introduction lessen 7 (learning)

3 types of (associative) learning

• classical conditioning

• operant conditioning

• observational learning

Pavlov

• classical conditioning = Pavlovian learning

• before: food = US and saliva = UR, bel = NS and no conditioned response

• during: food + bel → saliva = UR

• after: bel = CS and food = CR

rules of classical conditioning

• Effective conditioning is more likely if Unconditioned Stimulus (US) follows Conditional Stimulus (CS) shortly in time

• Repeated presentations of CS + US are called “training trials”/"acquisition trials”

• Presentations of CS without US (after the acquisition phase), are called “extinction trials”

• the stronger the US, the few acquisition trials to learn!

generalization

When one exhibits a conditioned response (CR) in response to a stimulus that is (perceptually/conceptually) similar to the conditioned stimulus (CS), but was never presented with the unconditioned stimulus (US)

latent inhibition

• If the conditioned stimulus (CS) for the acquisition phase has often been experienced without the unconditioned stimulus (US), the association is more difficult to achieve

• So: easier with ‘new ’ stimuli as CS

• Importance of first, good experiences with doctor, hospital, medication ....!

extinction

• Extinction = teaching a new, inhibitory association (CS – no US) that can inhibit the previously learned association (CS – US) with associated response generation (CR)

• (belangrijk is dat het hier gaat over INHIBITION en niet over het totaat afleren/verdwijnen van de conditionering!)

little albert experiment

• Watson

• ‘we can make everything of a person that we want’

• see how you can condition very young kids to generate anxiety responses to a wide range of stimuli

• problem: the extinction is way weaker than the acquisition

exposure therapy

• (based on extinction)

• repeated exposure to the stimuli and situations that elicit fear to extinguish the fear

• based on extinction learning

problems with exposure therapy

• spontaneous recovery: CR is suppressed but does not dissapear (different temporal context)

• extinction is sensitive to context = renewal (different physical context)

• reinstatement: CR comes back after succesful extinction

operant conditioning

• = instrumental learning

• the subject sets a behavior an receives a consequence (punishment or reward) → affects the future frequency of the behavior

• voluntary behavior (<-> involuntary, classical cond.)

Thorndike

• exp. cat in a box: cat would start associating the actions he needs to do in order to get out of the box and receive food (=reward)

• law of the effect

law of the effect

• ⊥ Thorndike

• If a behavior is followed by a satisfying consequence (like a reward), that behavior is more likely to be repeated in the future. If the consequence is unpleasant, the behavior is less likely to occur again.

Skinners box

= practical demonstration of Thorndike’s lw of the effect

⊥ operant conditioning

reward and punishment

→ reward leads to an increase in that behavior

→ punishment leads to a decrease in that behavior

discriminative stimulus (SD)

= stimulus die aangeeft dat een bepaalde reactie waarschijnlijk beloond of bestraft zal worden

behavior (R) is associated with SD

positive reinforcement

je voegt iets aangenaams toe → gedrag neemt toe

negative reinforcement

je neemt iets onaangenaams weg → gedrag neemt toe

positive punishment

je voegt iets onaangenaams toe → gedrag neemt af

negative punishment

je neemt iets aangenaams weg → gedrag neemt af

positive vs negative

positive = addition of a stimulus vs negative = removal of a stimulus

extinction (operant conditioning)

when the behavior (R) is no longer followed by the expected effect (O), the probability that the behavior decreases

shaping

= stapsgewijze operante conditionering

= procedure in which the target behavior is broken down into smaller steps that are succesively reinforced

learned helplessness

⊥ Seligman

= a characteristic of several negative affective states (very present in depression)

• ervaring met oncontroleerbare negatieve uitkomsten → leren dat hun gedrag geen effect meer heeft → lack of control → they stop trying

schedules of reinforcement

•Extinction: the response is not rewarded or punished

•Continuos: every response is rewarded or punished

•Intermittent: responses are not always rewarded or punished

intermittent

responses are not always rewarded or punished

• interval: after a certain time interval, the next response is reinforced/ punished

• ratio: responses are sanctioned/punished after a certain number of responses

interval

• after a certain time interval, the next response is reinforced/ punished

• fixed or variable

ratio

• responses are sanctioned/punished after a certain number of responses

• fixed or variable

fixed

the time interval or number of responses between endorsed responses does not vary

(post-reinforcement pauses)

variable

an average time interval or average number of responses between endorsed responses is recorded, but there is random variation around that average

(no post reinforcement pauses) → high response rate!!!

effective punishment

(risk = negative attention or punishment is associated with reward)

• intense, not ‘build up’, consistent, contiguity with behavior, no deferred punishment, no association with reinforcement

observational learning

= social learning

acquiring newy information by watching others perform actions

Bobo doll experiment

Bandura, observational learning

kids learn a behavior, just by the observation of someone doing this sort of behavior