Lecture 11 - Extinction

Question 1

What does lambda equal when there is no reinforcement (extinction)

Answer 1

Lambda = 0

Nothing is going to happen if there is no prior learning.

Question 2

What will sigma V equal if learning has already occurred?

Answer 2

Sigma V = 1

Question 3

Define extinction

Answer 3

Essentially extinction is unlearning.

Extinction is NOT unlearning, because prior learning survives extinction and responding can be restored.

According to Bouton, memories from extinction episodes are situation-specific.

Question 4

Partial reinforcement effect on 100%, 50% and 25% reinforced groups?

Answer 4

The 100% reinforced group is going to extinguish at a much faster rate than the 50% and 25% reinforced groups.

This is because the animals are more likely to notice food reinforcement stopping because they are used to receiving it after every trial; whereas the thee two groups weren’t receiving every time anyway.

R-W model has no way of dealing with or explaining this.

Question 5

Define spontaneous recovery

Answer 5

Extinguished responding can spontaneously recover after a waiting period.

Question 6

Define rapid reacquisition

Answer 6

After extinction, conditioned responding is reacquired faster than before.

However, there is a massive problem with relapse (quitting smoking).

Question 7

Define renewal

Answer 7

After extinction of responding in one context, responding can be renewed if the CS is presented in a different context.

Renewal is quite context specific and extinction doesn’t seem to transfer across contexts very well.

Conditioning seems to transfer very nicely across contexts, by doesn’t generalise back again very well.

Question 8

Define re-instatement

Answer 8

After extinction, responding can be reinstated if the US is presented alone.

Simple exposure can re-instate a lot of conditioning.