September 27, 2011 | 3
I’ve got a kind of odd talent. Once I’ve been to a place physically, I can’t get lost getting there. As far as super powers go it’s really pretty crappy (I couldn’t have had something like the power of instant good signal wi-fi? Maybe hair that never tangles?). Looking at the map won’t do it, and I can’t ride a bus there. I have to drive or walk, following directions. The first time, I suck at directions, and I’ll probably get hopelessly lost. But after that, I WILL find your house again. I usually take directions just in case, but I don’t need them. And I don’t even remember where I’m going on a conscious level. I just KNOW that turn is right. I don’t know why, it’s just the right one. And the memory persists for years.
I’ve often noticed something odd about this talent. Looking at the satellite images on google maps doesn’t help me. It’s not just a matter of remembering to turn right at the Wendy’s. Landmarks help, but I need to experience the space myself. Once I’ve done that, my memory is solid. And despite my love of my own talents, I’m sure I’m not unusual in this. I bet many, and perhaps most, other people have this talent, too. Why? Because the rats appear to have it down pretty pat, and can we be far behind?
Rowland et al. “A stable hippocampal representation of a space requires its direct experience” PNAS, 2011.
Let’s start with your hippocampus, an area of your brain which is about 3 inches directly in from the tops of your ears (ish). Named for looking kind of like a seahorse tail, in neuroanatomy we learn about it as looking like a cinnamon roll. On the inside, it’s got all these lovely curls. On the outside, to be honest, it looks a little like a large bean.
The hippocampus may look kind of small, but it’s heavily involved in a lot behaviors that…neuroscientists are interested in. I often blog here about the involvement of the hippocampus in things like depression, but when most people learn neuroanatomy, they learn that the hippocampus is most often involved in learning and memory. The hippocampus features cells that we like to call place cells, which fire in response to an animal being in a particular place. When an animal first enters and explores a new environment, place cells begin to fire, and the neurons which fire “remember” the place, and will fire again when the animal is re-exposed to the environment.
All this is well and good, but the question then becomes: what makes those memories, those place cell firings, “remembered” or permanent? Does the animal HAVE to experience the environment by being in it? Or is looking at the environment enough? After all, place cells do fire in response to an animal looking at an environment they are not in, like looking at a map without actually going there. So do the place cells take on permanent firing patterns just by looking at a space? Or do you need to actually explore it?
To test this, the authors took some rats and performed a really interesting experiment. It’s kind of complicated, but we’ll get through it, I promise. And they proved that in order for your place cells, and your hippocampus, and thus your rat to really remember a space, looking at it just isn’t enough. You gotta go there.
They started with rats, implanted with electrodes to track the firing of neurons in their hippocampi (hippocampus, hippocampae, hippocampi? Anyone?). They had the rats explore a box. But this box was a smaller box inside a LARGER box. It was of clear plexiglass so the rats could see out, and thus see outside the small box to the large box around it, but not actually GO into the larger box. They gave the rats plenty of time to learn the first enclosure. During this time, the place cells in the hippocampus of each rat began to fire in correspondance with parts of the space. Like so:
They then wanted to see if the place cells could tell the difference between a place that had been experienced (the small enclosure) as opposed to one that had only been viewed (the large one). To do this, they gave the rats a drug which blocked a receptor for a chemical called glutamate. When neurons fire, they communicate by releasing small packets of chemicals (called neurotransmitters) into the space between one neuron and another (for more on neurotransmission, see my SCIENCE: 101 post on the topic). The main “stimulatory” (meaning it tends to stimulate other cells to fire in turn) neurotransmitter in the brain is glutamate, a small amino acid.
But the interesting thing about neuronal communication is that a neurotransmitter can’t just bump into a neuron and expect to get it to do anything. No, the neurotransmitter has to hit a receptor, and the RECEPTOR will determine what will happen in the target cell. In pharmacology we always say that a drug, chemical, or neurotransmitter is only as good as its receptor. And most neurotransmitters have more than one type of receptor that they can hit. The one we are looking at today is a receptor for glutamate called the NMDA receptor. It’s the receptor that controls most of the activity of place cells in the hippocampus.
So, the rats had exposure to the small box, and they had seen the large box from inside the small box. Then the scientists allowed them to explore the LARGE box. But when they did, they gave some of them saline, and some of them a drug which blocks NMDA receptors, which means that it blocks place cell signaling. They then looked to see what happened to the place cells.
This is a series of recordings from a saline treated rat. Each column represents an exposure to the boxes, each row represents a place cell, and each box is…the box. The little yellow glows show that the particular place cells was firing when the rat was exploring that particular area of the box.
What we’re interested in here is the bottom five rows, specifically the third and fourth columns. These are place cells that represented areas of the box that the animals could only SEE before, but are now experiencing for the first time. They didn’t fire in the first two columns, when the animal was confined to the inner chamber, but when the animal explored the outer chamber, they fired, and 12 hours later, fired again, showing that the place cells now had a “memory” for the location.
Now what happens if you allow the place cells to fire when the rat LOOKS outside the box, but don’t allow them to fire when the rat experiences the new space?
This is a recording from a rat that received the drug to blog NMDA signalling, effectively stopping place cell activity, right before he EXPERIENCED the big box. He’d SEEN it before, just not experienced it. The reasoning goes like this: (1) if the place cells have encoded a firm memory from just seeing the outer box, then the EXPERIENCE of the outer box will not change the place cell firing. The place cell will fire just the same, and the NMDA blocker won’t make a difference, because new memories aren’t being formed, the rat is just confirming memories that are already there. On the other hand, (2) if place cells require MORE than just seeing the outer box, if they require the actual experience, then NMDA signaling cells will have to fire to CREATE new place cells. This means that if you block NMDA signaling, you block the creation of new place cells, and the place cells will change from time point to time point.
And that second option was what happened here. What you want to look at are the four rows, and from the third column on. We know each row is a place cell. If you look up at the saline treated rat, you’ll see that in these areas the place cells are firing when the rat goes into the same part of the space every time, showing a solid memory. But in the rat that got an NMDA-blocker, the place cells are NOT firing in the same place. They are shifting back and forth over time, with new ones each time the rat explores the box.
So what does this mean? It means that the rats had to explore the box to develop real place cell “memories” of it. It’s not enough to just look at the large box from inside a small one. While the saline treated rats then went into the large environment, and confirmed the place cells they had before, this required NMDA signaling, and new “memories” being laid down, and so rats treated with an NMDA blocker could not do it.
Staring at a goal, or staring at a map, won’t do it. You’ve got to go there.
I was discussing this paper at the bar with some friends last night (yeah, I’ve got nerdy friends, but of course I do! They are MY friends, after all), and one of them wanted to know if it was the same deal when navigating a virtual space, like in a video game. They said that many of the places they remembered from games they could navigate perfectly well (and a lot of these are like mazes) years afterward. I imagine the study would have to be done (though I bet most humans wouldn’t respond well to a pile of electrodes in the hippocampus), but I think it might be the same. If you’re “in” the space, and moving through it from a first person perspective, that might be enough experience to give you your solid place cells. But if you’re just looking at it from the outside, it’s not going to work. But then again, it could be another mechanism!
The take away message is the same. In rats at least, just looking at a space isn’t enough for you really to remember it. You have to experience it. You may not remember where you’re going, but you’ll always remember where you’ve been.
Rowland DC, Yanovich Y, & Kentros CG (2011). A stable hippocampal representation of a space requires its direct experience. Proceedings of the National Academy of Sciences of the United States of America, 108 (35), 14654-8 PMID: 21852575