Rewards and Punishments
A second type of conditioning is based on the principle that rewarding -or reinforcing- certain behavior encourages its repetition. Similarly, the absence of reward or the use of punishment tends to decrease the probability of behavior being repeated. This learning from consequences is known by psychologists as operant or instrumental conditioning. B.F. Skinner is the name most closely associated with operant conditioning.
The controversial Harvard psychologist and his colleagues have shown time and again that most behavior is related to the rewards and punishments it has produced in the past.
Take the child who is rewarded for being pleasing and charming-and compare her with her contemporary, who gets her way by being obnoxious and throwing temper tantrums. The parent of the second child tends to yield to aggressive behavior 'for peace', thus rewarding her by giving in to her demands. A 'learned' pattern of behavior is being set up- it can last well into adulthood.
We are all familiar with the application of operant conditioning in the animal world. The circus animal act is evidence enough.
Here the elephant, say, is consistently rewarded for his antics appeared sometimes to the monkey's left, sometimes to his right. It took a little time for the monkey to learn that green was the clue to success. Then Harlow changed the problem, and the monkey had to learn that a triangular lid, wherever it was, concealed the raisin. In the end the monkey needed only one trial to discover which lid covered the prize, even when new shapes or colors were introduced.
Quick results, and carefully rewarding the correct response, are the two chief lessons of operant conditioning research.
Practice and repetition, while essential, are not enough. Whether you are a novice pianist or about to tackle an advanced study by Bach; whether you are a newly-wed learning to cook for the first time or a Cordon Bleu learning a new-style cuisine, shaping is how you learn.
You start by learning the general rules or outline of what you want to do-the basic skill of finger movements, the sequence of notes, the method of combining ingredients or the special use of soya beans and quick frying. You monitor your progress carefully.
If you are a beginner you will probably have a teacher who rewards correct moves with praise. Your teacher should also extinguish false moves by ignoring them or by helping you to distinguish between right and wrong responses. Gradually, as your grasp of the new skill becomes better and you begin to see the next step, you should be able to reinforce yourself by recognizing which responses are correct and which are mistaken.
Making Your Mistakes Better
If you find yourself making mistakes when you are learning something new, stop. You are learning to incorporate the mistakes. In these circumstances, practice will make you worse, not better, and your mistakes will become 'perfect'. Start again, but practice very slowly and carefully, so that you do not make mistakes. Speed up again only in stages, keeping fault-free all the time. Whenever you make a mistake, drop a notch or two in speed. Do not try to reach target speed in a single attempt. Take a break and do something else when you are about halfway home.
This is so that satiation (a condition in which you are satisfied with your progress and will not continue because you have had enough of the effort involved in any learning task) and inhibition (suppression or restraint of your behavior) do not set in and interfere with good learning.
You can, however, use satiation to eliminate mistakes. For example, some concert pianists who play the same pieces many times find that mistakes gradually creep in. Practicing correctly is no good to them: they already know the music. So they practice the mistakes over and over again. After a while satiation sets in and the mistakes actually become more difficult to play than the correct version.
Attention, approval, affection...
'Flattery will get you anywhere,' we say flippantly. Yet, almost without realizing it, we are talking about the use of three powerful social reinforcers-attention, approval and affection. It is the application of operant conditioning to social life.
In an experiment some years ago, a college student was observed while engrossed in a supposedly informal conversation with the experimenter. Unbeknown to the student, the experimenter had decided in advance to reinforce all statements of opinion made by the subject, such as sentences beginning 'I think...' or 'I believe...'. The reinforcement was the experimenter's saying 'You're right,' 'I agree' and 'That's so'.
In another part of the experiment, extinction was carried out by silence following a statement of opinion.
The student's statements of opinion showed a marked increase in frequency when verbal reinforcement was carried out. They decreased following extinction. The student remained entirely unaware of the whole process.
Continue reading: Man or Machine