The imitation of aggressive behavior, modeled by another individual
Imitative aggression, or the imitation of aggressive behaviors modeled by another, serves an important developmental process. Specifically, imitative aggression may function to use others as models for appropriate behaviors in specific environments or circumstances. By imitating others’ behaviors in similar situations, individuals are able to learn vicariously through the experiences of those they model, decreasing their chances for experiencing negative outcomes and increasing their chances for experiencing positive outcomes in those situations. Thus, individuals who utilize imitative aggression are able to maximize their fitness in situations which they themselves have not yet experienced.
Imitative aggression research dates back to Bandura’s seminal works, assessing the effects of modeling aggression directed toward bobo dolls on children’s aggressive behaviors. These experiments suggested that demonstrating aggression to children made them more likely to imitate that behavior and increased the aggressiveness with which they played in the subsequent experimental sessions (Bandura 1965; Bandura et al. 1961, 1963a, b). These experiments also contributed to the development of social learning theory, which states that learning occurs through direct experience and observing others’ behavior. However, subsequent research has been unable to assess imitative aggression as rigorously as Bandura, due to ethical concerns (Barrett 2007). Instead, recent research generally uses exposure to media violence in place of a more direct and intense experimental manipulation, to assess these relationships.
Bandura’s Studies on Imitative Aggression
Bandura and colleagues’ first imitative aggression study randomly assigned children to an aggressive, nonaggressive, or control condition. Children in the aggressive and nonaggressive conditions were individually led into a room and provided with art supplies. After the child was settled in, the experimenter led a model into the room as well and had them sit on the opposite side of the room from the child. The model had some tinker toys and a mallet to play with, as well as an inflated bobo doll. In the nonaggressive condition, the model quietly played with the tinker toys when the experimenter left the room. However, in the aggressive condition, the model acted aggressively toward the bobo doll by sitting on the doll and punching it, striking it in the head with the mallet, and kicking it around the room. Additionally, the model also stated several distinct, verbally aggressive phrases while aggressing against the bobo doll. Control participants, on the other hand, did not have any interaction with the model at all. Then, the child was led into a new experimental room, filled with a variety of aggressive and nonaggressive toys, including a bobo doll and a mallet, and were given 20 min to play freely, while raters observed their play through a one-way mirror and coded it as imitative and aggressive, nonimitative and aggressive, or nonaggressive. Results of the study showed that individuals in the aggressive condition were more likely than either control or nonaggressive participants to verbally and behaviorally imitate the model’s actions. On the other hand, subjects in the nonaggressive condition were substantially more likely than subjects in either of the other groups to simply sit quietly, without playing with any of the toys at all. These results were the first to show that the observation of others’ behaviors is sufficient to elicit imitative behaviors which an individual almost certainly would not have exhibited otherwise (Bandura et al. 1961).
Subsequent experiments by Bandura replicated his original finding and extended the implications to other similar situations as well. For example, in one experiment, aggressive behavior toward a bobo doll was either demonstrated by a real-life aggressive model, in a paradigm identical to Bandura and colleagues’ original study (1961), with a film clip depicting a human model behaving in the same manner as the real-life model, using a cartoon film clip, in which a cartoon character behaves in a similar manner to the model in the above conditions or was not demonstrated at all. When children were allowed to play freely, participants who viewed the non-cartoon film clip displayed the most total and imitative aggression, followed by individuals in the real-life, cartoon film, and control conditions, respectively (Bandura et al. 1963a). Additionally, in a separate experiment, subjects were shown a film clip in which an aggressive child either gets punished for his aggressive behavior or gets rewarded for his aggressive behavior, and only children who observed the child get rewarded for his actions displayed increased imitative aggression during a subsequent play session; those who saw the child get punished showed similar levels of imitative and total aggression to control participants, who did not view a film clip (Bandura et al. 1963b). Finally, one last experiment showed that the inhibitory effects of viewing an aggressor be punished can be eliminated if the researcher subsequently offers the subject a reward for imitating his behavior. Specifically, Bandura (1965) noted that participants who viewed a model get rewarded for his aggressive behavior showed identical levels of imitative aggression as those who saw no consequences of his aggression, but those who saw him get punished were less likely to imitate his behavior. However, after the children in all conditions were offered rewards for imitating the model’s behavior, children who previously saw the model get punished showed identical levels of imitative aggression to those in the other two experimental conditions (Bandura 1965).
Social Learning Theory
When trying to determine possible boundary conditions for imitative aggression, informed by social learning theory, Bandura (1986) concluded that the following four elements must be present in order for imitative aggression to occur: First, the individual must pay attention to whomever is modeling their behavior. Without paying attention, subsequent steps of this process would be impossible, as the subject would have no knowledge of the behaviors they were going to imitate. Next, the individual must retain a memory of the aggressive behavior the model engaged in. In order to ensure that the subject retained a memory of the aggressive actions in his experiments, Bandura et al. (1961, 1963a, b) included both physically and verbally aggressive behaviors, as it would be virtually impossible for a subject to reproduce both types of behaviors spontaneously. Additionally, the model must be capable of imitating the actions of the model, because if they are unable to properly imitate the model’s actions for any reason, they will not be able to aggress in the same way that the model did. Finally, the individual must have the desire to imitate the aggression, and such desire is much more likely to be present if he saw the model rewarded for his aggressive actions than if he seems him punished for the same actions or if the subject is offered a reward for imitating the subjects actions (Bandura 1965, 1986; Bandura et al. 1963b).
Due to ethical concerns, recent research has not used procedures similar to Bandura et al. (1961) when assessing imitative aggression. Specifically, the Belmont Report states that one of the fundamental ethical principles for conducting research using human subjects is beneficence, meaning that one must “do no harm” to human participants (National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research and Ryan 1978). Since experimentally inducing subjects, especially children, to behave in an aggressive manner is considered harmful (Barrett 2007), contemporary research on imitative aggression does not use procedures similar to those used by Bandura (1961). Instead, researchers have used proxies which they may use in place of directly exposing participants to aggressive behavior, with the intention of having them imitate it.
Contemporary Research and Media Violence
The proxy most commonly used by contemporary researchers of imitative aggression is violence encountered by participants willingly, through their consumption of media, including movies, television shows, and video games. While a limited amount of research has found little to no effect of violent media and real-world aggression (e.g., Decamp and Ferguson 2016), the overwhelming majority of experimental research disputes this and suggests instead that increased exposure to violent media causes increased levels of hostile expectations (Hasan et al. 2013), increased aggressive cognitions and affect (Carnagey and Anderson 2004) short- and long-term increases in aggressive behavior (Huesmann, Moise-Titus, Podolski, & Eron, 2003), and an overall negative impact on the effected population (see Huesmann and Taylor 2006 and Wiedeman et al. 2015 for reviews). Thus, although the study design associated with research assessing aggressive or violent media cannot assess the imitative nature of the increased aggression as stringently as Bandura’s studies did, the tenets of social learning theory suggest that much of the increases in aggression caused by exposure to violent media may be due to increases in imitative aggression by those who are exposed to it.
Since Bandura’s initial studies, researchers have also been interested in how imitative aggression may affect male and female participants differently. Early work examining possible sex differences in imitative aggression largely showed that males tended to be more aggressive than females were. However, more recent research has not consistently replicated this finding. Specifically, Bandura’s seminal bobo doll studies showed that, in most cases, male participants displayed higher levels of both imitative and nonimitative aggression than females did (Bandura 1965; Bandura et al. 1961, 1963a, b). However, more recent research examining sex differences and the effects of exposure to media violence have been mixed. For instance, a study by Bartholow and Anderson (2002) showed that playing a violent video game caused an increase in aggression relative to a nonviolent game in both sexes but that this effect was larger in male participants than it was in female participants. On the other hand, other studies have found that there are no sex differences when examining the effect of exposure to violent video games (Hasan et al. 2013) or other violent media (Bushman 1995), on subsequent aggressive behaviors.
Bandura’s groundbreaking studies demonstrated that viewing specific, aggressive behaviors causes children to imitate those exact behaviors, especially when they are associated with receiving a reward or even a lack of punishment (Bandura 1965; Bandura et al. 1961, 1963a, b). Although contemporary research does not induce specific aggressive actions for ethical reasons (Barrett 2007), research on the effect of media aggression demonstrates that exposure to violent media causes increases in aggressive cognitions, affect, and behaviors (Carnagey and Anderson 2004; Hasan et al. 2013; Huesmann et al. 2003), which may be due, in part, to increases in imitative aggression. Future research on imitative aggression should more closely examine this relationship, to determine the extent to which the increases in aggression are due to direct imitation, rather than more general increases in aggression. Additionally, research should seek to more definitively determine whether men are more susceptible to the effects of media violence than women are or if this is merely an artifact of cultural influences.
- Bandura, A. (1986). Social foundations of thought and action: a social cognitive theory. Englewood Cliffs: Prentice-Hall, Inc..Google Scholar
- Barrett, M. (2007). Practical and ethical issues in planning research. Research Methods in Psychology, 24–48.Google Scholar
- Carnagey, N. L., & Anderson, C. A. (2004). Violent video game exposure and aggression. Minerva Psichiatrica, 45(1), 1–18.Google Scholar
- DeCamp, W., & Ferguson, C. J. (2016). The impact of degree of exposure to violent video games, family background, and other factors on youth violence. Journal of Youth and Adolescence, 1–13.Google Scholar
- Hasan, Y., Bègue, L., Scharkow, M., & Bushman, B. J. (2013). The more you play, the more aggressive you become: a long-term experimental study of cumulative violent video game effects on hostile expectations and aggressive behavior. Journal of Experimental Social Psychology, 49(2), 224–227.CrossRefGoogle Scholar
- National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research, & Ryan, K. J. P. (1978). The belmont report: ethical principles and guidelines for the protection of human subjects of research-the National Commission for the protection of human subjects of biomedical and behavioral research. US Government Printing Office, Washington, D.C.Google Scholar