L11 - Extinction Flashcards

1
Q

What happens when the CS is no longer paired with the US?

A

Extinction - the conditioned response (CR) stops

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What does the R-W model assume the value of lambda is for non-reinforcement?

A

0

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What happens according to the R-W model, when the CS is presented without the US?

A

It generates a -ve discrepancy (US is expected but absent)

Lambda = 0, sigmaV = 1 
(0-1) = error term
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

When the CS is presented without the US, what happens to deltaV and V?

A

DeltaV is -ve and V begins to decrease

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Why does lambda normally equal 1 in examples?

A

Due to convenience and ease of the model

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How do you work out the true value of lambda when presented with a graph?

A

Find the asymptote
e.g. if asymptote = 17, then lambda = 17

We take the value of 1 in examples for convenience

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Does the R-W model predict extinction?

A

Yes

Learning (V) decreases as CS-US pairing is extinguished

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

What is a partial reinforcement schedule?

A

The CS and the US are not always followed by each other - only partially reinforced

Under reinforcement schedule of 100% - every time CS is presented, followed by US

Under partial rft schedule of 50% - on 50% of time CS is presented, followed by US

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Does the R-W model predict partial reinforcement schedules well?

A

Yes - the way it accounts for it is similar to extinction

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Explain how the R-W model predicts partial reinforcement learning

A

R-W predicts a reduction in associative strength similar to extinction when the CS is not always followed by the US - this is because deltaV is -ve

Partial reinforcement schedules show a lower asymptote for learning

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What does the R-W model predict for different learning schedules when they are being extinguished?

A

It predicts the highest asymptote takes the longest to decrease (i.e. continuous takes longer than partial because 100% is higher than 50%).

This is incorrect

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is the partial reinforcement effect?

A

Using a partial reinforcement schedule results in resistance to extinction - the R-W model does not account for this

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

How does the R-W model explain extinction and what are the 4 contradictions of this?

A

Explains it as unlearning the CS-US pairing, as if we are erasing the link

However, there are conditions that produce a recovery in responding following extinction:

Spontaneous recovery
Rapid reacquisition
Renewal
reinstatement of responding

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

What is spontaneous recovery?

A

Responding recovers to the CS simply by the passage of time

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is rapid reacquisition?

A

After extinction, conditioned responding to the CS is reacquired faster than before

Demonstrates that the CS-US link has not been completely erased

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

What is renewal?

A

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

17
Q

What are the 3 different groups in the experiment that shows evidence for renewal?

Bouton and King (1983)

A

3 groups: All conditioned by a CS to a US

All learn in context A

First group: CS is extinguished in context A

Second group: CS is extinguished in context B

Control: not extinguished at all but is presented in context B

1-2 different in context
2-3 different in extinction

18
Q

What are the results for the different groups in the renewal experiment?

A

After the initial conditioning suppression is high after the tone.

Gradually throughout extinction group A become less scared

Group B: RECOVERY OF FEAR IN CONTEXT WHERE IT WASN’T EXTINGUISHED

Control scared the whole time

19
Q

What is reinstatement?

A

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

Reminder of the shock is enough to scare the CS

20
Q

What is the explanation for extinction referring to “masking”?

A

Extinction is new learning about the CS that masks the original learning (but otherwise leaves it intact)

21
Q

How does Bouton explain extinction?

A

Suggests animals acquire memories of the CS, not just a single memory but MULTIPLE

22
Q

What is the key difference between Bouton’s explanation for extinction and the R-W model?

A

RW see it as single, Bouton sees it as different associations

After extinction, animals have 2 different memories with conflicting context

  1. memory of CS-US
  2. memory of CS alone (enters unambiguous state - remember second memory in that context)
23
Q

What do animals use to decide which memory to retrieve (Bouton)?

A

They remember the second memory in that context.

The animals are using temporal and visual cues to decide which memory to retrieve

24
Q

Are memories from extinction episodes situation specific according to Bouton?

A

Yes, they rely on the time and environment.

Under different context you will retrieve appropriate memory.