Study Notes on the Transmission of Aggression Through Imitation

Authors: Albert Bandura and Dorothea Ross, with contributions from A. Ross, Stanford University.
The study discusses the phenomenon of identification linked to incidental learning, focusing on children's mimicry of aggressive behavior.

Previous studies (Bandura & Huston, 1961; Blake, 1958; Grosser, Polansky, & Lippitt, 1951; Rosenblith, 1959; Schachter & Elall, 1952) have illustrated that children imitate adult behaviors observed firsthand.
Experimentation indicated that mere observation of aggressive models can facilitate immediate aggressive behaviors in children.

Children exposed to aggressive models will imitate aggressive behaviors in a new situation absent of the model. This imitation should differ markedly from children who observed non-aggressive models or who had no exposure.
Predictions include:
- Greater likelihood of boys imitating aggressive behaviors than girls due to social reinforcement structures around gender-specific behaviors.
- Subjects exposed to same-sex models would show higher imitation than those observing opposite-sex models, particularly in aggressive behaviors.

Total of 72 children (36 boys, 36 girls) from Stanford University Nursery School.
Ages ranged from 37 to 69 months, mean age 52 months.
Subjects matched on baseline aggression levels using a 5-point rating scale assessing physical aggression, verbal aggression, aggression towards objects, and aggressive inhibition (tendency to inhibit aggression).

The study divided subjects into three groups: aggressive models, non-aggressive models, and a control group (no model exposure).
Each condition further divided by gender, with half of all subjects exposed to models of the same sex.

Initial Phase: Subjects played with non-aggressive materials (e.g., potato prints) with corresponding models.
Model Behavior:
- Aggressive Model: Assembled toys while displaying aggressive interactions with a Bobo doll (e.g., punching, kicking).
- Non-Aggressive Model: Engaged in calm toy play without aggression.
The differing model presentations aimed to ensure subjects observed distinct behaviors without direct prompts to imitate.

Subjects underwent a mild aggression arousal phase (e.g., sharing attractive toys) before being evaluated for imitation in a new room to delineate experimental results from natural interactions in nursery school.
Behavior was categorized into types of imitative aggression, including physical (e.g., hitting Bobo doll) and verbal (e.g., repeating aggressive phrases).

Aggression Scoring: Three imitative measures recorded including:
- Physical aggression (direct actions to Bobo),
- Verbal aggression (repetition of aggressive phrases),
- Non-aggressive interactions or quiet behavior.
Aggression behavior was rated for inter-rater reliability (Pearson product-moment correlation at .89).

Significant imitation of aggressive behavior was confirmed for subjects exposed to aggressive models compared to non-aggressive and control groups.
Results indicated:
- Male subjects showed greater physical aggression imitation than females when exposed to a male aggressive model.
- Both boys and girls displayed similar levels of verbal aggression imitation.

Males exhibited more imitative physical aggression following exposure to male models compared to female models.
Girls were more likely to replicate non-imitative verbal aggression.
Evidence suggests diverging constructs of behavior depending on the perceived appropriateness of gender roles in model behavior.

This research suggests that children mimic aggressive behavior based simply on observation, potentially weakening their own inhibition towards aggression.
The observed imitation followed distinctive behavior patterns linked to the gender of both models and subjects, illustrating societal influence on behavior modeling.

The authors cited multiple foundational studies underscoring the role of social learning theory in understanding child behavior, emphasizing contexts of aggression modeling.
Further investigations were recommended to analyze the relationship dynamics of models with children, particularly in assessing the broader social implications of such aggressive behavior emulation.