Analyzing Barnga Gaming Simulation Using an Agent-Based Model

Conclusions

Through a model simulation of the card game Barnga, it was shown that ABM can complement GS. ABM provides a new tool for expanding analysis of GS from different angles and viewpoints, and overcoming the difficulties of GS for what-if analysis. Through ABM implementation of the game, key points on how humans learn could be explored. Implementation of an agent’s learning process is based on the capability to correctly predict who the winner of the trick will be based on the game rules the agent believes that the table is governed by. It was observed that the use of simple evaluations to perform adaptation prevents appropriate learning, and induces misunderstandings within a cross-cultural group. Through the implementation of ABM, it was suggested that humans may increment the conditions for evaluations to perform learning. Also, humans were able to make use of an efficient selection of attributes that correctly informs and supports the learning of the rules that govern a game. Another major finding is that cultural diversity is reduced over time, even if players have different levels of flexibility.

Future research will consider: (1) investigations on the effectiveness and applicability of ABM in several gaming simulations, (2) further exploration of the Barnga model for the study of cross-cultural differences, and (3) searches for new techniques that can help in the study of behavioral science.