Deception Detection in Computer-Mediated Communication

In an ultimatum game, participants were randomly assigned to the role of allocator or recipient to interact either video-to-video (VtV) or audio-to-audio (AtA). The allocator was given money to divide and an opportunity to deceive or be truthful to the recipient who did not know how much money the allocator received. The VtV channel led to more truth accuracy and the AtA channel led to more lie accuracy. Channel richness did not help the overall accuracy, nor the presence or absence of demeanour cues. Instead, lower truth bias in AtA was the root cause of the differences between the two channels' accuracy rates. Deceptive allocators were more distressed in the VtV condition, but receivers did not pick up on this. Theoretical and practical implications are discussed.