1.4 What is Learned in Instrumental Learning Flashcards
what is the definition of learning?
the process by which new associations are formed between stimuli, thoughts, emotions, and behaviours
with thorndike’s law of effect, describe what happens when you go to a cafe
- you go to a cafe –> neural activity evoked
- you make numerous responses
- one response you make is to order a coffee. this is rewarded by the presence of a coffee.
- now, when you go to a cafe the likelihood of you ordering a coffee will increase
- ordering a coffee when at a cafe becomes a habit
what did hull say about reinforcement that was different? what was the ‘kind’ of theory of learning he came up with called?
reinforcement is a reduction in drive, so a reinforcer is a stimulus that reduces a drive. actions coinciding with drive reduction are reinforced
food –> increase in blood sugar, decrease in hunger –> drive reduces
S-R learning
what are the two problems with S-R theories of learning?
- you don’t need drive reduction to reinforce behaviour (e.g. strip shows)
- you don’t need a reinforcer for learning to occur (e.g. latent learning by Tolman)
outline and explain the results of tolman and honzik’s 1930 rat studies on latent learning
- 4 different groups (1: food @ end of maze, 2: no food @ end of maze, 3: food made available from 11th day, 4: food taken away from 11th day)
- group 3 originally showed little improvement in maze navigation without a reward. as soon as food introduced, errors reduced rivalling the group that was given food as a reward the whole way through.
RESULTS EXPLANATION:
* group 3 HAD learned the maze structure, but only DEMONSTRATED this knowledge when presented with food as motivation
* reinforcement provides impetus to PERFORM
* suggesting organisms can form cognitive maps of environment w/o reinforcement
what is contingency learning?
the process by which organisms learn the relationship between its behaviour and the consequences of this behaviour. i.e. how contingent an outcome is on an individual behaviour
to determine whether our responses cause an outcome, what two things do we need to know? how are these probabilities expressed?
- what is the probability that an outcome follows the response?
- what is the probability that the outcome will occur when the response is NOT made
- p(O|R)
- p(O|noR)
what is ΔP?
measure used to quantify the strength of the relationship between a behaviour and its consequence. helps determine how likely it is that a particular outcome will follow a specific behaviour.
ΔP = (O|R) - p(O|noR)
if ΔP is ANYTHING BUT 0, there IS A CONTINGENCY between the R and O
when does learning occur? (ΔP)
if there is a positive OR negative relation between response R and outcome O
what is the difference between if ΔP is positive or negative?
if ΔP is positive, your behaviour CAUSES something (putting the POSITIVE in positive reinforcement/punishment)
if ΔP is negative, your behaviour INHIBITS something (putting the NEGATIVE in negative reinforcement/punishment)
what is goal-oriented learning?
learning characterised by knowledge of what the consequences of your actions will be, and the CHOICE of the task that will maximise your reward. i.e. the drive towards maximising expected utility of your behaviour
motivation can be broken down into which logical form…
- you know that doing an action leads to an outcome
- you desire the outcome
- so you do the action
what is the process of goal-oriented and habit learning?
EARLY in learning is very goal-oriented, then transitions to habitual
e.g. started scrolling reels because it was fun and WANTED to have fun, now you do it all the time even when it’s not fun and you don’t necessarily need to have fun
