1. Myth: It is possible to produce everlasting memories. Even reputable researchers use the term permastore (see: Prof. Harry Bahrick). It is a widely-held belief that it is possible to learn things well enough to protect them permanently from forgetting. Fact: It is possible to learn things well enough to make it nearly impossible to forget them in lifetime. Every long-term memory, depending on its strength, has an expected lifetime. When the memory strength is very high, the expected lifetime may be longer than our own lease on life. However, if we happened to get extra 200 years to live, no memory built in present life would remain safe without repetition
2. Myth: We never forget. Some accelerated-learning programs claim that we never forget what we learn. Knowledge simply gets "misplaced" and the key to good memory is to figure out how to dig it out. Fact: All knowledge is subject to gradual decay. Even your own name is vulnerable. It is only a matter of probability. Strong memories are very unlikely to be forgotten. The probability of forgetting one's name is like the probability of getting hit by an asteroid: possible but not considered on a daily basis
3. Myth: Memory is infinite. Fact: Anyone with basic computational understanding of memory will call this claim absurd, but 50% of Americans still believe the earth was created by God less than 10,000 years ago. We cannot even hope to memorize Encyclopedia Britannica in lifetime. Because information is stored in synapses which are finite, memory storage is naturally finite too. Even worse, storing information long-term is not easy. Most people will find it hard to go beyond 300,000 facts memorized in a lifetime (with SuperMemo, 300,000 items is quite realistic though). For the other extreme see: Memory overload may cause Alzheimer's
4. Myth: Mnemonics is a panacea to poor memory. Some memory programs focus 100% on mnemonic techniques. They claim that once you represent knowledge in an appropriate way, it can be memorized in a nearly-permanent way. Fact: Mnemonic techniques dramatically reduce the difficulty of retaining things in memory. However, they still do not produce everlasting memories. Repetition is still needed, even though it can be less frequent. If you compare your learning tools to a car, mnemonics is like a tire. You can go on without it, but it makes for a smooth ride
5. Myth: The more you repeat the better. Many books tell you to review your materials as often as possible (Repetitio mater studiorum est). Fact: Not only frequent repetition is a waste of your precious time, it may also prevent you from effectively forming strong memories. The fastest way to building long-lasting memories is to review your material in a precisely determined moments of time. For long memories with minimum effort use spaced repetition (see SuperMemo)
6. Myth: You should always use mnemonic techniques. Some enthusiasts of mnemonic techniques claim that you should use them in all situations and for all sorts of knowledge. They claim that learning without mnemonic techniques is always less effective. Fact: Mnemonic techniques also carry some costs. Sometimes it is easier to commit things to memory straight away. The pair of words teacher=instruisto in Esperanto is mnemonic on its own (assuming you know the rules of Esperanto grammar, basic roots and suffixes). Using mnemonic techniques may be an overkill in some circumstances. The rule of thumb is: evoke mnemonic techniques only when you detect a problem with remembering a given thing. For example, you will nearly always want to use a peg-system to memorize phone numbers. Best of all, mnemonic tricks should be a part of your automatically and subconsciously employed learning arsenal. You will develop it over a long run time with massive learning
7. Myth: We cannot improve memory by training. Infinite memory is a popular optimist's myth. A pessimist's myth is that we cannot improve our memory via training. Even William James in his genius book The Principles of Psychology (1890) wrote with certainty that memory does not change unless for the worse (e.g. as a result of disease). Fact: If considered at a very low synaptic level, memory is indeed quite resilient to improvement. Not only does it seem to change little in the course of life. It is also very similar in its action across the human population. At the very basic level, synapses of a low-IQ individual are as trainable as that of a genius. They are also not much different from those of a mollusk Aplysia or a fly Drosophila. However, there is more to memory and learning than just a single synapse. The main difference between poor students and geniuses is in their skill to represent information for learning. A genius quickly dismembers information and forms simple models that make life easy. Simple models of reality help understand it, process it and remember it. What William James failed to mention is that a week-long course in mnemonic techniques dramatically increases learning skills for many people. Their molecular or synaptic memory may not improve. What improves is their skill to handle knowledge. Consequently, they can remember more and longer. Learning is a self-accelerating and self-amplifying process. As such it often leads to miraculous results.
8. Myth: Encoding variability theory. Many researchers used to believe that presenting material in longer intervals is effective because of varying contexts in which the same information is presented. Fact: Methodical research indicates that the opposite is true. If you repeat your learning material in the exactly same context, your recall will be easier. Naturally, knowledge acquired in one context may be difficult to recover in another context. For this reason, your learning should focus on producing very precise memory trace that will be universally recoverable in varying contexts. For example, if you want to learn the word informavore, you should not ask How can I call John? He eats knowledge for breakfast. This definition is too context-dependent. Even if it is easy to remember, it may later appear useless. Better ask: How do I call a person who devours information?. Now, even if you always ask the same question in the same context, you are likely to correctly use the word informavore when it is needed. For more on encoding variability and spacing effect see: Spaced repetition in the practice of learning 
9. Myth: Mind maps are always better than pictures. A picture is worth a thousand words. It is true that we remember pictures far better than words. It is true that mind maps are one of the best pictorial representations of knowledge. Some mnemonists claim that all we learn should be in the form of a picture or even a mind map. Fact: It all depends on the material we learn. One of the greatest advantages of text is its compactness and ease at which we can produce it. To memorize your grandma's birthday, you do not really need her picture. A simple verbal mnemonic will be fast to type and should suffice. In word-pair learning, 80% of your material may be textual and still be as good or even better than pictorials. If you ask about the date of the Battle of Trafalgar, you do not need a picture of Napoleon as an illustration. As long as you recall his face at the sound of his name, you have established all links needed to deduce relevant pieces of knowledge. If you add a picture of the actual battle, you will increase the quality and extent of memorized information, but you will need to invest extra minutes into finding the appropriate illustration. Sometimes a simple text formula is all you need
10. Myth: Review your material on the first day several times. Many authors suggest repeated drills on the day of the first contact with the new learning material. Others propose microspacing (i.e. using spaced repetition for intervals lasting minutes and hours). These are supposed to consolidate the newly learned knowledge. Fact: A single effective repetition on the first day of learning is all you need. Naturally it may happen, you cannot recall a piece of information upon a single exposure. In such cases you may need to repeat the drill. It may also happen that you cannot effectively put together related pieces of information and you need some review to build the big picture. However, in the ideal case, on the day #1 you should (1) understand and (2) execute a single successful active recall (such as answering the question "When did Pangea start breaking up?"). One exposure should then suffice to begin the process of consolidating the memory trace
11. Myth: Review your material next day after a good night sleep. Many authors believe that sleep consolidates memories and you need to strike iron while it is hot to ensure good recall. In other words, they suggest a good review on the next day after the first exposure. Fact: Although sleep is vital for learning and review is vital for remembering, the optimal timing of the first review is usually closer to 3-7 days. This number comes from calculations that underlie spaced repetition. If we aim to maximize the speed of learning at a steady 95% recall rate, most well-formulated knowledge for a well-trained student will call for the first review in 3-7 days. Some pieces must indeed be reviewed on the next day. Some can wait as long as a month. SuperMemo and other computer programs based on spaced repetition will optimize the length of the first interval before the first review
12. Myth: Learn new things before sleep. Because of the research showing the importance of sleep in learning, there is a widespread myth claiming that the best time for learning is right before sleep. This is supposed to ensure that newly learned knowledge gets quickly consolidated overnight. Fact: The opposite is true. The best time for learning in a healthy individual is early morning. Many students suffer from DSPS (see: Good sleep for good learning) and simply cannot learn in the morning. They are too drowsy. Their mind seems most clear in the quiet of the late night. They may indeed get better results by learning in the night, but they should rather try to resolve their sleep disorder (e.g. with free running sleep). Late learning may reduce memory interference, i.e. obliteration of the learned material by the new knowledge acquired during the day. However, a far more important factor is the neurohormonal state of the brain in the learning process. In a hormonal sense, the brain is best suited for learning in the morning. It shows highest alertness and the best balance between attention and creativity. The gains in knowledge structure and the speed of processing greatly outweigh all minor advantages of late-night learning
13. Myth: Long sleep is good for memory. Association of sleep and learning made many believe that the longer we sleep the healthier we are. In addition, long sleep improves memory consolidation. Fact: All we need for effective learning is well-structured sleep at the right time and of the optimum length. Many individuals sleep less than 5 hours and wake up refreshed. Many geniuses sleep little and practice catnaps. Long sleep may correlate with disease. This is why mortality studies show that those who sleep 7 hours live longer than 9-hour-sleepers. The best formula for good sleep: listen to your body. Go to sleep when you are sleepy and sleep as long as you need. When you catch a good rhythm without an alarm clock, your sleep may ultimately last less but produce far better results in learning. It is the natural healthy structure of sleep cycles that makes for good learning (esp. in non-declarative problem solving, creativity, procedural learning, etc.) 
14. Myth: Alpha-waves are best for learning. Zillions of speed-learning programs propose learning in a "relaxed state". Consequently, gazillions of dollars are  misinvested by customers seeking instant relief to their educational pains. Fact: It is true that relaxed state is vital for learning. "Relaxed" here means stress-free, distraction-free, and fatigue-free. However, a red light should blink when you hear of fast learning through inducing alpha states. Alpha waves are better known from cropping up when you are about to fall asleep. They are better correlated with lack of visual processing than with the absence of distracting stress. You do not need "alpha-wave machinery" to enter the "relaxed state". You can do far better by investing your time and money in ensuring good peaceful environment for learning, as well as skills related to time-management, conflict-resolution, and stress-management. Neurofeedback devices may play a role in hard to crack stress cases. However, good health, peaceful environment and loving family are your simple bets for the "relaxed state"
15. Myth: Memory gets worse as we age. Aging universally affects all organs. 50% of 80-year-olds show symptoms of Alzheimer's disease. Hence the overwhelming belief that memory unavoidably gets rusty at an older age. Fact: It is true we lose neurons with age. It is true that the risk of Alzheimer's increases with age. However, a well-trained memory is quite resilient and shows comparatively fewer functional signs of aging than the joints, the heart, the vascular system, etc. Moreover, training increases the scope of your knowledge, and paradoxically, your mental abilities may actually increase well into a very advanced age
16. Myth: You can boost your learning with memory pills. Countless companies try to market various drugs and supplements with claims of improved memory. Fact: There are no memory pills out there (August 2003). Many drugs and supplements indirectly help your memory by simply making you healthier. Many substances can help the learning process itself (e.g. small doses of caffeine, sugar, etc.), but these should not be central to your concerns. It is like running a marathon. There are foods and drugs that can help you run, but if you are a lousy runner, no magic pill can make finish in less than 3 hours. Do not bank on pharmiracles. A genius memory researcher Prof. Jim Tully believes that his CREB research will ultimately lead to a memory pill. However, his memory pill is not likely to specifically affect desired memories while leaving other memories to inevitable forgetting. As such, each application of the pill will likely produce a side effect of enhanced memory traces for all things learned in the affected period. Neural network researchers know the problem as stability-vs.-plasticity dilemma. Evolution solved this problem in a way that will be hard to change. Admittedly though, combination of a short-lasting memory enhancement with a sharply-focused spaced repetition (as with SuperMemo) could indeed bring further enhancement to learning
17. Myth: Learning by doing is the best. Everyone must have experienced the value of learning by doing. This form of learning often leads to memories that last for years. No wonder, some educators believe that learning by doing should monopolize educational practice. Fact: Learning by doing is very effective in terms of the quality of produced memories, but it is also very expensive in expenditure of time, material, organization, etc. The experience of a dead frog's leg coming to life upon touching a wire may stay with one for life (perhaps as murderous nightmares resulting from the guilt of killing). However, a single picture or mpeg of the same experiment can be downloaded from the net in seconds and retained for life with spaced repetition at the cost of 60-100 seconds. This is incomparably cheaper than hunting for frogs in a pond. When you learn to program your VCR, you do not try all functions listed in the manual as this could take a lifetime. You skim the highlights and practice only those clicks that are useful for you. We should practise learning by doing only then when it pays. Naturally, in the area of procedural learning (e.g. swimming, touch typing, playing instruments, etc.), learning by doing is the right way to go. That comes from the definition of procedural learning
18. Myth: It is possible to memorize Encyclopedia Britannica. Anecdotal evidence points to historical and legendary figures able of incredible feats of memory such as learning 56 languages by the age of 17, memorizing 100,000 hadiths, showing photographic memory lasting years, etc. No wonder that it leads to the conviction that it is possible to memorize Britannica word for word. It is supposed to only be the question of the right talent or the right technique. Fact: A healthy, intelligent and non-mutant mind shows a surprisingly constant learning rate. If Britannica is presented as a set of well-formulated questions and answers, it is easy to provide a rough estimate of the total time needed to memorize it. If there are 44 million words in Britannica, we will generate 6-15 million cloze deletions, these will require 50-300 million repetitions by the time of job's end (see spaced repetition theory), and that translates to 25-700 years of work assuming 6 hours of unflagging daily effort. All that assuming that the material is ready-to-memorize. Preparing appropriate questions and answers may take 2-5 times more than the mere memorization. If language fluency is set at 20,000 items (this is what you need to pass TOEFL in flying colors or comfortably read Shakespeare), the lifetime limit on learning languages around 50 might not be impossible (assuming total dawn-to-dusk dedication to the learning task). Naturally, those who claim fluency in 50 languages, are more likely to show an arsenal of closer to 2000 words per language and still impress many
19. Myth: Learning while sleeping. An untold number of learning programs promises you to save years of life by learning during sleep. Fact: It is possible to store selected memories generated during sleep by: external stimuli, dreams, hypnagogic and hypnopompic hallucinations (i.e. hallucinations experienced while falling asleep and while waking up). However, it is nearly impossible to harness this process into productive learning. The volume of knowledge that can be gained during sleep is negligible. Learning in sleep may be disruptive to sleep itself. Learning while sleeping should not be confused with the natural process of memory consolidation and optimization that occurs during sleep. This process occurs during a complete sensory cut-off, i.e. there are no known methods of influencing its course to the benefit of learning. Learning while sleeping is not only a complete waste of time. It may simply be unhealthy
20. Myth: High fluency reflects high memory strength. Our daily observations seem to indicate that if we recall things easily,  if we show high fluency, we are likely to remember things for long. Fact: Fluency is not related to memory strength! The two-component model of long-term memory shows that fluency is related to the memory variable called retrievability, while the length of the period in which we can retain memories is related to another variable called stability. These two variables are independent. This means that we cannot derive memory stability from the current fluency (retrievability). The misconception comes from the fact that in traditional learning, i.e. learning that is not based on spaced repetition, we tend to remember only memories that are relatively easy to remember. Those memories will usually show high fluency (retrievability). They will also last for long for reasons of importance, repetition, emotional attachment, etc. No wonder that we tend to believe that high fluency is correlated with memory strength. Users of SuperMemo can testify that despite excellent fluency that follows a repetition, the actual length of the interval in which we recall an item will rather depend on the history of previous repetitions, i.e. we remember better those items that have been repeated many times. See also: automaticity vs. probability of forgetting

The list of myths is by no means complete. I included only the most damaging distortions of the truth, i.e. the ones that can affect even a well-informed person. I did not include myths that are an offence to our intelligence. I did not ponder over repressed memories, subliminal learning, psychic learning, or remote viewing (unlike CIA). The list is simply too long.