AI Model Decodes Human Emotions with Accuracy.
Originally Published 2 years ago — by Neuroscience News
MIT neuroscientists have developed a computational model that can predict human emotions, including joy, gratitude, confusion, regret, and embarrassment, based on the prisoner’s dilemma game theory. The model uses factors such as a person’s desires, expectations, and whether their actions are being observed to predict emotional responses. The researchers used scenarios from a British game show called “Golden Balls” to train the model, which outperformed previous models in predicting emotions, signaling significant progress in emotional artificial intelligence.