Lecture 9 - Models of Associative Learning

Question 1

Mackintosh on attention to CS

Answer 1

Mackintosh argued that variations in conditioning reflect variations in attention to CS.

More limited and simpler than R-W model.

Animals have limited attention resource: must be shared; we can’t take everything from our environment in, we have selective attention.

Question 2

Rescorla-Wagner Model

Answer 2

R-W model is all about surprise. It assumes that learning occurs on a conditioning trial only if the US is surprising.

Describes in a formal way how animals learn associations between a CS and US, and how fast the learning occurs.

Question 3

Delta V

Answer 3

Change in CS-US association or ‘amount of learning’

Question 4

Alpha

Answer 4

Salience (intensity) of CS

Question 5

Beta

Answer 5

Salience of US

Question 6

Lambda

Answer 6

Experience of US presentation.

The upper limit of the curve - the asymptote.

Question 7

Sigma V

Answer 7

Expectation of US, based on total associative strength of all CSs present.