Operant conditioning
In operant conditioning, behaviour is shaped by its consequences. Behaviours followed by rewards become more likely; those followed by punishment become less likely.
Thorndike and Skinner
Edward Thorndike's law of effect (1898) stated that behaviours followed by satisfying consequences become more likely, based on cats learning to escape puzzle boxes. B. F. Skinner developed operant conditioning, studying animals in operant chambers, known as Skinner boxes, where rats pressed levers or pigeons pecked keys for food, with responses recorded precisely.
Reinforcement
Reinforcement increases a behaviour. Positive reinforcement adds something desirable after a behaviour, such as praise or money. Negative reinforcement removes something unpleasant, such as taking painkillers to stop a headache, or fastening a seat belt to stop a car's beeping. 'Negative' means removal, not that it is bad; both kinds strengthen behaviour.
Primary and conditioned reinforcers
Primary reinforcers are naturally satisfying, such as food, water, and relief from pain. Conditioned (secondary) reinforcers gain power through association with primary ones, such as money, grades, and praise. Token economies, used in classrooms and hospitals, give tokens for desired behaviours that can be exchanged for rewards.
Punishment
Punishment decreases a behaviour. Positive punishment adds something unpleasant, such as a scolding or a fine; negative punishment removes something desirable, such as taking away a phone or a time-out from play. Punishment can suppress behaviour quickly, but physical punishment is linked to increased aggression, fear, and poorer outcomes, and it teaches what not to do rather than what to do.
Schedules of reinforcement
Continuous reinforcement, rewarding every response, produces fast learning but quick extinction. Partial (intermittent) reinforcement produces slower learning but greater resistance to extinction. Fixed-ratio schedules reward after a set number of responses (piecework pay); variable-ratio after an unpredictable number (slot machines), producing high, persistent responding; fixed-interval after a set time (weekly pay checks); and variable-interval after unpredictable times (checking email).
Shaping
Shaping reinforces successive approximations of a desired behaviour. To teach a dog to roll over, a trainer first rewards lying down, then lying on its side, then rolling partly, and finally the full roll. Skinner famously trained pigeons to play a form of ping-pong. Shaping is used in animal training, teaching children skills, and rehabilitation.
Applications
Operant principles are used in applied behaviour analysis, including therapies for autism; in education, through immediate feedback and reinforcement; in workplaces, through incentives; in health, through rewards for exercise or quitting smoking; and in technology, where apps use variable rewards, such as unpredictable notifications and likes, which keep people checking frequently.
Worked example
A child whines for sweets at the checkout, and the parent gives in to stop the whining. The child's whining is positively reinforced (they get sweets), and the parent's giving in is negatively reinforced (the whining stops). If the parent gives in only sometimes, the child is on a variable-ratio schedule, making whining especially persistent. Consistently ignoring whining while praising polite requests would help extinguish whining and reinforce better behaviour.
Key takeaways
Thorndike's law of effect (1898) and Skinner's operant conditioning show behaviour is shaped by consequences, studied in Skinner boxes. Positive reinforcement adds something desirable; negative reinforcement removes something unpleasant; both increase behaviour. Primary reinforcers are innate; conditioned reinforcers such as money are learned; token economies use them. Positive punishment adds something unpleasant; negative punishment removes something desirable; both decrease behaviour, and physical punishment has harmful side effects. Schedules include continuous and partial: fixed-ratio, variable-ratio (most persistent), fixed-interval, and variable-interval. Shaping rewards successive approximations. Applications range from therapy to apps using variable rewards.
Flashcard
Rewarded behaviours become more likely
Check your understanding
Taking a painkiller to remove a headache, which makes you more likely to take painkillers again, is an example of:
