Task 4 Flashcards
Classical Conditioning
Classical Pavlovian Conditioning
- US, UR, CS, CR
- US = stimulus that naturally evokes some response
- UR = the response
- CS = if US is repeatedly and reliably paired by a natural stimulus, that neutral stimulus can become a CS
- CR = anticipatory response followed by CS
Classical Pavlovian Conditioning
- aversive/appetitive conditioning
- aversive = when US is an unpleasant, or negative, event
- appetitive = when the US is a positive event
Classical Pavlovian Conditioning
- eyeblink conditioning
- US = air puff; UR = reflexive blink; CS = e.g. a tone
- similar across species
- conditioning is a gradual process
Classical Pavlovian Conditioning
- extinction
= previously acquired association would have diminished through repeated presentation of the CS in the absence of the US
- combination of unlearning and learning a new, opposite response
Classical Pavlovian Conditioning
- tolerance
= decrease in reaction to a drug so that larger doses are required to achieve the original effect
Classical Pavlovian Conditioning
- blocking
- CS must provide valuable new information that helps animal predict the future
- even if given cues does predict US, it may not become associated with that US if its usefulness has been preempted (blocked) by a co-occurring cue that has a longer history of predicting the US
Rescorla-Wagner Model
- prediction error
- error-correction rule
- key idea: amount of change that occurs in the association between a CS and a US depends on a prediction error ( = difference between whether the animal expects the US and whether the US actually occurs)
- model is an error-correction rule because over many trials of learning, it reduces, or corrects, the likelihood of prediction errors
Rescorla-Wagner Model
- assumptions
(1) each CS has an association weight that describes the strength of association between that cue and the US
(2) expectation of the US is described by the sum of the weights of all the cues that are presented during a trial (compound cues)
(3) difference between outcome the animal expects and what actually occurs (prediction error)
Rescorla-Wagner Model
- equations
(1) Prediction error = Actual US (either 0 or 100) - Expectation of US based on CS
(2) Expectancy (weight) change = Prediction Error X Learning Rate
Rescorla-Wagner Model
- compound conditioning
- weights not rising as quickly as in single-cue conditioning –> but: having multiple (compound) cues should allow expectation of the US to increase faster, even while the changes in the individual weights rise more slowly
Modulation of CS Processing
- latent inhibition
= phenomenon of impaired learning following cues pre-exposure; formed an association to something predicting nothing
- critique on Rescorla-Wagner model
Modulation of CS Processing
- US modulation theory
= argues that the manner in which the US is processed determines what stimuli become associated with that US
- Rescorla-Wagner model
Modulation of CS Processing
- CS modulation theory
= propose that the way in which different potential CSs are processes determines which of them become associated with the US
- latent inhibition can only be explained by CS modulation theories
- Mackintosh theory –> previously conditioned stimulus derives its salience from its past success as a predictor of important events, and this happens at the expense of other co-occurring cues that don’t get access to you limited pool of attention
Cerebellum
- two main regions
- cerebellum is responsible for conditioned response only
- two main regions:
(1) cerebellar cortex –> Purkinje cells
(2) cerebellar deep nuclei –> Interpositus Nucleus
Cerebellum
- CS input pathway
(1) pontine nuclei
(2) mossy fibers that branch in two directions:
(2. 1) contacts with interpositus nucleus
(2. 2) projects towards cerebellar cortex (by way of Granule cells), across parallel fibers, then connecting to dendrites of Purkinje cells
