Reinforcement increases the likelihood that a behavior will happen again. This is the single most consistent finding in behavioral science, dating back more than a century: when a behavior produces a favorable outcome, the connection between that situation and that behavior strengthens, making repetition more probable. The size of this effect depends on timing, consistency, and the type of reinforcement involved.
The Basic Principle
In 1911, psychologist Edward Thorndike formalized what he called the law of effect after watching cats learn to escape puzzle boxes. His principle states that responses followed by satisfaction become more firmly connected to the situation that triggered them, so they are more likely to recur. Responses followed by discomfort have their connections weakened and become less likely to occur. The greater the satisfaction or discomfort, the greater the strengthening or weakening of that bond.
Modern behavioral science builds on this foundation with a more precise definition. To call something a reinforcer, you must observe a subsequent increase in the behavior it was tied to. If the behavior doesn’t increase, whatever followed it wasn’t actually reinforcing, regardless of how pleasant it seemed. This distinction matters because what works as a reinforcer varies from person to person and moment to moment. Praise reinforces some people’s work habits; for others, it has no measurable effect.
Positive and Negative Reinforcement
Both positive and negative reinforcement increase behavior, but they work through different mechanisms. Positive reinforcement involves the production of a stimulus after a behavior. A dog sits, then receives a treat. A student answers correctly, then hears applause. The new stimulus strengthens the behavior.
Negative reinforcement involves the removal of a stimulus after a behavior. You take a painkiller and the headache disappears, making you more likely to reach for that painkiller next time. A child cleans their room to stop a parent’s nagging. The key phenomena that define negative reinforcement are escape and avoidance: you either end something unpleasant that’s already happening or act to prevent it from starting. Both types share the same outcome, a behavior that’s more likely to repeat, but they arrive there through different routes. Importantly, negative reinforcement is not punishment. Punishment decreases behavior; negative reinforcement increases it.
What Happens in the Brain
Reinforcement has a biological basis in a brain network called the mesolimbic dopamine system, which plays a central role in motivated behavior, reward processing, and learning. When you experience something rewarding, neurons in a midbrain region fire in short bursts, releasing dopamine into areas involved in decision-making, memory, and motor control. This dopamine release promotes active seeking behaviors, essentially driving you to pursue goals and repeat actions that previously led to good outcomes.
This system responds to more than just pleasure. Dopamine is also released in response to stress and unpleasant events, which helps explain why negative reinforcement works too. Under both positive and negative emotional conditions, dopamine promotes a generalized behavioral arousal, sometimes described as seeking safety or seeking reward depending on the context. The system doesn’t just stamp in habits. It regulates how neural activity patterns flow through circuits connecting deeper brain structures to the cortex, shaping which actions get prioritized in a given situation.
Timing Changes Everything
The delay between a behavior and its reinforcer has a powerful effect on how strongly that behavior is learned. Research comparing short and long reinforcement delays found that when the reinforcer arrived quickly (within about 10 seconds), responding was consistently higher than when the reinforcer was delayed (around 40 seconds). This held true even when the total amount of reinforcement was the same.
When researchers matched the delay but varied how often reinforcement occurred, the difference in responding largely disappeared. In other words, how soon the reinforcer follows the behavior matters more than how frequently reinforcement happens overall. This has practical implications: if you’re trying to build a habit or teach a skill, delivering the reinforcer as close to the target behavior as possible produces the strongest connection. A compliment given right after a child shares a toy is more effective than one given at the end of the day.
How Schedules Shape Persistence
Not every instance of a behavior needs to be reinforced for that behavior to persist. In fact, behaviors that are reinforced only some of the time are more resistant to extinction (the gradual fading of a behavior when reinforcement stops) than behaviors reinforced every single time. This is one of the more counterintuitive findings in behavioral science.
Think of it this way: if you check your phone and always find a new message, you’ll stop checking quickly once messages dry up. But if messages arrived unpredictably, you’ll keep checking for much longer after they stop, because the absence of a message on any given check is normal. This is why slot machines, social media feeds, and other variable reinforcement systems are so effective at maintaining behavior. The unpredictability itself generates persistence.
Fixed schedules, where reinforcement arrives after a set number of responses or a set amount of time, tend to produce pauses right after reinforcement. You finish a project, get paid, and take a break before starting the next one. Variable schedules, where the number of responses or time between reinforcers changes unpredictably, produce steadier, more consistent responding with fewer pauses.
Reinforcement Versus Punishment
Punishment can suppress a behavior in the short term, but reinforcement tends to produce more durable changes. Punishment may increase obedience temporarily, but it also carries longer-term costs: it can increase aggression and damage the relationship between the person delivering punishment and the person receiving it. A child punished for lying may learn to lie more carefully rather than to tell the truth.
Reinforcement builds behavior by creating a positive association, which means the person is motivated to repeat the action rather than simply afraid of consequences. This is why reinforcement-based approaches dominate modern therapeutic practice. In applied behavior analysis for children with autism, for example, reinforcement strategies led to clinically meaningful improvements in adaptive behavior for 58% of children within 12 months. Children who started with the lowest skill levels showed the greatest gains, averaging a 9-point improvement in adaptive behavior scores over 24 months. These results came even though only 28% of children in the study received the full recommended amount of therapy.
When Reinforcement Backfires
There is a well-documented scenario where adding external reinforcement actually decreases a behavior over time. The overjustification effect occurs when you reward someone for doing something they already enjoy. Once the reward is removed, their interest in the activity drops below where it started.
In a classic study, college students were paid money for completing tasks they found interesting. After the payments stopped, their performance declined compared to a group that was never paid. Interestingly, students who received verbal praise instead of money actually showed increased performance after the praise ended. In another study, children who expected a “good player” award for an activity they liked spent less time on that activity after the award was discontinued. Children who received the same award unexpectedly showed no such decline.
The pattern is clearest when three conditions align: the activity already has high intrinsic interest, the reward is tangible rather than verbal, and the person expects the reward before performing the behavior. A meta-analysis confirmed that overjustification effects are most likely when baseline interest in the task is already high. From a practical standpoint, this means external rewards are most useful for building behaviors that aren’t yet self-sustaining. For activities someone already loves, verbal encouragement tends to be safer than material incentives.
Practical Takeaways for Shaping Behavior
Several principles emerge from the research that apply whether you’re training a pet, teaching a child, or trying to change your own habits. Deliver reinforcement as quickly as possible after the target behavior. Use variable, unpredictable reinforcement schedules when you want a behavior to persist long after active reinforcement ends. Rely on reinforcement rather than punishment when you want lasting change, because reinforcement builds behavior while punishment only suppresses it temporarily and often creates side effects.
Be thoughtful about reinforcing activities that are already intrinsically motivating. If someone already enjoys reading, paying them per book may undermine that enjoyment once payment stops. Verbal recognition or unexpected rewards carry less risk of this effect. And remember that what counts as a reinforcer is defined entirely by its effect on behavior, not by your assumptions. If a behavior isn’t increasing, the consequence you’re providing isn’t functioning as reinforcement, no matter how rewarding you think it should be.