3.7 Classical Conditioning

Behaviorism grew out of research on conditioning. Early behaviorists like John B. Watson argued that psychology should study only observable behavior, not hidden mental processes, because behavior can be measured objectively. Classical conditioning, discovered by Ivan Pavlov, was their first major tool. Learning, in general, is a relatively lasting change in behavior due to experience.

Ivan Pavlov noticed that his dogs began to drool before food ever reached their mouths, at sights and sounds that usually came before feeding. Studying this, he identified five parts of classical conditioning.

• Neutral stimulus (NS): something that doesn't trigger the response at first, like a tone before conditioning.
• Unconditioned stimulus (UCS): something that naturally triggers a response without learning, like food in a dog's mouth.
• Unconditioned response (UCR): the natural, unlearned response to the UCS, like salivating to food.
• Conditioned stimulus (CS): the formerly neutral stimulus that now triggers the response after being paired with the UCS, like the tone.
• Conditioned response (CR): the learned response to the CS, like salivating to the tone.

The UCR and CR are usually the same behavior (salivating). What changes is what triggers it: the UCS (unlearned) or the CS (learned). Acquisition is the initial learning stage, when the NS and UCS are paired. Order matters: learning works best when the neutral stimulus comes just before the UCS, so it predicts what's coming. You won't be tested on the names of different timing arrangements (such as delayed, trace, simultaneous or backward conditioning). After a response is acquired, several things can happen to it.

• Extinction: if the CS is repeatedly presented without the UCS, the CR weakens and eventually stops.
• Spontaneous recovery: after extinction and a rest period, the CS is presented again and the extinguished CR suddenly reappears, usually weaker.
• Generalization: responding to stimuli similar to the CS, like a dog salivating to a slightly higher tone.
• Discrimination: learning to respond only to the CS and not to similar stimuli that weren't paired with the UCS.
• Higher-order conditioning: a CS acts like a UCS to condition a new neutral stimulus. If a tone predicts food, and a light is then paired with the tone, the light can come to trigger salivation too.

Emotional responses can be classically conditioned. In Watson and Rayner's 'Little Albert' study (1920), a baby was shown a white rat (NS) while a loud noise (UCS) frightened him (UCR). Soon the rat alone (CS) made him cry (CR), and the fear generalized to other furry things. The study would be considered unethical today: it caused distress, and the fear was never removed. Similar learning helps explain some phobias, and therapies use counterconditioning, pairing the feared stimulus with a new, incompatible response like relaxation (5.5).

Taste aversion is a learned avoidance of a food after it's followed by sickness. John Garcia's research with rats showed two surprises. First, it's one-trial learning: a single pairing is enough, and it can form even when sickness comes hours later. Second, it shows biological preparedness: animals learn some associations more easily than others. Rats easily linked a taste with nausea but not a light or sound with nausea. This makes evolutionary sense, since tastes reliably predict food poisoning.

Habituation is a decrease in response to a stimulus that repeats or continues, like no longer noticing a ticking clock. It's the simplest form of learning and isn't the same as extinction, which requires a previously conditioned response.