User:Psych119a-wiki/sandbox

For other uses, see Social learning (disambiguation).

Observational learning is the learning that occurs through observing the behaviour of other people. Albert Bandura, who is best known for the classic Bobo doll experiment, discovered this basic form of learning in 1986. Bandura stressed the importance of observational learning because it allowed children especially, to acquire new responses through observing others behaviour. This form of learning does not need reinforcement for learning to occur, rather a model. A social model is a parent, sibling, friend, and teacher, but particularly in childhood a model is someone of authority or higher status. A social model is significantly important in observational learning because it allows one to cognitively process behavior, encode what is observed and store it in memory for later imitation. While the model may not be intentionally trying to instill any particular behavior, many behaviors that one observes, remembers and imitates are actions that models display. A child may learn to swear, smack, smoke, and deem other inappropriate behaviour acceptable through poor modeling. Bandura claims that children continually learn desirable and undesirable behaviour through observational learning. Observational learning suggest that an individual’s environment, cognitive and behaviour all integrate and ultimately determine how one functions.

Culture and environment also plays a role in whether observational learning will be the dominant learning style in a person or community. In some cultures, children are expected to actively participate in their communities and are therefore exposed to different trades and roles on a daily basis . This exposure allows children to observe and learn the different skills and practices that are valued in their communities . In communities where children’s dominant mode of learning is observation, it is seen that these communities rarely separate children from adult activities. This incorporation into the adult world at an early age allows children to use observational learning skills in multiple spheres of life. Culturally, they learn that their participation and contributions are valued in their communities and from this they start to feel as if it is their duty, as a community member, to observe contributions being made in order for them to gradually become involved and participate further. .

Stages of observational learning

Bandura's social cognitive learning theory states that there are four stages involved in observational learning:

1. Attention
2. Retention
3. Production
4. Motivation

Effect on behaviour

Observational learning can affect behavior in many ways, with both positive and negative consequences. It can teach completely new behaviors, for one. It can also increase or decrease the frequency of behaviors that have previously been learned. Observational learning can even encourage behaviors that were previously forbidden (for example, the violent behavior towards the Bobo doll that children imitated in Albert Bandura's study). Observational learning can also have an impact on behaviors that are similar to, but not identical to, the ones being modeled. For example, seeing a model excel at playing the piano may motivate an observer to play the saxophone.

Observational learning and effects on infants

Infancy is a period in life where children are quickly acquiring new skills. In one study utilizing 8-, 10-, 12-, 15- and 18-month-old infants, the researchers used an indirect object retrieval task, in which the object was not directly retrievable, to examine the development of observational learning in infants when learning a new skill. There was a control condition that acted as a baseline measure, in which the children were given the object for 30 seconds and their reaction to the stimulus was recorded. This was compared to the demonstration condition, where the infants received three demonstrations of a task relative to their age group’s ability, and two minutes after they had seen the last demonstration, they were given the object for 30 seconds to try and perform the task they had seen. Due to the difference in abilities between 8- and 18-month-old infants, the researchers had each age group perform the same relative task with a different absolute difficulty, reflecting what each age group can perform at that age. For example, 8-month-olds had a toy placed inside a box in front of them where only the two sides were open to retrieve the object, while 15-month-olds had a bottle with a small peg inside. Both tasks required the infants to retrieve the object, but the means by which they did it and the level of difficulty varied.

The researchers found that through observation, there was an increase in performance that occurred between 10- and 12-months of age. This means that once an infant reaches 12-months, they are able to perform significantly better on an object retrieval task if they have learned through observational learning than if they did not receive a demonstration of the object retrieval task. The authors took this to mean that observational learning starts to become an effective learning tool once an infant reaches 12-months of age. The authors mentioned that these results may also suggest that children younger than 12-months old may need more demonstrations to learn how to do a task than the older subjects in this study.

Observational Learning and Effects on Children

There has been much debate in regards to whether or not children are more prone to learn more from the same sex model or opposite sex model. A study examined this hypothesis, whereby 118 boys and girls in grades three and four were examined by cue cards. The cue cards displayed tools, household articles, and game. The child and/ or participant was tested by the same sex; and based on their response,the child was given a score which classified he or she into one of four conditions; masculine boy, feminine boy, masculine girl, feminine girl. Participants were then brought into a spare room, to watch a film involving an adult male and an adult female. After watching the film, the participants were asked to recall the adult models’ behaviour, which was then recorded by a checklist. The research found that the original hypothesis was supported. Results showed that boys demonstrated a stronger tendency to recall more of the male models behaviour, whereas girls recalled an equal amount of each adult’s behaviour. The author found that children might learn more from a male model as oppose to a female model because males are generally perceived as possessing a higher social powers. Therefore, while boys are not only being encouraged and rewarded for imitating a man role, they also view men as more powerful and therefore are highly motivated to learn from male models.

Age difference in observational learning

Albert Bandura highly stressed that developing children learn from different social models, meaning that no two children are exposed to exactly the same modeling influence. From infancy to adolescence, one is exposed to various social models. It was once believed that newborn children were unable to imitate actions until the latter half of first year. However a number of studies now report that infants in less than 7 days are able to imitate simple facial expressions. By the latter half of there first year, 9 month olds are able to imitate actions hours after first observing them. As they continue to develop, toddlers around age two are observing important personal and social skills by imitating a social model. An important developmental milestone of a two year old is referred to as deferred imitation. This means that children are not only constructing symbolic representations but can also recall information from memory Observational learning heightens around the age of elementary school children. Unlike toddlers, elementary kids are less likely to rely on imagine to represent an experience, instead they are able to verbally describe the models behaviour. Since this form of learning does not need reinforcement, it increases the likelihood of it occurring daily.

Compared to imitation

Observational learning differs from Imitation in that it does not require a duplication of the behavior exhibited by the model. For example, the learner may observe an unwanted behavior and the subsequent consequences, and thus learn to refrain from that behavior. For example, Riopelle, A.J. (1960) found that monkeys did better with observational learning if they saw the "tutor" monkey make a mistake before making the right choice.

Observational Learning and Peer Model Influences

Observational learning is very beneficial when there are positive, reinforcing peer models involved. Although individuals go through four different stages for observational learning: attention; retention; production; and motivation, this does not simply mean that when an individual's attention is captured that it automatically sets the process in that exact order. One of the most important ongoing stages for observational learning, especially among children, is motivation and positive reinforcement. Examples of this are scaffolding and guided participation where children are positively instructed on how they can improve a situation and where children actively participate alongside a more skilled person. <Schaffer, David et al. (2010). Developmental Psychology, Childhood and Adolescence. 284</ref>

Observational learning across cultures

Cultural variation can be seen in the extent of information learned or absorbed by children through the use of observation and more specifically the use of observation without verbal requests for further information. For example, children from Mexican heritage families tend to learn and make better use of information observed during classroom demonstration then European heritage children . Another example is seen in the immersion, of children in some Indigenous communities of the Americas, into the adult world and the effects it has on observational learning and the ability to complete multiple tasks simultaneously. This might be due to children in these communities having the opportunity to see a task being completed by their elders or peers and then trying to emulate the task. In doing so they learn to value observation and the skill-building it affords them because of the value it holds within their community. This type of observation would not be considered passive, but focus on the child’s intent to participate or learn within a community .

Observational learning in indigenous communities

Children observe elders, parents, and siblings completing tasks and learn to participate in them as they grow. Observational opportunities tend to be more prominent in indigenous communities because children integrate in adult activities. They are seen as contributors themselves and therefore they learn to observe multiple tasks being completed at once and can learn to complete a task, while still engaging with other community members without being distracted. The heightened value towards observation allows children to multi-task in actively engage in simultaneous activities. The exposure to an uncensored adult lifestyle incorporates children and allows them to observe and learn the different skills and practices that are valued in their communities .

Indigenous parent’s teaching styles are often shaped by their influence with western schooling. In traditional Mayan families, given the difference of the education levels of mothers, those with more years of formal education often prompt children to take turns in learning to solve a problem, while mothers with less education facilitate open ended discussion with the children . In a similar study among Mayan fathers and children showed that fathers with 0-3 years of education operated through a mix of observation and shared collaboration between adults and children, compared to the fathers with higher levels of education who structured a discussion .

Observational learning and children with autism

There are not a lot of studies done on the acquisition of knowledge through observation and there are none on observational learning in children with low-functioning autism according to Nadel, Aouka, Coulon, Grad-Vincendon, Canet, Fagard & Bursztejn (2011). This group of researchers set out to change this by looking at whether or not children with low-functioning autism are able to learn through observation only. There were two groups divided into two subgroups based on developmental age (24 or 36 months). The first group contained 20 children, aged four to nine diagnosed with Autism Spectrum Disorder (ASD), who were then divided by developmental age. These two subgroups were matched with 10 typical children for each developmental age group. The two groups received different tasks corresponding to their developmental age. The task involved a red box in which the children were trying to get the candy inside. This study lasted nine days in which all of the children were given the box on the first day and after trying to get the object out were then shown a demonstration video twice. The following day, the participants were only given the box with no video demonstrations to test for observational learning. This was repeated seven days later.

Nadel et al. (2011) found that typical children showed improvements after the first demonstration, with the younger children improving one week later and the older children improving after they had only seen the demonstration once. Children with autism improved after the second demonstration only. The authors believed that this means that children with autism progress the same as typical children, however, it takes longer for the children with autism to learn. They also believed this to mean that children with autism can form motor representations for a task without prior experience with that task and that they can correct motor representations after failure with a task. The researchers argued that the difference in the need for more demonstrations for children with autism was not due to a lack of attention but more as a result of an increased difficulty of creating a motor representation of an action that leads to a remote goal as opposed to an immediate goal.

In free play, children with autism often display self-stimulatory behaviours that actually work to inhibit appropriate behaviours and decrease their ability to learn new behaviours. In one study, the researchers were looking to determine the effect of observation of peers on appropriate toy-playing skills in autisticlike children. They were also examining the effects that training can have on autisticlike children in the training situation but also in a generalization setting. Before this study was done in 1986, there were no previous studies looking at the ability of autisticlike children to learn similar skills to the ones they learned in training through the observation of a peer in a different setting. This study wanted to provide insight into the generalization of skills of autisticlike children through observational learning. The participants included three autisticlike boys with a mean chronological age of 4.4 and a mean age of 2.5. Six other boys with a mean age chronological age of 4.3 and a mean age of 3.2 that had good receptive and language skills were chosen as the peer models for the experiment. Each participant took part in a pretest in order to determine ten different toys that the child did not play with appropriately, a baseline test of free play where children were allowed free play in which a peer and the 10 toys determined in the pretest were present, and two training sessions that were each followed by a generalization and maintenance condition. In the first training session, the participant would watch a peer that they did not see in the baseline test, who was correctly playing with a toy. The second training was the same as the first except that the participant was exposed to both a new peer model and a new toy. The amount of exposure to a modeled play task was the independent variable and it was also manipulated through additional tasks, models and settings. Tryon and Keane (1986) identified the dependent variables as the training task acquisition, generalization in free-play, and the frequency of both self-stimulatory behaviour and imitative play behaviours.

The authors found that all three of the autisticlike boys were able to learn to imitate the peer model and they learned to play with an unfamiliar toy in the training sessions through the observation of a peer model. In the following generalization and maintenance sessions, all of the boys learned to play with the unfamiliar toy that they had not be trained in. The researchers noticed that boys decreased their self-stimulatory behaviours as a result of the imitative play learned in training. The authors suggested that enhanced imitation of play behaviours may have been due to the use of multiple peer models. Therefore, the authors suggested that training autisticlike children with multiple peer models as examples may help to generalize their learning to new tasks that are similar to the observed task. It is also possible that the decrease in self-stimulation behaviours of autisticlike children may have actually just been that the self-stimulatory behaviours were simply replaced with more appropriate self-stimulatory behaviours. More research is needed on observational learning and children with autism in such areas as free play in the natural environment and whether or not observational learning interferes with how children spontaneously learn to play.

Bobo doll experiment

In 1965, Albert Bandura claimed that children learned by observing a social model. Therefore, to validate this claim, Bandura conducted a perhaps famous experiment at Stanford University called the Bobo Doll experiment. This experiment was based on a study of aggression, where he used an inflatable plastic toy that looked like a cartoon clown. These bottom-weighted toys were then given to nursery school children. Each of the school children watched a short film in which a social model demonstrated aggressive behavior towards the Bobo doll. These aggressive responses included hitting the doll and shouting "Bang bang!" and "Sockeroo!". The nursery children were divided into three experimental conditions. The three conditions consisted of model-rewarded, model-punished, and no-consequences. Children were then placed in a room with Bobo and other props used by the social model. Bandura discovered that children in the no-consequence and model-rewarded conditions imitated more aggressive actions than the children in the model punishment condition. Bandura did another experiment to investigate how much the children actually learned from observing. This time children were offered a juice box for reproducing the model's behavior. It was evident that all conditions learned the same amount by observing the model. This experiment is influential because it distinguishes between what a child learns through modeling and their willingness to perform these aggressive acts.

Other Experiments

Hummingbird experiment

Experiments with hummingbirds provided one example of apparent observational learning in a non-human organism. Hummingbirds were divided into two groups. Birds in one group were exposed to the feeding of a knowledgeable "tutor" bird; hummingbirds in the other group did not have this exposure. In subsequent tests the birds that had seen a tutor were more efficient feeders than the others.

Light box experiment

Children from about eighteen months to two years old appear to learn much from observing their parents. Pinkham and Jaswal (2011) did an experiment to see if a child would learn turn on a light box by watching a parent. They found that children who saw a parent use their head to turn on the light box tended to do the task in that manner, but children who had not seen the parent chose a more efficient way by using their hands.

Imitation or emulation: The ghost condition

There has been some disagreement as to whether young children’s observational learning is due to pure imitation of the model’s actions or rather, an imitation of the result of the model’s actions. In one study examining babies between 14 and 26 months old, the researchers noted whether or not the babies were able to learn through observation to manipulate an apparatus in a laboratory playroom in order to bring a toy within reach. The babies were categorized as younger (under 20 months) or older (over 20 months) and then put into one of three groups. Babies were either a) in a baseline condition where they were encouraged to attempt to retrieve the toy but not assisted in any way, b) in a modeling condition in which they watched a live model act upon the apparatus to retrieve the toy, or c) in a ‘ghost’ condition in which babies observed the apparatus moving without any apparent manipulation to bring the toy within reach. There was a ‘single-mat task’ in which the toy sat on a cloth; pulling the cloth resulted in the toy moving further away, whereas pushing the cloth brought the toy within the babies’ reach. A ‘double-mat task’ worked the same way except that the action had to be done to the cloth next to the one holding the toy. A concealed remote control operated the apparatus in the ghost condition. The researchers found that babies were more likely to perform the target action in the single-mat task if they were exposed to the ghost condition, whereas the modeling condition produced more target actions in the double-mat task. The authors argued that this means that young children learn difficult tasks (i.e. the double-mat task) better if they can observe a live model demonstrating the task. However, if it is a simpler task (i.e. the single-mat task), babies learn better merely by imitating the results of an action (i.e. in the ghost condition).

Peer Models On Swimming Skill Performance

When adequate practice and appropriate feedback follow demonstrations, increased skill performance and learning occurs. Lewis (1974) did a study of children who had a fear of swimming and observed how modelling and going over swimming practices affected their overall performance. The experiment spanned over nine days and included many steps. The children were first assessed on their anxiety and their swimming skills. They were then placed into one of three conditional groups and were exposed to these conditions over a few days. At the end of each day, there would be a group swimming lesson with every child involved. The first group was a control group where the children would watch a short cartoon video each day that was unrelated to swimming. The second group was a peer mastery group where the subjects would watch a short video of children similar in age who had very good task performances and high confidence statements. Lastly, the third group was a peer coping group where the subjects watched a video of children similar in age who progressed from low task performances and low confidence statements to high task performances and high confidence statements. The day following the exposures to each condition, the children were assessed. Finally, the children were also assessed a few days later for a follow up assessment. Upon reassessment, it was shown that the two model groups who watched videos of children similar in age had successful rates on the skills assessed because they perceived the models as informational and motivational.

See also

• Mirror neuron
• Cognitive imitation
• Learning
• Social cognition
• Inference
• Machine learning
• Educational psychology
• Educational technology
• Emulation (observational learning)
• Imitation
• Hypercorrection
• Social learning tools

Further reading on animal social learning

• Galef, B.G. & Laland, K.N. (2005). Social learning in animals: Empirical studies and theoretical models. Bioscience, 55, 489-499. Abstract
• Zentall, T.R. (2006). Imitation: Definitions, evidence, and mechanisms. Animal Cognition, 9, 335-353. (A thorough review of different types of social learning) Full text

Category:Social learning theory Category:Behavioral concepts Category:Learning