Participants (N = 73) in the present research were assigned to play a violent videogame (Super Smash Brothers) with either a supportive or an ambivalent friend. Orthogonal to this manipulation, participants were assigned to play the game either competitively or cooperatively. Subsequent aggression toward their friend was assessed by measuring participants' competitive or cooperative behavior in a Prisoner's Dilemma game. Results revealed no differences in aggression as a function of friendship type or game strategy, although means were in predicted directions. The influence of context on exposure to violent media is discussed, as are issues of power and sample type as possible reasons for the nonsignificant findings.

published_or_final_version / Asian Studies / Master / Master of Arts

The Prisoner's Dilemma is a two player game where playing rationally leads to a suboptimal outcome for both players. The game is simple to analyze, but when it is played repeatedly, complex dynamics emerge. Recent research has shown the existence of extortionate strategies, which allow one player to win at least as much as the other. When one player plays such a strategy, the other must either decide to take a low payoff, or accede to the extortion, where they earn higher payoff, but their opponent receives a larger share. We investigate what happens when one player uses this strategy against an ``evolutionary'' player, who makes small changes to her strategy over time to increase her score, and show that there are cases where such a player will not evolve towards the optimal strategy of giving in to extortion.

Thesis (Ph. D.)--University of Oregon, 2000. / Typescript. Includes vita and abstract. Includes bibliographical references (leaves 115-123). Also available for download via the World Wide Web; free to University of Oregon users.

Thesis (M.A.)--University of Hong Kong, 2001. / Includes bibliographical references (leaves 132-134).

Thesis--M. Phil., University of Hong Kong, 1978.

A variety of movements and social pressure have driven the need for an increase in environmental awareness, and subsequently fuels the need for individuals to reduce their ecological footprint. Firms are now trying to implement 'eco-friendly' technologies that both build and run their products. How these 'eco-friendly' products will perform in the market is strongly tied to a variety of consumer related influences and decisions, as well as personality type. This thesis presents a model of varied social influence on consumer markets. First we show how varied playing characteristics amongst opponents in the Iterated Prisoner's Dilemma yields a different distribution of strategies. Utilizing two variations of IPD, we map scores to edges based on the agents involved in each edge as one construct of influence. Other types of influence include a homogeneous influence, and a zero influence for comparison of results. We also introduce the Rate of Social Mobility as a basis for initializing random social movement in a network. We show that the social influence of the network in the consumer market plays a vital role in the dynamics of product adoption. In closing we discuss future model refinements, and advances.

Thesis (M.A.)--University of Hong Kong, 2001. / Includes bibliographical references (leaves 132-134). Also available in print.

Healthy social interactions are critical for children’s development and academic and life success. Relevant research is found in two key disciplines with different emphases: Developmental psychology focuses on individual thoughts, motivations and traits; and behavioral economics and game theory focuses on behavioral tasks. This project integrated these approaches by validating a game-theoretic task for children, the Repeated Prisoner’s Dilemma (RPD), and demonstrated how it can be used to elucidate the mechanisms underlying children’s social interactions. I developed a novel RPD with fixed-strategy partners in order to test specific hypotheses based on developmental theories of social interaction. Children between 9 and 11 years of age (N = 167) were tested on the RPD followed by questions about how they played and interpreted the task. Parents completed a questionnaire assessing their child’s reactive and proactive aggressive traits, a basis for predicting decisions in the RPD. Children also completed a Social Information Processing (SIP) task with novel positive scenarios in addition to standard negative ones. I hypothesized that: 1) children would interpret the RPD as a real social interaction and engage in strategic forms of play according to game theory; 2) children with different levels of reactive and proactive aggression would show different patterns of RPD play based on theories of aggression; 3) the SIP responses would predict different levels of cooperation in the RPD, and the positive scenarios would generate responses consistent with the general SIP theory. Results showed support for the first hypothesis with classes of motivations (interpersonal and strategic self-interest) predicting RPD behavior. The second hypothesis was partially confirmed: Children rated high on reactive aggression showed reactive responses in the RPD. This analysis also revealed an important novel finding that high-reactive children followed a game-theoretic strategy known as “Grim” – they did not return to cooperation after partner defection. The third hypothesis was partially confirmed: Responses for the positive scenarios were consistent with the SIP model but did not predict RPD play. These findings demonstrate the value of integrating theoretical and methodological approaches from developmental psychology and game theory in order to study the mechanisms of social interaction.

No description available.