Examining human behavior in video games: The development of a computational model to measure aggression

Soc Neurosci. 2018 Jun;13(3):301-317. doi: 10.1080/17470919.2017.1318777. Epub 2017 Apr 24.

Abstract

Video games with violent content have raised considerable concern in popular media and within academia. Recently, there has been considerable attention regarding the claim of the relationship between aggression and video game play. The authors of this study propose the use of a new class of tools developed via computational models to allow examination of the question of whether there is a relationship between violent video games and aggression. The purpose of this study is to computationally model and compare the General Aggression Model with the Diathesis Mode of Aggression related to the play of violent content in video games. A secondary purpose is to provide a method of measuring and examining individual aggression arising from video game play. Total participants examined for this study are N = 1065. This study occurs in three phases. Phase 1 is the development and quantification of the profile combination of traits via latent class profile analysis. Phase 2 is the training of the artificial neural network. Phase 3 is the comparison of each model as a computational model with and without the presence of video game violence. Results suggest that a combination of environmental factors and genetic predispositions trigger aggression related to video games.

Keywords: Computational modeling; aggression; latent class analysis; media; video games and violence.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Adolescent
  • Aggression*
  • Computer Simulation*
  • Databases, Bibliographic / statistics & numerical data
  • Female
  • Humans
  • Male
  • Models, Psychological*
  • Neural Networks, Computer
  • Probability
  • Video Games*