Cooper and Mackie’s 1986 study tested whether violent video games make children more aggressive. In a laboratory, 84 American children played or watched a violent game, a mild game or a maze task. Afterwards, the violent game made girls play more with an aggressive toy. Boys’ play with that toy hardly changed.
Key Takeaways
Aim
- To find out whether violent computer games affect children, and whether they affect boys and girls differently.
Method
- A laboratory experiment (a controlled test in a special room) with an independent measures design: each child took part in one condition only.
- The sample was 84 children (44 boys, 40 girls) from the fourth and fifth grades in Trenton, USA, in pairs of the same sex.
- In each pair, one child played for eight minutes while the other watched. The game was Missile Command (violent), Pac-Man (mild) or a maze task (the control).
- Each child then spent eight minutes in a toy room, secretly watched. One toy fired darts: the aggressive toy.
- Each child also pressed a buzzer for longer to punish an imaginary child more. This measured interpersonal aggression (aggression towards other people).
- The independent variables (what was changed) were the game and playing or watching. The dependent variables (what was measured) were toy play, the buzzer task and ratings.
Results
- Girls who played Missile Command spent about 80 seconds with the aggressive toy. Girls in the other two groups spent about 15 seconds.
- Boys spent more time with the aggressive toy than girls did, but the game they had played hardly changed their play with that toy.
- Watching had the same effect as playing. The game did not change how long any child pressed the buzzer.
- More boys than girls had a video game at home: 34 of 44 boys and 17 of 40 girls.
Conclusion
- The violent game changed girls’ play with the aggressive toy. It did not change boys’ play with it, or any child’s buzzer scores.
- The study is used as evidence for social learning theory (children learn behaviour by watching and copying).
Strengths
- Control: every player had eight minutes on the game, and every child had the same toys and tasks. Other researchers can repeat this standardised procedure.
- Random allocation: pairs were put in the games by chance, and the maze group gave a fair comparison with no violence.
- Covert recording (secret recording): children did not know their toy play was being recorded, so they were less likely to guess the aim and change their play (demand characteristics).
Limitations
- Ecological validity (how well results fit real life): children played in an experimental room with an adult present, so the results may not fit everyday life.
- Construct validity (whether a test measures what it claims to): aggression is complex, and a buzzer or a toy measures only one small part of it.
- Population validity (whether results fit the wider group): all the children were fourth and fifth graders from one American area, so the results may not apply to other ages or countries.
Ethical Issues
- Informed consent (agreeing after being told what will happen): parents agreed, but children may not have been told the aim. A sensible way of dealing with it: ask the children too.
- Deception (misleading people): children were given cover stories, which are made-up reasons for the tasks. A way of dealing with it: debrief them afterwards.
- Protection from psychological harm: the harm was probably small, but some girls disliked Missile Command. A way of dealing with it: let children stop at any time.
Exam Use
- In the exam you get marks for the method and design in the right order: laboratory experiment, then independent measures.
- In the exam you get marks for saying the games affected girls more than boys, especially Missile Command. Saying girls are more aggressive earns nothing.
- In the exam you get marks for saying the buzzer lacks construct validity (it does not measure aggression well). A better study could use a wider age group, an arcade and a longer time.
Background
Before this study, researchers had found that violent television could make children more aggressive. Most of that research used boys. The evidence for girls was mixed.
Cooper and Mackie thought girls might react more strongly than boys. Girls usually watch less violence than boys. People who watch a lot of violence get used to it (they become desensitised), so new violence excites them less.
So children who see the least violence should be the most excited by it. They should also be the most likely to act aggressively.
Video games were new and hugely popular. Most players were boys, so girls had less experience of them. Games also let researchers compare children who play with children who only watch.
Aim
The aim was to find out whether computer games that show a lot of violence affect children. The authors also wanted to know whether the effect is different for boys and for girls.
A hypothesis is a prediction that a study can test. The study had two.
- Boys and girls will react differently to video games.
- Aggressive video games will affect children’s behaviour more than non-aggressive video games.
Method
Design and Variables
Cooper and Mackie (1986) ran a laboratory experiment. This is a test in a controlled setting, where the researcher changes one thing to see its effect.
The design was independent measures. Each child took part in one game condition only, and as either a player or a watcher. Children never swapped roles. Later, the pair swapped rooms for a second round of measures, as the Procedure shows.
The independent variable is what the researcher changes. The dependent variable is what the researcher measures. This study had two independent variables.
- The game: a high aggression game (Missile Command), a low aggression game (Pac-Man) or a control condition (a pen-and-paper maze task).
- The child’s role: a player, or an observer who watched.
There were three dependent variables, all measured afterwards.
- Toy play: how long each child spent with each of four toys. This used observation, which means watching and recording behaviour.
- Interpersonal aggression: aggression aimed at another person. Here it was how long each child pressed a buzzer to punish or reward an imaginary child.
- Ratings: the children’s own answers on questionnaires. This is self-report.
The researchers also compared boys with girls. They could not give children a sex, so sex was not something they changed.
Sample
The sample was 84 children: 44 boys and 40 girls. They were fourth and fifth graders (two school years in the USA) from a school system in suburban Trenton, New Jersey. About half were in each grade.
Children were paired at random, with one rule. Both children in a pair had to be the same sex and the same grade. Parents gave their consent.
Materials
- The games: Missile Command and Pac-Man, played on an Atari (a home video game console). In Missile Command you destroy laser beams before they hit your cities. In Pac-Man you chase ghosts.
- The control task: children used magic markers to solve mazes with cartoon characters on them.
- The four toys: an aggressive toy (a Shogun Warrior figure with dart-firers), an active toy (a basketball set), a skill toy (a pinball game) and a quiet toy (building pieces).
- The buzzer panel: a red button for a “bad buzzer” and a green button for a “good buzzer”.
- Questionnaires: on video game experience and on what the children thought of the games.
Other children had rated the toys beforehand to check each fitted its label.
Procedure
- A week earlier, children with parental consent filled in a questionnaire about their video game experience.
- Each pair was put at random into the Missile Command, Pac-Man or maze condition. One child was chosen at random to play.
- The player played for eight minutes. The other child watched and could advise, but could not touch the controls.
- The child then went to a toy room. An adult said there would be a short delay and the child could play with any toy.
- The adult sat in a corner and seemed to be adding numbers. In fact she was secretly recording which toys the child used, and for how long, for eight minutes.
- Meanwhile, the partner was in a second room. Children were told their ideas would help stop other children doing “silly or bad” things. This was a cover story.
- The child chose the worst of three bad acts, such as hitting someone. They imagined another child doing it and chose a punishment.
- They pressed the red button for as long as the punishment should last, three times.
- A matching task followed for good acts and rewards, using the green button.
- The two children then swapped rooms for a second round, so each child did both parts. For half the pairs the order of the two rooms was reversed.
- Every child finished by rating the game for action and for violence.
Results
Toy Play
Girls who played Missile Command spent about 80 seconds with the aggressive toy. Girls in the Pac-Man and maze conditions spent about 15 seconds. So the violent game raised girls’ aggressive play.
Boys played with the aggressive toy more than girls did in every condition. The game they had played hardly changed this.
After Missile Command, girls played with the aggressive toy about as long as boys did.
Girls who played Missile Command also switched toys more often. They played less with the quiet toy and the skill toy.
Watching had the same effect on girls as playing.
The Buzzer Task
The game did not change how long children pressed the buzzer. This was true for boys and girls, and for players and observers.
Children who played Missile Command pressed the buzzer for no longer than children in the other groups. So the violent game did not make children punish more.
Boys did press the buzzers for longer than girls overall: about 32 seconds on average, against about 17.
Video Game Experience and Ratings
- Games at home: 34 of 44 boys and 17 of 40 girls had a video game system at home. In all, 61% of the children had one.
- Experience: most children had played a video game before, and experience did not explain the results.
- Ratings: children rated Missile Command as the most violent game on average and the maze as the least. Even Missile Command did not score above the middle of the five-point scale.
Conclusion
The authors drew five conclusions.
- Playing or watching video games affected girls’ behaviour. The violent game made them busier, as they switched toys more, and made them play more with the aggressive toy.
- Watching other children play had the same effect on girls as playing.
- Boys were not affected by the games. Their choice of the aggressive toy was already higher than the girls’.
- Neither playing nor watching changed any child’s scores on the punishment task. Toy play changed for girls but the buzzer did not, so the two measures gave different answers.
- There was no real difference between playing and watching.
Why Did the Game Affect Girls?
The authors gave two possible reasons.
- Arousal: girls have less experience of violence, so Missile Command may have excited them. Excitement makes aggressive behaviour more likely.
- Disinhibition: asking girls to play a violent game may have shown that aggressive play was allowed. So girls may have felt freer to play with the toy that fired darts.
Link to Social Learning Theory
In the exam, this is the research study for the social learning theory of criminality. The theory comes from Albert Bandura. It says people learn behaviour by watching others, as the children in his Bobo doll experiment did.
- Role models and identification: children look up to role models, such as parents or video game characters. They feel closest to models who are like themselves.
- Imitation: watching a crime builds a mental picture of it, so the child may copy it.
- Vicarious reinforcement: seeing a model rewarded, for example with money, makes copying more likely. Seeing the model arrested makes it less likely.
- Internalisation: the watcher takes in how to do the act, such as how to pick a pocket.
- Direct reinforcement: if the watcher does the act and is not caught, they get the reward themselves, such as the stolen item. They then carry on.
In this study, the video game stands in for a model the child watches. Aggressive toy play stands in for copying. Watching had the same effect as playing, which fits learning by observation.
The support is limited. The toy play result was found only for girls, and the punishment task showed no effect.
The theory has a criticism. It focuses too much on nurture, meaning what we learn from our surroundings, and ignores nature. Not everyone who watches crime becomes a criminal. Children of criminals may also share their biology. This is the nature and nurture debate.
Strengths
- Control and reliability: every player had eight minutes on the game, and every child had the same toys and tasks. Other researchers can repeat this standardised procedure, so the findings can be checked.
- Random allocation: pairs were put in the games by chance, and one child in each pair was picked at random to play. This makes the groups fair.
- A control condition: the maze task had no violence, so the researchers could compare the violent game with a task that had none.
- Counterbalancing: half the pairs did the toy room first and half did the punishment task first. The order made no difference.
- Covert observation: the adult seemed busy with numbers, so the children did not know their play was the measure. This reduces demand characteristics, which are clues that tell people the aim.
- Girls were included: earlier television research mostly used boys. Here, girls were the group that showed the main effect.
Limitations
- Ecological validity: how well results fit real life. Children played in an experimental room with an adult present, which is not how they normally play.
- Construct validity: whether a test measures what it claims to. Aggression is complex, but a buzzer narrows it to one score.
- Population validity: how well results apply to a wider group. All 84 children were fourth and fifth graders from one American area, so the results may not generalise (apply) elsewhere.
- Short-term effects: children were measured straight after eight minutes of play, so the study cannot show long-term effects.
- Social desirability: on the punishment task, children may have given the answer they thought an adult would approve of.
- Extraneous variables: these are other things that could change the results. The games differed in more than violence, such as how much action they had and how much children enjoyed them.
- Observer bias: this happens when what an observer expects changes what they record. One adult recorded the toy play, and the paper does not say whether she knew each child’s condition.
Were the Games the Cause?
In Pac-Man and the maze, girls scored no worse than boys. They still thought they had played worse. They thought this most after Missile Command, where boys did score higher.
Girls who thought they had done badly played more with the aggressive toy. So feeling that they had failed may have changed their play. The violence itself may not be the cause.
Ethical Issues
The study raises three ethical issues. Each has a way of dealing with it.
- Informed consent: agreeing after being told what will happen. Parents gave consent. The paper does not say the children knew the aim. A way of dealing with it: ask the children too.
- Deception: misleading people about the study. The adult said she needed time to “work some things out” while secretly recording. A way of dealing with it: debriefing, a full explanation afterwards.
- Protection from psychological harm: keeping people from distress. The harm was probably small, but some girls disliked Missile Command. A way of dealing with it: the right to withdraw (stopping at any time).
The paper does not say whether children were debriefed or whether their names were kept private (confidentiality).
Exam Questions
What the Exam Asks
Answers must be tied to the study. General points score little.
A point about the sample, such as “all the children were American”, earns marks only when you add why it matters. For example, “so the findings may not generalise”.
The practice questions below are examples, not the exam board’s own wording.
- Facts (1 mark each): multiple choice on the sample (children from the USA), the toys (pinball was the skill toy) and the method (an unstructured interview was not used).
- Name the method and design (2 marks): laboratory experiment, then independent measures, in the right order. Independent groups is also accepted.
- Outline a finding (2 marks): a brief finding (1 mark) and a detail such as a toy or a number (1 mark). More boys than girls had video games at home earns one mark.
- The buzzer (2 marks): a child presses a buzzer to show how much punishment to give. Link it to a type of validity (1 mark) and to the study (1 mark).
- Suggest an improvement (3 marks): a basic idea earns 1 mark, one linked to the study earns 2, and one that explains how it fixes the weakness earns 3.
- Social learning theory questions: these reward the theory, not the study. Students often mix up vicarious reinforcement and direct reinforcement.
Practice Question 1: Describe the Study (6 marks)
Question: Describe the procedure and findings of the video games study. [6 marks]
Model answer: The study was a laboratory experiment with an independent measures design. The sample was 84 children from the fourth and fifth grades in Trenton, USA, in pairs of the same sex.
Each pair was put at random into the Missile Command, Pac-Man or maze condition. One child played for eight minutes. The other watched.
Next came a toy room with four toys. An adult secretly recorded how long each child played with each one. Then each child pressed a buzzer for longer to give an imaginary child more punishment.
Girls who played Missile Command spent about 80 seconds with the aggressive toy. Other girls spent about 15. Boys’ play with the aggressive toy hardly changed. The game did not change the buzzer scores. Watching had the same effect as playing.
- Example marking guide: this is an example answer, not an official scheme. A fuller answer with accurate detail on both procedure and findings would earn more of the marks.
- Points to include: laboratory experiment, independent measures, 84 children, random pairs, eight minutes of play, a toy room, the buzzer and the girls’ result.
Practice Question 2: Explain Two Criticisms (4 marks)
Question: Explain two criticisms of the study. [4 marks]
Model answer: One criticism is ecological validity. The children played in an experimental room for eight minutes with an adult present. This is not how children normally play, so the results may not apply to real life.
A second criticism is construct validity. The buzzer measured aggression with one score, how long a button was pressed. Aggression is more complex than that, so the buzzer may not measure real aggression.
- Example marking guide: this is an example answer, not an official scheme. Give credit for naming each criticism and for a development tied to the study.
- Points the exam credits: ecological validity, construct validity and population validity. Short-term effects and social desirability are also reasonable.
Practice Question 3: Suggest an Improvement (3 marks)
Question: Suggest one improvement to the study and explain how it would help. [3 marks]
Model answer: The study measured aggression with toy play and a buzzer. This is a narrow measure, so it may lack construct validity. A repeat could watch children in a natural setting, such as an arcade, over several weeks. This would give a broader measure of aggression.
- Example marking guide: 1 mark for a basic improvement, such as a better sample. 2 marks for one linked to the study. 3 marks for one that explains how it fixes the weakness.
- Improvements the exam has credited: a wider age group, an arcade, and a broader measure of aggression than toy play, for example over a longer time.
The 13-Mark Question
No 13-mark question has yet named this study. If one did, you would judge a statement. You would say how far you agree.
You would describe the study for 6 marks. You would evaluate it for 7 marks, and end with a judgement.
To reach the middle level you must also use a study or theory from a different topic. A study from the same topic would not count.
Themes that fit this study include laboratory experiments, ecological validity, bias, and nature and nurture.
Later Research
Later research asks the same question with other methods.
Przybylski and Weinstein (2019) studied 1,004 British adolescents aged 14 and 15. The teenagers said how much they played violent games. Carers rated their aggression. It found no link between violent game play and aggression.
Prescott, Sargent and Hull (2018) pooled 24 studies that followed over 17,000 people for between 3 months and 4 years. They found a small link between violent games and later physical aggression.
Drummond, Sauer and Ferguson (2020) pooled 28 samples of about 21,000 young people. They found a very small link. In the best-designed studies, it was close to zero.
So researchers still disagree about long-term effects. None of these studies repeated the 1986 laboratory design, and none of the abstracts reports a girls and boys comparison.
References
Cooper, J., & Mackie, D. (1986). Video games and aggression in children. Journal of Applied Social Psychology, 16(8), 726-744. https://doi.org/10.1111/j.1559-1816.1986.tb01755.x
Drummond, A., Sauer, J. D., & Ferguson, C. J. (2020). Do longitudinal studies support long-term relationships between aggressive game play and youth aggressive behaviour? A meta-analytic examination. Royal Society Open Science, 7(7), Article 200373. https://doi.org/10.1098/rsos.200373
Prescott, A. T., Sargent, J. D., & Hull, J. G. (2018). Metaanalysis of the relationship between violent video game play and physical aggression over time. Proceedings of the National Academy of Sciences, 115(40), 9882-9888. https://doi.org/10.1073/pnas.1611617114
Przybylski, A. K., & Weinstein, N. (2019). Violent video game engagement is not associated with adolescents’ aggressive behaviour: Evidence from a registered report. Royal Society Open Science, 6(2), Article 171474. https://doi.org/10.1098/rsos.171474