[ Pobierz całość w formacie PDF ]
in common is the dreamer s feeling that he or she is unprepared. This is not unusual in life, especially in the lives of busy, competitive people. In dreaming, it may be related to the fact that we have a great deal of difficulty vocalizing our thoughts. In other words, another formal feature of dreaming is the inability to remember. It is rare in dreams to remember something; it is strikingly peculiar that, although dreams are loaded with memory fragments, we don t stop in the middle of the dream and say, Hey, that reminds me of something or remember, for example, that one of our dream characters has recently died. It is this defect in thought, added to the coupling of anxiety to one of its most common associations, the fear of failure, that may determine what an individual calls his recurrent exam dream. It may seem to the reader that we are attempting to explain away rather than explain recurrent dreams, but this is not the case. What is recurrent are certain emotionally salient themes that depend on certain formal properties of dreams, and these are deeply repetitive. Every dream is characterized by visual perception and strong emotion, most often elation, anger, or anxiety. With these emotions come our own historical experiences the experiences that are associated with these emotions are likely to appear in our dreams. Human learning and memory The distinction between learning and memory is particularly clear in the case of procedural tasks that expose individuals to sensorimotor challenges some of which are met entirely outside our awareness. On a task of visual discrimination the Karni Sagi task, for example individuals improve their performance without knowing how or why (as they learn the task); after sleep they then do better 111 Dreaming, learning, and memory when re-tested (again without knowing how or why). On the visual discrimination task (or VDT) individuals must fix their gaze on a symbol, an L or a T , at the centre of the screen and then indicate when an aberrant stimulus array (\\\ instead of ///) is flashed in another part of the screen. As is usual in cognitive science, reaction time is the measure used and, within an hour of training, most individuals become very fast at this subliminal recognition. Despite not knowing consciously how they do it, people get quite good at this task and improve markedly during training sessions. Improvement is measured by the achievement of a high percentage of correct judgements as the time of stimulus exposure is reduced. We don t think of this as memory (and it may have little or nothing to do with dreaming), but it is probably typical of most of the things that we learn and so it is quite important. In other words, we learn a myriad of procedures without being able to describe them verbally. Most learning is unconscious. We call this kind of learning procedural memory to distinguish it from episodic and semantic memory. The next day, when they are re-tested, Karni Sagi subjects performance correlates strongly with how they slept. If they have been deprived of REM sleep, they behave as novices showing no advantage of their previous exposure. If they sleep deeply early in the night and/or have long REM periods late in the night, they retain their learned skill and may even improve on it. The greatest improvements occur when both deep early night (NREM) and long late-night (REM) sleep are present. As can be seen in Fig. 10, the correlation between the product of the two sleep measures and improved performance is almost one. Suitable controls, for lapsed time and for sleepiness, show that it is sleep itself that confers the benefit of skills improvement. This result is important for several reasons: one is that it is robust; another is that it is highly replicable; a third is that, because it is 112 Dreaming 10. Visual discrimination task learning. (a) Correlation of learning with slow wave sleep (SWS) and rapid eye movement sleep (REM) across the night. For each quartile of the night, the Pearson correlation coefficient between SWS% and overnight improvement (filled squares) and between REM% and overnight improvement (open circles) was calculated. (b) Two-step model of memory consolidation. Improvement is plotted as a function of the product of the amount of SWS during the first quarter of the night and the amount of REM in the last quarter. Both amounts are quantified as percentages of the entire night. The strong correlation suggests a two-step consolidation process, including an early, SWS-dependent process and a late REM-dependent one. entirely unconscious, it cannot be faked; a fourth is that the learning probably takes place in a restricted area of the brain that can be specified the primary visual cortex which makes the experimental theory amenable to study using imaging techniques. Unfortunately, it is not a task that rats can learn so we cannot easily expose the cellular neurobiology of the sleep benefits. As for cats, forget it! Dreaming itself and learning People who never master the Karni Sagi task do not dream of taking the test in any way that relates to their competence. Are there waking learning experiences that are so strong that they are in evidence within the mental state/thoughts of sleepers? We have previously alluded to the reports of skiers and sailors who notice a return of the illusion of movement of skiing or sailing at sleep onset. We also know that mental experiences at sleep onset are dream-like. 113 [ Pobierz całość w formacie PDF ] |