classical conditioning Flashcards
classical conditioning
unlearned/innate reflex- reflex an animal already has- when food is present animal salivates (unconditioned stimulus)- salivates (unconditioned response). Neutral stimulus- bell, presented before reflex is triggered, after the pairing becomes a conditioned stimulus and because it is paired with unconditioned stimulus the dog salivated which is the conditioned response
quail sex conditioning
example of classical conditioning
male quail is put in a box, when the light turns on they pair that, light goes on door opens and there is a female quail. After a while when the light turns on the quail moves towards it. This is appetitive conditioning- learning helps predict something good is coming.
fruit fly fear conditioning
example of classical conditioning
shock fruit flies want to go away- US shock, UR fleeing the shock. Introduce a smell to the chamber, after pairing the odour with the shock, the odour makes fruit flies leave because they know shock is coming. Aversive conditioning- helps avoid something negative (harmful)
Eyeblink conditioning
puff of air- us blink- ur, tone is paired with the puff of air. UR is fast blink, CR is a slow anticipatory squint- averisve conditioning, learning to avoid a negative feeling. (humans and rabbits)
Conditioned Compensatory Response
extinction
Breaking the association between the CS and US can extinguish the new CS→CR reflex
Rules for classical conditioning
- Timing
- Blocking
- Latent inhibition
- Associative bias
Timing
if US comes first there will be no learning, the tone will not tell you anything and the association will not form. The best learning is if the US happens really quickly after the CS. Peak is less than 500 ms.
Delay conditioning
CS and US over lap- best learning
Trace conditioning
Longer gap- worse learning
Blocking
A previously learned association can block the formation of a new association.
Latent inhibition
first pre expose, then the pairing it will take a lot longer to learn the pairing because the animal originally learnt that the stimulus is meaningless.
Associative learning
pre disposed to make some associations over others.
Taste will become associated with poison tone will not even though both are being played- but associate that tone will be more related to taste. Vice versa tone and shock. A more natural association is food and getting sick and sound sand noise. Innate preferences for making certain associations. Even though both are being used these are prioritised because they are more likely to be casually related
Sign and goal tracking
sign goes to sign, goal goes to goal. E.g. quail experiment- sees light goes to light is sign tracking, sees light goes to door goal tracking. They realise a girl is going to come out.
backwards conditioning
Does not actually form an association (in most cases)
Blocking experiment
tone and light- footshock, afterwards you get a response to either or both
another group rate has just the light, then they were trained on the combination stimulus, the light produced the response and the tone did not- the tone is irrelevant, do not know if the tone is actually involved
Kamins Blocking Paradigm
Latent inhibition study
Control group- animal sits in a chamber and afterwards is exposed to a tone which is followed by a foot shock. Associate that tone means foot shock and start freezing when they hear tone. Experimental group- sit in chamber and listen to tone. Afterwards are exposed with tone then foot shock. Takes them a lot longer to freeze and associate tone and shock because originally learnt they do not correlate
Models of classical conditioning
CS Modulation (Mackintosh)
US Modulation (Rescorla-Wagner)
US Modulation Theory
figure out how likely is the us to occur, do we need to update our mental model how likely the us is to occur based on information. Theory based on learning from errors,learning is driven from where the prediction is wrong (prediction error), then you learn to update the prediction. the strength of a conditioned response can be significantly influenced by changes in the intensity or presentation of the unconditioned stimulus
Rescorla-Wagner Model
- Each stimulus has a weight (W) for predicting a US
- Weights go from 0.0 (no expectation; neutral stimulus) to 1.0 (certain)
- Sum of the stimulus weights is the animal’s prediction
- When the prediction is wrong, the weights are adjusted
- Error: difference between expected and actual outcome
- Prediction Error = Actual US – Expected US
- Adjustment: each active stimulus adjusted by a proportion of Error
- Change in stimulus weight = Learning Rate x Prediction Error
- When error is 0, no learning occurs (asymptote)
Rescorla-Wagner Model explaining blocking
CS Modulation Theory (mackintosh)
Mackintosh Model explaining latent inhibition
