Translated from the Portuguese original. Read the original (Português)
In Behavior Analysis, punishment is any consequence that makes a behavior decrease in the future. It is the opposite of reinforcement, which increases behavior. There are two types: positive punishment (something is added, such as a reprimand) and negative punishment (something pleasant is removed, such as losing screen time). The name only tells you whether something was added or subtracted — not whether it is “good” or “bad.” And, in ethical practice, punishing is almost never the first resort.
In Applied Behavior Analysis (Cooper, Heron & Heward), punishment is the contingency in which a stimulus is presented or removed immediately after a response, resulting in a decrease in the future frequency of that response. Like reinforcement, punishment is defined by its effect on behavior — not by the intention of the person applying it, nor by whether it seems unpleasant.
Reinforcement and punishment, positive and negative: the four quadrants
Four words organize almost all operant learning. Two axes, easy to tell apart:
- Reinforcement × Punishment = what the effect is. Reinforcement increases behavior; punishment decreases it.
- Positive × Negative = what happened to the environment. Positive = something was added; negative = something was removed. As in math, the sign does not mean “good” or “bad.”
Crossing the two axes produces the four basic contingencies:
The two types of punishment
As with reinforcement, “positive” and “negative” only indicate whether a stimulus was added or subtracted after the behavior. Explore each type:
Positive punishment — adding something
After the behavior, a stimulus (usually aversive) is presented, and as a result the behavior decreases. Examples: a verbal reprimand right after the action, or overcorrection (redoing and fixing what was done). Note that “positive” here does not mean “good” — it only means that something was added to the environment.
Negative punishment — removing something
After the behavior, a pleasant stimulus (a reinforcer) is removed, and the behavior decreases. The two classic procedures are time-out (stepping away for a while from whatever was reinforcing) and response cost (losing a quantity of something valuable, such as points or screen time).
Tap the cards to see everyday examples — and what changes in each one:
Don't confuse it with negative reinforcementNegative punishment and negative reinforcement both remove something — which is why they get confused. But they have opposite effects: negative reinforcement increases behavior (it removes an aversive, such as pain that goes away), while negative punishment decreases it (it removes a reinforcer, such as screen time). The reinforcement/punishment axis is always about the effect.
Why punishing is not the first resort
Technically, punishment reduces behavior, but it has important limits — especially positive punishment, which uses aversive stimuli:
- It doesn't teach what to do instead. Punishment suppresses a response, but it doesn't build an adequate alternative behavior.
- It can have side effects. The literature describes emotional responses, aggression, and escape or avoidance of the person or environment associated with punishment.
- Its effect tends to be temporary. Without teaching a reinforced alternative, the behavior tends to come back.
- It raises ethical questions. Professional codes of ethics call for prioritizing the least intrusive strategies, based on reinforcement.
That is why ethical, evidence-based practice begins by understanding the function of the behavior and invests in positive reinforcement of alternative behaviors. See also challenging behaviors and how to manage them.
The preferred pathBefore thinking about reducing a behavior, you teach and reinforce what you want to see in its place. Strategies such as differential reinforcement make the problem behavior lose ground to a better alternative — without relying on aversives.
Defined by the effect, not by the intentionIf you “punish” someone and the behavior does not decrease, technically no punishment occurred — even if the consequence was unpleasant. Worse: sometimes a scolding ends up reinforcing the behavior, because it becomes attention. It is the effect on frequency that defines what happened.
Test your understanding
Quick quiz
Three questions to reinforce the concept. You'll see the explanation right away.
Further reading
- Cooper, J. O.; Heron, T. E.; Heward, W. L. Applied Behavior Analysis. 3rd ed., 2020 — chapters on positive and negative punishment.
- Skinner, B. F. Ciência e Comportamento Humano. 1953.
- Azrin, N. H.; Holz, W. C. “Punishment”. In: Honig, W. K. (ed.). Operant Behavior: Areas of Research and Application, 1966 — on the effects and side effects of punishment.
- Catania, A. C. Aprendizagem: Comportamento, Linguagem e Cognição. 4th ed.
This entry is educational and does not replace assessment, diagnosis or treatment by a qualified health professional. Every person is unique — for a specific case, seek individual guidance.