182 CHAPTER 5
Why does the removal of an unconditioned stimulus lead to extinction of the condi-
tioned response? One theory is that the presentation of the CS alone leads to new learning.
During extinction, the CS–UCS association that was learned is weakened, as the CS no
longer predicts the UCS. In the case of Pavlov’s dogs, through extinction they learned to
not salivate to the metronome’s ticking, as it no longer predicted that food was on its way.
Look back at Figure 5.1. Once conditioning is acquired, the conditioned stimulus (CS)
and conditioned response (CR) will always come before the original unconditioned stimu-
lus (UCS). The UCS, which comes after the CS and CR link, now serves as a strengthener,
or reinforcer, of the CS–CR association. Remove that reinforcer, and the CR it strengthens
will weaken and disappear—at least for a while.
The term extinction is a little unfortunate in that it seems to mean that the original
conditioned response is totally gone, dead, never coming back, just like the dinosaurs.
Remember the definition of learning is any relatively permanent change in behavior. The
fact is that once people learn something, it’s almost impossible to “unlearn” it. People
can learn new things that replace it or lose their way to it in memory, but it’s still there. In
the case of classical conditioning, this is easily demonstrated.
After extinguishing the conditioned salivation response in his dogs, Pavlov waited
a few weeks, putting the conditioned stimulus (i.e., the metronome) away. There were no
more training sessions, and the dogs were not exposed to the metronome’s ticking in that
time at all. But when Pavlov took the metronome back out and set it ticking, the dogs all
began to salivate, although it was a fairly weak response and didn’t last very long. This
brief recovery of the conditioned response proves that the CR is “still in there” somewhere
(remember, learning is relatively permanent). It is just suppressed or inhibited by the lack of
an association with the unconditioned stimulus of food (which is no longer reinforcing or
strengthening the CR). As time passes, this inhibition weakens, especially if the original
conditioned stimulus has not been present for a while. In spontaneous recovery the condi-
tioned response can briefly reappear when the original CS returns, although the response is
usually weak and short lived. See Figure 5. 3 for a graph showing both extinction and spon-
taneous recovery.spontaneous recovery
the reappearance of a learned
response after extinction has occurred.
Figure 5.3 Extinction and Spontaneous Recovery
This graph shows the acquisition, extinction, spontaneous recovery, and reacquisition of a conditioned
salivary response. Typically, the measure of conditioning is the number of drops of saliva elicited by the CS
on each trial. Note that on the day following extinction, the first presentation of the CS elicits quite a large
response. This response is due to spontaneous recovery.