Dreaming LifeDreaming Life

The reverse learning theory of dreams was developed by Francis Crick and Graeme Mitchison and presented in an article in Nature in 1983. This theory says that the neocortex (the part of the brain that is involved with higher levels of thought) is a network in which neural connections are constantly being made. Dreaming eliminates unnecessary connections in the neocortex and so prevents it from becoming overloaded and malfunctioning.

The reverse learning theory says that as we learn and grow, connections are made in our neocortex in a semirandom way. As the number of connections increases, the network becomes less and less efficient. If there were no mechanism to control the number of connections, "parasitic" memories - memories that combine parts of real memories and are falsely associated with many different inputs - would develop eventually. Crick and Mitchison predicted that if this were to happen, people might develop bizarre thoughts (from mixed up memories), hallucinations (from memories being associated with the wrong inputs) or obsessions (from the same connections being made over and over).

According to the reverse learning theory, rapid eye movement (REM) sleep, the sleep stage that is associated with dreaming, prevents this from happening. During REM sleep, the neocortex's major connections to the external world are shut off. The brain stem then sends random stimuli to the neocortex. Crick and Mitchison hypothesized that this would cause some of the connections in the cortex to weaken, which would eliminate some of the thoughts and random associations that aren't useful to keep. They commented on the fact that dreams are often bizarre and illogical, like the expected outputs of parasitic memories.

Arguments Against Psychoanalysis

Crick and Mitchison pointed out that almost all mammals and birds - animals which have a neocortex or similar structure - experience REM sleep. Psychoanalysis can't explain why even tiny moles experience REM sleep.

They noted that in mammals, fetuses and newborns experience much more REM sleep than older animals. Psychoanalysis doesn't explain why a newborn baby or someone that hasn't even been born yet would need to dream, let alone spend more time dreaming than an adult. What unwanted thoughts could they be repressing? According to the reverse learning theory, however, this observation is explained by the fact that fetuses and newborns are developing neural connections at an extremely rapid rate.

In addition, Freud and his contemporaries did not know that we forget many more dreams than we remember. We normally only remember a dream if we happen to wake up during the dream, and if we don't make a conscious effort to continue to remember that dream, we will soon forget it. The function of REM sleep therefore must have more to do with what is happening in our brains when we are unconscious - when we are dreaming and forgetting - than with the few dreams that we remember.