Skip to main content

Verified by Psychology Today

Punishment

Common Misconceptions about Science VI: “Negative Reinforcement”

Negative reinforcement ≠ punishment!

Another one of my pet peeves is the fact that many people – civilians and scientists alike – use the phrase “negative reinforcement” to mean “punishment.” The two are not at all the same; in fact, they are polar opposites.

I have always known that many people think that “negative reinforcement” means punishment. But I was reminded of the enormity and pervasiveness of the problem when I watched a recent episode of The Big Bang Theory, in which the character of Sheldon Cooper makes the same mistake. In the show, Cooper is depicted as a real genius with an IQ of 187, a former child prodigy who graduated from college at the age of 11 and earned the first of his two Ph.D.s at the age of 16. He is generally depicted as having encyclopedic knowledge of all sciences. I suppose this means that the writers of The Big Bang Theory are not quite as conversant in all sciences as the character of Sheldon Cooper is portrayed to be, which is a surprise because they usually appear to be quite knowledgeable about science in general.

I was once a behaviorist (just as I was once a Marxist and a feminist), so this widespread confusion about what exactly “negative reinforcement” is really bothers me, even though behaviorism in psychology is pretty much dead now and thus it really couldn’t matter less any longer.

In the language of operant (Skinnerian) conditioning, which along with respondent (Pavlovian or classical) conditioning forms the twin pillars of behaviorism, positive means onset or commencement whereas negative means end or termination. Positive and negative have no other meanings in behaviorism; in particular, unlike in math, “negative” does not mean “opposite.” Reinforcement is a procedure that increases the probability that a response will occur, whereas punishment is a procedure that decreases the probability that a response will occur.

Positive reinforcement consists of the onset of a rewarding stimulus which increases the likelihood of a target response. So giving a child a dessert after he finishes the vegetables is an example of positive reinforcement. The dessert will make it more likely that the child will eat the vegetables again.

Negative reinforcement consists of the termination of an aversive stimulus which increases the likelihood of a target response. So reducing the prison sentence for good behavior is an example of negative reinforcement. The reduced sentence will make it more likely that the prisoner will engage in good behavior again.

Positive punishment consists of the onset of an aversive stimulus which decreases the likelihood of a target response. So giving a prison sentence for committing a crime is an example of positive punishment. The prison sentence will make is less likely that the convict will commit the crime again.

Negative punishment consists of the termination of a rewarding stimulus which decreases the likelihood of a target response. So taking away the dessert that has been served to a child after he hits his sister at the dinner table is an example of negative punishment. The lost privilege of eating the dessert will make it less likely that the child will hit his sister at the dinner table again.

Note that both positive reinforcement and negative reinforcement are reinforcement and thus increase the probability that a response will occur. Both positive punishment and negative punishment are punishment and thus decrease the probability that a response will occur. So even though many people use “negative reinforcement” to mean “punishment” (thinking that “negative” must mean “opposite” so “negative reinforcement” must mean “the opposite of reinforcement”), they are actually polar opposites and have opposite consequences. Negative reinforcement rewards the behavior by removing an aversive stimulus and increases the probability of its future occurrence, whereas punishment (both positive and negative) punishes the behavior and decreases the probability of its future occurrence.

Please honor the memory and work of B. F. Skinner and please do not desecrate it by using negative reinforcement and punishment synonymously!

advertisement
About the Author
Satoshi Kanazawa

Satoshi Kanazawa is an evolutionary psychologist at LSE and the coauthor (with the late Alan S. Miller) of Why Beautiful People Have More Daughters.

More from Satoshi Kanazawa
More from Psychology Today
More from Satoshi Kanazawa
More from Psychology Today