Models of human learning #2 Flashcards
Rescorla
1968
~ conditions is closely related to contingency
Kamin
1969
cue interactions effect a.k.a. competition
~ if two cues are both presented before response then the conditioned response is weak
When is CR learning blocked?
when light –> food and then light + tone –> food
When is CR learning mutually overshadowed?
when light + tone –> food
Beesley and Le Pelley
(2011)
~ experiment demonstrating blocking
~ most attention goes to established predictive cue
~ less attention goes to blocked cue
Theories of blocking
SUPRISE (Rescorla-Wagner, 1972) the outcome is not surprising as it is already predicted
ATTENTION (Mackintosh, 1975) the predictive cue grabs out attention, leaving little processing for blocked cue
- more evidence for attention
- stronger evidence would be found looking at overt biases in selective attention
Eye-tracking
~ useful for examining changes in attention
~ precise recording of gaze direction and duration
ASSUMPTION = eye movement reflects overt changes in attention
Associative models
cue competition is explained by models of animal conditioning
e.g. Rescorla-Wagner (1972)
Rescorla-Wagner
(1972)
~ if events are surprising (large error term) then the model updates it associative strengths
Propositional account
~ learning occurs from the operation of controlled reasoning processes
~ the processes result in beliefs about the world
Verbal instructions
~ linked to propositional account
~ if behaviour is related to reason, verbal instructions should directly influence behaviour
Lovibond (suggested ideas)
(2003)
~ propositional account occurs after trial-by-trial learning and then acquisition of propositional knowledge
~ associative account after trial-by-trial learning
Perruchet effect
(1985)
~ participant has to rate expectancy for air-puff
~ participants have a tendency to show a gambler’s fallacy effect
Lovibond’s claim
~ learning is based on reasoning and belief formation
~ verbal instructions directly influence behaviour, even in automatic responses