Thursday, 11 September 2014

Why We Dream: Part 4 - Consolidation of learning during sleep

Here is the final part of a new Human Givens Blog series, extracts from Chapter 7 of the book Why We Dream: The definitive answer.

Part 1 - Solving Problems: Creativity and Dreams
Part 2 - Stages in the creative process
Part 3 - Invention and dreams

Part 4: Consolidation of learning during sleep

Many studies have, however, shown that there is a connection between REM sleep and the effectiveness of memory recall for new learning, and some researchers believe that dreaming is primarily concerned with this. Much recently published work still reflects this stance.

However, the dreaming and learning effect is most notable in types of learning that involve ‘false starts’, such as memory for routine procedures, as demonstrated by Professor Peretz Lavie when he set rats to learn the path through a maze. More recently, Erin Wamsley and colleagues taught students to find their way through a complex computer maze. After learning to navigate the maze, half the students enjoyed a 90-minute nap while the rest watched videos. Five hours later, all tried the maze again. Those who had not slept showed no improvement; those who had slept but not reported dreaming about the maze showed slight improvement whereas those who dreamed about the maze performed 10 times better than the other sleepers. However, the dreams were not exact replays of navigating the maze – one sleeper even reported images of a bat cave he had once visited, mixed up with maze images.

The expectation fulfilment theory would predict these findings. It suggests that REM sleep allows the introspected false starts that were not acted upon to be deactivated. Once the metaphorical acting out of the introspected false starts has taken place, the autonomic arousal is switched off, allowing the ‘correct’ memory to be consolidated – hence the improvement in memory for certain types of learning after REM sleep.

Even the finding that, when rats are in REM sleep after learning a maze, the pattern of brain activity is almost identical to that experienced while learning the maze can be explained by our dream theory. Co-author of the study Matthew Wilson, based at the Massachusetts Institute of Technology’s Center for Learning and Memory, infers from his findings that, during sleep, the brain rein- forces skills learned during the day by repeating and consolidating memories. We, however, would suggest, that the equivalent brain patterns are likely to be indicators that the unexpressed emotional arousal (likely to be high anxiety) experienced while trying to learn a maze is now being discharged in a dream.

That dreaming is not primarily concerned with memory consolidation would seem to be evidenced by the fact that animals and humans denied REM sleep do not show deficits in task learning. As the brain is in a non-memory encoding state during sleep, consolidation at that time is unlikely. Other researchers have suggested that, as the hippocampus is reactivated during sleep after a learning task, and performance on the task is improved afterwards, dreams are a side-effect of the reactivation of this area, explaining why dream content is not directly related to the learned material. All this is more easily and economically explained by the expectation fulfilment theory of dreaming.

The REM state is the repository of our stored instinctive knowledge templates, and all new understandings are the result of a process of refinement and integration of our existing knowledge and new learning. That, in essence, is what the creative process is about, but we have more to say about it in the next chapter....


Knowing the basics of the expectation fulfilment theory of dreaming will enhance your enjoyment of this blog series, so if you are not already familiar with it, watch this talk, or read this page of the Why We Dream website.

Of course you could read the entire book, (The Kindle version is only £4.99!) where all references can be found.

No comments:

Post a Comment