the hippocampus varied as a function of speed.

What you can see here is a correlation between the power of the gamma oscillations of the place neurons, and the speed of the animals as they ran the maze. It's not an increase in the frequency of the oscillation, because that would just move it up the scale, and it would no longer be a gamma oscillation. Rather, it is an increase in power, more neurons firing in synchrony. They found that as speed increased, there was a strong correlation for the gamma oscillation to increase in power as well.

But was this a function of the animal's speed? Or of it recognizing different points in the maze faster? To figure this out, they simply put the animal in a different maze. The results were still the same, there was a correlation between the rat's speed and the gamma oscillation power in the hippocampus. So it's not just the result of knowing where you are. When you run fast, the gamma oscillations in your brain become more powerful, so your brain runs "faster" too.

But what does this mean? What is causing the increase in the gamma oscillations? Well, increased power of gamma oscillations has to be triggered by something. In this case, the authors hypothesized that excitatory input on to interneurons in the hippocampus, which provide inhibitory input to the place cells, might be determining the increase power. When they recorded from interneurons as well as place cells, the authors found that the interneuron firing rate correlated with the increase in gamma power. This suggests that the increased excitatory input stimulated interneurons to increase inhibitory input as the rat runs faster, which might increase synchronicity and might be driving the increased power in the gamma oscillations.

So running speed results in a higher excitatory drive to the interneurons in the hippocampus, resulting in increased neural oscillation in the gamma range. So basically, higher excitatory drive means that there is a higher likelihood of moref neurons firing together, in synchrony, at a specific frequency. Running speed increases firing rate of some neurons, which increases the oscillation of others. So running speed ends up correlating with the gamma oscillation power.

But what is the purpose of this increase in gamma oscillations? Since these are taking place in the hippocampus, they may well have something to do with learning and memory. As I mentioned above, neurons in the hippocampus encode things like place, they fire in response to where you are in an area. When they fire synchronously, they do so in the gamma range, resulting in these gamma oscillations (they also fire in the theta range, but that's for another time).

And this is an important thing to think about. Because, if you're running FASTER, you have to determine where you are faster as well! This means that you'll need to process your spatial information faster. How do you do that? Well, the authors propose two options. First, you could leave out bits of spatial recognition processing, skipping certain elements so you only get the most essential information. But you could also the whole spatial processing sequence faster. And the authors hypothesize that this second option may be the right one. The increased gamma oscillations that correlate with speed are how the animal keeps up with it's own feet, processing spatial knowledge faster as it runs faster. The faster transitions might allow the hippocampal cells to encode place no matter how fast you're going.

But of course, this isn't the end of the line. This paper shows that increased excitatory input to hippocampal interneurons increases the gamma oscillations, and this varies in speed as you run....but where is that excitatory input itself coming from? From the visual system? The vestibular system? Both? While this provides a cool and interesting piece of the puzzle, there's always still another step in knowing how we walk and run. And it's puzzles like that that stop a neuroscientist in their tracks.

NOTE: I'm seeing some stuff on twitter that makes me think people think that increased gamma oscillations while moving faster means you think "better" or "faster". That's not true. The increased power in the gamma oscillations here is basically to keep up with the sensory processing as you go faster, not to make you suddenly have bright ideas while trail running. There IS, however, some evidence that exercise can improve cognition in other domains. Just...not in this paper.

NOTE 2: As commenter Horrible Clarity pointed out, I made a mistake when talking about inhibitory interneurons. Interneurons in the hippocampus are GABA neurons and thus tend to inhibit the cells they impact. The excitatory input TO the interneurons would drive increased inhibitory input, which could increase synchronous firing in this case.

Ahmed OJ, & Mehta MR (2012). Running speed alters the frequency of hippocampal gamma oscillations. The Journal of neuroscience : the official journal of the Society for Neuroscience, 32 (21), 7373-83 PMID: 22623683