Lie to my face: An electromyography approach to the study of deceptive behavior

Abstract

Deception is present in all walks of life, from social interactions to matters of homeland security. Yet, reliable indicators of deceptive behavior remain elusive. We integrated biological and social approaches to show that facial muscle activity serves as an objective and participant-specific indicator of deception. Participants completed an interactive two-person deception task. We recorded facial muscle activity using novel dry screen-printed electrode arrays and applied machine-learning algorithms to identify lies. With an average accuracy of 73%, we revealed two groups of participants based on their indicators of deception—those who reveal their lies by moving their cheeks and those who move their eyebrows. Not only did participants lied more often with time, some changed their indicators by switching muscle groups. Although the classifier outperformed the human detectors, their performance was correlated, suggesting reliance on shared features. Our findings set the stage for new research on individual differences in deception expression.