Influence of Perspectives on Human Player Decisions in Board Games Online

For the board game of Go, most online servers provide both players with the same perspective, thus deviating from the in-person play mode. In addition, due to the symmetry in the board, there are eight natural perspectives of the same game situation. Yet, AI review software such as KataGo does not recognize or use different perspectives when searching for the best next move, often leading to self-conflicting recommendations. This paper explores the connection between the board’s perspectives and players’ decision-making experiences. First, we enumerated all six possible relations between any pair of distinct board perspectives of the same game. Next, we develop a software system that allows a board to be rotated and reflected during gameplay. Using our system, we create material for and design a user survey, which we conduct with Go players from several local clubs. A player is given numerous games, each shown in two perspectives. The player’s move for each perspective is recorded and compared to the other perspective of the same game in terms of their distance on the board, the difference in the loss of win rates, and the difference in the time taken to make the moves. From the survey, we perform analysis and interpretation that lead to a number of insights, from which we make several recommendations regarding the board’s perspectives on improving player experience during online play and review.