Facial recognition, in our classrooms?
PATENT CN109255739A A technology that collects and analyzes the facial expressions and movements of students inside a classroom and categorizes the behavior it detects as “good” or “bad.” After weighting these various behaviors, the algorithm will calculate the total “learning efficiency” of each classroom.
In the past 20 years, there’s been a bunch of research in computer science that tries to look for particular facial expressions that might indicate that learning is taking place. One study argues that wrinkling the mouth or furrowing the brow is correlated with learning. So perhaps these systems track " href="https://story.californiasunday.com/facial-recognition#block-3762766">students’ eyeballs and gazes, and look at whether their mouths are wrinkled or their eyebrows are furrowed. It will also look at students’ movements — either just the top half of their bodies, or how they’re moving around a classroom. Maybe if the student is sitting still and looking toward the front, the system takes it as a sign that " href="https://story.californiasunday.com/facial-recognition#block-3762768">they’re engaged.
Education is keen on generating feedback in real time; I could see this data being " href="https://story.californiasunday.com/facial-recognition#block-3762770">fed back to students and teachers, perhaps in the form of a dashboard in the classroom, with a “learning indicator” that goes from zero to 100. And if we’re being Orwellian, you could also see the school principal with a huge, real-time dashboard in their office, indicating which classrooms are doing well and which are not. And, in an educational system where administrators try to gather as many data points on students as possible, this data on classroom performance will likely be added to student records.
In an old-fashioned school, students who aren’t paying attention know to sit down, face the front, and not look too bored. Students would learn to game this system. They aren’t stupid — they’ll just figure out how to move in the right way: If I know not to move, then I won’t move. If I’m supposed to furrow my brow, I will furrow my brow.
A problem we talk about in education is educators teaching to the test, and this is a similar phenomenon: If the machine punishes me for getting students to move around, then we won’t move around. If I were a teacher, I’d be making sure my students are moving in the right way, with the right expressions on their faces. I might start coordinating all of it myself: telling students, “Move now. Stay still. Make this face.”
— NEIL SELWYN, PROFESSOR OF EDUCATION AT MONASH UNIVERSITY, AUSTRALIA
AI for Social Good?
Data Bite #02: The shift to off-campus, online study
You may republish this article online or in print under our Creative Commons licence. You may not edit or shorten the text, you must attribute the article to Monash Education Futures, and you must include the author’s name in your republication.
If you have any questions, please email email@example.com