How Did the Behavioral Study of Learning Develop?
The behavioral study of learning began with the work of Ivan Pavlov, a Russian physiologist who was studying the effects of salivation on digestion. He noticed that the dogs he was using as research subjects would begin to salivate before they had received any food. Pavlov decided to investigate this phenomenon and as a result elucidated the principals of classical conditioning. In classical conditioning, a neutral stimulus is paired with a stimulus that elicits an automatic reflexive response. In Pavlov's work, food reliably produces salivation, and because no learning is required for this response, the food is called an unconditioned stimulus (US) and the salivation is called an unconditioned response (UR). The neutral stimulus or conditioned stimulus (CS) is repeatedly presented with the US, and the animal learns, or becomes conditioned, to associate the CS with the US. After repeated CS/US pairings, the CS comes to produce the behavioral response by itself. This response is known as a conditioned response (CR), because it is occurring in response to the CS.

After reading this section, you should be familiar with the principals of classical conditioning, and you should be able to define the following terms: acquisition, extinction, spontaneous recovery, generalization, discrimination, and second-order conditioning. Classical conditioning has been demonstrated to play a role in the development of phobias. It also plays a role in drug addiction, by which the environment in which the drugs are used becomes conditioned with the effects of the drugs. This helps explain why recovered drug addicts can experience cravings for their drug when they are exposed to the environmental cues associated with their past drug usage. Pavlov believed that any neutral stimulus could become conditioned, an idea known as equipotentiality. However, work by John Garcia on conditioned taste aversion has shown that humans and animals are biologically prepared to become conditioned to certain types of stimuli. A cognitive model of classical conditioning has been proposed by Rescorla and Wagner. They believe that the strength of a CR can be predicted by the extent to which the US is unexpected or novel.

How Is Operant Conditioning Different from Classical Conditioning?
Another form of learning is operant conditioning. In operant conditioning, it is the outcome of a behavior that predicts whether the behavior is likely to occur again. The study of operant conditioning began with work by Edward Thorndike, who placed cats in puzzle boxes and measured how long it took the animals to escape from the boxes. As a result of this work, Thorndike developed his Law of Effect, which states that behaviors that are followed by a positive outcome are more likely to occur, while behaviors that are followed by a negative outcome are less likely to occur. These positive outcomes are known as reinforcers, and the negative outcomes are known as punishment. Reinforcement can be positive (a positive stimulus is administered) or negative (an aversive stimulus is removed). In both cases, the behavior is strengthened, i.e., reinforced. Likewise, punishment also can be either positive (an aversive stimulus is administered) or negative (a positive stimulus is removed).

Much of the early work on reinforcement and punishment was done by Harvard psychologist B. F. Skinner, and you should be familiar with his name and with the testing device that bears his name, the Skinner box. There are different ways in which reinforcers can be delivered to a person or animal. When reinforcement follows every targeted response, it is said to be a continuous reinforcement schedule. Partial reinforcement schedules can be based on the number of responses required for reinforcement ( ratio schedules) or on the time between reinforced responses (interval schedules), and these ratio and interval schedules of reinforcement can be fixed or they can vary. Biology and cognition also play roles in operant conditioning, and after reading this section, you should know what is meant by the terms genetic drift and latent learning.

How Does Watching Others Affect Learning?
Classical and operant conditioning are not the only ways through which humans and animals learn. We also learn by watching others, observational learning. Albert Bandura conducted a classic study on children's observational learning of aggression, and this study is described in the text. The authors make a distinction in the text between the acquisition of a behavioral response and the performance of the response. We can learn a response by watching others (acquisition) but we may not perform the response. One of the reasons we may not perform a response is because we have seen the consequences of that response; this is known as vicarious learning. Recent studies of neuronal activity suggest that the activity of mirror neurons may be involved in observational learning.

What Is the Biological Basis of Reward?
Research investigating the rewarding properties of intracranial self-stimulation (ICSS) has identified dopamine as the neurotransmitter involved in reward. Regions in the brain that support ICSS, known as pleasure centers, overlap with known dopamine systems in the brain. The dopamine system involved in reward is the mesolimbic dopamine system, which projects from the ventral tegmental area to the nucleus accumbens. It is dopamine activity in the nucleus accumbens that is involved with the experience of pleasure we have when we engage in behavior. This section of the text concludes with a discussion of different types of drugs and drug addiction. Individuals may become addicted to drugs because they have developed either a physical or a psychological dependence on the drug. The use of drugs is associated with both positive and negative reinforcement. The pleasure associated with drug use is positively reinforcing, whereas the escape from the negative symptoms of withdrawal or the escape from problems and stress is negatively reinforcing.

How Does Learning Occur at the Neuronal Level?
Psychological scientists have long believed that learning must be associated with changes at the neuronal level, or as Donald Hebb put it, that “cells that fire together, wire together.” Eric Kandel's work with the aplysia (a sea slug) has demonstrated that habituation and sensitization, simple forms of learning, are associated with changes in the function of the synapse. This work has led to the study of long-term potentiation, the process by which repeated stimulation of one neuron leads to an increased likelihood of firing by a connecting postsynaptic neuron. In examining learning at a number of different levels, psychologists have also developed computer models of neural networks. These networks are often referred to as connectionist models of learning because they are based on the idea that neurons are connected with one another. Perhaps the best known connectionist model is the parallel-distributed processing model.