Detecting betrayers in online environments using active indicators

Abstract

Research into betrayal ranges from case studies of real-world betrayers to controlled laboratory experiments. However, the capability of reliably detecting individuals who have previously betrayed through an analysis of their ongoing behavior (after the act of betrayal) has not been studied. To this aim, we propose a novel method composed of a game and several manipulations to stimulate and heighten emotions related to betrayal. We discuss the results of using this game and the manipulations as a mechanism to spot betrayers, with the goal of identifying important manipulations that can be used in future studies to detect betrayers in real-world contexts. In this paper, we discuss the methods and results of modeling the collected game data, which include behavioral logs, to identify betrayers. We used several analysis methods based both on psychology-based hypotheses as well as machine learning techniques. Results show that stimuli that target engagement, persistence, feedback to teammates, and team trust produce behaviors that can contribute to distinguishing betrayers from non-betrayers.