You are currently browsing the daily archive for May 22nd, 2008.
Jennifer Lin emailed this to the FBCC Ladies’ Group after we were discussing memorizing Bible verses last night:
Surrender to This Algorithm
by Gary Wolf
04/21/08 | 6:00 PM
SuperMemo is based on the insight that there is an ideal moment to practice what you’ve learned. Practice too soon and you waste your time. Practice too late and you’ve forgotten the material and have to relearn it. The right time to practice is just at the moment you’re about to forget. Unfortunately, this moment is different for every person and each bit of information. Fortunately, human forgetting follows a pattern. We forget exponentially. A graph of our likelihood of getting the correct answer on a quiz sweeps quickly downward over time and then levels off. This pattern has long been known to cognitive psychology, but it has been difficult to put to practical use.
Twenty years ago, Piotr Wozniak realized that computers could easily calculate the moment of forgetting if he could discover the right algorithm. SuperMemo is the result of his research. It predicts the future state of a person’s memory and schedules information reviews at the optimal time. The effect is striking. Users can seal huge quantities of vocabulary into their brains. But for Wozniak, 46, helping people learn a foreign language fast is just the tiniest part of his goal. As we plan the days, weeks, even years of our lives, he would have us rely not merely on our traditional sources of self-knowledge — introspection, intuition, and conscious thought — but also on something new: predictions about ourselves encoded in machines. They will be able to tell us when to wake, sleep, learn, and exercise; they will cue us to remember what we’ve read, help us track whom we’ve met, and remind us of our goals. Computers, in Wozniak’s scheme, will increase our intellectual capacity and enhance our rational self-control.
The reason the inventor of SuperMemo pursues extreme anonymity, asking me to conceal his exact location and shunning even casual recognition by users of his software, is not because he’s paranoid or a misanthrope but because he wants to avoid random interruptions to a long-running experiment he’s conducting on himself. Wozniak is a kind of algorithmic man. He’s exploring what it’s like to live in strict obedience to reason. On first encounter, he appears to be one of the happiest people I’ve ever met.
In the late 1800s, a German scientist named Hermann Ebbinghaus made up lists of nonsense syllables and measured how long it took to forget and then relearn them. He discovered many lawlike regularities of mental life. He was the first to draw a learning curve. Among his original observations was an account of a strange phenomenon that would drive his successors half batty for the next century: the spacing effect.
Wozniak couldn’t help noticing that within a few months of completing a class, only a fraction of the knowledge he had so painfully acquired remained in his mind. The problem wasn’t learning the material; it was retaining it. If students nonetheless manage to become expert in a few of the things they study, it’s not because they retain the material from their lessons but because they specialize in a relatively narrow subfield where intense practice keeps their memory fresh. As Wozniak later wrote in describing the failure of his early learning system: “The process of increasing the size of my databases gradually progressed at the cost of knowledge retention.” In other words, as his list grew, so did his forgetting.
The problem of forgetting might not torment us so much if we could only convince ourselves that remembering isn’t important. Facts can be looked up. When it comes to learning, what really matters is how things fit together. We master the stories, the schemas, the frameworks, the paradigms; we rehearse the lingo; we swim in the episteme. The disadvantage is that it’s false. “The people who criticize memorization — how happy would they be to spell out every letter of every word they read?” asks Robert Bjork, chair of UCLA’s psychology department and one of the most eminent memory researchers. Once we drop the excuse that memorization is pointless, we’re left with an interesting mystery. Much of the information does remain in our memory, though we cannot recall it. They were also curious about the paradoxical tendency of older memories to become stronger with the passage of time, while more recent memories faded.
Long-term memory, the Bjorks said, can be characterized by two components, which they named retrieval strength and storage strength. Retrieval strength measures how likely you are to recall something right now, how close it is to the surface of your mind. Storage strength measures how deeply the memory is rooted. Some memories may have high storage strength but low retrieval strength. Take an old address or phone number. Try to think of it; you may feel that it’s gone. But a single reminder could be enough to restore it for months or years. Conversely, some memories have high retrieval strength but low storage strength. Perhaps you’ve recently been told the names of the children of a new acquaintance. At this moment they may be easily accessible, but they are likely to be utterly forgotten in a few days, and a single repetition a month from now won’t do much to strengthen them at all.
One of the problems is that the amount of storage strength you gain from practice is inversely correlated with the current retrieval strength. In other words, the harder you have to work to get the right answer, the more the answer is sealed in memory. The most popular learning systems sold today — for instance, foreign language software like Rosetta Stone — cheerfully defy every one of the psychologists’ warnings. With its constant feedback and easily accessible clues, Rosetta Stone brilliantly creates a sensation of progress. “Go to Amazon and look at the reviews,” says Greg Keim, Rosetta Stone’s CTO, when I ask him what evidence he has that people are really remembering what they learn. “That is as objective as you can get in terms of a user’s sense of achievement.” The sole problem here, from the psychologists’ perspective, is that the user’s sense of achievement is exactly what we should most distrust.
The day I first meet Wozniak, we go for a 7-mile walk down a windy beach. Wozniak takes an almost physical pleasure in reason. He loves to discuss things with people, to get insight into their personalities, and to give them advice — especially in English. One of his most heartfelt wishes is that the world have one language and one currency so this could all be handled more efficiently. He’s appalled that Poland is still not in the Eurozone. He’s baffled that Americans do not use the metric system. For two years he kept a diary in Esperanto.
Wozniak’s chief partner in the campaign to reprogram the world’s approach to learning through SuperMemo was Krzysztof Biedalak, who had been his classmate at the University of Technology. The two men used to run 6 miles to a nearby lake for an icy swim. Biedalak agrees with Wozniak that winter swimming is good for mental health. Biedalak also agrees with Wozniak that SuperMemo produces extreme learning. But Biedalak does not agree with Wozniak about everything. “I don’t apply his whole technique,” he says. “In my context, his technique is inapplicable.”
“Piotr would never go out to promote the product, wouldn’t talk to journalists, very rarely agreed to meet with somebody,” Biedalak says. “He was the driving force, but at some point I had to accept that you cannot communicate with him in the way you can with other people.” The problem wasn’t shyness but the same intolerance for inefficient expenditure of mental resources that led to the invention of SuperMemo in the first place. By the mid-’90s, with SuperMemo growing more and more popular, Wozniak felt that his ability to rationally control his life was slipping away. Having turned over his mental life to a computerized system, he refuses to be pushed around by random inputs and requests. Naturally, this can be annoying to people whose messages tend to sift to the bottom. “After four months,” Biedalak says sadly, “you sometimes get a reply to some sentence in an email that has been scrambled in his incremental reading process.”
The Baltic Sea is dark as an unlit mirror. Wozniak and I walk along the shore, passing the wooden snack stands that won’t be open until spring, and he tells me how he manages his life. He’s married, and his wife shares his lifestyle. They swim together in winter, and though Polish is their native language, they communicate in English, which she learned with SuperMemo. Wozniak’s days are blocked into distinct periods: a creative period, a reading and studying period, an exercise period, an eating period, a resting period, and then a second creative period. He doesn’t get up at a regular hour and is passionate against alarm clocks. If excitement over his research leads him to work into the night, he simply shifts to sleeping in the day. When he entrusts his mental life to a machine, it is not to throw off the burden of thought but to make his mind more swift. Extreme knowledge is not something for which he programs a computer but for which his computer is programming him.
Wozniak gives close attention to the qualitative estimate of fatal risks. By graphing the acquisition of knowledge in SuperMemo, he has realized that in a single lifetime one can acquire only a few million new items. This is the absolute limit on intellectual achievement defined by death. So he guards his health. He rarely gets in a car. His advice was straightforward yet strangely terrible: You must clarify your goals, gain knowledge through spaced repetition, preserve health, work steadily, minimize stress, refuse interruption, and never resist sleep when tired. This should lead to radically improved intelligence and creativity. The only cost: turning your back on every convention of social life. It is a severe prescription. And yet now, it occurs to me that the cold rationality of his approach may be only a surface feature and that, when linked to genuine rewards, even the chilliest of systems can have a certain visceral appeal. By projecting the achievement of extreme memory back along the forgetting curve, by provably linking the distant future — when we will know so much — to the few minutes we devote to studying today, Wozniak has found a way to condition his temperament along with his memory. He is making the future noticeable. He is trying not just to learn many things but to warm the process of learning itself with a draft of utopian ecstasy.
Links mentioned in comments:
