Bargaining is a major alternative to violence in actual situations of human conflict but is rarely available to players of the computer and console games where our youth spend so much of their time. It has been pointed out that today's children, physically confined by urban conditions, are finding virtual play spaces in console and computer games for the sort of adventure, risk taking, challenges, and problem solving vital to identity formation. The role of, special purpose computational engines included in a computer or console game product (e.g., dynamics engines that provide game developers with ready models of simulated physical motion) is well established. In this project, the PI team will investigate the application potential and the feasibility of a "bargaining engine" that would simulate and assist bargaining behavior in computer games. To address the application potential aspects, the PI team will carry out a survey and analysis of the instances and forms (representation and interaction) of bargaining and proto-bargaining in contemporary computer/console games. They will use those results in a sample of interviews with game designers and marketers, who will be asked to project what a sophisticated bargaining model could contribute to game play, character, and plot. The result will be a report incorporating the survey, analysis, and interviews. The overall goal will be to develop an accurate picture of bargaining in games today and to assess the potential for bargaining in games of the future. To address feasibility aspects, the PI team will produce an MS Windows application that implements continuous-time (as opposed to turn-based) bargaining on a single divisible issue with perfect information. Continuous-time bargaining game theoretic models are a relatively recent development that may be particularly relevant to many of the newer computer games that involve continuous-time play. Bargaining behavior is represented by a Markov strategy that combines offers with a holding-time stochastic process to describe the chances of acceptance as a function of elapsed time. Players' preferences are classically represented by utility functions to be expressed before bargaining takes place. The members of the PI team bring complementary, interdisciplinary expertise and experience to this subject, and are uniquely qualified to undertake this research.
Broader Impact: The PI argues that adding bargaining to the repertoire of skills such virtual experiences can offer could enrich the value of interactive entertainment for children and widen their perspective on meeting real-life challenges. Yet, bargaining as a capability cannot simply be grafted onto existing game content, structure, or interaction. One must start by analyzing where even primitive forms of bargaining exist now, researching where the potential lies for expansion and depth, projecting what forms this might take, and identifying directions for research and experimental prototyping.
