Sleeping cements memories

People often say that when you've got a tough decision to make you should sleep on it. A good night's sleep seems to be critical for processing information and organising the facts in your mind. Indeed, experiments have repeatedly suggested that people perform better at recalling newly-learned things when they've had a chance to sleep in the interim. But is that better performance just because they're fresh, or is the sleep doing something special to the memory? One way to find out is to record what the brain is doing while we sleep, and this week scientists have reported on how a unique opportunity presented itself to a team at Northwestern University who were able to study patients who had electrodes implanted into their brains to study their epilepsy but it also meant that they Ken Paller could eavesdrop on how newly-formed memories were altering as they slept…

Ken - One of the interesting questions about memory is: how do we actually maintain enduring memories? Because there's so many things we learn every day and yet not everything sticks. So how do memories survive? We call that consolidation, a process where memories are stored in the brain effectively. And a new idea about that is that part of what makes that work is the processing of memories during sleep. We think that memories are revived and processed again during sleep without us knowing that it's happening. In the context of deep sleep, when memories seem to float up things that have happened recently and they accrue to related memories. And all this interactions between memories is part of what makes memories stick because each memory is not just isolated, but it's connected to other memories, other things you've learned earlier. And that's what makes a very solid memory. One that's well integrated with other things you know.

Chris - So what evidence did we have that that is the case?

Ken - Well, the early evidence was comparing people's memory for things they learned some hours earlier and manipulating whether they slept during that intervening time or stayed awake during that intervening time. Now we have a lot of other sources of evidence, including looking at the physiology of sleep that has made the case more strongly. And one of the methods we use in my laboratory is to provoke memory reactivation during sleep and then see what are the repercussions of that. And what we found is that when memories are reactivated, people wake up and they're better able to remember those things that were reactivated while they were sleeping.

Chris - What about actually physically getting inside the brain? Because this week you've published some new data where you've gone inside.

Ken - Yes. We want to understand more about the mechanisms of memory storage and how memories become well consolidated so that we can remember them later. And so we looked at patients who had electrodes inside their brains. And the reason the electrodes were there is because the surgeons were offering the patients perhaps a better therapy for their epileptic seizure disorders. And so while the electrodes are in there, we could also record electroactivity while they're awake and while they're asleep. And so what we'd done in this recent study by teaching each of the patients locations of objects and the objects came along with sounds, a distinct sound for each object like the cat picture came along with a meow sound. So they learn all these locations of objects and then we remind them of a subset of those during sleep by presenting the sound very quietly so we would be sure not to disturb their sleep. And when we presented the sounds, we could then look at the brain activity that's produced in response to the sound and the memories get better when we're reminded of them. So when we wake people up, they were more accurate at remembering the locations of objects that we had reminded them of during sleep. And so now we can see what's happening in certain brain areas in the course of memory change. As memories become stronger through this process,

Chris - How is that reflected in the electrical activity? Because obviously you can test the patients and you're saying they performed better when we asked them to remember where the picture of the cat was, but what happened in their brains when they were remembering the memory of where the cat picture is?

Ken - Yeah, well one of the things we do is we actually presented them two types of sounds. So half of the sounds related to the pictures they had seen in this memory task before they went to sleep. And half of the sounds hadn't been there. So we were able to compare, well what's different about how those two sounds are processed? And we saw larger responses in certain brain areas for those sounds that were connected to the memories they had learned. And furthermore, some of those memories got really a lot better. And we could compare as a function of how much better the memories got for some sounds versus other sounds and see differential activities. Some of the electrodes were in or near an area called the hippocampus. It's known to be very important for memory function because people who have brain damage to this area have serious memory problems as a result. In that area, we would see larger responses to these meaningful sounds that related to their prior memory and predicted how much better their memory would be when they woke up.

Chris - When during their sleep did you do this? And does it matter when during the sleep you do this? Because as we know, sleep is not a static thing. You don't go to sleep, switch off and wake up eight hours later, fully rested. You go through phases of dreaming and not dreaming, deeper sleep, lighter sleep. How did this apply in your patients?

Ken - Yeah, it's a fascinating side of sleep that actually there are different stages. REM sleep for rapid eye movements and non REM sleep of a couple varieties that range from light sleep to deep sleep. And our results were focusing on non REM sleep. So we are looking at deep sleep. It's also called slow wave sleep. And that's when a lot of results have converged to show that this part of sleep, this timing, is very important for memory.

Chris - What are the implications then for as we age and the amount of sleep that we take and the proportion of sleep that's genuine, restful, deep sleep, may change. What are the implications for forming new memories and the function of our memory overall?

Ken - Yeah, that's very important because sleep does change in interesting ways with aging. And one of the changes that starts even in the twenties is that we have less slow wave sleep. And that could be a reflection of sleep not working quite as efficiently. And there are many ideas about how that could be important and be related to age related memory troubles that people have. The general question is: what is good sleep? One of the usual answers is, Well, you have to have enough sleep, seven or eight hours of sleep. And that's a starting point. And then you might say, Well, we also have to measure how much of that sleep is REM sleep versus slow wave sleep. And we're even going further than that because we think, well good sleep quality is about not just how many minutes, but how efficient is the processing and what's happening during that processing. And I think further from the mental side, what kind of memories are you bringing up? Because every day things happen to you and every night you revisit some of those memories. Well, which memories do you revisit? If you're spending some of your time being depressed during the day or maybe having a lot of anxious thoughts, perhaps those are the type of memories that are coming up again during your sleep and exacerbating the issues. So we want to think carefully about good quality sleep and think about, well can we push around sleep to make it better for people so that they wake up, you know, on the right side of the bed. And now that we know it's a little bit malleable, that we can push it around with sounds we present. There's a lot of opportunity here for making sleep better.