Reading B

In his social cognitive theory (also called social learning theory), Stanford psychologist Albert Bandura (1977, 1986, 1989, 2000, 2006) claims that humans are cognitive beings whose active processing of information plays a critical role in their learning, behavior, and development. Banudra argues that human learning is very different from rat learning because humans have far more sophisticated cognitive capabilities. He agrees with Skinner that operant conditioning is an important type of learning, but he notes that people think about the connections between their behavior and its consequences, anticipate the consequences likely to follow from their behavior, and often are more affected by what they believe will happen than by the consequences they actually encounter. Individuals also reinforce or punish themselves with mental pats on the back and self-criticism, and these cognitions also affect behavior. Bandura prefers to call his position social cognitive theory rather than social learning theory for a reason: to distance himself from behavioral learning theories like Watson’s and Skinner’s and to emphasize that his theory is about the motivating and self-regulating role of cognition in human behavior (Bandura, 1986).

By highlighting observational learning as the most important mechanism through which human behavior changes, Bandura made his cognitive emphasis clear. Observational learning is simply learning by observing the behavior of other people (called models). By learning from and imitating other people, children can master computer skills and math problems, as well as how to swear and smoke. Observational learning is regarded as a more cognitive form of learning than conditioning because learners must pay attention, construct and remember mental representations (images and verbal summaries) of what they saw, retrieve these representations from memory later, and use them to guide behavior.

In a classic experiment, Bandura (1965) set out to demonstrate that children could learn a response neither elicited by a conditioned stimulus (as in classical conditioning) nor performed and then strengthened by a reinforcer (as in operant conditioning). He had nursery school children watch a short film in which an adult model attacked an inflatable “Bobo” doll: hitting the doll with a mallet while shouting “Sockeroo,” throwing rubber balls at the doll while shouting “Bang, bang, bang,” and so on. Some children saw the model praised, others saw the model punished, and still others saw no consequences follow the model’s violent attack. After the film ended, children were observed in a playroom with the Bobo doll and many of the props the model had used to work Bobo over.

What did the children learn? The children who saw the model rewarded and the children in the no-consequences condition imitated more of the model’s aggressive acts than did the children who had seen the model punished. But interestingly, when the children who had seen the model punished were asked to reproduce all of the model’s behavior they could remember, it turned out they had learned just as much as the other children about how to treat a Bobo doll. Apparently then, through a process termed latent learning in which learning occurs but is not evident in behavior, children can learn from observation even though they do not imitate (perform) the learned responses. Whether they will perform what they learn depends partly on vicarious reinforcement (or punishment), a process in which learners become more or less likely to perform a behavior based on whether consequences experienced by the model they observe are reinforcing or punishing. You are more likely to imitate someone who is rewarded for their behavior than someone who is punished.

So important is observational learning in human development that children everywhere show a tendency to overimitate what they see, even then it is pretty useless. In a study demonstrating this (Nielsen & Tomaselli, 2010), children in both urban Australian and the Kalahari bush of Africa watched an adult demonstrate how to open a box. Children in both cultures readily imitated the adult model’s action of pushing a stick down on a knob on the front of the box, even though a more direct and effective way to open the box was to simply pull the knob with one’s hand. More surprisingly, children also imitated completely irrelevant actions such as waving the stick three times over the box like a magic wand before opening the box. Such overimitation is not as evident among chimpanzees, who imitate only actions that directly help them reach their goals (Horner & Whiten, 2005). Why do humans overimitate then? Possibly, Nielson and Tomaselli (2010) suggest, because it has proven adaptive and helps them learn the many – and often arbitrary rituals and rules of their culture.

Observational learning may be even more important in traditional societies than it is in modern ones (Rogoff et al., 2003). There, children learn not in schools where they are segregated from adults and given formal instruction, but through participation in everyday activities in which they actively observe and listen to their elders, learning skills such as weaving and hunting without the adults intentionally teaching them (Rogoff et al., 2003). In an interesting cross-cultural study (Correa-Chavez & Rogoff, 2009), Mayan children living in Guatemala, especially those whose mothers had had little Western schooling, were much more attentive while their siblings were taught to use a new toy than European American children from the United States were – and they learned more as a result. The Mayan children were used to learning by watching what was going on around them; the European American children seem to wait for adults to arrange learning experiences directed toward them personally.

In his later work, Bandura (2000, 2006) moved beyond the study of observational learning to emphasize ways in which people deliberately exercise cognitive control over themselves, their environments, and their lives. From the time they are infants recognizing that they can make things happen in their worlds, people form intentions, foresee what will happen, evaluate and regulate their actions as they pursue plans, and reflect on their functioning. These cognitions play a real casual role in influencing their behavior. For example, individuals develop a high or low sense of self-efficacy, or belief that they can effectively produce a particular desired outcome. Whether you try to lose 10 pounds or get an A on a test, your success depends greatly on whether you have a sense of self-efficacy concerning your ability to achieve your goal.

Watson and Skinner may have believed that people are passively shaped by environment to become whatever those around them groom them to be, but Bandura does not. Because he views humans as active, cognitive beings, he holds that human development occurs through continuous reciprocal interaction among the person (the individual’s biological and psychological characteristics and cognitions), his or her behavior, and his or her environment – a perspective he calls reciprocal determinism (Figure 2.3). As Bandura sees it, environment does not rule, as it did in Skinner’s thinking: people choose, build, and change their environments; they are not just shaped by them. And people’s personal characteristics and behaviors affect the people around them, just as other people are influencing their personal characteristics and future behaviors.

Like Watson and Skinner, however, Bandura doubts that there are universal stages of human development. He maintains that development is context specific and can proceed along many paths. It is also continuous, occurring gradually through a lifetime of learning. Bandura acknowledges that children’s cognitive capacities mature, so they can remember more about what they have seen and can imitate a greater variety of novel behaviors. Yet he also believes that children of the same age will be dissimilar if their learning experiences have differed considerably. Obviously there is a fundamental disagreement between stage theorists such as Freud and Erikson and learning theorists such as Watson, Skinner, and Bandura (see Table 2.3 for a summary of the three men’s distinct contributions to the learning perspective on development). Learning theorists do not provide a general description of the typical course of human development because they insist that there is no such description to give. Instead, they offer a rich account of the mechanisms through which behavior can change, using principles of learning that are universal in their applicability to explain how each individual changes with age in unique ways (Goldhaber, 2000). We imagine what Watson, Skinner, and Bandura would say about school refusal in Exploration Box 2.3.

Pavlov’s and Watson’s demonstrations of classical conditioning, Skinner’s work on operant conditioning, and Bandura’s modern social cognitive theory with its highlighting of observational learning have contributed immensely to the understanding of development and continue to be influential. Learning theories are precise and testable. Carefully controlled experiments have shown how people might learn everything from altruism to alcoholism. Moreover, learning principles operate across the life span and can be used to understand behavior at any age. Finally, learning theories have incredibly important applications; they have been the basis for many highly effective behavioral and cognitive behavioral techniques for optimizing development and treating developmental problems.

Still, behavioral learning theories, and even Bandura’s more recent social cognitive theory, leave something to be desired as explanations of human development. For instance, it has been demonstrated that reinforcing 3-month-old infants with smiles and gentle rubs on the chin whenever they happen to make babbling sounds, such as “bababa,” causes them to babble more often than infants who receive the same social stimulation randomly, rather than only after each babbling sound they make (Weisberg, 1963). But does this mean that infants normally begin to babble because babbling is reinforced by their caregivers? Not necessarily. All normal infants, even deaf ones, babble around 4 months of age. Moreover, no matter what learning experiences are provided to new-borns, they will not be maturationally ready to babble. We must suspect, then, that the maturation of the neural and muscular control required for babbling has more than a little to do with the onset of babbling during infancy.

This example highlights two criticisms of learning theories as theories of human development. First, learning theorists rarely demonstrate that learning is responsible for commonly observed developmental changes; they show through their experiments that learning might have resulted in developmental change, as in the case of reinforcement increasing the frequency of babbling. Some critics wish that learning theorists would provide a fuller account of normal changes across the life span. Second, learning theorists, even Bandura, probably put too little emphasis on biological influences on development, such as genetic endowment and maturational process. We may well learn to fear snakes, for example. However, probably because snakes were a threat to our ancestors, we have evolved so that we are biologically prepared to be wary of these critters. Thus, we learn to fear snakes far more easily than we learn to fear bunnies or flowers (Ohman & Mineska, 2003). Today’s learning theorists appreciate more than Watson and Skinner did that factors such as genetic endowment, previous learning, personality, and social context all affect how humans react to their learning experiences (Mineka & Zinbarg, 2006).