Psychology › Section L · Social psychology › Topic L.6
Aggression
Why do people hurt each other, and why do onlookers sometimes fail to step in? This topic covers the biological, emotional and learned roots of aggression, bullying online and off, and the bystander effect, including the true story behind its most famous example.
What counts as aggression
Psychologists define aggression as behaviour intended to harm someone who does not want to be harmed. Intent is the key: a dentist who causes pain is not aggressive, while a punch that misses still is. Hostile aggression is driven by anger and aims at causing pain, as in a fight that starts with an insult. Instrumental aggression is a means to another goal, such as a robbery or a deliberate foul to stop a breakaway. Aggression can also be direct (hitting, shouting) or indirect, also called relational aggression: damaging someone's relationships or reputation through rumours or exclusion.
Biological influences
The amygdala helps detect threats and trigger anger, while the prefrontal cortex helps restrain impulses, and damage to it is linked to impulsive aggression. Higher testosterone is associated with somewhat more aggression, but the link is modest and runs both ways, since competing also raises testosterone. Alcohol weakens self-control and is involved in a large share of violent incidents. Across countries, men commit the large majority of homicides. Evolutionary psychologists explain male physical aggression as competition for status and mates; this account is debated, and women's aggression is more often indirect.
Frustration and aversive events
The frustration-aggression hypothesis of John Dollard and colleagues (1939) held that being blocked from a goal always produces an urge to aggress. Leonard Berkowitz revised it in 1989. Frustration, like pain, heat, crowding or insults, produces negative feelings, and those feelings make aggression more likely. Studies have linked hot weather with higher rates of violent crime, for example. The revision explains why frustration often leads to sadness or problem solving instead. It also explains displaced aggression: taking anger out on an innocent target, such as snapping at a sibling after a bad day at work.
Learning to be aggressive
Aggression is also learned. In Albert Bandura's Bobo doll studies (1961), children who watched an adult hit and kick an inflatable doll were more likely to copy those actions, often exactly, than children who watched a calm adult. In later work, children who saw the model rewarded imitated more than those who saw the model punished. This is observational learning, covered in the Learning section. Whether violent video games increase real violence is contested: they are linked to small rises in aggression on laboratory measures, but in 2020 the American Psychological Association judged the evidence insufficient to link them to violent crime. Venting anger by hitting a pillow tends to increase later aggression, not reduce it.
Bullying and cyberbullying
Bullying, as defined by Dan Olweus, is repeated aggression in which a more powerful person or group targets someone less able to defend themselves. Repetition and a power imbalance separate it from a one-off fight. It can be physical, verbal or relational. Being bullied is linked to anxiety, depression, poorer school performance and, in some cases, self-harm. Cyberbullying uses phones and online platforms. It can be anonymous, reach a wide audience, continue around the clock and leave a permanent record, so a target may find no safe place.
The bystander effect
In a 1968 experiment by John Darley and Bibb Latané, students in separate booths heard over an intercom what seemed to be another student having a seizure. Of those who believed they were the only listener, about 85% went for help before the seizure ended; of those who believed four others could also hear, only about a third did. The presence of others made each person less likely to help: the bystander effect. Latané and Darley described helping as five decisions: notice the event, interpret it as an emergency, accept responsibility, know how to help and decide to help. Diffusion of responsibility, the sense that someone else can act, blocks the third step. Pluralistic ignorance, reading others' calm as a sign that nothing is wrong, blocks the second.
The Kitty Genovese story: what really happened
Bystander research was inspired by the murder of Kitty Genovese in Queens, New York, in March 1964. Two weeks later The New York Times reported that 38 respectable citizens had watched the attack for over half an hour without calling the police. Later investigations, including a 2007 analysis by Rachel Manning, Mark Levine and Alan Collins, found that story largely inaccurate. Few witnesses saw much of the attack, which happened in two stages and partly out of sight. At least one neighbour shouted at the attacker, at least one call to the police was reported, and a neighbour, Sophia Farrar, held Genovese as she died. In 2016 the Times acknowledged that its account had been exaggerated.
The bystander effect today
The parable was wrong, but the laboratory effect replicates. A meta-analysis by Peter Fischer and colleagues (2011) confirmed it, but found it weaker in clearly dangerous emergencies, perhaps because danger is unambiguous. A 2020 study by Richard Philpot and colleagues analysed CCTV footage of public conflicts in three countries and found that at least one bystander intervened in about nine of every ten, with intervention more likely when more bystanders were present. Larger groups can make each person less likely to act while still making it likely that someone does.
Worked example: classifying aggression
Classify each act. 1. A footballer, furious after a bad tackle, shoves the opponent: hostile aggression, driven by anger, direct and physical. 2. A defender deliberately trips a striker to stop a certain goal: instrumental aggression, harm used as a means to win. 3. A student spreads a rumour to push a rival out of a friendship group: indirect, relational aggression. 4. A nurse gives a painful injection: not aggression, because there is no intent to harm. 5. Check: classify by the motive and intent behind the act, not by how much damage it causes.
Worked example: a group chat pile-on
In a class group chat of 40 students, several start mocking one classmate with edited photos, day after day. 1. Bullying test: it is repeated, intended to humiliate and backed by group power, so it is cyberbullying, not banter. 2. Noticing: everyone sees the messages, so the first step is passed. 3. Interpreting: silence from others suggests it is 'just a joke', which is pluralistic ignorance. 4. Responsibility: with 40 members, each assumes someone else will object, which is diffusion of responsibility. 5. Remedy: one student messaging the target privately and another reporting the posts each take personal responsibility.
Common mistakes to avoid
First, defining aggression by the harm caused rather than by the intent to harm. Second, assuming frustration always leads to aggression: Berkowitz's revision says it raises the risk through negative feelings. Third, overstating the biology: testosterone's link with aggression is modest and two-way. Fourth, retelling the Kitty Genovese story as 38 people watching and doing nothing: that account was largely inaccurate. Fifth, concluding that crowds never help: in real conflicts someone usually intervenes. Sixth, calling every unkind act bullying: bullying is repeated and involves a power imbalance.
Key takeaways
Aggression is behaviour intended to harm. It can be hostile (anger-driven) or instrumental (a means to a goal), and direct or relational. Brain systems, testosterone and alcohol play a part. Frustration and other aversive events raise aggression through negative feelings (Dollard, 1939; Berkowitz, 1989), and aggression is also learned by observation (Bandura). Bullying is repeated aggression with a power imbalance; cyberbullying can be anonymous and hard to escape. The bystander effect, driven by diffusion of responsibility and pluralistic ignorance, replicates but is weaker in dangerous emergencies. The Kitty Genovese story that inspired it was largely inaccurate.
Flashcards
Flashcard
Behaviour intended to harm someone who does not want to be harmed
Test your understanding
Question 1 of 5
A hockey player calmly and deliberately trips an opponent to stop a likely goal. This is best described as:
Question 2 of 5
In Darley and Latané's intercom study, how did believing that four others could also hear the seizure affect helping?
Question 3 of 5
What did later investigations conclude about the 1964 report that 38 witnesses watched Kitty Genovese's murder and did nothing?
Question 4 of 5
According to Berkowitz's revision of the frustration-aggression hypothesis:
Question 5 of 5
Which feature distinguishes bullying from a one-off fight between equals?
