The Law of Effect
In operant conditioning, behavior is shaped by its consequences, and the schedule of reward controls how fast and how stubbornly the behavior persists. · 12 min
Classical conditioning explains reflexes — responses your body gives to a signal. But most of what you do is not a reflex. You study, or you do not. You check your phone, or you do not. These are actions, and actions are governed by what follows them. Edward Thorndike put it plainly in his law of effect: responses followed by satisfaction are repeated; responses followed by discomfort are not. B. F. Skinner turned that law into a science of consequences — the study of how rewards and costs shape behavior over time.
Guess before you learn
Two machines stop paying out. One had always paid every single time you used it, like a vending machine; the other paid only now and then, unpredictably, like a slot machine. At which machine will people keep trying LONGER after the payouts quietly stop?
People persist far longer at the unpredictable machine. When a reward came every single time, its absence is obvious and quick to notice — so you stop. When rewards were already irregular, a dry spell feels normal, and you keep going. Psychologists call this the partial reinforcement effect, and it is why unpredictable rewards build the most stubborn habits. You will rank the schedules yourself in a moment.
9–12
3–5
Your actions have results, and the results teach you. A reward makes you want to do the thing again. A cost makes you want to stop. A reward can be getting something good, like praise, or escaping something bad, like a chore going away. Either way, the action that paid off is the one you repeat.
6–8
Operant conditioning shapes voluntary actions through their consequences. Reinforcement makes a behavior more likely; punishment makes it less likely. Each comes in two flavors. Positive means something is added; negative means something is removed. Positive reinforcement: you get praise. Negative reinforcement: an annoying alarm shuts off. Positive punishment: you get a fine. Negative punishment: your phone is taken away.
Notice that 'negative' never means 'bad' here — it means subtracted. The only question that fixes the word is whether the behavior went up (reinforcement) or down (punishment).
9–12
Operant conditioning turns on a 2-by-2. One axis is direction: reinforcement raises a behavior's future rate, punishment lowers it. The other axis is operation: positive adds a stimulus, negative removes one. Crossing them gives four cases — positive and negative reinforcement, positive and negative punishment — and every consequence you meet falls into one of them.
To classify any case, ask two questions in order. Did the behavior become more or less likely? That fixes reinforcement versus punishment. Was something added or taken away? That fixes positive versus negative. Answer both and the label is forced.
K–2
Do something and get a treat, you do it again. Do it and lose a toy, you do it less. What happens right after you act teaches you whether to do that thing again.
Undergrad
Thorndike's law of effect became Skinner's operant framework: behavior is selected by its consequences much as traits are selected by fitness. Reinforcement and punishment are defined functionally, by their effect on response rate — not by whether the outcome seems pleasant. A stimulus is a reinforcer if and only if it raises the behavior it follows: circular in appearance, but testable in practice.
Skinner's deeper move was the three-term contingency: discriminative stimulus, then response, then consequence. Behavior is not emitted at random but under stimulus control — a signal marks when a response will pay — which is how consequences build context-sensitive action rather than blind reflex.
Postgrad
The functional definition invites Herrnstein's matching law: on concurrent schedules, relative response rates match relative reinforcement rates, so an organism allocates behavior in proportion to the reward each option returns. This reframes choice as continuous allocation rather than discrete decision, and grounds behavioral economics, where reinforcers trade off along demand curves with their own elasticities.
Punishment is theoretically asymmetric with reinforcement, not its mirror image. It suppresses without teaching an alternative, its effects are often temporary and context-bound, and it can condition fear of the punisher. Applied behavior analysis therefore leans on differential reinforcement of an alternative behavior — a design choice that follows directly from the theory.
reinforcement
Any consequence that makes the behavior it follows more likely to recur. It is defined by that effect, not by whether it happens to feel good.
The labels trip almost everyone at first, because 'negative' sounds like 'bad.' Hold to the two questions: more or less likely fixes reinforcement or punishment; added or removed fixes positive or negative. Work one all the way through before you trust yourself on the rest.
Classify a consequence — the steps fade as you master them
More likely — so this is reinforcement
The beeping was removed
Negative reinforcement
Why is this true?
Why is switching off an alarm called reinforcement, when nothing pleasant was handed over?
Because reinforcement is defined by its effect, not its feel. Buckling up became more likely, so the consequence reinforced it. Taking away something unwanted — the noise — is exactly what makes it negative reinforcement.
Consequences shape whether a behavior happens. Schedules of reinforcement shape how it happens — how fast and how persistently. Reward can arrive on a ratio (after a number of responses) or an interval (after an amount of time), and either can be fixed (a set amount) or variable (an unpredictable one). A factory paying per hundred parts uses a fixed ratio. A slot machine uses a variable ratio: it pays after some unpredictable number of pulls. Variable schedules produce the highest, steadiest rates and the greatest resistance to extinction — because when reward is already irregular, its absence is hard to notice. That is the engine under gambling, and under a phone that only sometimes brings news.
Consequences are doing quiet work all day. The habits that stick hardest are rarely the ones rewarded every time — they are the ones rewarded just often enough, and never on a schedule you can predict. Seeing that is the first step toward arranging your own consequences on purpose, instead of letting them arrange you.
Note
The next folio turns from what shapes an action to what feels like the push behind it — emotion and motivation.
Practice — new ink and old, interleaved
1.After watching several news reports about plane crashes, Maya feels flying is more dangerous than driving and takes the car on a long trip. Which shortcut is steering her?
2.Match each part of the neuron to its job.
3.In one sentence, explain why a variable-ratio schedule builds such persistent behavior.
4.Which of these is operant conditioning, not classical conditioning?
5.You put on sunglasses and the painful glare disappears, so you reach for sunglasses more often on bright days. Which consequence is at work?