Think about how what happens when you recall a memory from an episode in your life. It doesn’t all come to you at once. First you think of the event – right now I’m thinking of my last birthday party. I remember who was there at the start, and the decorations we put up. Then those memories trigger others, and I can “see” the scene unfolding: who I talked to, how many glasses of wine I had… and the memory spreads back and forward in time until I can see the whole event.
These episodic memories are special somehow; they are complex, vivid, rich in detail and – amazingly – learned after just a single experience of an event. Compare that to memorizing a list of words in a foreign language, or a phone number, which have to be practiced over and over again.
For a long time we thought we knew which brain regions were responsible for forming memories, because patients with brain damage in the medial temporal lobe, including the hippocampus (like the famous patient HM), have amnesia. But now we know that although those regions are still important, our memories are actually much more widespread. Two recent studies investigated how this phenomenon occurs in the brain.
In the first study, Bradley Buschsbaum and colleagues investigated the way that episodic memories are stored. They asked people to watch videos while in an MRI scanner, and trained a computer algorithm to recognize the patters of brain activity associated with each video. Then they asked the same people to recall the videos in as much mental detail as possible, and found that the brain activation patterns overlapped partly with the patterns of activity that occurred when they were watching the videos themselves. There was not a complete overlap, though, suggesting they may have also recruited some different brain regions when they were recalling the videos (like the hippocampus and the anterior prefrontal cortex). However, their results definitely point to a lot of overlap between the systems used for watching and remembering.
The authors suggested that remembering the videos in vivid detail might be allowed to happen due to “pattern reinstatement”. So when you remember an episode of your life you can literally “see” it with some of the areas of the brain you used to actually see it the first time (and hear, and touch and smell, too, presumably!).
In the second study, Aleena Garner and colleagues investigated whether we can actually manipulate these scattered memories. They wanted to know whether they could create false memories by manipulating the specific cell populations associated with a particular memory, in this case the memory of a white room, in mice. They used a type of molecule known as a DREADD (designer receptor activated by a designer drug – catchy, right?). Using the DREADD, which they switched on for just a short time, they labeled all the neurons involved in the memory of the white room. Most importantly, they could activate those neurons again whenever they wanted to just by delivering the designer drug. The next day, the mice were introduced to a new context, a checkered room, but this time they received a mild electric shock that caused them to freeze up. Mice that have received a shock in a room usually freeze up whenever they go into that room, even if they don’t receive another shock, because they are anticipating something bad happening in there.
The important part of this study was that they also fired the neurons representing the memory of the white room while the mice were getting shocked in the checkered room. Then they tested whether the mice had formed a memory of the shock by putting them in the checkered room and seeing whether they froze or not. The mice only remembered the shock when they were in the checkered room and the neurons representing the white room were being fired. Essentially the scientists had given the mice a false memory; a sort of mixture of the white room and the checkered room. They only remembered the shock when they had the mixed memory.
What do these two studies mean? Well, they demonstrate just how complex our memories are. It now seems as if our memories are stored by a network or community of neurons, which all fire at the same time when you remember. This might mean that reconstructing the complex architecture of memory in patients such as people with amnesia as a result of Alzheimer’s disease or traumatic brain injury may not be as easy as we once hoped. But they also show us that if it is possible to map the particular patterns of cells associated with each memory, we will be able to learn more about how memories are stored. And the idea of “pattern reinstatement” might also help us to understand what happens when the line between memory and reality is blurred. Understanding what happens when these two patterns fail to separate may help us to treat patients who experience vivid hallucinations.