Learning

Question 1

Purpose of learning

Answer 1

Need a representation of the world it can’t be innate, we have to build that representation by learning what goes with what associations

Question 2

Salient stimuli

Answer 2

Lead to better learning CS examples: bright light vs. dim light, loud noises vs. quiet noises. US examples: high valued food vs low value food. Describes how meaningful and attention giving a stimulus is e.g., loud horn is played then you are given a thousand dollars you will learn to associate the sound with the reward faster than if someone sniffed and then you received a chocolate

Question 3

Surprising outcomes

Answer 3

The US gets less surprising each time we come to expect the US, the CS and the US become associated

Question 4

Importance of surpise

Answer 4

A mis match between our knowledge and what we observe tells us what we need to learn

Question 5

Model of learning

Answer 5

Learning is a product of salience and surprise

Question 6

What happens to surprise across training?

Answer 6

Surprise decreases, and as it decreases so does the rate of learning

Question 7

Learning explained graph format

Answer 7

Learning is the change in expectation across the trail. The line straightens as out learning stops as we have no more surprises

Question 8

What happens the more salient the CS is

Answer 8

The more salient a CS is the faster we learn about it the asymptote (a value that you get closer and closer to, but never quite reach) is the same

Question 9

The US determines the asymptote:

Answer 9

the larger the US the higher our expectation, takes the same number of trials to reach asymptote

Question 10

Omission of the US:

Answer 10

Extinction if the US is more than expected, expectation increases. If the US is omitted, we get less than we expect. This means out expectation decreases.

Question 11

More than one CS:

Answer 11

if we want to test whether a tablet makes us feel good, we would usually test in isolation but there are many other things that can change how we feel such as: sleep and diet. Need to adjust model to predict the effects

Question 12

Three examples of CS interactions:

Answer 12

overshadowing
blocking
protection from extinction

Question 13

overshadowing

Answer 13

In a doubles match of tennis for example learn less about each player than we would if they were playing alone.

Question 14

Mackintosh (1976) overshadowing

Answer 14

Conditioned rats a light and a shock measured “fear”
Less fear when the light conditioned in compound with a noise. The effect was smaller when the noise was quieter. It’s as if the noise and light competed to be learned about

Question 15

When does learning stop for overshadowing?

Answer 15

When the summoned expectation reaches asymptote, the amount of learning about each CS depends on their relative salience

Question 16

Blocking:

Answer 16

The noise reached asymptote before the light is added, because there is no surprise on the trials with light there is no learning.

Question 17

The Rescorla-Wagner model: (1972)

Answer 17

One of the most influential models in psychology, making a lot of predictions many of which have subsequently been observed. Based on classical conditioning, asserts that amount of associative change (ΔV) or fear that is learned to a CS on a given fear conditioning trial is determined by the salience of the CS

Question 18

Exposure therapy for phobias:

Answer 18

Presenting a CS without consequence, (extinction) only in cases where a phobia is the result of conditioning.

Question 19

Problems with exposure therapy

Answer 19

Therapist becomes a safety signal so the phobia returns once the therapist is no longer there, so need to do therapy in the most realistic environment as possible.

Question 20

Extinction depends on context?

Answer 20

Id the feared CS is presented without the safety signal the phobia returns, prediction states that exposure therapy will be most effective when the context is varied