Improving Human Deception Detection Using Algorithmic Feedback

Arbeitspapier

Can algorithms help people detect deception in high-stakes strategic interactions? Participants watching the pre-play communication of contestants in the TV show Golden Balls display a limited ability to predict contestants' behavior, while algorithms do significantly better. We provide participants algorithmic advice by flagging videos for which an algorithm predicts a high likelihood of cooperation or defection. We find that the effectiveness of flags depends on their timing: participants rely significantly more on flags shown before they watch the videos than flags shown after they watch them. These findings show that the timing of algorithmic feedback is key for its adoption.

Object Details

• Language: Englisch

References and Relationships

• Bibliographic citation: Series: CESifo Working Paper ; No. 10518

Classification and Topics

• Classification: Wirtschaft ; Search; Learning; Information and Knowledge; Communication; Belief; Unawareness ; Micro-Based Behavioral Economics: Role and Effects of Psychological, Emotional, Social, and Cognitive Factors on Decision Making‡ ; Noncooperative Games ; Design of Experiments: Laboratory, Individual
• Subject: detecting lies ; machine learning ; cooperation ; experiment

Contributors, Places and Time

• Event: Geistige Schöpfung
• (who): Serra-Garcia, Marta ; Gneezy, Uri
• Event: Veröffentlichung
• (who): Center for Economic Studies and ifo Institute (CESifo)
• (where): Munich
• (when): 2023

Further information

• Handle: http://hdl.handle.net/10419/279268
• Last update: 10.03.2025, 11:42 AM CET