We use cookies to personalize content and to analyze our traffic. Please decide if you are willing to accept cookies from our website.

The Education Futures Blog


With the free release of Invisible Learning (Aprendizaje Invisible), I am pleased to share the original English version of the epilogue, penned by Roger Schank.

The full Spanish-language text of Invisible Learning may be downloaded directly from http://www.invisiblelearning.com/download


Epilogue: Real learning; Real memory

by Roger Schank

What do people need to learn and how can they learn it?

Every curriculum committee and every training organization has at one time or another convened a committee to answer this question. Their answers are always given in terms of telling about subjects: “more math,” “leadership,” “risk management,” “company policies.” But subject matter is far less important in learning than one might think.

Consider medicine. What should a doctor learn? Doctors take courses in anatomy and immunology and so on, and certainly we want any doctor who treats us to know about these things. But, what skill do we want him to have above all? We want a doctor to make a proper diagnosis of our problem.

Now consider a car mechanic. We want him to understand how an engine works and such. But what do we want him to know more than anything? We want a mechanic to make a proper diagnosis of our problem.

The same is true of business consultants, architects, financial planners, and most other professions. We want people who can do diagnosis. But, when do we teach diagnosis? Typically we teach it within the confines of a particular subject, way at the end, after all the theories and facts have been explained. This is exactly backwards.

What is harder to learn, proper diagnosis of an illness or the names and functions of all the body parts? Most anyone can learn body parts, but diagnosis is a seriously important skill. You would never choose a doctor based on their ability to name the body parts quickly.

But, if diagnosis is difficult to learn, that implies that one needs a lot of practice in doing it. And, if it is important to learn, that implies that one ought to be practicing it very early on in life.

Other critical skills include determining causation, making predictions, making plans, and conducting experiments.

How can we learn these skills?

People learn diagnosis by doing diagnosis. This means that learning occurs when people have to do diagnosis. They might have to do diagnosis in order to figure out why they are losing a video game or why they always eat too much. While diagnosis is, unfortunately, not a subject in school, it is a process that everyone practices. They practice it without help most of the time and unless they have a parent who can help they may well be lost and might not get better at it.

Consider experimentation. We think of this as being something scientists do, when in fact, two year olds do it constantly. They try out experiments about what is good to put in their mouths, what annoying behaviors they can get away with, and what happens when they smash a favorite toy.

When we assess someone’s intelligence we can forgive lack of subject matter knowledge much more easily than we can forgive lack of diagnostic ability. Here is a Sarah Palin supporter responding to a question about Palin’s foreign policy:

I don’t know much about her foreign policy but the state that she did govern was right across the street from Russia. You know so I’m not saying that she ever had to deal with Russia but I’m sure she had boundaries issues she had to deal with. We have boundary issues right now with Mexico now.

Clearly this man has no ability to make an effective diagnosis. He does not understand causation either. In short, he seems stupid not because he doesn’t know about Palin’s foreign policy, but because he has diagnosed “illegal immigration” as something one would certainly be an expert on if one had governed Alaska. The critical issue in learning is learning to think more clearly.

How can technology play a role in teaching diagnosis and in teaching thinking in general? Or, to put this another way, why is it that courses rarely work the way I am suggesting (diagnostic issue first, facts and theories later)?

When you teach a course in a classroom, it is not so easy to start with a diagnostic problem. Such problems require real thought, hard work, recovery from errant hypotheses, and mentoring focused on creating new ways of looking at a problem. In other words, teaching diagnosis is facilitated by one-on-one interactions between teacher and student. We can do this easily online (or at home with our children), but it is very hard to do in the classroom. One value of technology is to enable one-on-one teaching in a world where people can no longer afford personal tutors. And, of course, we can model physical situations virtually. These situations can be richly elaborated and allow for exploration and discovery. It is much better to diagnose a virtual patient (or a business or an electrical problem) than a real one.

To understand why learning needs to happen this way it is important to realize that all human beings have a dynamic memory, one that changes in response to new experiences. The popular conception of memory is a static one, more like a library in which what one puts in stays there unchanged until it is needed again. This popular conception of memory causes schools to try to pour in information and test to see if it is still there. And, it causes parents to worry if their child doesn’t seem very good at either acquiring information or retaining it.

Human beings do not have static memories. They can change their internal classification systems when their conception of something changes, or when their needs for retrieval changes. For the most part, such changes are not consciously made.

Despite constant changes in organization, people continue to be able to call up relevant memories without consciously considering where they have stored them. A dynamic memory is one that can change its own organization when new experiences demand it. A dynamic memory is by nature a learning system.

People use the knowledge structures created by this memory, the ways of organizing information into a coherent whole, in order to process what goes on around them. What knowledge structures does a child have and how do they acquire them? They have knowledge structures about their own worlds: what the people they know are likely to do, how the stores and parks around them function, and they ask questions endlessly to find out more.

Understanding how knowledge structures are acquired helps us understand what kinds of entities they are. A script is a simple knowledge structure that organizes knowledge we all know about event sequences in situations like restaurants, air travel, hotel check in, and so on. We know what to expect and interpret events in light of our expectations.

If something odd happens to us in a restaurant, how do we recall it later? We would recall it if we entered the same restaurant later on, or if we had the same waitress at a different restaurant, or if we ate with the same dinner companions (assuming we ate with them rarely.), or if the food was extraordinary, or if we got sick. An incident in memory is indexed in many ways. Those indices are about actions, results of actions, and lessons learned from actions.

People can also abstract up a level to organize information around plans and goals. To put this another way, if the waitress dumped spaghetti on the head of someone who offended her, you should get reminded of that event if you witness the SAME KIND OF EVENT another time. The question is, what does it mean to be the same kind of event? Whatever this means, it would mean different things to different people. One person might see it as an instance of “female rage” and another as an instance of “justifiable retribution.” Another might see it as a kind of art.

The key issue is to learn from it. Any learning that occurs involves placing the new memory in a location in memory whereby it adds to and expands upon what is already in that place. So, it might tell us more about that waitress, or waitresses in general, or women in general, or about that particular restaurant, and so on, depending upon what we previously believed to be true of all those things. New events modify existing beliefs by adding experiences to what we already know or by contradicting what we already know and forcing us to new conclusions. Either way, learning is more than simply adding new information.

A child’s mind is acquiring and abandoning scripts. A child is wired to create patterns by expecting something to happen after something else because that is the way it happened last time. A child is set up to make generalizations, have them fail because his expectations were not met, and then create a new generalization.

And then, there is school. No actual experiences, except those about school itself, are had. So a child easily learns how one is expected to behave in school and how school functions, but he may not want to behave that way or function in that way. Reading, writing, and arithmetic, actual skills, can be taught because they are the new experiences the child is wired to seek. But other subjects, ones that are not themselves experiences, i.e., scripts that can be practiced, are much harder for a child to learn because they are not offered up by schooling, typically.

As a child gets older, he begins to understand implicitly that it is his goals, and his plans to achieve those goals, that drive his learning. While the child seeks to make his script base larger and to clarify the expectation failures he has had and to find new stories to tell or hear stories that will help him make sense of his world, the school takes a passive, librarian’s view of knowledge as something you can just deposit.

In school, all children are seen as the same, and the goal is teach them all the same stuff. But, a child processes new information in terms of the memory structures he already has. Since those are different than those of the child sitting next to him, he literally will not hear the same thing that a teacher is saying, in the same way.

The people who are in charge of schools completely misunderstand the inherently experiential nature of learning.

Students who are wired to learn from experience will have a hard time learning from static information that does not clearly relate to goals they have. Curiously, little children learn very well until they meet up with school and its arbitrary standards. They have experiences and they learn from them. The more varied their experiences, the more they can be said to know. The more they have interesting people to discuss their experiences with, the more excited and comprehending they become about their own knowledge.

Not only does school ignore what we know about how human memory and learning work, it is also concerned with teaching subjects that have nothing to do with everyday life. So students learn the wrong stuff in the wrong way.

young men grow up such blockheads in the schools, because they neither see nor hear one single thing connected with the usual circumstances of everyday life

That was written by Gaius Petronius in the Satyricon although it is just as true today.

We need to re-think our very conception of learning. What we have now simply doesn’t work. It’s time for a new model.


Dr. Roger Schank is the CEO of Socratic Arts and Managing Director of Engines for Education (a non-profit). He was Chief Education Officer of Carnegie Mellon West and Distinguished Career Professor in the School of Computer Science at Carnegie Mellon University from 2001-2004. He founded he renowned Institute for the Learning Sciences at Northwestern University in 1989 where he is John P. Evans Professor Emeritus in Computer Science, Education and Psychology. From 1974-1989, he was Professor of computer science and psychology at Yale University, Chairman of the Computer Science department, and Director of the Yale Artificial Intelligence Project. He currently works with La Salle University in Barcelona on developing new online degree programs.