digest
Short Summary
Main recommendation: Do not rely on eye contact, blinking, or gut feeling to spot a liar, because people guessing in the ball-throwing demo were only right about 50% of the time, and studies show professionals like judges, social workers, and police perform no better than chance. The video instead presents research showing that lying can be detected through measurable facial blood flow changes and brain activity, though these methods are still experimental.
Key Points
- In a study of children told to throw balls into a basket for a $5 reward, 80% of kids over age four lied when left alone, and viewers guessing who lied were right only about 50% of the time.
- The common belief that liars avoid eye contact or stop blinking was disproven by Dr. Ken Lee's research with children, who looked directly into the experimenter's eyes while lying.
- Studies show judges, social workers, and police officers, who are expected to detect lies, perform no better than chance, and even parents cannot reliably tell when their own kids are lying.
- Transdermal optical imaging uses video of the face to track blood flow changes, and machine learning can separate lying patterns from truth-telling patterns with about 85% accuracy in one study.
- Functional near-infrared spectroscopy (fNIRS) shows that lying increases activity in the prefrontal cortex and that the brain becomes more disorganized after a lie.
Details Worth Noting
- Concrete instructions: The video gives no step-by-step instructions for detecting lies. The only actionable guidance is to avoid relying on common-sense cues such as eye contact or blinking, and to recognize that the lab methods described (recording faces, analyzing blood flow, or using fNIRS) are research tools, not something a viewer can apply directly.
- The facial blood flow pattern found in lying children: blood flow on the cheek goes down while blood flow in the nose goes up.
- FaceReader software tracks emotions such as joy, anger, surprise, fear, contempt, and disgust from a webcam or recording, but the speaker notes it does not help much with detecting lies.
- fNIRS works by shining light into the head and measuring how much light is absorbed between emitter and detector ports, which reveals blood oxygen levels and brain activity.
- The speaker imagines hypothetically using such lie-detection tech in a presidential election, but cautions it would need to be perfected first so it isn't misused.
Bottom Line
Human lie detection is largely unreliable, but emerging technologies like facial blood flow analysis and brain imaging show promise, though they remain research tools that still need refinement before real-world use.
