Previous U. of California Professor Paul Dourish: the Sexiest Man Alive? |
Next U. of California Researchers Hold Wikipedia Authors Accountable |
June 27, 2008, 03:43 PM ET
Facial-Recognition Software Could Give Valuable Feedback to Online Professors
Many professors who teach online complain that they have no way of seeing whether their far-away students are following the lectures — or whether the students have fallen asleep at their desks. But researchers at the University of California at San Diego say they have a solution. They recently tested a system that can detect facial expressions of online students and determine when they find the material difficult, so that cues could be sent to the professors telling them to slow down.
Jacob Whitehill, a doctoral student at the university working on the research, presented results from the experiment this week at the Intelligent Tutoring Systems 2008 conference in Montreal.
In the experiment, eight subjects were shown short video clips of lectures while a Web cam tracked their facial expressions — looking for smiles, blinks, raised eyebrows, and the like. The subjects were then asked to report how difficult they found each section, and to take a quiz on the material. Mr. Whitehill says that the system correctly detected when students were having trouble (the most reliable indicator: students blinked less when they were struggling to understand).
The system could be used to give valuable feedback to professors teaching online, says Mr. Whitehill. “It’s not going to be perfect by any means,” he says, but it’s better than no student feedback at all. “Professors say that they can’t see the students. This could do it for them automatically.” —Jeffrey R. Young
Categories: Teaching
Add Your Comment
Commenting is closed.