New teaching software can sense human emotions

Teaching software that recognizes cognitive and emotional states of its users has been developed by U.S. scientists, Science Daily reported Friday.

“AutoTutor” is an Intelligent Tutoring System (ITS) that simulates the teaching and motivational strategies of human tutors. It gauges the students’ knowledge by asking questions and analyzing the answers, proactively identifying and correcting students’ misconceptions.

As it interacts with the students, the program senses their frustration or boredom through facial expressions and body posture. It then changes its strategies to help the student cope with those negative emotions.

“Most of the 20th-century systems required humans to communicate with computers through windows, icons, menus and pointing devices,” said Professor Sidney D’Mello of Notre Dame University, a human-computer interaction and artificial intelligence specialist who participated in the research

He said the new technology “provides information regarding the cognitive states, motivation levels and social dynamics of the students” which enables humans and computers to communicate with each other through nonverbal cues, like the way humans interact with each other.

“Much like a gifted human tutor, AutoTutor and Affective AutoTutor attempt to keep the student balanced between the extremes of boredom and bewilderment by subtly modulating the pace, direction and complexity of the learning task,” D’Mello said.

Studies have shown that one-on-one tutoring is far more effective then typical classroom environments, which AutoTutor tries to simulate.

Test results of AutoTutor on more than 1,000 students show that it produces learning gains of approximately one letter grade -- gains that have proven to outperform novice human tutors.


From news reports