I remember vividly that January day, more than 20 years ago, when I was sitting in my faculty office and in walked the chair of the Sociology Department.

"Ted," she asked, "when you were in college, did you take coursework in sociology?"

"Well, no," I said, "I was an English major."

"Well," she said, "maybe as a graduate student, you took courses in research?"

"Oh yes," I said. "All of us at Michigan had to do a research sequence."

"Well," she said, "one for two may be close enough. I just got a phone call from our colleague who teaches the required statistics course for sociology majors and she'll have to take a medical leave this spring. Could you stand in and teach that course for us?"

Through a hurried sequence of events, ten days later I stood in front of that class, supposedly ready to teach statistics. I was, at the time, nothing if not conscientious; I read all the textbooks and chose what I thought was a good one, prepared a careful syllabus, and worked hard on my lectures. During the semester, I think I can say, I was respectful toward students; I tried to get everybody involved; I kept my office hours; I got the homework back on time; we covered every chapter; my tests were tough but fair.

That summer, two or three weeks after the end of the semester, I remember the department head again coming into my office and this time saying, "Ted, you must be an excellent teacher!"

"How's that?" I asked.

"Well," she said, "the computer center just sent me a printout of all the student evaluations for the department's courses and your course had a 4.7 on a 5-point scale . . . the second-highest mark of any course in the whole department!"

And I can hear her again — it was music to my young ears! — saying, "Ted, you must be an excellent teacher."

Reflecting back on that experience, I'd have to tell you now that of the 28 students in that class, all but two passed the exams, but maybe only half of them ever really got it. If you gave them all my final exam six months later, perhaps three of them would have passed. It turns out that the next course they took in their department's sequence was one on research methods; but the person who taught that focused on qualitative research and field-based studies, so none of the parametric statistics I taught came to be practiced or applied. If you ask me now about "quantitative reasoning" or "statistical literacy" as possible outcomes, the answer is, 20 years ago I'd never heard of those things.

What these students were good at — and I played right into this — was feeding back correct answers; they had mastered the arts of short-term memory and recall. The whole class was a wonderful example of what the British call "surface learning." But very little "deep learning" — which comes with time, depth, practice, and reinforcement — seems in retrospect to have occurred.

So I return to the question, Is Ted an excellent teacher? Today, were I up for promotion, I could put together one heck of a portfolio based on that course. I'd have, first of all, letters from students saying how much they enjoyed my course. I'd have a letter from my chair, who'd vouch again, Ted is an excellent teacher. And I'd have that big 4.7, an irrefutable number.

Today I have doubts, as you sense, about whether I did right by those students. But this is my point: I don't think you can answer the question — Is Ted an excellent teacher? — without having an idea about the kind of learning that was appropriate from that course. If surface learning (and student satisfaction) will do, the answer might well be "yes"; if a different, deeper character of learning were supposed to occur, then "no."

This insight is the starting premise of my presentation. That is, in making judgments about programs or curricula, about teaching or teachers, the ideas we hold about learning itself will and should be decisive. In making choices about what to teach and how to assess it, clarity about the character of learning we intend for students must be an essential bedrock for what we do.

To take an assessment example, should we use a standardized instrument or short-item questionnaire to find out about our students' learning? Or should we put our chips behind direct faculty readings of student work?

Are the forms of "authentic assessment" that Grant Wiggins urges upon us . . . portfolios and the like . . . worth the time and effort?

My intention here, then, is to talk about what usually gets left out — learning itself — as a way of helping you think about the choices you make in enacting and assessing undergraduate education. What I propose to do might be thought of as a little tour de terrain, an all-too-brief look at a whole string of conversations about learning that are going forward today across several disciplines.

There is high danger in such an ambition; you will immediately know that I am not a member of most of these disciplines. I am a writer, editor, and sometime teacher. What I'm attempting to provide is a selective, user's-end view of intriguing research, without claiming expertise in any of the things I'm about to report.

Indeed, if I might paraphrase J.L. Austin's famous remark about philosophers for my own role here, I'd say of any oversimplification that follows, "One might be tempted to call it the occupational disease of keynoters, if it were not their occupation."

The Teacher's Wisdom
For most of our professional lives, we've had two ways of knowing about learning.

The first is through a wisdom literature, in which famous teachers tell us how to instruct students. This literature is as old as written history itself. To take three examples, a traditional Chinese proverb tells us, "Teachers open the door. You enter by yourself." Aristotle echoed this when he said, "What we have to learn to do, we learn by doing." Galileo admonished that "You cannot teach a man anything; you can only help him find it within himself."

Reading admonitions like these, you wonder whether 20th-century social science has busied itself proving what's already known.

There's a related but more rigorous literature, of course, in philosophy, about knowledge and knowing, to which I'll return. Here let me add a particular insight from ancient Greece, attributed to Aristotle, namely that there is a difference between knowledge, understanding, and wisdom. If knowledge implies basic comprehension, understanding means comprehending well enough to be able to do the thing, and wisdom reflects dispositions to act wisely.

Two years ago at this AAHE Conference on Assessment & Quality, the systems theorist Russell Ackoff spread knowledge, understanding, and wisdom across a spectrum, and to the left of them data and information. His point was that ever so much of what transpires in undergraduate education today has to do with the left and middle portions of that spectrum — with data, information, and knowledge — and that the goal of reform ought to be to move the locus of attention further to the right, toward greater emphasis on understanding and wisdom . . . a point I take as wise.

Wisdom literatures have brought us important insight over the years. Who thought more deeply about teaching and learning than Alfred North Whitehead? I reread his short book The Aims of Education, published in 1929, every two or three years. I think also of the wonderful books on teaching from Gilbert Highet and Kenneth Eble. And, good as any of these, Parker Palmer's The Courage to Teach, due later this year.

Psychological Studies
In the present century, psychologists have developed a huge literature on learning. Over these years, psychology has been dominated by behaviorist models, then developmental ones — think of Thorndike, Skinner, and Piaget; more lately it has been cognitive and constructivist (led often by Jerome Bruner). Sadly, perhaps, psychologists and their social science kinfolk have been dismissive of thinkers working from wisdom or humanistic traditions; but they, in turn, are marginalized by humanists (as reductionist) and by "hard" scientists (as "soft"). And when it comes to the practice side of psychological research — educational research — then, I'm afraid, we're into true dismissiveness.

All of which is unfortunate in that it can brush aside important insight. There are indeed good reasons not to get carried away with the corpus of ed-psych research, not least because it has too often been theory-free, consumed by measurement, and in love with the trivial. A more important thing to note may be this: Often what it has to say about teaching is based on quite thin ideas about learning itself . . . witness all the studies that took student performance on final exams as the outcome criterion.

Having said this much, now let me say that cognitive psychologists — in psych departments and ed schools alike — have achieved understandings over the decades that meet high tests of rigor and generalizability, and that all who teach must heed. Who among us can doubt the importance to learning of feedback and reflection? Of intrinsic motivation? Or doubt the difficulty of teaching for transfer?

One of the most important findings from cognitive psychology surely has been the whole matter of prior beliefs and mental models; let me focus here on that one finding. The insight is simple enough: It begins with the innate need of humans to make meaning out of their experience of the world. So we develop, at quite early ages — as five-year-olds, for example — basic sets of ideas about how the world works, what's dangerous, who's friendly, about right and wrong, what to like and how to behave, and so on. The scary part is that these childhood versions of reality tend to get pretty hard-wired into the brain and prove quite resistant to change: Once we think we've figured out some corner of the world, we tend to see what we want to see and hear what we want to hear, bending subsequent experience into confirmation. I say "scary" because the existence of prior beliefs can be a major impediment to subsequent learning:

The beliefs, after all, may be objectively wrong, or bigoted, or dysfunctional, and block fair and open encounter with the new or different. Very significantly, prior beliefs turn out to be especially impervious to classroom-based instruction, and especially to teaching as telling.

There's a chilling example of this in a video made at Harvard several years ago, called A Private Universe. The heart of it consists of interviews of Harvard graduates on the day of graduation. Picture the Harvard Yard, gowns and bands, proud parents with cameras, and a wandering interviewer asking the new graduates a seemingly simple question: Why is it warmer in summer than in winter?

Our graduates are nothing if not confident. "Well," they say, "let's see, in summertime the Sun comes a lot closer to the Earth and, well, you know, the closer it gets the more heat you're going to have . . . right?"

Now, mind you, these are students who've taken units on the solar system back in junior high school, who've had three years of high school science, who took physics and/or astronomy at Harvard, and who probably got an "A" in them every time out. Yet they persist in a pre-Copernican view of how the world works. The point of the video is to show how mental models, once formed, are incredibly resistant to change, especially from conventional instruction.

A larger-scale example of this appears this fall in an American Journal of Physics article by Indiana University physicist Richard Hake. It looks at the impacts of instruction on student beliefs across some 62 first-level physics courses, 6,549 students in all, from rural high schools through Ivy universities. At the start and finish of each of these courses, students were administered an instrument called the Force Concept Inventory, a well-developed diagnostic instrument that assesses student orientation toward mechanics . . . it basically looks to see whether people's view of mechanics — of how the physical world works — is pre-Newtonian or post-Newtonian.

Many of us, like the students in the 62 courses, tend to carry in our heads naive, commonsensical ideas about mechanics that are pre-Newtonian. Wrong as they may be, these views turn out to be ever so resistant to change by conventional instruction. What the study shows is that in traditional physics courses — lecture-based, recipe labs, right-answer quizzes and tests — the impacts on deeper beliefs about the physical world are small . . . about .22 on the instrument's measurement scale. Strikingly, the small-impacts finding held whether the instructor was an experienced "teacher of the year," a brand-new instructor, or anything in-between. Gravity, light, motion, feather vs. pellet in a vacuum, whatever these students believed beforehand was pretty much what they believed at the end . . . notwithstanding all the right answers they produced for exams.

Hake's study is compelling because his set of 62 courses included 41 that were taught in quite a different way, emphasizing more active forms of learning ("heads-on"), problem-based labs ("hands-on"), and immediate feedback through discussion with peers and/or instructors. On the same assessment scale, the gain among students was on the order of .52 — quite a striking difference, though still a less-than-full victory in terms of changing mental models, which (again) remain ever-so-tough to alter. The good news here, of course, is that smart instruction does work . . . the unhappy thought is that such approaches are hardly yet the norm.

I want to conclude these remarks on wisdom literatures and cognitive psychology by pointing out that both have historically been focused on the teacher, not on learning itself. Even in classic experimental designs — where you set up a treatment, the teacher teaches a certain way (as in Hake's courses), then you observe student performance as the outcome — you're basically treating the mind as a black box. At best, the method allows for inference about what's happening in student brains . . . potentially valuable, as when we infer the operation of mental models, but an inference all the same.

The Neurosciences
What's new this decade is the emergence of a robust, exploding set of literatures in the neurosciences. I use the plural here because brain research now attracts the attention of scientists from at least half a dozen specialties: neurophysiologists, molecular biologists, neuroanatomists, certain chemists and medical researchers, and so on. The big breakthrough, especially this decade, has been the availability of PET scan and functional MRI devices that allow scientists to observe mental activity directly, to take a "picture" of the brain at work. So exciting are the possibilities here that there's been an outpouring of federal funding in support of neuroscientific inquiry. It is the "Decade of the Brain."

With this activity, a veritable flood of discoveries has come forward on the functioning of the brain. These have excited hopes among educators that soon, at last, we'll learn what really happens inside all those student heads and have a scientific basis for teaching.

But, alas, not yet . . . and maybe not soon. When you get neuroscientists together in a room with educators, one of the first things you learn is that there have been a tremendous number of findings, yes, but the meaning of many of them remains in dispute. Another circumstance is that for all the findings, there's precious little theory to connect or interpret them . . . and there's nothing so useful as a good theory, Kurt Lewin taught us. No scientist has yet come up with a coherent set of ideas about how the brain works that would be persuasive and usable for those of us who teach.

For the past 18 months I've been a participant in a series of Wingspread conversations sponsored by John Abbott's 21st Century Learning Initiative, which has brought together scientists with educators to try to make sense out of these new literatures. One of the things that I've observed at these meetings, and that I'm glad for, is that neuroscientists are reluctant to generalize from the findings they have so far, to tell us as teachers what we should be doing.

One reason for this is that much of the brain research that's gone on has been done as an aspect of larger projects on the "high-dollar diseases" . . . studies of brain functioning among Alzheimer's patients, for example, or alcoholics; these aren't studies of college sophomores. And, thankfully, good scientists are reluctant to make prescriptive leaps from disease studies to college classrooms.

An interesting thing to me is that the more you get to know the newer brain literatures, how few surprises there are. So many of the findings seem to confirm what we've already known, or at least theorized.

To take one tiny example, if I look out the window and see a tree, the commonsense idea about what's happening is that a picture of the tree comes in through the retina and an image of it forms straightaway on some screen inside our head, just like photography. Well, of course, it doesn't quite work that way . . . there is no "screen" and, more importantly, the mind's image of that tree is far from a simple reproduction of an external reality: 80 percent of what winds up in the brain's image comes from information, ideas, and feelings that are already in the brain, just 20 percent from outside. The learning here is that when we look at a tree, or another person, or hear an idea, the sense we form of it is highly colored by a whole range of prior experiences and emotional dispositions. What we have from the brain researchers, then, seems just to confirm what we knew before about the power of the mental models we carry around in our head.

The University of Oregon's Robert Sylwester argues that we shouldn't be so surprised when neuroscientific findings parallel what we've found as teachers or educational researchers. If I have 28 students in my statistics class, I have a semester-long opportunity to observe 28 brains in operation . . . the inferences I'd draw just from watching what works with them reflect a form of "brain research." It may not be science and it does have its limits, but there's a "wisdom of practice" teachers develop that warrants respect.

None of this is to say, however, that there's nothing new coming out of the neurosciences. I'll present a few summary findings next. Here I want to note that brain science provides us with new ways — and vocabularies — for talking about learning. As educators, for example, we've long spoken about how student development in college is a function of the intellectual and affective . . . but these "domains" of an earlier psychology are not the way neuroscientists describe things. They pay little heed, too, to the nice distinctions we educators want to make between younger, college-age, and adult learners, or to our preoccupation with "learning styles"; I've not found one of them who thinks of "intelligence" as a unitary, fixed characteristic of individuals, or who thinks of the brain as an "empty vessel" or computer-like machine. To the neuroscientist, learning is a whole-person/whole-brain activity that confounds received categorizations.

Cognitive Science
Right alongside the neurosciences, new, hybrid forms of cognitive science have emerged. Lacking a tidy definition, let me proceed with an example, albeit one that conflates a string of actual experiments.

Imagine an experiment in which rats are being raised in a series of five boxes. In the first box, you have a single rat, raised the usual (sterile) way. In box two, you have a rat raised the same way, except that it is given toys to play with. In box three, same idea, except that the rat's toys are changed every week. Box four, same idea, changed toys, but there are several rats growing up together. In box five, you have several rats, rich toys, but each rat is removed from the cage every day and lovingly stroked for 15 minutes. At the end of a time period, all these rats are given learning tasks to accomplish: pushing levers for food, finding their way through mazes, and so on. The finding, when you look at their respective abilities to learn these tasks, is a learning curve that goes up steadily from the first box through the fifth . . . a 25 percent gain in "rat intelligence," if you will, attributable to differences of upbringing. The new cognitive scientists buttress these observations by measurements of brain weight and cortical development in the different rats, and with counts of cells and synapses. What we have in today's cognitive sciences, then, is a new blend of psychology and biology.

At this point you might be thinking, well, we're still talking rats, not students; but it would be hard to get your university's human-subjects committee to approve an experiment that requisitioned sophomore brains for counting and weighing. On the other hand, rats and humans have about 95 percent of their genetic material in common. The real news in these experiments is not just the importance of rich, social environments and of nurture in upbringing but the brain's plasticity, its ability to realize new capacities in response to experience. Again, none of the neuroscientists I speak with thinks of "intelligence" as an innate capacity fixed at birth. Indeed, the best news (at least for adults my age!) is the evidence coming forth of the brain's plasticity across the lifespan, of human abilities ever to learn, to "effloresce" in creativity in the right conditions of challenge and safety. Early experiences and genetic inheritance are very important; yet all kinds of people are capable of incredible feats of learning through decades of their life. Just how "new" an understanding we should think this is I'm not so sure; but it is a valuable counter to our academic folk wisdom that wants to categorize people early and keep them there.

In the Wingspread meetings I mentioned earlier, there have been a few (brave) souls willing to help construct summary lists of learnings from the neuro- and cognitive sciences; let me share two such here in abridged form.

The first is from Dee Dickinson, head of Seattle's New Horizons for Learning project, and is based on the work of Marian Diamond of UC-Berkeley. Four of Dickinson's items in summary form are:

• the brain is remarkably plastic across the lifespan;
• powerful learning is prompted when all five senses are engaged;
• adequate time is needed for each phase of information processing (input/assimilation/output); and
• emotional well-being is essential to intellectual functioning, indeed to survival.

Little on this list comes as a surprise. Experiments with the teaching of language, for example, have shown how quickly and lastingly a new tongue is learned when students can hear, live, speak, act, and sing it. Good teachers have always known that speed in producing answers isn't a good indicator of inherent capacity in students. And readers of Daniel Goleman's recent book, Emotional Intelligence, know well the cortex's ties to a powerful limbic system and the social-emotional origins of thinking.

Here is a second list, this one by Geoffrey Caine, an Australian living in San Diego . . . he and his wife, Renate Caine, have made a specialty of translating these literatures for K-12 educators:

• body, mind, and brain exist in dynamic unity;
• our brain is a social brain;
• the search for meaning is innate;
• the brain establishes meaning through patterning;
• emotions are crucial to patterning;
• learning involves conscious and unconscious processes;
• complex learning is enhanced by challenge, inhibited by threat; and
• every brain is uniquely organized, with resulting differences of talent and preference.

One of the things scientists have established pretty clearly is stated next to last on this list: When humans confront a situation they perceive as threatening, their brain "downshifts" . . . higher-order cortical functioning is supplanted by the more elemental limbic . . . the emotions come to rule. The point to ponder — the Caines make it — is the high reliance in American classrooms on sticks and carrots, on competition and scarce rewards, an ethos that can engender a mix of student attitudes somewhere between grudging compliance and sullen disengagement.

High challenge, yes, they say; high anxiety, no.

Evolutionary Studies
My next disciplinary excursion is into the companion fields of evolutionary biology and evolutionary psychology, the latter often associated with the more recent work of Jerome Bruner. There was a wonderful review of Bruner's latest book (questions and all) by Clifford Geertz in the New York Review earlier this spring. Also this spring, Stephen Jay Gould, as you may have seen in the New Yorker, attacked evolutionary biology as an immature, overreaching form of science, as sociobiology in new clothes.

So these fields are not without controversy. But I take the "finding" that follows as at least provocative. It is that natural dispositions and ways of knowing can be identified in the human species, approaches to learning that proved evolutionarily successful over the ages and are all but in our gene pool. In effect, there's a "natural" way of learning for humans. What is it? you say, surely with interest. The answer: apprenticeship.

How the evolutionists get there is by looking back at how humans learned over the course of the 1,000 or so generations of knowable (recorded) history. All prior societies, for example, have had the very problem we confront today, that of how to bring young people up into valued adult roles. The approach of a whole range of societies, across a tremendous number of generations, was apprenticeship. The way people became midwives, stonecutters, artists, shamans, masons, hunters, a hundred other occupations, was through socially organized apprenticeships. That was true right up until the 1840s in England and until the Civil War in the United States, when universal schooling became the new way of preparing young people for adult roles.

It's only in the last six generations or so in the industrial democracies that we've had formal mechanisms of schooling. But for the figurative 994 generations that came before, young people learned what they learned through apprenticeship.

Another way to confirm this observation is through anthropological studies that look cross-culturally at how societies today that do not have universal schooling prepare their young; again the answer is through forms of apprenticeship. Studies have been done, for example, of how girls or young women become midwives in rural Yucatan or Zimbabwe; they learn that role by attaching themselves to experienced practitioners.

As Jean Lave outlined in her 1991 book Situated Learning, there are many forms of apprenticeship across time, occupations, and cultures, yet commonalities, too. These common elements came to be fleshed out in a flurry of literature that appeared in the early 1990s. Apprenticeship has typically been a cohort activity. That is, there were often two or three masters and a whole set of apprentices, rather than simply a one-to-one arrangement. The master was both taskmaster and mentor. Among the masters and apprentices there was always rich conversation about what it is they were learning . . . the important knowledge was tacit, seldom written down, and had to be learned by doing and talking. Very importantly, too, care was taken that the young person always understood the context, the real-life meaning of each lesson or step . . . the classic example is from stonecutting, where the apprentice knew the stone had to be absolutely square to fit just so in the wall of the cathedral.

Finally, there was a notion in apprenticeship that today's scholars call "scaffolding." The idea here is that in the earliest years of apprenticeship, the tasks were highly structured and supervised. But gradually that oversight and support were removed, so that by the time the apprentice was an adult — which in most societies meant age 16 — he or she would be able to do it alone and be free-standing in the trade or craft. The young adult would be, in today's expression, an independent learner.

Interest in apprenticeship has been keen in the K-12 reform conversation in this country. One of the things people notice, when they look at that model, is how utterly teacher-dependent American education has become. Even at the college level, you and I might notice, we have an almost entirely teacher-driven system. We preach the goal of preparing independent learners, but you can go to any college bookstore and watch yourself as second-semester seniors file through to buy their assigned texts and notepads for the prepackaged courses they'll take.

Situated Learning
Our next set of literatures — closely related in their findings to the evolutionary studies we've just discussed — derive from workplace studies. These have been essentially ethnographic inquiries into questions such as, How do new employees enter a work setting and learn to contribute? What does it take to become a plumber . . . or a physicist? To make the transition from novice to expert? How is human cognition manifest in everyday worklife?

A fresh set of answers to these questions came in a famous 1989 journal article by John Seeley Brown, Allan Collins, and Paul Duguid, colleagues at the Institute for Research on Learning, in Menlo Park, California. In it they argued that the most important knowledge to performance is tacit, that such knowledge resides within and helps define the relevant "community of practice," and that newcomers become contributing practitioners through rites of entry they term "legitimate peripheral participation."

What this last piece of argot means is that you "gotta be there" to learn the most important parts of any job. You don't become a football player, for example, by memorizing the rules and playbook . . . the important knowledge is tacit, it resides in practice; you learn the sport by getting out there on the field with experienced players and doing the thing, over and again. The same, they claim, is true for any craft or profession. You're not suddenly a physicist because you've memorized the formulas in a textbook; the community of practice called physics consists of expert practitioners who have their own, internal rules and store of tacit knowledge, an unwritten, sixth sense that tells them what constitutes an interesting question, a good experiment, or a viable theory. Book knowledge has a role, but it's only part of the story: The most important learning is always "situated in practice."

The "learning by doing" aspects of this, you'll sense, hearken back to John Dewey, and indeed the "situated learning" insight has epistemological dimensions. Schools and colleges proceed on the belief that some mix of facts and conceptual representation constitutes best preparation for life's work; indeed, the "merely practical" (and the community itself) are held at arm's length, if not in disdain, by academics. To Brown, Collins, and Duguid, however, important knowledge cannot be abstracted from the situations in which it is learned and used; knowledge is ever a part of a particular activity, context, and culture. In effect, they propose an epistemology of knowledge that puts activity and perception before conceptual representation — not the other way around, as it is in classrooms.

One source for this argument comes from findings that student newcomers to the workplace want to solve problems by reasoning from laws . . . they hearken to the text. Experienced workers, on the other hand, reason from stories — there's a whole set of "narratives" at play in any workplace that allow people to pass on information, share discoveries, and know how things really get done. Experts also draw on stories but they reason with causal models, which they are always testing and refining. Students are trained to produce right answers with fixed meanings; experienced workers and experts produce negotiated meaning and socially constructed understanding.

To enact their vision of how important learning occurs, Brown, Collins, and Duguid propose "cognitive apprenticeships" that will "enculturate students into authentic practices through activity and social interaction" and that support "learning in a domain by enabling students to acquire, develop, and use cognitive tools in authentic domain activity."

The authors think the key to more powerful learning lies in building connections between learners and to the workplace.

There is a profound critique of schooling and academic life, of course, in the situated learning thesis. Not all of it, though, is at odds with practices in higher education. The upper reaches of graduate education (at its best) constitute a form of cognitive apprenticeship:

Doctoral students undertake "legitimate peripheral participation" for entry into a community of practice.

Studies of Students
As I said earlier, one way of looking at the American literatures on "teaching and learning" is that they are mostly about teaching, and teachers. In Europe, on the other hand, perhaps because researchers have been loathe to intrude on the lecture halls or send questionnaires to Herr Doktor Professor, there's been a far greater emphasis on studies of students . . . ethnographic inquiries into how they study, how they approach the taking of courses and doing of assignments, how students approach exams, and so on.

In 1976, two Swedish investigators, Ference Marton and Roger Säljö, published the first of what since has become a stream of studies on student approaches to study. What they reported was that Swedish university students time and again did not get the point of what they were studying "simply because they were not looking for it." What were they doing? Cramming facts into their heads. Why were they doing that? Because they knew that's what they'd be tested for. So what was missed? The meaning of the texts they studied, and (especially) how those texts might relate to their own thinking about the world. In Marton and Säljö's observations, most students focused on short-term memorization of facts, formulas, and concepts — as their instructors seemed to ask; the traditional teaching they encountered prompted superficial forms of engagement with subject matter, a "surface" approach to learning.

Only a small percentage of students, Marton and Säljö found, either by disposition or by the demands of the course, undertook what the researchers called a "deep" approach to learning, in which students seek meaning in study, reflect on what they read and hear, and undertake to create (or re-create) their personal understanding of things. Relating this concept to issues raised earlier, we could say that deep learning is requisite to a student's movement from knowledge to understanding, and certainly to any recasting of the mental models he or she uses to view the world.

In 1982, Scotland's Noel Entwhistle replicated these studies with British university students, finding that some 90 percent of the studying going on was of the surface variety. These studies have since been replicated in Canada and Australia, with basically the same findings. To my knowledge, there's been no parallel inquiry in the United States. (One could guess the result.)

Provocative findings have emerged as this line of research was elaborated. One is that when the same instruments are deployed across the high school and college years, there is a steady falling off in deep approaches to learning as students progress up the educational ladder — a depressing finding, indeed. (As W. Edwards Deming used to preach, every school in time will drive the joy from learning.) The second finding is that the same student will take a surface approach in some courses, a deep in another . . . short-term memorization in science, for example, but simultaneous deep engagement with a particular literature course.

This latter finding has led researchers to suspect that the approach students take is more a function of instructor demand than of student disposition.

The courses obsess over coverage; the huge amount of material they cram in precludes opportunity to pursue topics in depth; students have little choice about what and how to study; and such courses have a threatening, anxiety-riddled instructional and examining environment.

In a latest twist, Entwhistle last year published findings showing that

anxiety, fear of failure, and low self-esteem are associated with surface approaches to learning. Further, students are more likely to engage in active forms of learning when they believe that their own effort, rather than external factors beyond their control, determines success. He urges university faculty to build up students' sense of control over their work and get them to exercise responsibility for their own learning.

Let me add that the "deep and surface" line of inquiry posits not two, polar-opposite traits but a continuum of behaviors. Nor should it be taken as denigration of "mere facts." When facts become the be-all and end-all of education, of course we're all in trouble. But a sound base of available knowledge, indeed of socially shared knowledge, is an indispensable platform for shared work and democratic living. In light of this, Entwhistle now talks of "knowledge seekers" and "understanding seekers," with good parts of the former (knowing what or how) necessary to the latter (knowing why).

As you sense, I find this whole line of inquiry provocative. I think the language of deep and surface learning is highly useful . . . preferable, even, to U.S. locutions such as "active vs. passive learning." "Active learning" has the ring of a slogan; "passive learning" is an oxymoron. In contrast, the deep vs. surface formulation is evocative; it captures something important; the words say what they mean. We'd be on a right track if we set as our aim deep experiences of learning for every student in every course.

If "deep vs. surface" as a concept has yet to take hold among researchers here, the phenomenon of superficial approach to task, and of what to do about it, has indeed attracted the attention of American psychologists. Let me mention two recent books on the theme, then a transnational study.

The first is a book that appeared this April, The Power of Mindful Learning, by Harvard psychologist Ellen Langer. Like Marton and Säljö, she notices that even the best student may produce right answers but fail utterly to get their point or meaning; like Ramsden, she's keen to the dangers of rote learning, canned assignments, and hurried coverage. The mindless learner, in effect, is on autopilot. Though Langer's book isn't directed at teachers, it is so masterful in adducing insight from a series of small, telling experiments that any teacher will find lots to chew on.

Mindfulness, in her view, is promoted by the continuous creation of new categories, openness to new information, and an implicit awareness of more than one perspective.

A different slant on these matters is offered by Chicago psychologist Mihaly Csikszentmihalyi, author of the best-selling books Flow and Creativity.

Csikszentmihalyi uses "flow" as a metaphor to describe the sense of effortless action we feel in times of peak action, moments when heart, will, and mind are as one in an activity . . . that time when our tennis serve was really grooved, or when we lost ourself at the easel or in our garden, or when we got a terrific piece of work done against an impossible deadline.

Flow happens, he says, when there's a clear set of goals requiring an appropriate response; when feedback is immediate; and when a person's skills are fully involved in overcoming a challenge that's high but manageable . . . when these three conditions are met, attention to task becomes ordered and fully engaged.

In Finding Flow, just out this year, Csikszentmihalyi presents data on how often and where people in the United States and other countries experience flow . . . not often in school, it seems. Why not? we might well ask ourselves.

Then there is the Third International Mathematics and Science Study (TIMSS), perhaps the single largest endeavor in the history of educational research: tens of millions of dollars for parallel studies and testing in some 50 countries, 498 curricula and 628 textbooks analyzed, all in search of cross-national insights about science and math learning in schools. The first TIMSS report, we may note with interest, is called "Characterizing Pedagogical Flow," and you may recall the headline that resulted last year, "Japanese Outdo U.S. Kids in Math." But now the more fine-grained analyses are coming forth, and the story isn't what we may have imagined.

That is, Japanese eighth graders do not do better because of better-prepared teachers, smaller classes, a longer school year, less TV, and more homework.

The differences trace right to the classroom, where the learning goals are clear, the topics per year are few but treated in depth, and where students learn to understand and apply through real-world problem solving and verbalization for meaning.

A third TIMSS volume is now out ("A Splintered Vision"), with explicit recommendations for mathematics educators that have resonance across disciplines:

• Focus on powerful, central ideas and capacities.
• Pursue greater depth, so content has a chance to be meaningful, organized, and linked to the student's other ideas, and to produce insight and intuition rather than rote performance.
• Provide rigorous, powerful, meaningful content-producing learning that lasts.

Earlier I observed, perhaps cryptically, that 20th-century social science busied itself proving what philosophers knew right along. I observe here that most of the findings and comment in TIMSS can be found in Whitehead's The Aims of Education, published in 1929.

Archaeology
In my search for new literatures on learning, it came as a surprise to me that archaeologists were weighing in on the matter. I'd lost track of developments in that discipline. I didn't know, among other things, that archeologists have developed quite elaborate databases encompassing every prehistoric skull found around the world, with data on its characteristics and on the surround of artifacts found with it; and that from these data it was possible to construct hypotheses not only about how earlier humans lived but about the capabilities of thought apparently available to them.

My text for this section is a wonderful book that appeared last fall, Steven Mithen's The Prehistory of the Mind.

The big breakthroughs of 30,000 to 60,000 years ago — agriculture, religion, art — were to Mithen prompted by a cognitive breakthrough in which the separate "side chapels" of intelligence became accessible to one another, creating a new "cathedral-like" modern brain in which thinking across domains became possible.

I don't want to overdo the Mithen thesis; it is bold, but also a leap, and much at the mercy of new evidence. I report it in the spirit of showing, again, that there are intriguing new ways of looking at learning. I recommend Mithen's book because the subject is fascinating, the writing accessible; the extended footnotes are a treasure trove of leads to related research in a dozen fields. Mithen's concept of a "modular" mind fits well with the "multiple intelligences" postulated by Howard Gardner and Robert Sternberg. Indeed, as you may know, Gardner recently embraced an eighth intelligence, a natural history one, that happens to be a main emphasis in Mithen's thesis.

Let me call your attention to another new book, published in 1995 but just out in paper, called Lessons From an Optical Illusion, by psychiatrist Edward Hundert, of the Harvard Medical School. Hundert's book is valuable because it weaves the philosophic with the neuroscientific. It considers enduring epistemological questions such as those we began with — What is knowledge? How do humans know? How does the self discover and construct the world? What is the contribution of nature and nurture? What is the mind's relationship between thoughts and things? — and takes the reader through history's conversation about these matters: from Plato and Aristotle to Descartes, up through Locke, Hume, Kant, and Hegel to Wittgenstein and Russell, and on to moderns such as Hilary Putnam and W.V. Quine. Hundert then tours psycho-biological literatures relevant to the question, spending a lot of time on Freud and Piaget, dipping into linguistics, genetics, and artificial intelligence, and laying out the latest insights from brain science. I do bare justice to a learned (and witty) treatise when I tell you that a key conclusion in the philosophers' running conversation — the intersubjectivity of knowledge — coincides remarkably with scientists' emerging understanding of how the brain functions.

There's more to the book than this, let me say, and plenty to argue with, too; when it comes to learning, though, we're all too short on synthesis, so this author's "lessons" are appreciated. I must have marked this book in ein hundert places. On the nature vs. nurture debate, for example,

Hundert explains that heredity and environment both turn out to be more determinative than we imagined, then cites Gregory Kimball's observation that asking which is more important "is like asking whether the areas of rectangles are determined by their height or their width."

What Good Teachers Know
To Anton Checkov, the artist "observes, selects, guesses, and synthesizes." I claim no "art" for this paper, but it has been selective and open to a second-guess. There's so much more to add: about "intelligence," for example, or gender-based and non-Western ways of knowing. Let me press on, though, to Checkov's fourth charge, synthesis.

In the interests of synthesis, I introduce a still-further way of looking at learning, this from a new breed of researchers working in K-12 settings in search of a "wisdom of practice," that is, for what best teachers in real classrooms do with actual students. The basic idea here is that good teachers, like "reflective practitioners" in other professions, constantly test, adjust, and reframe their models of practice on the basis of experience and reflection . . . study what they do, and you'll create new knowledge for effective practice more widely.

But how might the "craft knowledge" of teachers stand in summary of the neuroscientific and related findings recounted in this paper? It can't fully, of course. Yet the two ought to be similar and close in spirit, since (as noted earlier) every perceptive teacher is in fact an observer of brain functioning, ever testing and adjusting on the basis of what works with students.

In fact, I'd be suspicious of any neuroscientific theory of teaching that was much at variance with what best teachers already knew and did.

All of this is by way of introducing an exercise I conducted with university faculty earlier this year that attempted to capture a "wisdom of practice" out of the "powerful pedagogies" that have sprung up on campus in recent years. You know these pedagogies — some old, some new, all of them with a following, some of them with a research basis:

Virtually all of these approaches were fashioned by classroom teachers as a response to real problems with real students; they weren't made up by researchers. So the question becomes, What are the common (if tacit) assumptions these pedagogies make about learning? Looking at the bunch of them, I asked in the exercise, What can we infer about their view of how students learn? What trees do they bark up in common?

The answers were these: The more a teacher can emphasize . . .

• learner independence and choice
• intrinsic motivators and natural curiosity
• rich, timely, usable feedback
• coupled with occasions for reflection and
• active involvement in real-world tasks
• emphasizing higher-order abilities
• done with other people
• in high-challenge, low-threat environments
• that provide for practice and reinforcement

. . . the greater the chances he or she will realize the deep learning that makes a difference in student lives.

In Sum: Cycles of Learning
The topic of this conference is assessment and quality. How might all this talk about learning color the way we think about those topics?

Most of you will be familiar with David Kolb's "learning cycle," which he introduced in the early 1980s. It's a diagram with arrows around a circle, showing that learning occurs when an act or experience is followed by feedback, which is then reflected on, leading to new understanding, and from there to revised action. In a lifelong, continuous learner, that cycle — act, feedback, reflection, act again — is recursive and never-ending.

Kolb's model, of course, doesn't capture the whole of learning . . . it doesn't by itself get into the kind of "double loop" learning that Chris Argyris showed as necessary for the transformation of mental models, for example. That said, the model nonetheless captures an important insight about learning as a process.

To elaborate with an example, I doubt anybody will ask me to give this presentation again. Nevertheless, if I had that chance and wanted to do it better, what I would need is feedback, which would have to be concrete, usable, and credible, from parties or sources I trusted. But feedback alone does not change behavior, a point proved by a hundred studies. What I'd need to do is to engage that feedback, to reflect upon it; and the most powerful kind of reflection I could do would be with other people, because that broadens the wisdom, sharpens judgment, and enhances my commitment. Even there the cycle doesn't stop, because I need the intent and resolve to improve, to use my new understanding in that next presentation.

The whole aim of the cycle, then, is Ted's learning, that the second time around I'll be smarter and better at this talk.

Kolb's learning cycle is an interesting, persuasive representation of individual learning. It is also a representation of assessment itself.

Assessment is a process in which rich, usable, credible feedback from an act — of teaching or curriculum — comes to be reflected upon by an academic community, and then is acted on by that community — a department or college — within its commitment to get smarter and better at what it does.

All of which is to say, assessment is more than data gathering. It also encompasses essential functions of meaning-making, action, and commitment to improve. Absent any of these elements, the doing of assessment becomes hollow. When you go to an assessment session at this conference and hear a presenter say, "We did this great study . . . now we're figuring out how to disseminate it," you know something is wrong. What's wrong is that the community was left out. Assessment as learning is not a third-party research project or someone's questionnaire; it is a community effort or nothing, driven by a faculty's own commitment to reflect, judge, and improve.

Kolb's learning cycle was conceived with individual learning in mind. But how does an organization learn? How does my department or college get smarter and better over time at prompting appropriate learning in students?

And the answer is, we learn as I learn. That is, a faculty gets smarter and better at its tasks by systematically collecting feedback, reflecting on it, and using the resulting understanding to enact next cycles of work.

Those of you with a background in quality management (CQI) will immediately get the point. Kolb's cycle is quite the same thing as the "plan, do, check, act" cycle of continuous improvement developed by Walter Shewhart at Bell Labs in the late 1920s . . . different words, same idea. (Shewhart, by the way, knew John Dewey, the intellectual father of so many of the ideas in this paper and in whose writing you'll find the idea of cycles of learning. Dewey, I'm told, got the idea from an essay written in 1904 by philosopher Charles Sanders Peirce, Dewey's predecessor in pragmatism.)

All of quality management's injunctions, as Deming and Peter Senge have pointed out, are on behalf of organizational learning. A key element to continuous improvement is the "systematic gathering, interpretation, and use of information for purposes of improvement." In academic settings, that statement also defines assessment. Thus, I say to the quality-management proponents in this audience: Think assessment. And to the assessors here: Think continuous improvement. CQI and assessment are alike in this regard: They are acts of learning in themselves, and the key to prompting the deep learning we want for the students we serve.

Bibliography

Abbott, John. Synthesis. Washington, DC: 21st Century Learning Initiative, 1997.

A Private Universe. (Video). Harvard Smithsonian Center for Astrophysics, 1987. (Distributed by the Annenberg/CPB Project, Washington, DC, and by Anker Publishing Co.: Bolton, MA.)

Brown, G., J. Bull, and M. Pendlebury. Assessing Student Learning in Higher Education. London: Routledge, 1997.

Brown, J.S., A. Collins, and P. Duguid. "Situated Cognition and the Culture of Learning." Educational Researcher, vol. 18, no. 1 (1989): 32-42. (See also a special issue of Educational Technology, March 1993.)

Bruner, J.S. Toward a Theory of Instruction. New York: Norton, 1966.

Caine, Renate N. and Caine, G. Making Connections: Teaching and the Human Brain. Reading: Addison-Wesley, 1994.

Csikszentmihalyi, Mihaly. Finding Flow: The Psychology of Engagement With Everyday Life. New York: Basic Books, 1997.

Dewey, J. How We Think. Boston: Heath, 1933.

Eble, K. The Craft of Teaching. San Francisco: Jossey-Bass, 1988.

Edelman, G.M. Bright Air, Brilliant Fire. New York: HarperCollins, 1992.

Entwhistle, N. "Motivational Factors in Students' Approach to Learning." Learning Strategies and Learning Styles, edited by R. Schmeck. New York: Plenum Press, 1996.

——— , and P. Ramsden. Understanding Student Learning. London: Croom Helm, 1983.

Goleman, Daniel. Emotional Intelligence: Why it Can Matter More Than IQ. New York: Bantam Books, 1995.

Hake, R.R. "Interactive Engagement vs. Traditional Methods: A Six Thousand Student Survey of Mechanics Test Data for Introductory Physics Courses." American Journal of Physics (in press).

Highet, G. The Art of Teaching. New York: Knopf, 1950.

Hundert, E.M. Lessons From an Optical Illusion. Cambridge: Harvard University Press, 1995.

Kolb, D.A. Experiential Learning: Experience as the Source of Learning and Development. Englewood Cliffs, NJ: Prentice-Hall, 1984.

Kotulak, R. Inside the Brain: Revolutionary Discoveries of How the Mind Works. Kansas City, MO: Andrews and McMeel, 1996.

Langer, E. The Power of Mindful Learning. New York: Addison-Wesley, 1997.

Lave, Jean, and Wenger, E. Situated Learning: Legitimate Peripheral Participation. New York: Cambridge University Press, 1991.

Marton, F., and R. Säljö. "On Qualitative Differences in Learning: 1. Outcome and Process. British Journal of Educational Psychology, 46(1976): 4-11. (See also National Teaching & Learning Forum, vol. 5, no. 1 (1995): 1-4.)

Mithen, Steven. The Prehistory of the Mind. London, Thames and Hudson: 1996.

Palmer, P. The Courage to Teach. San Francisco: Jossey-Bass, in press.

Ramsden, P. Learning to Teach in Higher Education. London: Routledge, 1992.

Senge, P. The Fifth Discipline: The Art and Practice of the Learning Organization. New York: Doubleday, 1990.

Shulman, L. "Knowledge and Teaching: Foundations of the New Reform." Harvard Educational Review, vol. 57, no. 1 (1986): 1-22.

Sylwester, R. A Celebration of Neurons: An Educator's Guide to the Human Brain. Alexandria, VA: Association for Supervision and Curriculum Development, 1995.

Third International Mathematics and Science Study. Successive Reports from Kluwar Academic Publishers, Boston/Dordrecht/London. (See also the websites www.ed.gov/NCES/timss/index/html and www.ustimss.msu.edu/.)

Whitehead, A.N. The Aims of Education. New York: Macmillan, 1929.

Marchese, Theodore J. "The New Conversations About Learning: Insights From Neuroscience and Anthropology, Cognitive Science and Work-Place Studies." In Assessing Impact: Evidence and Action, pp. 79-95. Washington, DC: American Association for Higher Education.