A General Feedback Theory of Human Behavior: Part II

INTRODUCTION The model described in Part I is only a part of our general theory-the part which organizes our more general ideas about human behavior and human nature. To conceive of human organization as following that of our hierarchical array of FBCS (externally fed-back Feedback Control Systems) implies a certain attitude toward behavior, different in some important respects from traditional psychological viewpoints. Some of these differences we began with, but most of them took form only as we went back and forth between modifying our organizational model and observing people behaving. One of the most puzzling, and in our opinion critical, aspects of human behavior is that behavior appears multiordinal. The same behavior can be described in a number of apparently equally-valid ways, from the particular ro rhe general. Usually this representation of human behavior at varying levels of abstraction is put aside during a scientific smdy, and one particular level is chosen as the most interesting, or sometimes as the only "proper" one. But for us this multiordinality raised a critical question: is it due to the way in w h c h behavior is observed, or is it somehow a significant property of the behaving system? The answer we have arrived at is, "Both." One must never forget that the person observing human behavior is a system like the one he is observing. If we accept that our model represents behavioral organization, particularly in the FBCS aspects, then it is pmception which gives form to behavior. Behavior will make sense to E only if E knows what perceptaal variables the behavior is maintaining at some reference-level. If an organism is producing behavior as a means of controlling a several-times-abstracted variable, then E has no hope of seeing order in this behavior unless he is capable of learning to select out of his experiences the relevant elementary sense-impressions and then can combine them in the same way that S is combining them to make a perceptual variable. If S and E are both FBCS, then even in a varying environment requiring widely-varying physical action, S will be able to maintain abstract variables at reference-levels, and E, if he resembles S, will be able to perceive that S is doing so.