Happy, excited or sad: MIT researchers develop machine that can detect your emotions

MIT researchers have made a machine that can read your feelings through radio waves.

Machines can now tell how you feel, even if you put on your best poker face. Researchers at MIT’s Computer Science and Artificial Intelligence Lab (CSAIL) have found a way to detect human emotions purely through radio signals. Using a device called the EQ-Radio, which emits and captures reflected radio frequency (RF) waves, the researchers bounced waves off subjects to measure their breathing patterns and heart rates.

That data was then run through an algorithm programmed to match physical factors with various emotional signifiers, which then categorised the person’s emotion as one of four states: sadness, anger, pleasure and joy.

The MIT team took 12 participants and instructed the subjects to recall a personal experience that made them feel a certain emotion. That established a 'ground truth', or an emotion that the researchers could test to see if the machine would catch it.

The EQ-Radio would use one of two methods to predict that subject’s feelings: It could either use that person’s ‘ground truth’ as a baseline for future predictions, or use the baselines gathered from the other 11 participants’ reported emotions. Each subject then recalled different experiences that triggered each of the four emotions the machine could detect, and the device measured heart and breathing rates in twominute sequences for each emotion.

The device’s predictions were startlingly accurate, especially when the EQ-Radio based its predictions on one person’s baseline. In that case, the machine’s emotional predictions were correct 87 per cent of the time. When the machine based its predictions on other people’s emotions, they were 72 per cent accurate. Researchers also found that heart rates, rather than breathing rates, were the bigger indicator of a person’s emotional state.