Learning Approach (APPROACHES 1/6) Flashcards
Behaviourism (classical and operant conditioning) and social learning theory
three theories
behaviourist approach
- classical conditioning (Pavlov)
- operant conditioning (Skinner)
social learning theory (Bandura)
assumptions
behaviour influenced by nurture, not nature
should study observable behaviours only
born as ‘blank states’ with the capacity to learn - genetics and biology do not affect behaviour
humans are only animals and should not be treated as more complex (animal research relevant)
classical conditioning
Pavlov, 1927
learn through association of stimuli in environment
consciously or sub-consciously learn to produce a reflex to a stimulus that would not normally cause this response
Pavlov’s research aims
originally investigating dogs’ digestive system
turned to investigating whether reflex response of salivation could be conditioned
Pavlov’s procedure
lab experiment on 35 dogs of different breeds
in sealed, sound-proof lab to prevent other stimuli
collected saliva - number of drops was measured
paired neutral stimulus, metronome, with unconditioned stimulus of food around 20 times
Pavlov’s results
found that neutral stimulus did not initially elicit a salivation response whereas the unconditioned stimulus of food did so immediately
salivation commenced 9 seconds after the metronome sounded, 45 drops of saliva collected
Pavlov’s conclusions
environmental stimuli, through repeated pairing could trigger a salivation reflex response
through process of association, the conditioned stimulus (metronome), leads to conditioned response (salivation)
Pavlov’s research strengths
controlling sounds and extraneous variables increases internal validity (IV had effect on DV)
sure that it was the metronome rather than other prompts
Pavlov’s research weaknesses
unethical
harness limited movement, limits credibility
reduced ecological validity, unnatural behaviour in artificial environment
before conditioning
dogs see food - UCS
naturally causes automatic response of salivation - UCR
metronome has no significance to the dog and so does not cause a response - NS
during conditioning
conditioning the dog to learn that food will immediately be presented after a metronome
NS and UCS repeatedly paired together or associated
dog’s salivation is still UCR - pairing of two stimuli means we can’t be sure that the metronome alone caused salivation
after conditioning
metronome is no longer neutral - has meaning to the dog - become conditioned stimulus
salivation has become a learned response to conditioned stimulus, and is now a conditioned response
unconditioned –> conditioned
unconditioned stimulus (food) –> unconditioned response (salivation)
neutral stimulus (metronome) + unconditioned stimulus (food) –> unconditioned response (salivation)
conditioned stimulus (metronome) –> conditioned response (salivation)
UCS
unconditioned stimulus
original stimulus that causes a reflexive reaction
naturally occurring and unlearned
UCR
unconditioned response
the original reflex action
NS
neutral stimulus
a stimulus that causes no reaction
CS
conditioned stimulus
a NS causing a learned, reflex reaction after associated with the UCS
CR
conditioned response
the learned response elicited by the CS
stimulus
any change in the environment that we register
reflex
an automatic response to a stimulus
immediate with no thought
can be learned and conditioned
response
changes in our behaviour due to a stimulus
extinction
if the CS is not paired with the UCS occasionally after conditioning then the CR will die out
stimulus generalisation
classical conditioning response seen with other stimuli similar to the original CS
the more similar, the stronger the response
spontaneous recovery
occurs after extinction
CR returns when there is no pairing of the NS and UCS and soon becomes extinct again
Watson and Rayner’s research - Little Albert
study using classical conditioning principles to see if they could cause a human baby to develop a fear he did not have previously
test whether classical conditioning worked with humans
Watson and Rayner’s research aims
whether they could condition fear of an animal by simultaneously presenting the animal and striking a steel bar to make a loud noise to frighten the child
whether fear would be transferred to other animals and objects
the effect of time on the conditioned response
operant conditioning study
Thorndike conducted an experiment where he put a cat in a cage with a latch on the door and a piece of salmon outside the cage
cat accidentally hit latch and door opened
with repetition, amount of time and effort spent reaching and scrabbling decreased and the latch was released earlier
analysis of behaviour - learning puzzle as consequence of food which was reinforcement
first took 5 mins –> after 10 trials became 5 seconds
trial and error learning
law of effect - of effect is desirable, behaviour is learnt and repeated
learning takes place because of what happens after the action
operant conditioning
law of effect enabled Skinner to develop the theory of operant conditioning, which explains hoe we learn new voluntary and complex behaviour and is known as ‘learning through consequence’
if we are rewarded/reinforced for a behaviour then we are more likely to repeat that behaviour, and if we are punished, then we are less likely to repeat it
reinforcement
reward to encourage repetition of behaviour
POSITIVE - receiving something we want
primary satisfies a basic need (hunger, thirst)
secondary - receiving something that allows us to get an object that will satisfy a basic need (money)
NEGATIVE - something undesirable is taken away
sometimes reinforcement is unintentional - lead to odd or undesirable behaviour
punishment
does not teach new behaviour
weakens or stops undesirable behaviour
POSITIVE - presents something unpleasant or painful whenever behaviour is shown
NEGATIVE - removing something pleasant
behaviour shaping
used to develop complex unnatural behaviours by reinforcing behaviour resembling target behaviour
reinforcement becomes more selective (successive approximations) until desired response achieved
behaviour broken into stages, reinforcement at each one until complex behaviour has been achieved
Skinner’s box
recorded behaviour of animals in response to different consequences
contained lever for an animal to press for food
had speaker and lights used to trigger a behaviour
shock generator deliver punishment
Skinner’s box positive reinforcement
rewarded with food pellets after pressing lever to encourage behaviour
Skinner’s box negative reinforcement
subjected rats to electric shock but rewarded them by removing it if they pressed the lever
would sometimes switch a light on before an electric shock - learnt to press lever when light was on to avoid shock
Skinner’s box punishment
punished with electric shocks if they pressed the lever, so they learned not to press it, behaviour weakened
Skinner’s box strengths
standardised procedure
good applicability
standardised procedure - all rewarded and punished in the same way, all confined to same Skinner box
other researchers can replicate easily
check for consistency
reliable
good applicability
applied in schools and amongst parents to encourage positive behaviours in children and discourage others
useful in many real life settings and simply - giving children sweets to encourage repetition)
increase understanding of how we learn and can be used for greater good in society
Skinner’s box weaknesses
ecological validity
generalisability
lacks ecological validity
unfamiliar and artificial environment - not accurate representation of natural behaviour, don’t know if rewards and punishment are relevant in a natural environment
findings difficult to generalise to humans - rats have smaller and less complex brains - difficult to make comparisons
cannot assume humans will learn in the same way - limits usefulness of findings
social learning theory
Bandura
learn through observation and imitation of role models
behaviour is not random and do not wait for rewards to determine whether we will continue a behaviour
observational learning - two way process with environment and individual interacting (reciprocal determinism)
imitation
acquired through imitation of behaviour and attitudes modelled by parents and significant others - quicker than conditioning
determinants of imitation
characteristics of model
observer’s percieved ability to perform that behaviour
observed consequences of behaviour
identification
refers to extent of which individual relates to a model and feels that they are similar enough to experience same outcomes
increases likelihood of imitating behaviour
characteristics increasing imitation
same sex
similar age or older
admired/respected/high status
modelling
someone must model (carry out) behaviour for it to be learned
live models - parents, teachers, peers
symbolic models - characters on TV, people on media
behave in certain way and this can be observed by an individual and later reproduced (imitation)
vicarious reinforcement / punishment
children more likely to imitate a model if they were rewarded for behaviour and less likely to imitate the model if they are punished
mediational processes between observation and imitation
observer forms mental representations of behaviour displayed by model and thinks about probable consequences
suggests there is more conscious thought as to whether to perform a behaviour
four mediational processes (ARRM)
ATTENTION - observer must carefully watch the model perform a behaviour and consequences by paying attention
RETENTION - observer must retain what they have observer to be able to later repeat it - imitation is not always immediate
REPRODUCTION - only imitated if observer has ability to carry it out
MOTIVATION - rewards and punishment that follow behaviour are observed
(intrinsic - how it makes you feel, extrinsic - physical reward)
if percieved rewards outweigh costs, more likely behaviour is imitated
skinner’s research strengths
high reliability
standardised procedure - specific controls in Skinner box
rewarded and punished each rat in the same way for performing the same behaviours
increases replicability
easily check for consistency - whether rats continue to press lever when rewarded and whether they stop when punished
good applicability
findings applied in schools and amongst parents to encourage more positive behaviours and eliminate negative ones in children
useful in many settings and in simplistic ways
increase our understanding of how we learn
used for greater good in society
Skinner’s research weaknesses
poor generalisability
used rats and pigeons - different brain structure and conscious awareness/decision-making to humans
can’t generalise to humans and suggest that humans learn through rewards and punishment in the same way
humans have different motivations to animals, so performing same procedure on humans might not yield same results
low ecological validity
Skinner box in a lab environment with strict controls and conditions
can’t be sure that they would repeat desirable behaviour in a more natural environment
lack validity - only know how behaviour is in an artificial environment
social learning theory strengths (supporting evidence)
aggression, dogs, monkeys
support by Bandura, Ross and Ross found that children were more aggressive after observing an aggressive role model, which supports that people learn through observation and imitation - theory is accurate and valid
support for idea of learning through observation - Kubinyi found that when pet dogs watched their owners push the handle of a box to release a ball, dogs were able to reproduce this behaviour after observing 10 times.
findings cannot be fully generalised to humans, as typical characteristics of role models are hard to apply to animals and dogs may not go through same cognitive processes
observation, imitation and modelling support by Cook and Mineka research where monkeys were made to to watch a video of another monkey reacting to a real snake and a toy snake. After multiple exposures, the monkeys demonstrated a fear of snakes after watching the model monkey show fear.
supports the idea of vicarious punishment - experiencing model monkey’s punishment as fear due to being exposed to snakes and imitating this behaviour.
social learning theory weaknesses
limitations in testing theory as behaviour observed may not be reproduced immediately e.g. a child observing someone being scared of a plane may not display this fear until they themselves go on a plane.
difficult to test for observational learning
experiments only show specific behaviours at a specific time, so results are limited, and lack of validity in findings at the time the study is carried out.
issues in claiming that observational learning shown by animals is also true of humans e.g. dogs in Kubinyi quickly imitated behaviour of their owners; this does not mean that humans learn through imitation in the same way or as quickly as dogs did.
humans have a greater understanding of moral and social norms meaning that greater thought process used in making a decision whether to reproduce the observed behaviour, not likely to be the case with animals, limiting generalisability
social learning theory weaknesses (different theory)
SLT involves cognitions (unobservable thought processes) - mediational processes refers to attention and retention
moves away from behaviourism which only studies observable behaviours
theory less scientific and more based on inferences
lacks credibility, not based on scientific methods
social learning theory strengths (application)
useful in explaining behaviour such as aggression and gender development and can be applied as a therapy
observational learning used alongside rewards - if OCD patient observed someone they trusted carrying out a more desirable behaviour with no unpleasant consequences, they may be inclined to imitate e.g. washing hands after touching things
has good applicability to real world
encouraging sufferers to improve mental health
Bandura, Ross and Ross aims
transmission of aggression through imitation of aggressive models
aims to investigate whether exposure to a real-life aggressive model increased aggression in children
Bandura, Ross and Ross hypotheses
children exposed to aggressive models would imitate the aggression shown
children exposed to non-aggressive models would not show high levels of aggression
there would be a gender difference - boys expected to show more imitative aggression than girls
Bandura, Ross and Ross method
tested 72 children (36 male, 36 female, between ages of 3 and 5)
a male adult and a female adult acted as models and one female experimenter conducted the study
participants divided into 8 experimental groups of 6 children, with remaining 24 forming a control group
AGGRESSIVE MODEL CONDITION
6 boys - male model
6 boys - female model
6 girls - male model
6 girls - female model
NON-AGGRESSIVE MODEL CONDITION
6 boys- male model
6 boys - female model
6 girls - male model
6 girls - female model
children matched for physical and verbal aggression from ratings - groups were on similar terms of aggression
individually brought in to a room with the model
the child sat at a table, where were shown how to design a picture
model was taken to opposite corner containing a table, chair, tinker toy, mallet and inflatable bobo doll
the child watched the model and could not play with the toys
Bandura, Ross and Ross aggressive conditions
model played with tinker toy for a minute then began to act aggressively towards the bobo doll for 9 mins
e.g. (repeated 3 times)
- laid the doll, sat on it, punched on the nose
- raised the doll, struck on head with mallet
- tossed it up and kicked around the room
verbal statements used between behaviours
- ‘sock him in the nose’
- ‘hit him down’
- ‘throw him in the air’
- ‘kick him’
- ‘pow’
non-aggressive comments
- ‘he keeps coming back for more’
- ‘he sure is a tough fella’
Bandura, Ross and Ross non-aggressive conditions
model sat in the corner, quietly playing with the tinker toys and ignoring the bobo doll
Bandura, Ross and Ross after 10 minutes
child was taken to another room and given toys to play with for 2 mins
toys were taken away as a control to provoke mild aggression arousal and ensured they were in a equally frustrated mood before being observed
Bandura, Ross and Ross after 2 minutes
child taken to final experimental room, allowed to play with aggressive and non-aggressive toys
- aggressive toys: bobo doll, mallet and peg board, dart guns, tether ball
- non-aggressive toys: tea set, crayons, ball, dolls, bears, cars, plastic animals
had 20 mins of free play whilst being observed through a one-way mirror
Bandura, Ross and Ross behavioural categories
behaviour observed at regular time intervals and scored according to three types of imitative behaviour
- imitative physical aggression - acts displayed by model
- imitative verbal aggression - repeating model’s phrases
- imitative non-aggressive verbal responses
other categories
imitated essential components of model’s behaviour but did not perform complete act or was aggressive to another toy - scored as partially imitative behaviour e.g. mallet aggression, non-imitative physical and verbal aggression, aggressive gun play
Bandura, Ross and Ross results
children exposed to an aggressive model displayed significantly more aggression than children exposed to non-aggressive model
boys and girls displayed more imitative aggression after observing aggressive model
this effect was stronger after watching a same-sex aggressive model
most aggressive group was male children who watched an aggressive male model
least aggressive was female children who watched a non-aggressive male model
watching an aggressive model had a greater effect on boys than girls
male models copied most by male children
boys were not more aggressive in all instances - females exposed to a female models were more aggressive than boys exposed to a female model
Bandura, Ross and Ross conclusion
boys are more influenced than girls when watching a male model
girls are more influenced by verbal aggression than physical
observing aggression incites both imitative and non-imitative aggression
girls more likely to copy aggression from a male model than a female model - boys were more influenced by a same sex model
Bandura, Ross and Ross strengths
good generalisability
large sample size of 72 children
findings generalised to target population of children
sample limited as children were aged 3-5, cannot generalise to adults and whether they imitate aggression easily
good population validity
high validity
covert experiment, no demand characteristics, one way mirror
participants wouldn’t adjust behaviour, unaware they are being observed
highly reliable
used standardised procedure - all children observed for 10 mins, played for 20, same toys
lab setting with controlled conditions, easily replicated for consistency
Bandura, Ross and Ross weaknesses
ethnocentrism (carried out in 1 country)
cannot be sure findings can be applied to children in other parts of the world
lacks generalisability
lacks ethical credibility
limited protection from harm as children should not have been exposed to aggression
removed toys to anger children - distressing
lacks ecological validity
strange and unusual setting - big room alone, one way mirror
results invalid as not accurate representation of behaviour in an everyday setting
