This book represents an attempt to organize nearly the entire field of psychology within a single new theory, based upon only one very simple assumption about neuronal functioning. The gestation period for this theory has been very long. The first parts were developed 15 years ago at the University of Cincinnati, when I was examining some very old ideas, dating back at least to the last century (Wundt, 1874), about optimal levels of stimulation for producing pleasure. Optimal levels of stimulation implied optimal levels of neuronal activity. Because there was an obvious connection between pleasure and reward or reinforcement, this suggested that optimal level of neuronal activity might in some way be related to reinforcement. Reinforcement is anything that increases the probability of a particular pathway of neurons being able to fire again in a similar situation. Therefore, it seemed possible that there might be an optimal level of firing for a neuronal pathway that would cause the connections within this pathway to become stronger. Conversely, levels of firing greatly different from this optimal rate might cause the connections to become weaker.
The idea of an optimal level of firing causing reinforcement was not particularly new. It is likely that much of my thinking at this stage was influenced by William Dember and his hypothesis that an optimal level of complexity was rewarding (Dember, 1965; Dember &Earl, 1957). Although I had little direct contact with him at the University of Cincinnati, his ideas permeated throughout the Psychology Department, and it would have been impossible not to have been influenced by them.
The next step in the development of the present theory took it out of the mainline of psychology and into neurology. I was examining the newly discovered phenomenon that electrical stimulation of certain brain areas produced positive reinforcement (i.e., that rats would bar press for such intracranial stimulation). One possible reason for this appeared to be that stimulation of these particular areas might produce inhibition of the pathways involved in motor acts such as bar pressing. But how could inhibiting a pathway make it become stronger? First, I went back to the idea of the optimal level of firing. If the level of firing were initially too high, inhibition could bring it closer to the optimal level and therefore be reinforcing.
A breakthrough came, however, when I tried viewing this relationship in the opposite way; that is, seeing if reinforcement from inhibition could explain why . . .