Evolutionary pressures leading to REM sleep

It was professor Michel Jouvet who first suggested that REM sleep. which is so closely linked to dreaming and trance states, may have evolved to permit more freedom in the expression of instinctive behaviour.  To understand how central this is to our understanding of the complexity of human psychology and the link between dreaming and hypnosis we need to take a brief look at the evolutionary pressures that led to the emergence of REM sleep.

The ability of mammals and birds to keep a constant internal temperature conferred great advantages on them in terms of mobility. But there was a great price to pay - more than a fivefold increase in basic energy expenditure over cold-blooded creatures. Such a massive increase in metabolism was compensated for by a corresponding increase in the requirement for energy. It made no sense to meet this requirement by simply extending the time spent looking for food. What was needed was a matching increase in productivity, a more productive return between the energy expended in seeking food and the energy gained by acquiring it. One part of this came from cutting down on wasteful time, for example, sleeping when prey or other sources of food were not available. This can be seen as one of the functions of 'slow wave' sleep. But, if one reduces energy expenditure by cutting out non-productive time, although this does conserve energy it doesn't actually provide any.

Since a massive increase in energy gain is required to compensate for the demands of being warm-blooded there was clearly a need to develop the ability to employ this new, high-powered energy system in more productive ways. Essentially the animal needed to become more intelligent, hence the expansion of the neocortex in mammals. This enables the animal to inhibit certain drives if their attempted expression is deemed unlikely to be successful, and thus avoids wasting precious energy. Equally, if the neocortex's analysis shows that a certain course of action is likely to bring results, it can promote such action by stimulating expectation. This, of course, is the function ascribed to the greatly expanded neocortex in mammals. MacLean summed this up well when he wrote: A remarkable feature of the neocortex is that it evolved primarily in relation to systems receiving and processing information from the external world, namely the exteroceptive, visual, auditory and somatic systems. It was as though the neocortex was designed to serve as a more objective intelligence in coping with the external environment.

For this objective intelligence to operate, it must have a detailed knowledge of, or access to, information about the instinctive programmes. If these instinctive programmes are to allow for individual and environmental variation, then this involves incompletely specified models for which sensory analogues have to be identified in the environment. It seems that the function of REM sleep in the foetus and neonate is the programming of these genetically anticipated patterns of stimulation. And, because they are necessarily incomplete, they can only be expressed metaphorically. The programming must carry the instruction to find the matching environmental stimuli to complete the template - be it language, sexual stimuli or prey for food etc.

Why we evolved to dream
Nature still had one more problem to solve. This is that, once an instinct driven pattern is activated, it can normally only be deactivated by the actual carrying out of the programmed behaviour by the central nervous system, and this clearly does not give animals the flexibility they need to survive. We can see this easily in our own lives. When you get annoyed with your children, for example, and tell them off or smack them this usually dissipates your anger. But if you bottle it up the anger is still retained in the nervous system. If every time we were emotionally aroused we were to act out those emotions it would be disastrous. There would be continuous violent and sexual mayhem for a start. Furthermore, if we could only inhibit arousals without dealing with them in some way, we would require an enormously larger brain than would be feasible. 

So, animals needed to evolve the ability to inhibit the arousals when necessary and deactivate them later when they could do no harm. That is why we evolved to dream. During REM sleep activated instinct drive patterns 'left over' from waking are vicariously run out, deactivating them and releasing the data processing potential of the neocortex to deal with the emotionally arousing contingencies of the next period of being awake.

Thus we can see the beautiful economy of nature. The same process that programmes instinctive behaviour - the genetically anticipated patterns of stimulation - is also used to deactivate 'left over' anticipated patterns of stimulation from waking - the activated instinctive drive patterns. The instinctive frames of reference programmed in REM sleep don't have  sensory content until they are matched up with their environmental counterparts The anticipated or introspected stimulation which gives rise to dreaming on the other hand, does contain a sensory description, and thus its analogical processing in REM sleep – dreaming.  So nature accomplishes two essential functions with the same process.