A PDP-11/45 program for playing n-person characteristic function games

NPER II is an on-line PDP-11/45 computer program for studying bargaining and coalition formation processes within the framework of n-person games in characteristic function form. The extensive vocabulary of the program and its mode of operation are presented nontechnically, followed by an examination of a protocol from a sample four-person game. Improvements over a previous version of the program and applications to other experimental paradigms investigating coalition formation and bargaining are discussed in the last two sections of the paper.     

Integrative bargaining in a competitive market

The behavioral decision theory literature was used to identify the determinants of negotiation success in an integrative bargaining, free-market exercise. This study provides a novel methodology for studying negotiation. Specifically, buyers and sellers were allowed to engage in negotiations with as many competitors as possible in a fixed time period. The results suggest that integrative bargaining behavior increases and the market converges toward a Nash equilibrium as negotiators gain experience. In addition, the results suggest that (1) positively framed negotiators (“What will be my net profit from the transaction?”) complete more transactions than negatively framed negotiators (“What will be my expenses on this transaction?”), (2) negotiators who are given moderately difficult profit constraints in order to be allowed to complete a transaction achieve more profitable transactions than negotiators without such constraints, and (3) both framing and the existence of constraints affect the total profitability of the negotiator.     

EMOTLAB: Software for studying emotional signaling in economic bargaining games

EMOTLAB software creates a virtual social environment in which individuals interact via computer with a virtual interaction partner in a series of economic bargaining games. The virtual partner appears on the participant's computer screen as a digital image (e.g., video or picture file) during each trial. A key feature of EMOTLAB software is its ability to control both the strategic behavior and the emotion signaling behavior (e.g., anger vs. embarrassment) of the virtual interaction partner. By simply editing a series of text files that control the subroutines governing the different features of the experiment (payoff structure, number of trials, etc.), EMOTLAB can generate an essentially infinite number of different social bargaining situations in which participants earn monetary payoffs contingent upon their decisions. This paper provides an overview of this software and how one can edit various subroutines to generate a typical experimental session in which research participants encounter a virtual interaction partner who displays different emotional signals.     

Different action patterns for cooperative and competitive behaviour

The aim of the present study is to elucidate the influence of context on the kinematics of the reach-to-grasp movement. In particular, we consider two basic modes of social cognition, namely cooperation and competition. In two experiments kinematics of the very same action - reaching-to-grasp a wooden block - were analyzed in two different contexts provided by a cooperative task and competitive task. For the 'cooperation' tasks two participants were required to reach and grasp their respective objects and to cooperate to join the two objects in specific configurations in the middle of the working surface. For the 'competition' tasks, the two participants had to compete to place their own object first in the middle of the working surface. Results revealed specific kinematic patterns for cooperation and competition which were distinct from similar actions performed by each participant in isolation. Further, during the cooperation tasks, a high level of correlation between key kinematical parameters of the two participants was found. In accordance with evidence from neuroimaging, developmental and social psychology our results suggest the existence of motor patterns which reflect the intention to act in a social context.     

Behavioural consequences of regret and disappointment in social bargaining games

Previous research on the role of negative emotions in social bargaining games has focused primarily on social emotions such as anger and guilt. In this article, we provide a test for behavioural differences between two prototypical decision-related negative emotions-regret and disappointment-in one-shot social dilemma games. Three experiments with two different emotion-induction procedures (autobiographical recall and imagined scenarios) and two different games (the ultimatum game and the 10-coin give-some game) revealed that regret increased prosocial behaviour, whereas disappointment decreased prosocial behaviour. These results extend previous findings concerning differences between regret and disappointment to interdependent (social) situations.     

Switching pauses in cooperative and competitive conversations

The effects of three variables (conversation type, turn completeness, and sexual composition of the interacting dyad) on switching pause durations was investigated. Conversation type was represented by two categories of conversation: cooperative and competitive. A cooperative conversation is one in which participants cooperate in floor apportionment, while a competitive conversation is one in which the participants compete for the floor. Turn completeness has two levels and refers to the presence or absence of a turn-yielding signal in the utterance immediately preceding a switching pause. Three dyad compositions, male-male, female-female, and male-female were used, with six dyads in each category. The subjects participated in four conversations, two friendly chats and two arguments, on various topics. The switching pauses in competitive conversations were shorter than those in cooperative conversations. Furthermore, conversation type had a significant effect on the switching pause length following imcomplete utterances.     

The relationship between cooperative,competitive, and individualistic attitudes and differentiated aspects of self-esteem1

Four studies to clarify the relationship between attitudes toward styles of social interdependence and ways of deriving self-esteem were conducted among 821 white, middle-class, secondary school students in a midwest suburban community. Correlations between attitudes toward three styles of interdependence and four ways of deriving self-esteem indicate distinctive and theoretically predictable patterns of relationship. Students who indicate a cathexis for cooperative relationships report patterns of higher self-esteem related to freedom of personal expressiveness and feelings of personal well-being; students indicating a cathexis for competitive or individualized patterns of interdependence report greater vulnerability on dimensions of self-esteem reflecting sensitivity to experiences of success, approval, and support of others. The magnitudes of correlations between a global measure of personal worth and attitudes toward the three types of interdependence reflect the extent to which positive social reinforcement is available in these settings. Specific correlations that reflect these findings are significant, p < .01.     

Effect of playing violent video games cooperatively or competitively on subsequent cooperative behavior

Research on video games has yielded consistent findings that violent video games increase aggression and decrease prosocial behavior. However, these studies typically examined single-player games. Of interest is the effect of cooperative play in a violent video game on subsequent cooperative or competitive behavior. Participants played Halo II (a first-person shooter game) cooperatively or competitively and then completed a modified prisoner's dilemma task to assess competitive and cooperative behavior. Compared with the competitive play conditions, players in the cooperative condition engaged in more tit-for-tat behaviors-a pattern of behavior that typically precedes cooperative behavior. The social context of game play influenced subsequent behavior more than the content of the game that was played.     

Popular and unpopular children's interactions during cooperative and competitive peer group activities

Popular and unpopular fourth-grade boys were videotaped as each attempted to gain entry into a cooperative and a competitive task involving two classmates who were average in popularity. During the competitive procedure, the unpopular entry children were more likely than their popular counterparts to break rules, emit silly noises, and appeal to authority. Children average in popularity directed more positive behaviors toward their well-liked classmates and more derisive and dominating behavior toward unpopular peers. Unpopular children exhibited less negative and immature behavior in the benign, tension-free atmosphere of the cooperative project and their peers were more tolerant toward them than during the competitive game. The findings suggest that contextual factors influence the social skills exhibited by the unpopular child. Implications for the treatment of peer relationship problems are discussed.This article is based on a doctoral dissertation by the first author under the direction of the second author. The contributions of Neil Olliviera, Kathleen Sengstock, Virgil Sheets, and Jerry Sullivan in collecting and coding data are gratefully acknowledged. Appreciation is extended to Ira Firestone for his assistance with the statistical analyses. We would also like to thank the children, teachers, and administrators of the Hazel Park, Livonia, Southfield, and Farmington Hills, Michigan, schools who participated in the project.     

The power of the glance: Desire to see and be seen in cooperative and competitive situations

Heider (1958) has suggested that perception aides control and power over that which is perceived. Individuals' belief in this assumption was tested in an interpersonal situation. Subjects were given the expectation that they would be either cooperating or competing with another person in a game. Half of the subjects were lead to believe they would be interviewing their partner or opponent before the game, and half expected to be interviewed. Subjects who expected to interview the other could choose whether or not to see the other person through a one-way mirror. Subjects who expected to be interviewed could choose whether or not they wanted to be seen by the other. As predicted, the tendency to choose to allow perception was strong in all conditions except the condition in which subjects expected to compete with the other and might be seen by him.     

Apes' and children's understanding of cooperative and competitive motives in a communicative situation

Chimpanzees (Pan troglodytes) and bonobos (Pan paniscus) (Study 1) and 18- and 24-month-old human children (Study 2) participated in a novel communicative task. A human experimenter (E) hid food or a toy in one of two opaque containers before gesturing towards the reward's location in one of two ways. In the Informing condition, she attempted to help the subject find the hidden object by simply pointing to the correct container. In the Prohibiting condition, E held out her arm toward the correct container (palm out) and told the subject firmly 'Don't take this one.' As in previous studies, the apes were at chance in the Informing condition. However, they were above chance in the new Prohibiting condition. Human 18-month-olds showed this same pattern of results, whereas 24-month-olds showed the opposite pattern: they were better in the Informing condition than in the Prohibiting condition. In our interpretation, success in the Prohibiting condition requires subjects to understand E's goal toward them and their behavior, and then to make an inference (she would only prohibit if there were something good in there). Success in the Informing condition requires subjects to understand a cooperative communicative motive - which apparently apes and young infants find difficult.     

Popular and aggressive boys' initial social interaction patterns in cooperative and competitive settings

Thirty-two third- and fourth-grade popular and aggressive boys were observed individually attempting to enter a game being played by a mixed-status unfamiliar-peer dyad in both competitive and cooperative game settings. Consistent with previous findings, popular entry children were accepted more readily into the game situation than were aggressive children. Popular children tended to approach the situations using more prosocial questions while their aggressive peers tended to use demands. More differences in entry strategies were noted in the competitive than in the cooperative game setting. While most entry children directed their first entry bid toward the aggressive game host, they reported having liked the popular game host best when the game was finished. Additionally, the quality of game interactions was observed to become more positive following a popular child's entry, while becoming more negative following an aggressive child's entry. The results are discussed in terms of leadership roles and implications for both assessment and intervention settings and strategies.The authors wish to acknowledge the support and cooperation of the Greensboro, North Carolina, public and parochial schools, and the assistance of William Livingston.This paper was presented at the 19th Annual Convention of the Association for the Advancement of Behavior Therapy, Houston, Texas.     

More equal than others: Equity norms as an integration of cognitive heuristics and contextual cues in bargaining games

Behavior in one-shot bargaining games, like the Ultimatum Game (UG), has been interpreted as an expression of social preferences, such as inequity aversion and negative reciprocity; however, the traditional UG design limits the range of possible psychological interpretation of the results. Here, we employed three different designs for ultimatum games, finding support for a more comprehensive theory: behavior is driven by cognitive factors implementing rules such as equal splitting, speaking up for the idea that equity works as a cognitive heuristic, applicable when the environment provides no reason to behave otherwise. Instead subjects deviate from this rule when environment changes, as, for instance, when personal interest is at stake.Results show that behavior varies systematically with contextual cues, balancing the self-interest with the automatic application of the equity heuristic. Thus, the context suggests the rule to be applied in a specific situation.     

When three is not always two against one: Coalitions in experimental three-person cooperative games

Four models of coalition formation are tested in a computer-controlled experiment in which three players negotiated to form coalitions in order to gain rewards for themselves. Formation of the grand coalition of all three players as well as any of the three possible 2-person coalitions is permitted. The results indicate that the grand coalition is the most likely to form, followed by the coalition between the two strongest players. The individually rational bargaining set model of coalition formation is supported over an egalitarian model, the Shapley value, and the coalitionally rational bargaining set model. Various measures of the progress of negotiations are introduced, examined, and discussed in an effort to provide an increased data base for the establishment of a behavioral theory of bargaining and coalition formation.     

The pyeTribe: Simultaneous eyetracking for economic games

The recent introduction of inexpensive eyetrackers has opened up a wealth of opportunities for researchers to study attention in interactive tasks. No software package has previously been available to help researchers exploit those opportunities. We created “the pyeTribe,” a software package that offers, among others, the following features: first, a communication platform between many eyetrackers to allow for simultaneous recording of multiple participants; second, the simultaneous calibration of multiple eyetrackers without the experimenter’s supervision; third, data collection restricted to periods of interest, thus reducing the volume of data and easing analysis. We used a standard economic game (the public goods game) to examine the data quality and demonstrate the potential of our software package. Moreover, we conducted a modeling analysis, which illustrates how combining process and behavioral data can improve models of human decision-making behavior in social situations. Our software is open source.