A collaboration with Prof. Yael Hanein’s lab. People’s ability to detect lies during social interactions is at chance level, and even commonly used techniques, like the polygraph, suffer from poor performance and reliability. It was longtime hypothesized that tiny facial expressions reveal when a person is lying; in this study, we use novel electromyography electrodes to detect the facial expressions associated with telling (successful or unsuccessful) lies. By combining common signal analyses with state-of-the-art machine learning algorithms, we can outperform widely-used lie-detection techniques such as the polygraph. We also examine features of speech that can contribute to lie detection.