Learning Theories Flashcards
Classical conditioning
Learning by association, occurs when a neutral stimulus is repeatedly paired with an unconditioned stimulus, the neutral stimulus eventually produces the same response as the unconditioned stimulus. Discovered by Pavlov
Unconditioned stimulus (UCS)
A stimulus that produces a response without any learning talking place
Unconditioned response (UCR)
An unlearned response to an unconditioned stimulus
Neutral stimulus (NS)
A stimulus that does not produce the target response, it becomes a conditioned stimulus after being paired with the UCS
Conditioned stimulus (CS)
A stimulus that only produces the target response after it has been paired with the UCS
Conditioned response (CR)
The response elicited by the CS (the UCR)
Extinction
When the CS and UCS have not been paired for a while the CR is extinguished and the CS no longer elicits it. Shows that our learning is flexible
Spontaneous recovery
An extinct response which activates again without new pairings with the UCS. Generally these responses are weaker than the initial response
Stimulus generalisation
When an individual that has acquired a CR to one stimulus begins to respond to similar stimuli in the same way
Process of classical conditioning
UCS produces UCR, NS does not produce this response
UCS and NS are paired
NS now produces the same response as the UCS
Now called the CS and CR.
Classical conditioning - strengths
Pavlov (1927)
Watson and Rayner (1920)
Application to aversion therapy - an uncomfortable response is paired with an unwanted behaviour to prevent the occurrence of this behaviour
Classical conditioning - weaknesses
Rescorla (1968)
Incomplete explanation of learning - can only explain a limited range of behaviours (simple reflex responses). For example, could explain how we come to fear dogs but not the maintenance of this fear over time
Rescorla (1968)
Found that contingency (extent to which the NS reliably predicts the UCS) is more important than contingency (how close in time they occur) in the pairing of the NS and UCS. Opposite to what Pavlov thought
Pavlov (1927) - aim
To explain the role of conditioned reflexes in the eating behaviour of dogs
Exploring how salivation becomes associated with new stimuli apparently unrelated to food and the properties of this association
Pavlov (1927) - procedure
Lab study, saliva measured by no. drops or volume in a cannula, took place in a soundproof chamber
Measured baseline salivation in response to NS (metronome), then paired NS and UCS (food) around 20 times varying the presentation before and after the UCS. Variations used to look at extinction, the CS was presented several times without the UCS producing extinction
Pavlov (1927) - findings
The NS did not initially elicit a salivation response, whereas the UCS did immediately. After forward pairings the NS did elicit the response but not when paired backwards and it only became associated if the dog was alert and undistracted. Extinction could be seen and there was also spontaneous recovery
Pavlov (1927) - conclusion
A link is likely to be made between a UCS and an NS that occurs just before the UCS.
Pavlov (1927) - strengths
High internal validity - controlled extraneous variables (soundproof room, collection of saliva in a cannula and carefully tested NS)
Application to eating problems - Jansen et al (2003)
Pavlov (1927) - weaknesses
Lack of generalisability to humans - humans have a larger cerebral cortex compared to dogs, enabling conscious choice
Jansen et al (2003)
Overweight children have acquired very strong associations between cues that predict the arrival of food and the salivation response
Operant conditioning
Learning that occurs when a behaviour is followed by reinforcement or punishment that will increase or decrease the chance of the behaviour occurring again.
Skinner (1948)
Conducted research on animals kept in a chamber. It contained a supply of food pellets that could be released as reinforcement when the animal learnt to do something (pull a lever). Some also contained electric floors that could be used as punishment.
Reinforcement
A behaviour is said to be reinforced when an event that takes place after it increases the probability of the behaviour being repeated. This can be positive reinforcement or negative reinforcement. It could also be primary or secondary
Positive reinforcement
When something nice is introduced to the individual following a behaviour, that increases the probability of the behaviour being repeated
Negative reinforcement
When something unpleasant is removed from the individual following a behaviour, increasing the probability of the behaviour being repeated
Primary reinforcement
When the thing acting as the reinforcer has biological significance, such as food
Secondary reinforcement
When the thing acting as the reinforcer has become associated with something that has biological significance, for example money, which can be used to buy food
Punishment
When an event following a behaviour makes the repetition of the behaviour less likely. It can be positive or negative.
Positive punishment
When something unpleasant is introduced to the individual following a behaviour, decreasing the chance of the behaviour being repeated
Negative punishment
When something nice is removed from the individual following a behaviour, decreasing the probability of that behaviour being repeated
Operant conditioning - strengths
Chase et al (2015)
Skinner (1948)
Application to education and childcare - used in school e.g. detentions or gold stars
Chase et al (2015)
Revealed brain systems that relate to reinforcement in humans using brain scans
Operant conditioning - weaknesses
Incomplete explanation of learning - can only explain how existing behaviours are strengthened or weakened and not how they originated.
Schedules of reinforcement
A plan of how often and when reinforcement will be provided.
Fixed ratio reinforcement schedule
A reinforcement is given after a specified number of behaviours
Variable ratio reinforcement schedule
A reinforcer is given after an unpredictable number of behaviours that vary around a mean value
Fixed interval reinforcement schedule
The time between reinforcers is kept constant
Variable interval reinforcement schedule
The time between reinforcers is varied around a mean length of time
Behaviour modification
A form of therapy in which reinforcement is systematically used to increase the number of instances of desired behaviour
Behaviour shaping
Shaping is used to encourage complex behaviours by reinforcing successive approximations of the behaviour.
Schedules of reinforcement - strengths
Latham and Dossett (1978)
Application to treating inappropriate behaviour - lovaas therapy involves intensive reinforcement using shaping, to normalise behaviour in children with ASD
Latham and Dossett (1978)
Found that mountain beaver trappers responded to variable ratio pay (4 dollars if they correctly guess the colour of a marble) than fixed ratio pay (1 dollar per animal)
Schedules of reinforcement - weaknesses
Does not account for intrinsic motivation, therefore not a complete explanation.
Social learning theory (SLT)
Learning behaviours through the observation and imitation of role models
Stages of social learning
Bandura (1977) suggested that four cognitive processes mediate whether observation of a model will lead to the imitation of the behaviour
Social learning - stage 1
Attention - focusing on the modelled behaviour
Social learning - stage 2
Retention - cognitive process of storing the modelled behaviour in memory
Social learning - stage 3
Reproduction - the recall and enactment of the modelled behaviour in result to appropriate circumstances
Social learning - stage 4
Motivation - the force that drives a person to enact the behaviour
Modelling
When one individual displays a particular social behaviour in the presence of another, with the individual demonstrating the behaviour being the model. Models are most likely to be imitated if they are similar to the observer
Observation and imitation
An active process in which the observer chooses to focus their attention on the modelled behaviour. Imitation is then copying the observed behaviour
Vicarious reinforcement
Reinforcement that is not directly experienced but occurs through the observation of somebody else being reinforced for a behaviour.
Social learning theory - strengths
Bandura (1961-65)
Application to violent behaviour - explains the effect of media violence on aggression particularly in children.
Social learning theory - weaknesses
Kendler et al (2015)
Kendler et al (2015)
Showed that identical twins are more similar in their levels of aggression than non-identical twins, suggesting that individual differences in social behaviour are genetically influenced and not due to different modelling experiences
Bandura et al (1961) - aim
To see whether aggressive behaviour could be acquired through the imitation of aggressive models
Whether they selectively imitated same-sex models
Whether boys were more prone to overall aggression
Bandura (1961) - method
36 boys and 36 girls aged 3-6 were selected from Stanford university nursery. The children were matched on aggressiveness using one of their teachers and an experimenter that knew them well. Placed in threes so that gender was also controlled.
Aggression group - observed an aggressive adult model punching, kicking and shouting at a large bobo doll
Non-aggression group - non-aggressive model assembling mechanical toys
Control group - no model present whilst the children were playing
Bandura (1961) - procedure
Step 1: children were taken into a room with an observation window and allowed to play with toys (with or without model present)
Step 2: children were taken into another room and deliberately frustrated by being shown new toys and told they were for other children
Step 3: children were taken into a playroom containing a range of toys including a bobo doll, behaviour was observed through a one-way mirror and a second observer was present for half the participants to determine inter-rater reliability.
Behaviours were categorised as imitative aggressive, partially imitative and non-imitative aggressive.
Bandura (1961) - findings
Children that witnessed the aggressive model were more likely to completely or partially imitate the aggression and those that did not observe the aggressive model were less likely to show any aggressive behaviour (none in 70%). Children in the aggressive group were more likely to engage in non-imitative aggression. Boys were more likely to imitate a same-sex model and physical aggression over verbal aggression
Bandura (1961) - conclusion
Social behaviour such as aggression an be acquired through the imitation of models. Imitation is more likely when the modelled behaviour is gender typical and when the model and observer are of the same gender
Bandura (1961) - strengths
High internal validity - high control, matched for aggression, non-aggressive group, children one at a time
Inter-rater reliability - second experimenter was present in half the observations
Application to learning aggressive behaviour - know the risk posed to children with violent parents
Bandura (1961) - weaknesses
Low ecological validity - laboratory setting, children would not normally display aggression when playing with a doll compared to with other children
Bandura (1963) - aims
Whether a filmed model would have the same effect as a live model on the children’s aggression
Whether cartoon aggression would have the same effect as filmed aggression
Bandura (1963) - procedure
Independent groups design, 48 boys and 48 girls aged 39-52 months, 4 conditions
Live aggression - watched an adult aggress towards a bobo doll
Filmed realistic aggression - watched the same behaviour displayed on screen
Cartoon aggression - watched on tv a model dressed as a black cartoon cat perform the same aggressive behaviours
Control - did not watch aggression
Other procedures were then the same as the original study
Bandura (1963) - findings
All three experimental groups showed increased aggression
Mean total of aggressive acts was 83 (live), 92 (filmed realistic) and 99 (cartoon).
Differences between the groups were not significant however they were all significantly larger than the control (54).
Bandura (1963) - conclusion
Exposure to live or filmed aggression increases the likelihood of aggression in response to frustration
Bandura (1965) - aims
Whether reinforcement and punishment of an aggressive model would influence the aggression displayed by the observers in response to frustration
Bandura (1965) - procedure
Same procedure , 33 boys and 33 girls aged 42-71 months, three conditions
Model-rewarded - saw a second adult praise the model for their aggression, giving them a drink and a chocolate
Model-punished - second adult scolded the model and spanked them with a rolled up magazine
No-consequence - model was neither reinforced or punished
All three groups were later offered attractive rewards to aggress towards the doll
Bandura (1965) - findings
Children in the model-punished condition were significantly less aggressive than the other two groups. However, introducing the reward wiped this out significantly increasing the score for all groups.
Bandura (1965) - conclusion
Vicarious punishment reduces imitated aggression, however the promise of reinforcement has a larger impact
Bandura (1963&1965) - strengths
Control of extraneous variables - children taking part were matched for aggression, non-aggressive conditions controlled for spontaneous aggression, observing one at a time reduced the likelihood of one child imitating another
Application to the Sabido method - using telenovelas to tackle specific social problems around the world as viewers identify with popular characters that become role models
Bandura (1963&1965) - weaknesses
Risk of demand characteristics - children may have believed that they were expected to aggress towards the doll
Two-process model
Mowrer (1960)
Explains the acquisition of phobias through classical conditioning and their maintenance through operant conditioning
Acquisition of phobias
The NS initially creates no fear, however when paired with the UCS, that already does, it can become the CS with the CR of fear. This was demonstrated by Watson and Rayner (1920)
Maintenance of phobias
Responses acquired through classical conditioning tend to weaken over time, however phobias are persistent.
When a person avoids the fear-provoking stimulus they are negatively reinforcing the avoidance behaviour, maintaining the phobia.
Cook and Mineka (1989)
Suggested that social learning theory could also explain the acquisition of phobias.
Infant rhesus monkeys that watched adults display fear to fear-relevant stimuli such as toy snakes also developed a fear for those toys
Learning theories explanation for phobias - strengths
Watson and Rayner (1920) showed that a fear response could be conditioned in humans
Application to treating phobias - systematic desensitisation is based on classical conditioning, therefore it is likely that classical conditioning plays a role in the acquisition of phobias
Treatments for phobias
Systematic desensitisation
Flooding
Systematic desensitisation
A therapy designed to reduce phobic anxiety through classical conditioning. Eventually a new relaxation response is triggered to the stimulus, curing the phobia as it is impossible to be relaxed and scared at the same time (reciprocal inhibition). There are three main stages: anxiety hierarchy, relaxation and exposure
Systematic desensitisation - anxiety hierarchy
Constructed by the client and therapist, it is a list of situations related to the phobic stimulus arranged in order from least to most scary.
Systematic desensitisation - relaxation
The therapist teaches the client to relax as deeply as possible, potentially involving breathing techniques or mental imagery.
Systematic desensitisation - exposure
The client is exposed to their stimulus whilst in a relaxed state, starting at the bottom of their hierarchy. Once they can remain calm in the presence of the stimulus they move up the list.
Systematic desensitisation - strengths
Gilroy et al (2003)
Gilroy et al (2003)
Compared clients receiving systematic desensitisation for a spider phobia with a control group. After both 3 and 33 months the SD group were less fearful than the control group
Systematic desensitisation - weaknesses
Not suitable for all cases - SD only tackles behavioural symptoms, clients with multiple phobias or symptoms linked to a history of trauma may want to try a different therapy.
Flooding
Involves immediate exposure to the clients phobic stimulus, without the gradual build up in hierarchy.
Flooding stop phobic responses quickly as without the option of avoidance the client learns that it is harmless, extinguishing the fear as it is no longer paired with the UCS. e.g. (dogs with being bitten).
Flooding is not unethical but it is important that the client gives fully informed consent
Flooding - strengths
Quicker and as effective as alternatives, this makes the treatment cheaper and rids the client of their phobia as quick as possible
Flooding - weaknesses
May not be effective for all phobias - social phobias have no cognitive aspects, for example they do not experience anxiety but think unpleasant thoughts about the social situation
Watson and Rayner (1920) - aim
To demonstrate that simple emotional responses such as fear could be classically conditioned
Watson and Rayner (1920) - method
Healthy baby boy , 9 months old, little Albert, tested his baseline response to a range of objects including a rabbit, dog, white rat, wooden blocks, cotton wool, showing no fear response. Each session was filmed.
Watson and Rayner (1920) - procedure - session one
When Albert was 11 months and three days old he was taken to a ‘lab’ and a white rat was presented to him (NS), when he reached towards it the bar was struck loudly behind his head (UCS)
Watson and Rayner (1920) - procedure - session two
A week later. Exposed five times to the NS and UCS pairing. Was then tested with the wooden blocks which he showed no fear to (not getting more scared generally)
Watson and Rayner (1920) - procedure - session three
Five days later. His responses to the rat and a range of other objects were assessed, including wooden blocks, a rabbit and a dog
Watson and Rayner (1920) - procedure - session four
Five days later. Albert was taken to a new environment, placed on a table and once again assessed for his responses to the objects
Watson and Rayne (1920) - procedure - session five
Tested again one month later at 12 months and 21 days old.
Watson and Rayner (1920) - findings
At baseline testing Albert did not respond to any of the objects but did respond to the loud noise (startled and his lips trembled).
Watson and Rayner (1920) - findings - session one
He reacted again to the loud noise (cried)
Watson and Rayner (1920) - findings - session two
Started being more cautious towards the rat, not reaching out and pulling away once the rat nuzzled him. After further conditioning he began to cry and tried to rapidly crawl away
Watson and Rayner (1920) - findings - session three
Albert showed generalisation of the fear to white furry objects (rat and rabbit), crying, a mild fear to the dog and no fear to any other objects
Watson and Rayner (1920) - findings - sessions four and five
Albert’s fear reaction to furry objects remained but was less severe after time and in a different environment
Watson and Rayner (1920) - conclusion
Relatively easy to condition an emotional response to a neutral stimulus, just two sessions of pairing was enough to elicit a response
Watson and Rayner (1920) - strengths
Good experimental controls (increasing internal validity) - Albert carefully selected for his emotional stability, well controlled room, response to wooden blocks to see if his general anxiety increased, change in environment
Application to acquiring phobias - shows that phobias can be acquired through classical conditioning, leading to the development of effective therapies
Watson and Rayner (1920) - weaknesses
Some aspects not well controlled - reluctant animals placed in front of or pushed towards Albert, may have elicited fear response instead of the animal itself
Poor generalisability - only one participant, may have been unusual (participant variables)
Becker et al (2002) - aims
To investigate the impact of Western attitudes on the eating behaviours and related attitudes of teenage girls.
Becker et al (2002) - procedure
One group of girls questioned in 1995 (as television was introduced to Fiji) and another three years later. Initial sample of 63 ethnic Fijian girls with a mean age of 17, later 65 of the same age. Participants were interviewed and completed the EAT-26 questionnaires individually at home, with a translator present to aid with any communication difficulty, the participants height and weight was also measured.
The 1998 group was asked additional questions such as ‘Have you ever tried to change your diet in order to change your weight?’ and ‘Do parents or family ever say you should eat more?’
Girls who scored over the threshold for dysfunctional eating on the EAT-26 were given further recorded interviews
Becker et al (2002) - findings
Weight did not differ between the two groups with respective BMIs of 24.5 and 24.9
Television viewing did increase with 41% having a TV at home in 1995 compared to 71% in 1998
Dysfunctional eating scores increased with 29.2% scoring over the threshold of 20 in 1998 compared to 12.7% in 1995.
11.3% reported purging in 1998 , opposed to none in 1995 with 74% saying that they felt too large or fat and 69% having dieted (alien to Fijian culture).
Interviews suggested that they considered TV characters to be role models
Becker et al (2002) - conclusion
Women in Western television programmes became role models for the Fijian girls, leading them to desire thinner bodies than was the precious norm. This led to a rise in dysfunctional eating and a decline in body image
Becker et al (2002) - strengths
Reliable measurements - Rivas et al (2010)
Application to treating eating disorders - shows that modelling in the mass media can impact body image and eating behaviours, suggesting the introduction of ‘larger models’ may help treat eating disorders
Rivas et al (2010)
Found good internal reliability for the EAT-26 survey with a correlation of +0.9
Becker et al (2002) - weaknesses
Low validity of the EAT-26 - Nunes et al (2005) found that it did not predict eating disorders in a group of Brazilian women
Low generalisability - cannot generalise sample to other countries as Fiji has traditionally distinctive attitudes to body type (high BMIs attractive), therefore the culture clash is much larger than it would be for other countries
