Bandura

Transmission of Aggression Through Imitation of Aggressive Models (1961)

Learning Approach

looks at observational learning, one of the main learning theories

Psychology

studies the social learning theory

social learning theory

comprised of attention, retention, reproduction, and motivation; can happen in two different ways - first-hand and by proxy

attention

must be paying attention to role model to learn in the first place

retention

need to store observations in long-term memory

reproduction

need to imitate the stored behavior

motivation

if rewarded for imitation, you want to repeat behavior, if punished, you want to stop behavior

first-hand social learning

in presence of person completing behavior

social learning by proxy

watched behavior through some type of media

Pavlov

(Background)

studied classical conditioning in dogs

John B. Watson

(Background)

applied Pavlovian principles to humans

Skinner

(Background)

taught us about learning through consequences

Bandura

(Background)

believes early learning theories are reductionistic

previous studies

(Background)

• proved children will imitate and copy behaviors

• behavior happened immediately in the same environment with role model still there

• never took gender into account

Aim

to investigate if social behaviors, particularly aggression, can be acquired through observation and later imitated

Sub-Aim #1

to test if we can learn to imitate a behavior when the role model is no longer present

Sub-Aim #2

to observe for any great differences in learning aggression

Hypothesis #1

If a behavior is observed, it will be imitated

Hypothesis #2

If a behavior is not observed, it cannot be imitated

Hypothesis #3

Boys will copy a male model more than a female model, and girls will copy a female model more than a male model

Hypothesis #4

Boys will be more predisposed than girls to imitating aggression

Independent Variables

conditions (aggressive model, non-aggressive model, and control group), gender of model, gender of child

Dependent Variable

number of aggressive behaviors out of a 240 maximum

Method

• laboratory experiment

• independent measures: can’t take out the behavior once it’s been placed there

• data collected in the behavioral direction and direct observation

• matched groups between non-aggressive and aggressive groups for pre-existing levels of aggression

Room 1 (Modeling) Apparatus

• potato prints and stickers

• tinker toys

• table and chair

• mallet

• 5 foot Bobo doll

Room 2 (Frustration-Aggression) Apparatus

• fire engine

• locomotive

• dolls

• top

Room 3 (Observation)

• potato prints and stickers

• tinker toys

• table and chair

• mallet

• 3 foot Bobo doll

• one-way mirror

• 2 dart guns

Participants

72 children (36 male, 36 female) ranging from 37 to 69 months in age, with a mean age of 4.4 years, randomly sampled from Stanford University Nursery School; likely higher on the socioeconomic scale and have at least one college-educated parent

Control #1

Model will always do the same behaviors in the same sequence for the same amount of time

Control #2

The toys in rooms 1 and 3 will always be in the same places at the start of the study

Control #3

The observations will always be done by 2 independent observers for the same duration (20 minutes), and each will take data every 5 seconds

Control #4

The levels of pre-existing aggression will be assessed by both the researcher and the nursery school teacher

Step 1 (Procedure)

Every child is assessed for their pre-existing levels of aggression, then each child between the aggressive and non-aggressive group will be matched.

Step 2 (Procedure)

Each child will be individually brought by the experimenter into Room 1, where there will be a model for the aggressive and non-aggressive groups. The child will be allowed to settle in, while the model plays in the corner. In the non-aggressive model, everyone continues playing, and in the aggressive model, the model will punch the Bobo doll, hit it with the mallet, throw it, and kick it, vocalizing their actions as they perform them

Step 3 (Procedure)

After 10 minutes in Room 1, the child will be taken into Room 2. The child will be allowed to play with the toys for 2 minutes, and then the child is told the toys are “only for the best children.”

Step 4 (Procedure)

After the child is frustrated, they are brought into Room 3, where they will be left with the toys for 20 minutes, and their behavior will be observed through the one-way mirror. The child is sent back to the nursery school afterward

Data

quantitative - behavioral checklist

qualitative - in Room 1, researchers wrote down comments made by the children

imitated aggression

exact reproduction of model’s behavior

partially imitated aggression

similar reproduction of model’s behavior

non-imitated aggression

something the model didn’t do

Finding #1

Children who saw the aggression model were significantly more aggressive than the non-aggressive and control groups

Finding #2

There is very little difference in aggression levels between the non-aggressive and control groups

Finding #3

Boys were significantly more likely to imitate an aggressive male model

Finding #4

Boys were significantly more physically aggressive

Finding #5

Children in all groups played aggressively with the dart guns

Conclusion #1

Observed behavior is more likely to be imitated

Conclusion #2

All hypotheses were supported

Confidentiality (Ethics)

maintained; pictures of the children were published with the study but it was allowed by parents

Ethics

Loco parentis would have been required in the use of children, and informed consent was likely given by the nursery school teacher, not the parents. Protection from harm wasn’t upheld; children were exposed to dart guns and the mallet, it is unknown how long the effects of the study last, and positive replacement was not used for the aggressive behavior. Debriefing did not occur

Strength #1

There is high inter-scorer reliability in the pre-existing aggression (0.89) and the observations (0.9)

Strength #2

The use of quota sampling provides comparable data

Strength #3

Matched groups controlled for aggression and as an independent variable

Strength #4

There was a high level of controls, so it is highly likely only the aggressive model made a difference in the outcome of the study

Strength #5

High standardization of the procedure allows for replication

Weakness #1

There is low ecological validity in a laboratory setting

Weakness #2

The task of a child playing with toys while an adult plays with kids’ toys lacks mundane realism

Weakness #3

The children coming from higher socioeconomic backgrounds and having a small age range makes generalization difficult

Weakness #4

It is unknown whether the children were compliant to please the adults around them

Weakness #5

The aggression was taken out on an inanimate object

Weakness #6

The long term effects of the aggression are unknown

Weakness #7

It is unknown whether the novelty of the toy made a difference in aggression

Application

(Issues & Debates)

• to TV networks limiting aggressive programming and promoting pro-social programming

• helping parents pick appropriate programming for their children

Nature vs. Nurture

(Issues & Debates)

Matched pairs nullified nature affecting the outcome

Use of Children (Issues & Debates)

Positive

• honest/lack of a social filter

• demand characteristics aren’t an issue

Negative

• may have been distressed

• exposed to psychological harm through intentional frustration