July 28, 1961
AUGMENTED HUMAN INTELLECT PROGRAM
Taking a fundamental approach to the problem of making a human more effective at his professional problem-solving tasks by capitalizing upon the joint capabilities of the man and automatic symbol-handling equipment. Consider a human approaching a complex problem--our interest is in increasing the effectiveness with which he can do such as (1) gain comprehension of the problem situation, (2) isolate the significant factors involved in the desired solution, (3) conceive good solution, (4) reasonably predict the consequences of the solution, and (5) repeat the process on problems subsequently associated with his doing what is needed to get the solution implemented. Our approach is to assure that we are studying a complex system (what we call an Individual Symbol Manipulating System) that has four principal functional components: the human, his language (i.e., the way he divides his universe into concepts and the symbols he attaches to them), his artifacts (the man-made devices he uses), and his methodology (his methods, procedures, algorithms, heuristics, etc.). We assume that introduction into this system of such as a digital computer as a real-time working artifact will be very fruitful (and our approach is oriented thereby), but we feel that all four of these components must be considered in a co-ordinated fashion in order to realize really significant improvements in the performance of this system.
We are developing a program in which we can simulate hypothesized artifacts for real-time human usage, and experiment with various innovations in the symbol-manipulating system whose capability we are trying to improve. We are aiming first at augmenting humans in practical but limited intellectual tasks, and later we will proceed to more general capabilities as we learn more about the system and its possibilities.