학술논문

Truth-Telling in a Sender–Receiver Game: Social Value Orientation and Incentives.
Document Type
Article
Source
Symmetry (20738994). Aug2022, Vol. 14 Issue 8, p1561-1561. 26p.
Subject
*VALUE orientations
*SOCIAL values
*INCENTIVE (Psychology)
*COMPUTER algorithms
*MONETARY incentives
Language
ISSN
2073-8994
Abstract
Previous research has discussed the effects of monetary incentives and prosociality on deceptive behavior. However, research has not comprehensively investigated the relationship between these two factors. In the current research, we introduce a repeated two-player sender–receiver binary choice task, where players in the role of senders or receivers receive asymmetric information regarding payoffs, offering the opportunity to explore the effects of economic incentives to lie according to the players' prosociality. In Experiment 1, players are paired to play the game as a sender or receiver online. We find that economic incentives determine the likelihood of deception from senders and the likelihood that receivers will deviate from the received suggestions. Moreover, prosociality is related to players' behavior: Prosocial senders send less deceptive messages and prosocial receivers choose options that benefit senders more. Furthermore, senders display consistent behavior when interacting with receivers, and they do not change their deceptive behavior even if detected by receivers. Experiment 2 further investigates how the players' behavior corresponds to their understanding and interpretation of the other players' actions, by pairing players with computer algorithms that display consistent probabilistic behaviors. We observe that senders deceive receiver algorithms by sending truthful messages when they expect the message not to be followed, and receivers follow the received messages by choosing the option that benefits "honest" sender algorithms. While we find a consistent result that prosocial senders send fewer deceptive messages than they should when telling the truth is costly, prosocial receivers are less considerate of sender payoffs in algorithms' interaction. [ABSTRACT FROM AUTHOR]