Operant conditioning

A rat presses a lever, a food pellet drops, and the rat presses again. B.F. Skinner built a whole science on that loop — and the NCE turns it into its single most notorious trap: the difference between negative reinforcement and punishment. Get the grid straight and you own the points.

Operant conditioning is learning by consequences. Unlike classical conditioning's reflexes, operant behavior is voluntary and emitted by an active learner. Thorndike's Law of Effect set the stage: behaviors followed by satisfying results are repeated. The whole system rests on two axes — are you adding something (positive) or removing something (negative), and does the behavior increase (reinforcement) or decrease (punishment)?

Cross those axes and you get four cells. Positive reinforcement: add a pleasant stimulus to increase behavior (praise a child, they study more). Negative reinforcement: remove an aversive stimulus to increase behavior (the seatbelt beep stops when you buckle, so you buckle). Reinforcement always increases behavior — negative reinforcement is not punishment. Positive punishment: add an aversive stimulus to decrease behavior (a speeding ticket). Negative punishment: remove a pleasant stimulus to decrease behavior (taking away a phone — this is response cost / time-out logic).

Two more testable pieces. Shaping builds a new complex behavior by reinforcing successive approximations — rewarding closer and closer steps toward the target. Extinction in operant terms means stopping the reinforcement that maintained a behavior; watch for an extinction burst, a temporary spike in the behavior before it fades.

Clinical Pearl: "Positive/negative" means add/remove, NOT good/bad; "reinforcement/punishment" means increase/decrease. Negative reinforcement removes something to increase behavior — it is the classic distractor for punishment.