Lesson 7 of 10 · 18 min
Predicting aggression with the GAM
Use the general aggression model as a tool to predict a person's internal state, appraisal and action in a single episode, and how repeated episodes build an aggressive personality.
What you will be able to do
- Predict the outcome of a single episode from its person and situation inputs.
- Predict how a change to one input changes the outcome.
- Explain how multiple episodes build aggression-related knowledge structures and an aggressive personality.
- Explain why psychology cannot give definitive solutions to antisocial behaviour.
The syllabus asks you to make predictions with the GAM. That is a skill, and it has a method: work through the model in order, using only what the scenario tells you, and say how likely aggression is and why. The GAM predicts likelihood, not certainty: it tells you which way each input pushes.
Predicting a single episode
List the person factors
Traits, beliefs, attitudes, goals, scripts. Say whether each pushes towards or away from aggression.
List the situation factors
Provocation, frustration, heat or pain, aggressive cues, alcohol, incentives, and anything that reduces them (an apology, a calm friend).
Predict the internal state
Affect (anger?), cognition (hostile thoughts? an aggressive script?), arousal (high?).
Predict the appraisal
Will the immediate appraisal be hostile? Does the person have the time and capacity to reappraise, and does the outcome matter enough?
Predict the action
Impulsive or thoughtful, and aggressive or not.
Predict the feedback
How the encounter goes, and what that does to the inputs for next time.
Put the stages of one episode of the general aggression model in order.
Worked example
Make a GAM prediction for one episode
Liam, 19, believes that if you let people push you around they will keep doing it. He has had a long, hot shift at work and two drinks at a party. A stranger spills a drink down Liam's shirt and laughs.
Use the general aggression model to predict how Liam will respond. Justify your prediction.
Step 1: Person inputs
Liam's belief that you must not let people push you around makes aggression seem justified and useful.
Step 2: Situation inputs
Provocation (the spill and the laugh), discomfort (heat, fatigue) and alcohol.
Step 3: Internal state
Affect: anger at being laughed at. Cognition: hostile thoughts ("he did that on purpose") and his "don't let them push you" script. Arousal: raised by the provocation, heat and alcohol.
Step 4: Appraisal
His immediate appraisal is likely to be hostile. The outcome matters to him, but alcohol and fatigue reduce his cognitive resources, so reappraisal (perhaps it was an accident and the laugh was embarrassment) is unlikely.
Step 5: Prediction
Liam is likely to respond with impulsive aggression, such as shouting or shoving the stranger.
Every input pushes the same way and the resources needed to reappraise are low, so the model predicts an impulsive, aggressive response as the most likely outcome.
Step 6: Feedback
If the stranger backs down, Liam's belief that aggression works is reinforced, making aggression more likely next time.
Try it yourself first, then check the working one step at a time.
Answer
With a pro-aggression belief, provocation, heat, fatigue and alcohol, Liam is likely to feel angry, think the spill was deliberate and be highly aroused; with few resources to reappraise, the GAM predicts an impulsive aggressive response, which could reinforce his belief.
Method
Person inputs, then situation inputs.
Affect, cognition, arousal.
Immediate appraisal; can the person reappraise?
Impulsive or thoughtful action, and the feedback.
Watch out for
Predicting "he will hit him" with no reasoning through the model.
Forgetting that alcohol and fatigue act on the resources for reappraisal.
Changing one input. Many exam questions change one detail and ask what the GAM predicts. Work out which part of the model the change acts on.
- Make the situation hotter, louder or more frustrating: more hostile affect and arousal, so a hostile appraisal is more likely.
- Add a weapon or a violent image: aggressive thoughts are primed (cognition).
- Add alcohol, fatigue or time pressure: fewer resources, so less reappraisal and more impulsive action.
- Add an apology or a clear accidental cause: less hostile cognition, so a benign appraisal is more likely.
- Give the person a belief that aggression is wrong, or a well-practised calm script: person inputs push away from aggression.
Two people in a long queue on a hot day are both pushed in front of. Ella has just been told she has plenty of time before her bus; Mikaela is late, tired and has a bus leaving in two minutes. Both have an immediate hostile appraisal. What does the GAM predict?
Multiple episodes. Each episode is also a learning trial. When a person repeatedly experiences or practises aggression (by being provoked often, being reinforced for aggression, or repeatedly watching or playing violent media), the GAM predicts learning, rehearsal and reinforcement of aggression-related knowledge structures:
- aggressive beliefs and attitudes ("aggression works", "violence is normal");
- aggressive perceptual schemata: seeing ambiguous events as hostile (the hostile attribution bias);
- aggressive expectation schemata: expecting others to be aggressive;
- aggressive behaviour scripts: well-practised, easily triggered sequences of aggressive action;
- aggression desensitisation: less emotional reaction to violence, so less discomfort at using it.
Together these make up an aggressive personality. That changes the person inputs of every later episode. It can change the situation inputs too: an aggressive person may lose some friends, attract more conflict and choose aggressive peers.
Multiple episodes in the GAM: each episode makes the next one more likely to be aggressive.
Each episode adds to the knowledge structures. Watch how they change the person inputs of the next episode.
Worked example
Predict the effect of repeated episodes
Zara, 14, spends about three hours a day playing an online game in which players earn points and status by attacking other players, and taunting is common in the chat. Over a year, her parents notice she argues more, reacts angrily when her brother accidentally bumps her, and seems unbothered by violent news.
Use the general aggression model to explain how repeated episodes could produce these changes, and predict how Zara is likely to respond the next time someone at school bumps into her. Note one limit on your prediction.
Step 1: Identify the repeated episodes
Each session is an episode with violent content, reward for aggression (points and status) and provocation (taunts). Three hours a day for a year is extensive rehearsal and reinforcement.
Step 2: Name the knowledge structures built
Aggressive beliefs and attitudes (aggression earns status); aggressive perceptual schemata (her angry reaction to an accidental bump suggests she now sees ambiguous acts as hostile); aggressive expectation schemata; aggressive behaviour scripts (well-practised responses to provocation); desensitisation (unbothered by violent news).
Step 3: Link to personality and inputs
Together these increase her aggressive personality, which becomes a set of person inputs she carries into every new episode, and her arguing may also change her situations (more conflict at home).
Step 4: Predict the next episode
When bumped at school, her hostile perceptual schema makes a hostile immediate appraisal likely, her aggressive script comes to mind quickly, and she is likely to respond aggressively (snapping or pushing back), especially if she has little time to reappraise.
Step 5: Limit the prediction
The GAM gives likelihoods. Other inputs (a supportive family, school rules, her own values, a friend nearby) could lead her to reappraise. The research shows a small association between violent games and aggression, not certainty, and the changes could also have other causes.
Try it yourself first, then check the working one step at a time.
Answer
Repeated rewarded, violent play rehearses and reinforces aggressive beliefs, hostile perceptual and expectation schemata, aggressive scripts and desensitisation, building an aggressive personality. As a person input, this makes a hostile appraisal and an aggressive response to the next bump more likely, though not certain.
Method
Describe what is repeated and how it is reinforced.
Name the knowledge structures, with evidence from the stimulus.
Show how they become person (and situation) inputs.
Predict the next episode through appraisal to action, and state the limits.
Watch out for
Explaining only the single-episode effect (arousal during play) when the question asks about change over a year.
Claiming the game will make her violent. Predict increased likelihood of aggression.
After years of conflict at home, Noah now assumes that when anyone bumps into him, they meant to. Which knowledge structure does this show?
Note
The limits of psychology
The GAM predicts what is more or less likely, not what a particular person will do. Aggression comes from many interacting causes, people differ, and the most harmful violence is rare, which makes it hard to study. Ethics rules out experiments that provoke real violence, so laboratory studies use stand-ins such as how loud a noise blast a participant gives an opponent. So psychology can reduce the harmful expression of antisocial behaviour (anger management programs that teach reappraisal, reducing aggressive cues, building calmer scripts) but cannot offer a definitive solution that works for everyone.
Exam tip
Making a prediction earns marks for the reasoning
A one-word prediction ("aggressive") earns little. Name the inputs from the scenario, say which route each acts on, state the appraisal and whether reappraisal is possible, give the predicted action, and for multiple episodes name the knowledge structures and how they change the next episode.
Summary
Predict a single episode in order: person inputs, situation inputs, internal state, appraisal, action, feedback.
Heat, provocation and aggressive cues push towards aggression; alcohol, fatigue and time pressure remove the resources needed to reappraise.
Multiple episodes build aggressive beliefs and attitudes, perceptual and expectation schemata, behaviour scripts and desensitisation: an aggressive personality.
An aggressive personality changes the person inputs (and often the situation inputs) of later episodes, so aggression becomes more likely each time.
The GAM predicts likelihood, not certainty: psychology can reduce antisocial behaviour but cannot give a definitive solution.
Finished the lesson?
Try the 3 quick checks above and the lesson completes itself, or mark it complete now.