Philippa Nottingham – U1652980
HIB 2004 – Cognition; Brain and Behaviour
Anderson and Bushman (2002) conducted research into violent media which suggested that aggressive cognitions (schema) and increased arousal can be triggered by short term exposure to violent media. Within this experiment, it was examined whether short term exposure to fantasy violent media posed an impact on the internal state in the General Aggression Model (GAM). An independent design with a total of 81 undergraduate students, 43 male and 38 female, was used. Participants were randomly assigned to watch either a fantasy violent film clip (Starship Troopers) or a nonviolent film clip (waterfall). EDA transducers were attached to the participants’ fingers and the Biopac MP35 Acquisition Unit was used to measure the participants pulse. The participants then filled out the Buss and Perry Aggression Questionnaire (1992). Results suggested that the total aggression score was significantly higher, with a mean difference of 19.82, with the violent clip and that the pulse score was also significantly higher with the violent film clip with a mean difference of 7.15.
Anderson and Bushman (2002) defined aggression as behaviours which are directed towards other individuals, with immediate “intent to cause harm’. When the General Aggression Model (GAM) was formed as it describes interactions that individuals have with their environment which could increase interpersonal aggression likelihood. A door was opened for other theories to be better integrated as it provided a more substantial explanation based on aggression with ‘multiple motives’ like passive or instrumental aggression (Anderson & Bushman, 2001).
The centre of the GAM (Anderson &Bushman, 2002) is that an individual has a present internal state down to two input variables, one being person factors and the other being situational factors. An example for personal factors would be an individuals own attitudes when referring to aggression and situational factors would be the violent media they are exposed to or the violence they see around other people. In terms of the GAM, cognition, affect, and arousal is representative of the present internal state. Exposure to violent media is a situational variable which will lead to the probability of increased aggressive behaviour. Therefore, short-term exposure to violent media is something that could cause an individual to form primed aggressive cognitions such as an aggressive perceptual schemata, later resulting in possibilities of both impulsive and thoughtful aggressive actions due to an aggressive affective state and arousal increase.
Research on the effects of fantasy media on aggression mainly focusses on video games and the violent exposure they give to young people. Anderson et al., (2010) concluded that its strongly suggested that exposure to violent video games is a causal risk factor for increasing generalised aggressive behaviour, aggressive cognition, aggressive affect, and a general lack of empathy and prosocial behaviour characteristics. Berkowity and Geen (1966) reported that men who had witnessed a violent fight scene were more likely to administer more shocks to a person. There was an association made with witnessed aggression as cues were heightened for aggressive responses leading to participants evoking stronger volumes of aggression. This means that exposure to violent media leads to aggressive knowledge structures in a persons memory being activated leading to priming (Anderson & Bushman 2001). Its further suggested that the cognitive route of the GAM can be linked to violence in media content. While it is argued that violence in video games is affective the cognitive route of the GAM, it’s also shown how competitive non-violent video games can subsequently lead to aggression in a person’s internal state due to excitement within the game (Anderson & Dill, 2000). Through this it is learnt that differentiations in situational variables impact a person’s internal state but Anderson and Bushman (2001) argue that a development in primed aggressive cognitions, like schemas or scripts, is down to the continuity in exposure to violence in media.
Barlett et al., (2009) resulted in the indication that aggressive feelings, thoughts, behaviour increased following violent gameplay, alongside peoples’ heart rate which also initially increased after violent video game play. This supports that physiological arousal has a scientific contribution when referring to the GAM (Anderson & Bushman 2002) as it tends to be characteristic of increased aggression.
The experimental aim of this research was to attempt a replication of Anderson and Bushmans’ (2002) findings of which support the GAM which they developed through their findings. The aim is to expeller whether a fantasy violent film clip is functional as stimuli in affecting the components of the GAM which are both arousal and cognition.
For the experiment there are two research hypotheses; the first hypothesis is that the total aggression score (Buss & Perry, 1992) will be significantly higher with the fantasy violent film clip (starship troopers) in comparison to the nonviolent film clip (waterfall), the second is that the score for the pulse will be significantly higher with the fantasy violent clip compared to the nonviolent clip.
Participants or Subjects
A total of 81 undergraduate students was used to conduct this research and from this sample 43 participants were male and 38 were female. The method for getting participants was an opportunity sample, as they were students at the University of Huddersfield, meaning that the mean age was 21.
The participants were randomised into the two film conditions and it was run on a voluntary basis. In each lab session there were about 8-10 participants exposed to one film clip, each session running a different clip.
Materials and/or Apparatus
The materials used were the 2 film clips as each lab was showing a different one (3 min each): the fantasy violent clip (Starship troopers) and the nonviolent clip (waterfall). Then the model of the Biopac was Biopac MP35 Acquisition Unit as they were used to measure the physiological responses. The other materials were Biopac Student Lab pro which was the software that was used to analyse the data and Signa Gel which was used for the electrodes.
Buss, A. H., & Perry, M. (1992). The aggression Questionnaire. Journal of Personality and Social Psychology. 63, 452-459 was one of the cognition measures, being the questionnaire used to measure four factors: Physical aggression (PA), Anger (A), Hostility (H), and Verbal aggression (VA) and in order to find the Total Aggression Score, these four factors were added together.
The two physiological measures were the Pulse rate – Beats per minute (BPM) and then the Data for the baseline condition and experimental condition was taken.
For this research an independent design was used as each participant only engaged with one independent variable . There was one independent variable having the two levels of fantasy violence or nonviolent film clips, then the dependant variables were pulse (BPM) and Total aggression score.
Firstly Instructions were presented to the participants as they were also told about ethics prior to the experiment and then commanded to attach pulse electrodes. EDA transducers were attached to the participants fingers with Signa gel. The baseline condition was established by exposing participant to blank screen for 3 minutes, then the film began for another 3 minutes. Individuals were made to remain seated throughout the clip as their physiological responses were recorded to prevent any other confounding variables (being movement) from effecting the result. Finally, participants were then told to fill the Buss and Perry Aggression questionnaire.
An independent-samples t-test was used to evaluate the hypotheses that fantasy violent media causes short-term effects on cognition (aggression score) and on physiological arousal (Pulse). The independent variable (IV) was the film clip and whether or not it contained fantasy violence, and the dependent variables (DV) were pulse and the aggression scores. All parametric assumptions were met for an independent-samples t-test to be conducted. A two-tailed test was used and normality assumptions were met and Levenes test for homogeneity of variance was not significant (p >.05), ensuring homogeneity of variance (Field, 2006). There was no missing data or data cleaning needed in this experiment and the results are normally distributed.
Means and standard deviations for total aggression scores for each condition is presented in Table 1. While means and standard deviations for the pulse scores are in Table 2.
It was found that there was a significant difference on the media content and total aggression score, t(79) = 24.12, p