Introduction & Social Explanations of Aggression

  • It is defined as a hostile or destructive tendency or behaviour.
  • There are several possible reasons:
  1. Legacy of an evolutionary past
  2. Biochemical/ neural causes
  3. Learnt behaviour
  4. Lack of cognitive
  • This was proposed by Bandura and Walters (1963)
  • According to Bandura, children learn their aggressive behaviour through observation, vicarious reinforcement and mental representation. Which is then imitated and then maintained or ceased based on their personal reinforcement.
  • Children watch or observe a behaviour and they may imitate it if they see the person being reinforced or punished. This is vicarious reinforcement. If when they imitate the behaviour and they are rewarded they are likely to maintain the behaviour. This is maintenance through direct experience.
  • Research support for SLT comes from Bandura et al (1961) and the Bobo Doll Studies.
  • The study involved male and female children between 3 and 5 watch the modelled behaviour of adults.

Half were exposed to an adult aggressively interactive with a Bobo Doll in a room, involving physical and verbal aggression. The other half were non-aggressive.

Following the exposure, the children were put in a room with attractive toys that they were not allowed to play with. This induced frustration in the children. They were then led to a room with other toys and the Bobo Doll.


  • The aggressive modelled group showed aggressive behaviour with 1/3 imitating the verbal noises made by the adults.
  • Children of the other group showed virtually no aggression towards the doll.
  • Boys reproduced more physical aggression but there were no differences found in verbal imitation between genders.


Although the study tells us that children do learn aggressive behaviour/ responses as a result of observation, it doesn’t explain why they’d imitate the behaviour without the presence of the doll.

In a later study, Bandura & Walters (1963) found that the children who saw rewards for aggressive behaviour showed high levels of aggression in their own play.

Those who saw the model punished showed low levels and those who saw neither a reward or sanction were somewhere in between.


It’s possible that the studies findings (Bandura) may be due to demand characteristics. The children may have known what was expected of them meaning that they were therefore just acting as told rather than because of the factors of the study. Noble (1975) reported that one child said  “Look Mummy, there’s that doll we have to hit” on arrival. The studies also use a doll rather than a real person, lacking mundane realism. In response to these criticisms, Bandura released a video using a live clown, which produced similar results.

Ethical issues make testing SLT experimentally hard. Exposing children to aggressive behaviour and then having them possibly replicate the behaviour raises ethical issues, such as protecting them from physical and psychological harm. As a result, there are few studies like Bobo and they would no longer be allowed to take place, meaning that it’s difficult to test the experimental hypothesis of SLT and aggressive children. Consequently, it is difficult to establish scientific credibility to the theory by this means.

The majority of studies involve children and therefore don’t explain adult behaviour; it is reductionist. However, Phillips (1986) found that daily homicide rates in the US almost always rise in the week following a major boxing match. This suggests that adult viewers imitate the aggressive behaviour viewed on their TV. Therefore this shows that Social Learning is evident in both adults and children.


Proposed theory by Zimbardo (1969) but it is based on the ‘classic crowd’ theory by Le Bon, 1895.

The situations of anonymity, large crowds and possibly a state of altered consciousness from drugs or alcohol will lead to reduced self awareness, decreased concerns of evaluation by others and less social inhibition. Zimbardo (1969) used an experiment with female undergraduates. He had them administer shocks to people in a ‘learning experiment’ Half of the participants wore their normal clothes and were referred to by name whereas the other half were in lab coats and their identity was hidden. The disguised condition p’s shocked for twice as long as the normal condition. Demonstrating that in a deindividuated state, where anonymity is strong, they will act differently.

Mullen (1986) analysed newspaper cuttings of lynchings and found that the bigger the crowd, the increased savagery in which they killed their victims.

Prentice-Dunn et al (1982) offer an alternative perspective to Zimbardo. They claim that it is reduced self-awareness rather than simply anonymity that leads to deindividuation. If an individual is self-focused they focus and act on their own internalised attitudes and morals, reducing the likelihood of antisocial behaviour. If they submerge themselves in a group they lose this focus and adopt the groups attitudes and morals and are more likely to engage in the behaviour as they are less privately aware and are less able to regulate their own behaviour.


There is research support for anonymity as a process of deindividuation. Rehm et al (1997) found that when German school children wore a uniform whilst playing sports they played more aggressively. However, Postmes and Spears (1998) found that there was insufficient support for the major claims of Deindividuation. They found that disinhibition and anti-social behaviour are not more common in large groups and anonymous settings. Therefore, evidence for the Deindividuation theory is mixed.

A benefit of the theory is that it can explain real world examples of aggression. Mann (1981) used Deindividuation to explain the ‘Baiting Crowd’, an aspect of collective behaviour. Mann analysed 21 suicide leaps during the 60s/70s and found that in 10 out of 21 cases, the jumpers were encouraged to jump by the mob/ baiting crowd. He also found these incidents tended to occur at night with a large crowd that was at a distance from the jumper. The events happening at night added to the anonymity. Collectively, these features were likely to produce a state of Deindividuation in members of the crowd, which made them behave in a way that they ordinarily may not.

There is a gender bias in the theory of Deindividuation. Cannavale (1970) found that male and female groups responded differently under Deindividuation conditions- displaying a gender bias. Diener et al (1973) found greater disinhibition of aggression in males, which agrees with Cannavale et al’s findings. In conclusion, this suggests that males maybe more prone to disinhibition of aggressive behaviour when deindividuated than females.  

A problem with the Deindividuation theory is that it doesn’t always lead to antisocial behaviour, rather it can increase pro-social behaviour. Spivey & Prentice- Dunn (1990) found that Deindividuation is dependent on situational factors for whether it leads to pro-social or antisocial behaviour. In the presence of pro-social environmental cues there were more altruistic acts performed by deindividuated participants and few antisocial acts compared to a control group. Therefore, evidence suggests that it cannot be certain that Deindividuation will lead to anti social behaviour.

  • Deindividuation as an explanation of aggression
  • Essay Plan ( 8 + 16 marks)
  • AO1
  • Zimbardo (1969) but based on ‘classic crowd’ by Le Bon 1895
  • Anonymity, large crowds and state of altered consciousness lead to
  • Reduced self awareness, decreased concerns and less social inhibition
  • Zimbardo 1969: female undergrads, electric shocks, half normal, half disguised. P’s in disguised condition, shocked x2 as long
  • Mullen (1986): newspaper cuttings of lynchings (US 1899-1946). Bigger crowd, increased savagery.
  • Prentice-Dunn et al (1982): alternate perspective to Zim. Reduced self-awareness, not anonymity. In a group, less privately aware, less able to regulate their own behaviour.
  • AO2
  • P: Supporting research from Rehm et al (1997)
  • E: When German school children wore uniform: more aggressive in sports
  • E: However, Postmes & Speares (1998): insufficient support, disinhibition and antisocial behaviour not more common.
  • E: Mixed evidence for theory
  • P: Real world examples explained (IDA)
  • E: Mann (1981): 21 suicide leaps, 60s/70s. 10 in 21 had a crowd. Also at night and from a distance. Adds to the anonymity
  • E: Deindividuation in the crowd- act in a way not normal to them
  • P: Gender Bias
  • E: Cannavale (1970): female and male groups respond differently to deindividuation conditions
  • E: Diener eta l (1973): greater disinhibition in males
  • Suggest males maybe more prone to disinhibition than females
  • P: Doesn’t always lead to antisocial behaviour, can increase pro-social behaviour
  • E: Prentice- Dunn (1990): dependent on situational factors
  • E: Cues influence behaviour
  • E: Therefore, it cannot be certain that deindividuation will lead to antisocial behaviour.