Effects of morality and reputation on risk-taking behaviors on behalf of robots

Abstract

Abstract The relationship between robots and humans is becoming increasingly intimate and will continue to become an integral part of life, with robots working together with human beings. In teamwork, individual decision making for others is influenced by morality and reputation. However, the effect of individuals’ decision making for others in human-robot collaborative teams remains unclear, and individuals may have different degrees of consideration for morality and reputation when making decisions for robots or humans. This research conducted three experiments based on Gneezy and Potters' investment decision-making experimental paradigm and compared the influence mechanisms of morality and reputation on decision making for others in human teams and human-robot teams. Experiment 1 verifies that when the individual decision-making objects were robots and human beings, the decision making for others was different, When the decision object is a robot, the risk preference of the decision making for others’ behavior is higher .Experiment 2 verifies that morality and reputation are the influencing factors of the differences in the decision making for others. Experiment 3 proposes the mechanism of the difference between the decision making for others in a human team and a human-robot team. That is, in a human-robot team, reputation is a mediating variable of the influence of morality on the decision making for others, while in a human team, there is no such mediating effect. The research results show that people's decision making for others is influenced by their cooperation partner. When the cooperation partner is a robot, the risk preference of the decision making for others is higher because of the low constraint of social morality. However, if a good reputation is maintained for morality considerations, then the risk preference of the decision making for others will be reduced.