dinal data from systematic testing programs, that tell us what is happening in our schools. The data measure at least the basic trends, showing whether there is improvement or deterioration in the cognitive skills that children are acquiring in society.

Now, that is a long way from knowing the reasons behind the trends. It requires other kinds of social science research to disclose what the reasons might be if Johnny isn't reading as well as he did 10 years ago. Who is the culprit? Is it the schools? Is it television? Is it Johnny, or whom shall we blame? Here is a prime area for social science research.

One of the remarkable things in our society over the past 20 years has been a vast change in our treatment of minorities and of groups that have been underprivileged. I don't mean that we have solved all these problems. In a society, such problems are never fully solved. I mean that most of us would feel that there has been a great change in our society and, on the whole, we have made substantial progress in these matters, although different ones among us would evaluate that progress differently. Here is as sensitive a topic as we have. What is the role of social science in a topic of this kind?

Early in the postwar period, a distinguished Swedish economist and social scientist, Gunnar Myrdal, came to this country to organize and head up a study of the American race problem, and out of that study came a very thick book which many of us are familiar with. The book was called "An American Dilemma." I think the title, itself, is instructive because the book did two things.

First, it made a very careful factual examination of our racial institutions and problems, with suggestions of possible steps that we might take to extricate ourselves from a dilemma of values which we ourselves felt. That is, if there was one thing we Americans were agreed about, in an area where there was immense conflict and disagreement about specific policies, it was that we were faced with a dilemma, a conflict between some of our basic values and the social practices in our society.

The Myrdal report, "An American Dilemma," is an excellent example, I think, of how the social sciences work, not by proposing specific solutions, specific measures, but by providing us with a mirrorin that case a rather accurate mirror-for ourselves. It is an excellent example of the doctrine of knowing ourselves, and with all of the progress and steps forward and backward we have taken on the problems of race and minorities, we have undoubtedly handled them with a great deal more skill over the past 20 years as a result of having some facts about where we were, some facts about what the problem was and is.

Today, we can monitor our progress toward solving these problems because we do have statistics in our society-economic statistics, for example, about the relative progress of blacks, of women and of other underprivileged groups in the job market, and statistics about their occupational status today. We don't have to debate about it. Well, we do have to debate it, but we don't have to guess about the basic facts. We did get some facts and analyzed them carefully. We did debate the controversial aspects of the statistics, but we are not just proceeding on pure imagination, pure conjecture, or our own personal experiences which may be very atypical of the experiences of the society as a whole.

I would like to add a footnote to what was said this morning about the area which is my particular area of interest and concern in research now, and has been for some years, the area called cognitive science.

We are making enormous progress today, and have been for perhaps 20 years, in our understanding of the mechanisms of human thinking, in understanding what goes on in the human head when a human being solves a problem. We don't know much in terms of neurology yet. We are still a little far from the squid's nervous system in this kind of research, but we have a growing understanding of the information processes that have to go on in the human brain when it thinks. We hope ultimately, maybe in my lifetime or the lifetime of my children, this knowledge can be linked up with the neurological research that was described earlier.

But even before that linkage takes place, we have now reached a point where we can, for example, characterize rather specifically and accurately the differences between the behavior of an expert and the behavior of a novice in solving a physics problem. I mean, we can say something more about it than that the expert will solve the problem while the novice scratches his head. We are able to characterize rather specifically the strategies in the programs that the expert uses and the contrast between these and the strategies that the novice uses. And although we are talking about basic research here-understanding the phenomena yet there is a rather direct and obvious set of steps that need to be taken as our knowledge develops, in order to apply our understanding of the differences between expert and novice behavior for practical ends.

As we understand the basic phenomena, we should be able to improve our methods of teaching physics. We should be able to pinpoint very much better the kinds of difficulties students are having in learning We should be able to analyze, in detail, the chapters of a textbook and to say with some assurance on the basis of our understanding of the process, what parts of a chapter are satisfactory for facilitating students' learning and what parts need to be revised or expanded.

That gets me back to my opening theme: that the social and behavioral sciences can and must make important contributions to the productivity of our society by making contributions to the efficiency of our most valuable resource, our own human behavior and human thinking, and problem solving processes.

I tried to emphasize in my remarks here the processes of application of the social sciences which, I think, are rather different than the processes of application of engineering and medical science. The social sciences don't invent pills, by and large. We don't invent wonder drugs. We invent basic knowledge-no, I guess invent is the wrong word, isn't it? We discover basic knowledge about ourselves, about our society. The most important channel of application of that new knowledge is its broad diffusion through public channels, until it becomes part of the knowledge of almost all of us.

Several times it has been mentioned this morning that the social sciences are often discounted because much of what they learn seems to be common sense. Well, it is common sense today to say that if you drop a feather and a rock together in a vacuum, they will fall at the same pace. It wasn't commonsense before Galileo. In a democratic society, which has to make its own decisions about what it wants to be,

one of the basic aims of the social sciences must be to take knowledge that comes out of the laboratory-knowledge that may be stated in language that is hard to understand-and make that part of the commonsense of our society. That is the most important goal of application in the social sciences.

Finally, we need to remind ourselves as social scientists, and remind others who are involved in the venture, that we should not over-claim what the social sciences can do because, in any society there are built-in deeply many sources of conflict of interests. There are many issues in which we have different values among us and we want to go in different directions. Different public policies will affect us quite distinct ways, and we should not expect the social sciences, somehow or another to lead the lion and the lamb to lie down together. We should not expect the social sciences to resolve the many conflicts of interest that must exist in any society.

But, I think we can expect the social sciences to help in understanding what those interests are and what the range of possible directions for the society are that can partially accommodate that whole collection of interests. For that reason, the social sciences simply should not-I won't say cannot, obviously, they can-stay away from important issues in our society simply because those issues might be sensitive issues, might involve a conflict of values.

Moreover, we need to remind ourselves that social science research cannot repeal the laws of nature. Engineers don't try to design bridges by inventing a gravity shield so that the bridge won't be pulled down by gravitational forces. They work with the system. They accommodate themselves to the laws of nature. Social science research, too, does not have, as its goal, changing human beings into some other kind of species with quite different characteristics than we have now.

Social science research is concerned with understanding ourselves with all of the warts on, with all of the quirks of our behavior, understanding why those quirks aren't really quirks but are part of us, and understanding how societies can operate with human beings as they really are.

Well, let me stop, at this point so that before hunger overcomes us all, we have time for questions.

Mr. BROWN. Thank you very much.

You raised an interesting parallel in your mention of physical scientists working with the laws of nature and not trying to overcome them. It leads me to the thought that there are, in the social sciences, certain fundamental strictures with very high importance governing the ways in which human societies operate. These are long ingrained cultural attitudes. They are religious or have other very strong bases and, in effect, they become sort of the laws of gravity for those who are bound by them and, yet, they are very special kinds of laws of gravity. They vary from culture to culture.

Of course, part of the problems of social science research is that it presumably becomes aware that the cultural laws of gravity are situational, that they vary from culture to culture. Yet, this doesn't solve the fact that it creates political problems to do research in these areas.

The point that Dr. Atkinson made in his editorial was that social science is either considered to be irrelevant, in which case we ought not

to fund it, or it is relevant and tries to change these laws of gravity and hence, we shouldn't fund it for political reasons.

I wonder if you could help us to overcome this problem? How do we approach this situation where social science research is perceived as trying to change the law of gravity?

Dr. SIMON. First, I think we have come a long way in this respect in the past 20 or 30 years, and again, I think it is a matter of not just a few people understanding the potential role of the social sciences but of our society understanding that role.

With all of the ups and downs and occasional checks, I perceive around me in the universities and in the Federal Government's activities in these fields a greater tolerance for examination of human institutions, for objective description of those institutions in all of their variations, than I saw 20 or 30 years ago.

I think we really have come a long way in this respect. I think social scientists can help by being very careful, when they touch on areas that have important policy applications and particularly on sensitive areas, to distinguish the facts that they are able to validate the hard facts from the interpretations of the facts, and to distinguish interpretations from the policy conclusions that they would draw from the facts.

It always has been part of the training of social scientists to have some understanding of that distinction, but perhaps we could improve that training so that we are less often misinterpreted, or correctly interpreted, as making policy prescriptions when we think we are doing social science. I don't mean that social scientists have to withdraw from the world. I mean that they need to make their role crystal clear as to whether they are prescribing or whether they are trying to understand.

I think, also, and your comment on differences between cultures is important here, I think, also, that within social science we haven't always been clear as to when we are dealing with the laws of nature and when we are dealing with cultural facts.

Sometimes, it is said as a half-joke that psychological theory is a theory of the American college sophomore, because most of the experiments have been run with college sophomores as our human subjects. We might get some surprises if we used juniors or people who hadn't gone to college or people from another culture as subjects.

But today, in cognitive science and psychology we are learning what anthropologists knew a long time ago: That there are certain basic invariants of human behavior. There is no way we know in which you can stretch your short-term memory so that you can hold in it three telephone numbers instead of one. So, that is kind of a basic invariant that will turn out probably to be a biological invariant. On the other hand, we are learning that between such invariants and the actual behavior, there are layers of strategy, and those strategies are modifiable. What we call culture is a piece of those strategies.

Mr. BROWN. Dr. Simon, I want to raise a subject which is of considerable current interest but may not be in an area that you care to comment on.

In the macroeconomic field, there have been some indications lately and some of them based on economic research, that our models of the economic system have failed to adequately factor in certain factors

which can only be described as psychological. That is, the anticipation of certain constituencies of a certain thing is going to perhaps lead them to modify their economic behavior. This may be true of labor unions anticipating inflation. In fact, one of the explanations given for continued inflation is that the expectation of continued inflation continues to fuel it.

Do you feel that there is a possibility that the interdisciplinary research and the social sciences can contribute to a better understanding of this macroeconomic problem that we seem to have at the present time?

Dr. SIMON. There, you have touched a very sensitive button. You asked in the beginning of my testimony how I could be an economist and a psychologist. That is a difficult question, indeed, because really those two disciplines, until recently, haven't been talking to each other very much.

The work on experimental economics we heard described here is quite novel and unusual in that respect. I do, indeed, think that the current crisis in macroeconomic theory and business cycle theory, in which there is now quite a diversity of views, can be attributed in considerable measure to our lack of a theory of how human beings form expectations. I don't mean there isn't enormous speculation about that in economics. As a matter of fact, a lot of the current excitement in economics is about something called "rational expectations theory,' which is a very elegant idea, very elegantly developed.

The only thing we don't know about rational expectations theory is whether it describes any real human behavior or not. Where economics and psychology need to bed down together, much more closely than they have in the past, is in finding ways to verify-by direct observation of human behavior, and by human laboratory experiments of the sort that were described here—which of the rather large set of theories of expectation formation that are now current really do describe how human beings go about predicting the future.

Probably before we get done, we will have to have the historians and the anthropologists in the act also, because expectation formation too is something that probably changes from one culture to another and over time. I am sure that if we had good data 15 years ago about the extent to which people took price trends into account in their economic planning and we had comparable data for today, we would find that there has been a great change in the public view about whether this variable is worth paying attention to or not.

That kind of shift-the economists have a name for it, they call it shift in structure-is not anything that is easily accommodated in the kinds of economic theories that we have now, nor does economics today have very much skill in detecting empirically when shifts take place. much less in anticipating them.

Mr. BROWN. That leads me to a more specific question. I am inclined to feel that the Foundation is emerging from a period in which it perhaps focused too narrowly on the interesting problems in the social sciences to a recognition of the importance of a good interdisciplinary approach within a somewhat broader framework than it has in the past. I would like to ask you to comment, if you could, on how sound a contribution the Foundation makes to the support of research that does broaden our understanding of the complexities of some of these areas?