What is the rescorla-wagner model?
mathematical expression of surprise
- learning will occur only when the subject is surprised - that is, when what happens is different from what the subject expected to happen
What is blocking?
Participants are not surprised because they’ve already learned that the light predicts the shock
- doesn’t produce a stronger conditioned stimuli but refers to the lack of responses to the second stimulus in phase 2
Explain the steps involved in blocking
Phase 1.
Blocking: L–> US
Control: nothing
Phase 2.
Blocking: L & T –> US
Control: L & T –> US
TEST:
Blocking: no CR
Control: CR
Why is the blocking experiment important?
Shows that:
- conditioning is not an automatic result of CS-US pairings
- for conditioning to occur, the CS must be informative and US surprising
What can we use to quantify the level of surprise?
deltaV = Vmax - Vn
where:
V = associative strength between CS and US
Vmax = maximum associative strength
delta V = change in associative strength on each conditioning trial
When does the most amount of learning occur?
delta V
- at the beginning of conditioning
- -> more surprised by the US at this time
Why do we care how surprising the US is?
So we can predict what is biologically meaningful
- and we can learn to avoid or work for these
Learning curves differ in terms of?
Vmax and rate of acquisition
How does Vmax differ in learning curves?
- asymptote level can change - determined by magnitude of US
- more incentive = higher Vmax
How does rate of acquisition vary between learning curves?
- gradual = little learning over certain amount of time
- steep = learning is almost immediate (e.g. food avoidance)
What is alpha related to?
the importance of the CS
What is beta related to?
the importance of the US
Which values change with conditioning?
Vn and deltaV
- Vn gets bigger
- delta V gets smaller
Formula for acquisition of a conditioned response
deltaVn = alphabeta(Vmax-sumofVn)
What is the extinction of a conditioned response
the weakening of a conditioned response when a CS is presented by itself
- Vmax = 0
What type of associative strength do conditioned inhibitors have?
negative
Explain blocking using Rescorla-wagner model
- when 2 CSs are used, the association or expectation at the beginning of the trial would be the sum the strengths of each stimuli present
- therefore the amount of conditioning on a compound trial in which A and B occur together would be:
= alphabeta(Vmax-sumVofA&B)
- the shock stays the same, therefore so does Vmax in compound conditioning (V= sum of light and tone = 1)
- change in learning as a result of extra conditioning trial with new tone but same light, new learning = 0
In blocking - what if the US is different in compound conditioning?
Then learning would happen - if US is bigger = Vmax will be bigger
Explain the over-expectation effect
- tone –> shock
- light –> shock
- tone + light –> shock
- US doesn’t change, animal will predict something that doesn’t happen (over-expected shock to occur)
Assume only a few trials were given before the compound trial, what is the result?
the model predicts an increase in associative strength for both T and L when presented during the compound trial
What if there is extensive training prior to the compound trial?
the model predicts a decrease in associative strength for both T and L when presented during the compound trial
Explain compound training in rats (tone and light –> shock)
Phase 1.
experimental group: tone + light –> shock
control: nothing
Phase 2.
both groups: tone –> shock; light –> shock
= much more fear associated with light
Why is there more fear associated with shock and light rather than shock and tone?
More natural to associated foot shock with a light
What would the result be in the US is not contingent with the CS?
conditioning will be strong to contextual cues but not to the CS
