Rosalind Picard published a technical report in 1995 that defined the modern field of affective computing. Her work at MIT established the core goal of giving machines emotional intelligence.
Systems achieve an average reported accuracy of 70 to 80 percent in research from 2003 and 2006. This performance outperforms average human accuracy which sits around 60 percent.
Paul Ekman proposed six basic emotions in 1972 based on cross-cultural research with Fore Tribesmen in Papua New Guinea. He expanded his list in the 1990s to include amusement, contempt, contentment, embarrassment, excitement, guilt, pride, relief, satisfaction, sensory pleasure, and shame.
Affective computing provides AI-based learning support benefiting cognitive and emotional outcomes for learners. Distance education systems lack emotional incentive between teachers and students due to separation of time and space so this technology addresses negative effects on learning.
Physiological sensors measure skin temperature and galvanic resistance directly. These passive sensors gather data analogous to cues humans use to perceive others including speech patterns and facial expressions captured by video cameras or microphones.