Sleep deprivation has been shown to lead to reduced attention and short-term or working memory. This in turn influences what gets saved as long-term episodic memories, but it also impacts the performance of higher-level cognitive functions such as decision-making and reasoning (it should come as no surprise, then, that accidents and industrial injuries tend to increase with sleep deprivation). Interestingly, sleep has been shown in experiments to play a major role, not only in memory consolidation after learning, but also in preparing the memory for encoding BEFORE a learning experience.
Several studies have shown how sleep facilitates long-term memory processing, both the conversion of short-term memories into long-term ones, and also the reconsolidation of existing long-term memories. Recent indications are that the REM sleep late in the sleep period particularly benefits procedural memory (the memory of how to do things), while the earlier slow-wave sleep benefits declarative memory (memory of facts and events) more. In addition, motor learning seems to depend more on the amount of lighter stages of non-REM sleep, while certain types of visual learning are more dependent on both deep slow-wave sleep and REM sleep. Also, REM sleep has been shown to particularly facilitate the retention of emotionally-laden (as opposed to more emotionally-neutral) information.
Neuronal and synaptic activity in the brain during sleep has been shown to be significantly greater in the same areas where learning took place during the day, and the content of most dreams tends to revolve around recent day-to-day events, all of which is consistent with a “replaying” of events during dreams and reconsolidation of the memories and learning recently acquired. Study after study has shown that tasks learned during one day are performed better the next day after a good night's sleep, and significantly better than after a night of no sleep at all.
There is also evidence to suggest that REM sleep and dreams play a role in “weeding out” or “pruning back” memories, deleting redundant or unnecessary synaptic connections in the brain, so that some of the less important experiences with which we are inundated over the course of the day are discarded, while the more important memories are retained and consolidated.
Interestingly, species of fish that tend to school (and so, it is argued, have little need for complex, higher-level information processing) are the very species of fish that appear to exhibit notably little, or even no, sleep, which supports to some extent the learning facilitation theory. However, it does not seem to follow that the animals with the largest and most complex brains also sleep the most, suggesting that memory and learning consolidation cannot be the sole reason for the evolution of sleep in the animal world.
As well as its effects on memory, though, sleep deprivation also has a detrimental effect on mood and emotion, including increased instances of rage, fear or depression, although the exact neural and cellular mechanisms for this are still poorly understood. Alterations in mood also appear to affect our ability to acquire new information and to subsequently remember it. Indeed, some researchers believe that sleep exists principally in order to allow us to dream, which is an essential part of the process of resolving emotional problems (see the section on Dreams).
Some studies have suggested that sleep, particularly REM sleep, can facilitate creativity, flexible reasoning and higher level “insights” (sudden gains of understanding or explicit knowledge). Dreams have long been thought to be instrumental in artistic creativity and idea-forming, although largely based on anecdotal accounts. However, it should be noted that other studies have muddied the waters somewhat, and one study has even positively associated creative children with insomnia and disturbed sleep.