The behavioural lens

The lens in brief

Behaviourism began with John B. Watson in 1913 and was developed most fully by B. F. Skinner. It studies observable behaviour and explains it through the environment. Its central idea is operant conditioning: behaviour that is followed by reinforcement becomes more frequent, and behaviour followed by punishment becomes less frequent. Behaviourists avoid explanations based on thoughts or feelings, because these cannot be observed.

The Bobo doll study grew out of this tradition and used its methods. Its hypotheses are even written in behaviourist language: the authors expected copying because children "had learned imitative habits as a result of prior reinforcement" (p. 575). So the behavioural lens is not an outsider's view. It is the view of the family the study came from, and it reads the result more sceptically than the paper's discussion did.

How this lens reads the study

Bandura's headline claim was that the children learned new behaviour without any reinforcement. A strict behaviourist would accept the data and reject that interpretation. There are four places where reinforcement could be hiding.

1. The children's past. By age four, every child has been praised many times for copying adults: waving goodbye, saying "thank you", stacking blocks the way a parent showed. Skinner wrote in 1953 that imitation "develops in the history of the individual as the result of discriminative reinforcements", and that the habit becomes so well developed that "its origins are forgotten". A behaviourist calls the result generalised imitation: copying itself has become a reinforced habit, so the child now copies new acts too. The paper's own first hypothesis assumed exactly this. The study did not measure the children's learning history, so it cannot rule it out.

2. The doll itself. A Bobo doll is built to be hit. When you punch it, it wobbles and springs back up. That bounce is an interesting result in its own right, and an interesting result can work as reinforcement. Behaviourists call this automatic reinforcement, meaning the behaviour produces its own reward. Once a child copied one punch, the doll's movement may have kept the child going.

3. The absence of punishment. Children have learned that hitting usually means being scolded. In the experiment, the adult model hit the doll and nothing bad happened. In the test room, the experimenter sat at her desk and said nothing. Bandura called this disinhibition. A behaviourist would describe it more simply: the room signalled that punishment was not coming. Signals that tell us which consequences are available are called discriminative stimuli.

4. The frustration. Just before the test, each child had attractive toys taken away. Behavioural researchers showed in the 1960s that when animals suddenly stop receiving an expected reward, they often attack a nearby animal or object (Azrin, Hutchinson & Hake, 1966). They would treat the frustration step as a powerful cause in its own right.

What this lens values in the study

The study is a model of behavioural method. It defines aggression as specific acts, counts them in five-second intervals, checks agreement between two observers, and holds everything constant except the model's behaviour. A behaviourist would admire that. Because the acts were defined so precisely, other researchers could repeat the study, and they did.

What it would question

Bandura's discussion argued that imitation can "short-cut the acquisition of new behaviors without the necessity of reinforcing successive approximations as suggested by Skinner" (p. 580). A behaviourist would reply that the short cut was built by reinforcement earlier in life.

• The claim that no reinforcement was involved. As shown above, reinforcement may have come from the child's history, the doll, or the safe room.
• The use of an invisible process, memory, to explain the result. A strict behaviourist would say we only know that behaviour changed after exposure to a model.
• The single short test. Behaviourists study behaviour over many sessions to see whether a change lasts and what keeps it going.

Evidence from this tradition

Two lines of behavioural research speak directly to the study.

In 1967 Donald Baer, Robert Peterson and James Sherman worked with children with severe developmental disabilities who did not imitate at all. The researchers reinforced copying, one action at a time. After many trials the children began to copy new actions that had never been reinforced. This suggests that imitation can itself be learned through reinforcement, as the behavioural reading claims.

Bandura's own 1965 study, described on the Why it matters page, found that children copied less after seeing the model punished. When they were later offered rewards for copying, the differences disappeared. A behaviourist reads this as proof that consequences still control what children actually do. Bandura read it as proof that learning and doing are different. Both readings use the same data, which is a useful example for your journal.

A study this lens would design

A behavioural researcher would study a few children closely over several weeks, using a single-case design. The steps would be:

1. Record each child's aggression in the ordinary nursery for a week. This is the baseline.
2. Introduce a model, and vary what happens to the model: praise, a scolding, or nothing.
3. Vary what happens when the child copies, again praise, scolding or nothing.
4. Remove the change and see whether behaviour returns to baseline, then add it again.

If aggression rises and falls each time the consequences change, the researcher has strong evidence about what controls it in that child. The measure would be real behaviour towards other children and objects in the nursery, not a doll built to be hit.

Try this

In a replication, a child copies the model's punch, the Bobo doll swings back and taps the child's arm, and the child laughs and punches it again eleven more times.

Which of the four behavioural explanations fits this best, and what would you change in the study to test it?

Show the reasoning

This fits automatic reinforcement. The doll's movement, and the fun of it, follow each punch, and the punching speeds up. The first punch may be imitation, but the next eleven could be kept going by the consequence, not by the model.

To test this, compare two dolls that look the same: one that bounces back and one that is weighted so it barely moves. If children who watched the aggressive model still punch the still doll as often, imitation is doing the work. If punching drops sharply with the still doll, the doll's own reinforcing movement was a large part of the effect.