Whether it's autocorrect learning what words you use most, Netflix learning which movies to suggest, or Google identifying which ads you're respond to, I'm sure we're all familiar with technology that figures out what we might want or need next.
OpenStax wants to use similar tech to craft textbooks which adapt to serve the needs of individual students. "The technology is already here, in the sense that most of us use it online every day," said Daniel Williamson, OpenStax managing director. "However, the full potential of this technology has yet to be realized for education. The project will allow us to demonstrate that this technology is effective and can be used in the classroom to improve both students’ and teachers’ return on effort."
The new textbooks will track each student's work and pinpoint areas where the student needs more assistance
That work is now being used to train machine-learning algorithms that give OpenStax's biology and physics textbooks the ability to adapt to individuals. If a reader seems to be struggling with a particular topic – acceleration, say – the book will slot in additional explanations and practice questions, and increase emphasis on related subjects, such as centripetal force, that could otherwise trip that person up.
The adaptive textbooks also incorporate a learning method called retrieval practice, in which material that students have already learned pops up again in occasional quizzes. This method has been shown to enhance students' ability to retain material, and the algorithmic textbooks will be able to decide when to ask questions based on past exercises.
This is going to be at least a two year project, with the goal of creating proof-of-concept, fully personalized textbooks for high school AP biology and physics classes. Luckily OpenStax won't be starting from scratch; instead they plan to work from 2 existing OpenStax College titles — Concepts of Biology and College Physics.
"The same sort of algorithms that might predict which songs or books you’ll like can be used to deliver a personalized experience to every child in a classroom," said OpenStax Founder Richard Baraniuk, Rice University's Victor E. Cameron Professor of Engineering.
OpenStax is best known for the CC-licensed college textbooks it has published over the years. Using philanthropic grants, OpenStax has published eleven freshman textbooks (with 3 more on the way) that can be bought as print textbooks for around $34 each. The digital textbooks can be downloaded as PDF or Epub, and a handful of titles are also available in iBooks as interactive textbooks.
That print price is far cheaper than the typical bio 101, chem 101, etc textbook, which often reach a retail price of $200 or more, and the ebook price is simply incomparable. OpenStax reports that their first seven books have already saved students more than $13 million.
image by francisco_osorio