Lie detection using facial muscle monitoring and machine learning, with Dino Levy

A talk with Dino Levy about his team’s lie detection research, which used monitoring of facial muscles and machine learning to detect lies at an impressively high 73% success rate. Their paper was titled “Lie to my face: An electromyography approach to the study of deceptive behavior.” Topics discussed include:

• The setup of the study, and the theoretical causes of the findings
• How this method compares to polygraph technology (lie detector machines)
• Applications of the technology
• Thoughts on ideas of the universal nature of emotion-related behavior
• Speculations on using these findings to analyze poker players

Episode links:

• iTunes (embedded below)
• Spotify
• Google Play
• YouTube

Other resources related to or mentioned in our talk:

• BBC article about Hanein, Levy et al’s research and other lie detection tech
• Dino Levy’s university page
• Dino Levy’s Twitter
• Criticisms of polygraph technology
• Wikipedia entry about Ekman’s “truth wizards” study