Charles Darwin (1872) once postulated that “…natural selection acts only by taking advantage of slight successive variations…” (p. 156). It is through these slight, successive variations, Darwin hypothesized, that some species evolve into other species and some species become extinct over time. In the same way, beginning with Pavlov’s Conditioned Reflexes and John Watson’s Behaviorist Manifesto, behaviorism has endeavored to understand and consequently control behavior through the mechanism of slight, successive variation. Behaviorism has sought to wield that most powerful instrument of natural selection, which brought the dinosaurs into existence and then subsequently wiped them from the face of the earth, in an effort to manipulate and predict behavior. Nonetheless, incumbent on the understanding of behaviorism is a contrast of the Watsonian and Skinnerian perceptions of stimulus and response, with the Tolmanian concept of learning, in light of the effect that all three viewpoints have had on the current understanding of learning and behavior.
James B. Watson famously summed up the conclusion of his behaviorist model when he said, “Give me a dozen healthy infants…and I’ll guarantee to take anyone at random and train him to become any type of specialist I might select…” (Goodwin, 2005, p. 303). Watson’s understanding of behavior represents a complete shedding of the traditional introspective approach to psychology, while at the same time forsaking the belabored functionalist/structuralist paradigm. No, Watson did not see consciousness as incremental, adaptive, or comprehensive; rather, Watson threw off the shackles of consciousness altogether and sought to understand behavior wholly through the mechanisms of stimulus and response. Watson’s predictions about behavior found their full personification in an investigation later dubbed the Little Albert study, carried out at John Hopkins and published in the Journal of Experimental Psychology under the title “Conditioned Emotional Reactions”. During this study, an 11 month-old child was presented with a white rat, originally an unconditioned stimulus (UCS), because the child showed no aversion to the animal. However, as part of the investigation, when the child actually touched the animal the researchers struck a bar, causing a loud noise and scaring the child (UCR). After several pairings of tactile contact with the rat, now a conditioned stimulus (CS), with the loud noise (UCS) the child began to fear the rat itself, now a conditioned response (CR). The strong association that the infant eventually formed between the fear response, usually evoked by the loud noise, and the rat itself, is at the heart of Watson’s understanding of stimulus and response. In fact, Watson believed that in a complete system of psychology, “…given the response the stimuli can be predicted; given the stimuli, the response can be predicted” (Goodwin, 2005, p. 278). What logically flows from Watson’s assertion is that if the stimulus is controlled the response can be manipulated, and as a matter of course behavior can be manipulated.
As to the abovementioned example of natural selection in the first paragraph, Watson sought to manipulate involuntary behavior through slight, successive variations in the stimulus. In addition, the effects of Watson’s work can be seen today in the areas of child development, mass marketing, and even the ultimate fulfillment of psychology as a scientifically rigorous discipline. There is little doubt that Watson’s ardent support for his theories helped herald in over three decades of behaviorist dominated psychology in America.
As Watson’s academic career was coming to an end, a young, auspicious psychologist by the name of B.F. Skinner stepped up to the plate to carry the mantle of behaviorism well into the latter part of the twentieth century. Skinner believed that psychology should concern itself completely with, “the prediction and control of behavior, both nonhuman and human” (Goodwin, 2005, p. 329). In this way, Skinner offered a logical next step from, rather than contrast with, Watsonian behaviorism. However, Skinner believed that behavior should be explained in completely observable terms, unlike Watson whose theory was dependent, at least in part, on unseen associations. Skinner believed that the consequences of an action have a feedback effect on an organism, which can be utilized to predict and control future behavior. Whereas Watson molded the response through the implementation of a conditioned stimulus, Skinner increased the probability of a response (operant) through the use of reinforcement, after the fact. Operant conditioning, as Skinner called his theory, is utterly dependent on natural responses (operant) in order to function, since reinforcement can only occur after the operant rather than before or concurrent with the operant. Furthermore, the future probability of an operant can be manipulated through the use of reinforcement, both positive and negative. Positive reinforcement entails adding or presenting stimuli after an operant, and conversely, negative reinforcement involves the withdrawal of a stimulus, in order to increase the frequency of said response. In this fashion, operant conditioning is more concerned with the future prediction and manipulation of voluntary behavior; whereas, Watson’s behaviorism seems content to fret about involuntary behavior.
In the well-known article, “An Experimental Analysis of Behavior” Skinner detailed by what means he empirically supported his theory. He conducted an investigation in which he isolated a pigeon in a box. When the pigeon brought its head above a certain point the food tray inside the box would be opened. Through the natural response of the pigeon raising its head (operant), Skinner was able to subsequently introduce food, through the opening of the food tray (positive reinforcement), which made the likelihood of the animal raising its head much higher in the future. Moreover, an example of negative reinforcement, not mentioned in Skinner’s paper but nonetheless applicable, would be if the food tray were closed only when the pigeon brought its head below a certain point. In this way, the operant of the pigeons head being above a certain point would be negatively reinforced by the removal of a stimulus, represented by closing the food tray. Through the mechanisms of positive and negative reinforcement, Skinner believed that a naturally occurring operant could be acted upon to inhibit or encourage future responses, in other words, that an operant is emitted, not elicited.
B. F. Skinner’s contribution to psychology, and indeed many other areas of science, cannot be overstated. The influence of operant conditioning stretches into childhood education and discipline, political philosophy (Walden Two), space exploration, and of course comparative psychology. In relation to natural selection, operant conditioning can be understood as slight, successive variations acting on a response to either encourage or inhibit the future probability of that response. In reference to Watson, Skinner’s operant conditioning wrote the next chapter in the scientific illumination of human and animal behavior.
If Watson’s behaviorism represents a sound denial of the established models of functional and structural introspection and Skinner’s operant conditioning constitutes an evolution beyond Watson’s unseen associations, then Tolman’s theory of learning embodies a critique of everything thus mentioned. Tolman represents a bridge between the behaviorism of the first half of the 20th century and the social and learning cognitive influence of the latter half of the 20th century. His theory of latent learning, exemplified in an article in the Psychological Review entitled “Cognitive Maps in Rats and Men”, put forth an understanding of behavior which separated learning from performance. The theory of latent learning dictates that learning occurs outside the realm of reinforcement and association. In which case, reinforcement and associations affect performance rather than learning. In other words, both Watson’s behaviorism and Skinner’s operant conditioning affect speed and accuracy (performance), but not learning directly. Tolman believed that learning occurred as the result of cognitive maps that the animals learned without the benefit of reinforcement. To prove his theory of learning he conducted an experiment in which three sets of rats were tested in a maze, a no reward group (NR), a reward group (R), and a group that was initially unrewarded but later rewarded (NR-R). The NR group was never given food but was observed to see how many times the rats would complete the maze naturally (i.e. the control group). The R group was positively reinforced from the commencement of the study by the introduction of a food reward at the end of the maze. The NR-R group was allowed to complete the maze naturally, like the NR group; however, at the end of 10 trials food was introduced, like positive reinforcement, at the end of the maze. Surprisingly, the number of errors while completing the maze was less for the NR-R group than the R group by the end of trial 12. However, the NR group consistently scored more errors throughout the study than the other two groups, while completing the maze. This study offers convincing evidence that learning does occur without reinforcement, but that without the motivating factor of reinforcement learning does not decrease the number of errors.
In light of these observations it is clear that even though reinforcement improves performance over no reinforcement, reinforcement is not largely involved with learning; but rather takes advantage of learning to further performance. Tolman theorized and later showed empirically that animals develop cognitive maps, as a means of learning, when presented with a new environment. It is through these cognitive maps that the rats in the aforementioned study were able to increase their performance so quickly after the introduction of food as a reward. In the absence of cognitive maps, the animals would not have learned the maze without the food reward and would, therefore, lag in performance without the mediating factor of reinforcement.
Tolman’s most significant contribution to current-day psychology is his work on comparative cognition. His learning theory and work on cognitive maps helped bring psychology out of the enlightenment, so to speak, into the industrial revolution; making psychology not only scientific but relevant. In reference to Darwin’s theory of natural selection, latent learning anticipates the effect that slight, successive variation will have on performance as a mediating factor to the practical implementation of, but not as a direct correlate to, learning.
In conclusion, the evolution of psychological thought throughout the 20th century is epitomized in the progression from Watsonian behaviorism to Skinnerian operant conditioning, to Tolmanian learning theory, right into the modern-day investigation of social and learning cognition. Additionally, Watson’s behaviorism uses the mechanism of natural selection as a means to an end, operant conditioning uses slight, successive variation as an end capable of affecting the means, and the theory of latent learning uses variation as a means to performance for the benefit of the end, learning. Likewise, Watson’s behaviorism and Tolman’s learning theory seem preoccupied with involuntary psychological phenomena; whereas, Skinner was concerned primarily with voluntary, observable behavior and its precursors. In all, the contributions of these three people brought psychology out of the shadows of introspection, through the blinding light of scientific discovery, into the shade of pragmatism and applicability.
Darwin, C. (1872). On the origin of species (6th ed.). London: John Murray.
Goodwin, C. J. (2005). A history of modern psychology (2nd ed.). Hoboken, NJ: Wiley.
- Examine the roots in early philosophy that lead to modern psychology.
- Identify major philosophers in the western tradition that were primary contributors to the formation of psychology as a discipline.
- Explore the development of the science of psychology during the 19th century.