What matters is not how much we remember, but how we remember. As I see it, intelligence is closely related to creativity, to noticing something new, to making unexpected connections between disparate facts. Isaac Newton’s genius consisted of realizing that what makes an apple fall from a tree is the same force that keeps the moon in its orbit around the earth: gravity. Centuries later, in his general theory of relativity, Albert Einstein uncovered another astounding relationship when he noted that the effect of the force of gravity is indistinguishable from the acceleration of a spaceship in outer space or the tug we feel in an elevator when it starts to move.
Attempting to memorize facts by rote does nothing more than distract our attention from what really matters, the deeper understanding required to establish meaning and notice connections—that which constitutes the basis of intelligence. The method of loci does nothing to help us understand the things we memorize; it is just a formula for memorization that, in fact, competes against comprehension. As we saw in the previous chapter, Shereshevskii was able to memorize a list effortlessly using the method of loci, but was incapable of grasping its content enough to pick out the liquids from the list or, on another occasion, to realize that he had memorized a sequence of consecutive numbers. Using the method of loci to store these lists left Shereshevskii no room to make any of the categorizations that we perform unconsciously (person, animal, liquid, etc.) or to find basic patterns in a list of numbers. To be creative and intelligent, we must go beyond merely remembering and undertake completely different processes: we must assimilate concepts and derive meaning. Focusing on memorization techniques limits our ability to understand, classify, contextualize, and associate. Like memorization, these processes also help to secure memories, but in a more useful and elaborate way; these are precisely the processes that should be developed and encouraged by the educational system.
We have seen the importance given to memory in antiquity, especially as a tool for oratory. We have also seen that, today, its importance is much more relative. Curiously, however, memorization is the ability most trained and rewarded by our current educational system, as though we are bent on equipping students to be senators in ancient Rome. We move abruptly from subject to subject and are constantly quizzed on the ability to repeat facts that we will inevitably forget a few days later. We are told dates, places, and the names of a multitude of founding fathers. We repeat this information until it is committed, however temporarily, to memory, regurgitate it in an exam, and move on to learning the names and locations of the main rivers and mountains of South America or the names and definitions of the different types of syllogisms. Not only are we given an overwhelming amount of information to memorize, we are evaluated precisely on our ability to do so. Test-prep courses and tutoring centers that promise to improve performance through the use of memorization techniques only exacerbate the problem. We learn to memorize, not to reason. Attempting to remember so much is akin to rowing upstream against the inevitability of forgetting; it steals resources from our ability to think. This is far from what I consider to be “learning.” We should evaluate—and value—the ability to process data, not merely to repeat it.
Richard Andersen, a professor at the California Institute of Technology (Caltech) and one of my mentors in neuroscience, said once that a lecture should convey at most one or two general messages. Richard does not study memory,14 but, as a gifted speaker, he has come to understand that attempting to communicate more than a couple of messages does little but confuse the audience and reduce the likelihood that anything will be remembered. Of course, during a one-hour talk, we must do more than simply recite our one or two messages (which would take seconds). The content of the talk must aim for its goal by developing those one or two ideas richly and memorably. The secret of good public speaking, in my opinion, lies in knowing well what these ideas are and communicating them in a way that ensures the audience will recall them a week, a month, or even a few years later. One may adorn the talk with vivid details—maybe some of these, if they strike a particular audience member as noteworthy, will be remembered in the future—but these details should reinforce the main ideas, not compete against them.
These are, of course, my personal opinions, not absolute truths. I am certainly not the first person to have such views; after all, it has become somewhat of a cliché to say that schools should teach pupils to think, not to memorize. Perhaps the greatest contribution neuroscience can make to this debate is the discovery that the human brain has a very limited ability to process and retain information. A teacher does his best to make it through the year’s curriculum because he wants his students to learn the subjects in full. What he may not know is that, no matter how hard they try, his students will not be able to remember much of what they learned a while later. If he teaches many topics, one after another, he will cover a wide program of study, but almost nothing will stay in the students’ memories for long. It may be much more effective to select a few subjects and flesh them out repeatedly, instead of jumping from topic to topic. Perhaps, as in a presentation, he can add details and related content, but he should always keep the core concepts he has decided to focus upon front and center, and come back to them time and time again, since these are what his students will remember.
As we saw in Chapter 4, Ebbinghaus showed in the late nineteenth century that repetition helps consolidate memory. However, the sort of repetition I refer to, going over and over the same topics, is very different from repetition as an aid to memory. In fact, I propose the exact opposite of requiring students to repeat the same facts from memory again and again. Instead, I argue that the same topics should be covered many times, but with different nuances, in different contexts, through different associations. It is precisely these contexts and associations that consolidate memories in a much sturdier and deeper way than that afforded by rote memorization. Recall my earlier mentions of Ravenna’s Phoenix: I did not need to check the internet or go through my books to know for sure that it had been published in 1491. Neither did I use the method of loci or some other mnemonic aid to remember the date. I had placed the date in context: it was the year before Columbus discovered the Americas. This association makes the date nearly impossible for me to forget; what’s more, the connection itself is much more useful than some mnemonic rule involving the four digits. Any future connection I make to Columbus’s voyage will further consolidate the date of his discovery of the Americas and help make it one of the pillars of my memory, around which I will build a web of associations.
As William James—and, long before, Aristotle in his De memoria et reminiscentia—argued, associations are a powerful mechanism for the consolidation of memories. If I generate associations and contexts, I may not remember a specific fact, but I can start by remembering some other related fact and arrive by association at the one I am looking for. James wrote:
If we have not the idea itself, we have certain ideas connected with it. We run over those ideas, one after another, in hopes that some one of them will suggest the idea we are in quest of; and if any one of them does, it is always one so connected with it as to call it up in the way of association . . . The “secret of a good memory” is thus the secret of forming diverse and multiple associations with every fact we care to retain. But this forming of associations with a fact, what is it but thinking about the fact as much as possible?15
Notably, the problems I have outlined with the educational system apply mostly to the humanities and soft sciences. The teaching processes and evaluation methods practiced in the hard sciences are more adequate, since there would be little point in testing whether a student remembers by heart a given formula. On the contrary: knowledge of the hard sciences is usually tested with problems, and between classwork and exams, students are required to solve many of these, using the same formulas in different situations. As they apply the same formula to different problems, the students go beyond repeating information and begin to understand its meaning; they learn that what matters
is not performing a computation well or remembering the value of a constant, but, rather, knowing when and how to use the formula—setting up a problem based on its statement, understanding what is being asked and how to arrive at a result. This is the hardest task for a child: understanding that “4 × 8” is the same problem as asking how much money is earned monthly by four siblings, each of whom makes eight dollars a month. To solve problems like the latter, the child has to carry out the same processes of abstraction and meaning extraction that we’ve seen are vital to learning and memory.16
In this chapter I have discussed several topics, but the lesson is the same: the brain has a limited capacity, and we should focus its resources on processes of comprehension and thought, not on memorization. New technologies are always a mixed blessing. The principle that lifts a commercial airliner is the same that allows a bomber to fly; the same atomic reaction that keeps a city lit at night can also destroy it in seconds. The internet and our twenty-first-century gadgets are no different. On the one hand, these technologies let us delegate memories and menial functions in order to focus on more important thoughts. However, they also impose on us a frenzied bombardment (but not assimilation) of information that is detrimental to our capacity for thought. These technologies guzzle our free time, those periods of boredom or seeming unproductivity that may well be the genesis of our most creative moments. Yet it is also true that a wiser use of the internet is well within our reach: it is up to us when we turn our smartphones on and off; it is we who decide with our scrolling fingers how fast to read or browse online. These technologies may supplement our understanding but do not replace it; we must learn to be their masters, not their slaves. There is a balance to be struck in the amount of information we receive: too much of it saturates our brain, leaving no room for thinking, while too little makes for a poor platform on which to develop our thoughts. We must attain this same balance in education, where focus on solidifying a relatively small set of ideas will allow them to become pillars around which students weave a tangle of associations and contexts. We must also avoid the overwhelming and superficial treatment of one topic after another, a practice that rewards only rote memorization and prevents those sturdier pillars of true knowledge from forming.
Chapter 7
TYPES OF MEMORY
In which we present the different classifications of memory, the multi-store memory model, the case of H.M., and the difference between declarative and procedural memory
Iremember how to ride a bike, drive a car, and compute integrals; I remember the bars of Beethoven’s Fifth Symphony, my last birthday, my mother’s name, and what I would like to write about in the introduction to this chapter. Are all these memories fundamentally the same? Do they all involve the same processes and areas within the brain? As we shall see shortly, the answer is no.
The different classifications of memory can be found in any textbook on the subject:1 We have semantic memories, episodic ones, visual, auditory, short- and long-term, emotional and working memories, and so on. An exhaustive description of these memory types lies beyond the scope of this book, but I would like to give at least a general idea of the most important differences between them. In Chapter 4 we described how Ebbinghaus distinguished between short- and long-term memory. Short-term memory lasts for a few seconds and enables us to be aware of the stream of events taking place in the present that, in general, do not become part of our past experiences. Long-term memory lasts for minutes, hours, or years, and stores our experiences; it enables us to bring a past event back into the present and to be aware that we have lived it before. We also saw how repetition consolidates memories, turning short-term memories into long-term ones. Most of our short-term memories will quickly fade into oblivion, but as we will see, the most dramatic memory loss happens earlier.
In 1960, American psychologist George Sperling published the results of a series of simple but clever experiments.2 Sperling first gave participants a fleeting glimpse of an array of letters (displaying, for example, twelve letters in a three-by-four table for fifty milliseconds) and then asked them to recall as many as they could. Subjects were able to remember three or four letters. In a second test, Sperling asked the participants to recall the letters from just one of the three rows. They were told they would hear a high-, medium-, or low-pitched tone immediately after the table disappeared from view to indicate whether they should recall the top, middle, or bottom row. Since they did not know which row they would be asked about as they looked at the table, one would in principle expect the subjects to remember just one or at most two letters—a third of what they remembered before—but, surprisingly, they were again able to recall three or four. From this result, Sperling deduced that the participants at first, and only momentarily, stored in memory an image of the whole table. This led him to hypothesize the existence of a sensory memory that precedes short-term memory and enables a person to retain information for vanishingly brief time intervals. This sensory memory, this image of the table in the brain, was erased during the time it took to repeat three or four letters; this is why subjects could often remember all four letters of a particular row when asked and yet only the same number of letters when asked to recall the entire twelve-letter table.
Figure 7.1
Letter table of the kind used by Sperling to study sensory memory
Based on Sperling’s experiments, we can infer that sensory memory turns into short-term memory via attention mechanisms: once the subjects heard the cue tone that signaled a particular row, they could focus on that row and discard the rest. Sperling played the cue tones after different time intervals and found that the ability to repeat the letters in a particular row decreased significantly as the delay between viewing the table and hearing the tone increased, demonstrating that sensory memory lasts only for a split second.3 In other words, sensory memory gives us a very brief window in which to retain whatever we pay attention to, and this will go on to form our short-term memory, the stream of thoughts that constitutes our present. In turn, those things that we revisit and consolidate become engraved in long-term memory and go on to become our awareness of the past. This is the basis of what is known as the Atkinson-Shiffrin model.4
Figure 7.2
Atkinson-Shiffrin three-component model of memory storage
We thus have a first general classification of memories based on their duration: sensory, short-term, and long-term. To those types one can add nuances such as working memory, the one we use to store temporary information as needed—for example, to perform a mental calculation. (If I multiply 17 × 3 in my head, I can start by computing 7 × 3, store this result temporarily, and then compute 10 × 3 and add the two results to obtain the answer, 21 + 30 = 51.) But the most important distinction between different types of memory came from the study of a single, unique case.
Henry Molaison began suffering from epileptic seizures at age ten, after he sustained a serious blow to the head. The seizures worsened during adolescence, and in September 1953, in a last-ditch effort to control them, neurosurgeon William Scoville surgically removed Molaison’s hippocampus—a seahorse-shaped structure often linked to the onset of epileptic seizures—and adjacent zones from each hemisphere of his brain. The surgery, which indeed stopped his seizures, also radically changed the history of neuroscience and our knowledge of memory, while unfortunately transforming Henry Molaison (known as H.M., from his initials) into the most famous patient in the history of science.
Following the surgery, H.M. appeared at first to be recovering normally, but soon a terrible deficit revealed itself: he could not recognize the hospital staff or remember daily events. H.M. had become incapable of forming new memories.5
Figure 7.3
Photograph of H.M. shortly before his surgery and a depiction of the hippocampus, an area located about an inch into each brain hemisphere, at approximately the same height as the ears
During a psychological test carried out more than a year and a half after the operation, H.M. es
timated the date to be March of 1953 (it was 1955) and stated that he was twenty-seven years old (he was in fact twenty-nine). He was unable to grasp the meaning of new words or recognize people he had met after the surgery. He was barely aware of having been operated on at all. On the other hand, his visual perception and his capacity for reasoning (as long as it required no memory use) were normal. He had no problem carrying on a conversation, which showed that his short-term memory was working adequately because, without it, we cannot form sentences, speak coherently, or understand what someone else is saying. In fact, H.M. could repeat sequences of six or seven numbers and remember for brief moments something he was told, but the only way he could prolong these memories was through constant repetition, and he would lose them the moment he turned his attention to something else.
H.M.’s case provides unquestionable evidence that the hippocampus is crucial to the formation of long-term memories. But H.M.’s contribution to our understanding of memory went far beyond this. Canadian psychologist Brenda Milner—who before each session had to introduce herself as though she were a complete stranger—had been studying H.M. for years when she decided to test his ability to learn a new skill. She asked him to draw a line along a contour between two concentric stars (but here is the tricky part) while looking only at the reflection of his hand and the drawing in a mirror. Over multiple sessions, H.M.’s performance at this task improved, surprising everyone, including H.M. himself, who each time could not recall having ever performed the task before. How was he improving with practice he could not remember?
The Forgetting Machine Page 8