Lie Detection from Multiple Cues: A Meta‐analysis

Abstract

Summary Despite the importance of judgments of veracity in many settings, research suggests that it is difficult to detect lies. In this meta‐analysis, we assess the detectability of lies from constellations of multiple cues, with a particular focus on whether lie detectability increases as the conditions approach real‐life, forensic settings, as some critics of laboratory research have argued. We synthesized 144 samples, including 9380 liars and truth tellers providing a total of 26,866 messages. We examined the accuracy with which deception could be predicted on the basis of multiple behavioral cues and to what extent lie detectability was moderated by the motivation of the sender, the presence of strong emotion, the content of the lie, the context in which the lie was told, and the demographics of the senders. The findings show that lies can be detected with nearly 70% accuracy. This level of detectability is stable across settings. Copyright © 2014 John Wiley & Sons, Ltd.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 210 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.