- Advanced
- About 16 min
- 5 practice questions
Conditioning: Pavlov and Skinner compared
Analyse classical and operant conditioning cases precisely, separate negative reinforcement from punishment, and predict the effect of reinforcement schedules.
If something looks wrong, please tell us.
Behaviourist theories explain learning as connections between stimuli, responses and consequences. Two of them, Ivan Pavlov's classical conditioning and B. F. Skinner's operant conditioning, produce some of the hardest questions in teaching aptitude papers, because the options use near-identical terms. This lesson works at the level of precise analysis: labelling every part of a case and spotting the classic traps.
Classical conditioning: learning by association
Pavlov, a Russian physiologist, noticed that his dogs salivated not only when food was in the mouth but also at the sight of the attendant who brought it. In his experiments a neutral sound, such as a bell, was presented just before food several times. Eventually the sound alone produced salivation. The four labels are: unconditioned stimulus (UCS), the food, which naturally causes the response; unconditioned response (UCR), salivation to food; conditioned stimulus (CS), the once-neutral sound; and conditioned response (CR), salivation to the sound alone.
| Process | What happens |
|---|---|
| Acquisition | CS is paired with UCS until CS alone produces the CR |
| Extinction | CS is given repeatedly without UCS, so the CR weakens and disappears |
| Spontaneous recovery | After a rest, an extinguished CR briefly returns |
| Generalisation | Stimuli similar to the CS also produce the CR |
| Discrimination | The learner responds to the CS but not to similar stimuli |
Classical conditioning explains how involuntary, emotional responses become attached to situations, such as fear of a subject after repeated humiliation in its class.
Operant conditioning: learning by consequences
Skinner studied rats and pigeons in a box where pressing a lever or pecking a key could bring food. Here the learner operates on the environment, and the consequence decides whether the behaviour becomes more or less frequent. Reinforcement always increases behaviour: positive reinforcement adds something pleasant (praise, a grade), while negative reinforcement removes something unpleasant (a nagging alarm stops when the seat belt is fastened). Punishment always decreases behaviour. Skinner also developed shaping, rewarding successive approximations to the target behaviour, and programmed instruction, small steps with immediate feedback.
Reinforcement, positive or negative, strengthens behaviour; punishment weakens it. 'Negative' means something is taken away, not that the outcome is bad.
Schedules of reinforcement
With continuous reinforcement every correct response is rewarded; it is best for establishing a new behaviour but the behaviour extinguishes quickly when rewards stop. With partial reinforcement only some responses are rewarded, giving slower learning but stronger persistence. A fixed-ratio schedule rewards after a set number of responses; variable-ratio after an unpredictable number, giving high, steady responding that is the most resistant to extinction; fixed-interval rewards the first response after a set time, producing a pause after each reward and a burst before the next; variable-interval after unpredictable times, giving slow, steady responding.
Worked example 1
A child was once bitten by a dog and now feels afraid when she sees any four-legged animal, even a calf. Label the case. Working: the bite (UCS) caused fear (UCR). The dog, paired with the bite, became the CS, and fear of dogs is the CR. Fear spreading to similar animals is generalisation. Answer: classical conditioning with stimulus generalisation.
Worked example 2
A teacher stops giving extra homework to any student who completes class work neatly, and neat work increases. Is this punishment? Working: extra homework is unpleasant; it is removed; the behaviour (neat work) increases. Removal plus increase is negative reinforcement. Answer: negative reinforcement, not punishment. Shortcut: first ask whether the behaviour goes up or down; up is always reinforcement.
Key takeaways: label UCS, UCR, CS and CR before answering; extinction comes from CS without UCS; reinforcement raises behaviour and punishment lowers it; continuous reinforcement teaches fast, partial reinforcement lasts longer, and variable-ratio is the most persistent.
Check your understanding
-
1. In Pavlov's experiment, salivation produced by the sound alone after training is the:
- Unconditioned response
- Conditioned stimulus
- Conditioned response
- Unconditioned stimulus
Show answer
Answer: C. Salivation to the once-neutral sound is learnt, so it is the conditioned response. Salivation to food is the unconditioned response.
-
2. After conditioning, the bell is rung many times without food and salivation gradually stops. This is:
- Spontaneous recovery
- Generalisation
- Extinction
- Shaping
Show answer
Answer: C. Presenting the CS repeatedly without the UCS weakens the CR until it disappears: extinction.
-
3. A student's headache goes away when he takes a tablet, so he takes the tablet more readily in future. This is:
- Negative reinforcement
- Positive reinforcement
- Punishment
- Extinction
Show answer
Answer: A. Something unpleasant (headache) is removed and the behaviour (taking the tablet) increases: negative reinforcement.
-
4. Which schedule of reinforcement produces behaviour that is the most resistant to extinction?
- Continuous
- Fixed interval
- Variable ratio
- Fixed ratio
Show answer
Answer: C. Under variable ratio the learner never knows which response will be rewarded, so responding stays high and persists longest when rewards stop.
-
5. Rewarding a shy child first for standing near a group, then for saying hello, then for joining the game is an example of:
- Extinction
- Discrimination
- Spontaneous recovery
- Shaping
Show answer
Answer: D. Rewarding successive approximations to a target behaviour is shaping, a technique from Skinner's operant conditioning.
Sign in to keep track of the lessons you finish.