Adaptive learning could reshape higher ed instruction (essay)
“It’s not the strongest of the species that survives,” Charles Darwin once observed, “but the one most responsive to change.”
If only it were true in higher education.
It’s interesting to observe, isn’t it, how much higher education is still driven by a “brute force” model of delivery? As much as we might wish it were otherwise, postsecondary courses and degree programs are still largely delivered in a one-size-fits-all manner, and those students who can’t keep up are simply left behind, sometimes irretrievably so – the higher education equivalent of natural selection, some might say.
(I once had lunch with a colleague, for example, who told me with no small amount of pride that he only taught to the 10 percent of the class who “got it.” The others, it seemed, were not worth his effort.)
But surely anyone – teacher, student, or otherwise – who has ever sat in a classroom has seen glaring evidence of the fact that not all students move at the same pace. Some are prepared to move more quickly than the majority while others require greater attention and more time to master the same material as their classmates. The limits of mainstreaming diversely skilled students are obvious to all and yet we largely persist in the vain hope that greater numbers of students will learn to move at “class pace” if only we underscore their responsibility to do so in syllabuses and first-class lectures.
Of course, when teachers face classes of 20 or 40 or 200 students, personalized instruction isn’t much of an option. It’s simply too expensive and impractical – until now, perhaps.
Witness the countervailing perspective emerging these days that the curriculum is the thing that needs to change pace. Indeed, after a number of years of quiet experimentation we may now be on the cusp of an evolutionary moment – one that promises greater personalization, deeper engagement, and stronger outcomes for students of many types. And it may even be affordable. In fact, it may even be cost-efficient, by virtue of allowing instructors to use their time more judiciously.
Welcome to the emerging realm of adaptive learning – an environment where technology and brain science collaborate with big data to carve out customized pathways through curriculums for individual learners and free up teachers to devote their energies in more productive and scalable ways.
What promises to make adaptive learning technologies an important evolutionary advance in our approaches to teaching and learning is the way these systems behave differently based on how the learner interacts with them, allowing for a variety of nonlinear paths to remediation that are largely foreclosed by the one-size-fits-all approach of traditional class-paced forms of instruction.
To put it simply, adaptive systems adapt to the learner. In turn, they allow the learner to adapt to the curriculum in more effective ways. (See this recent white paper from Education Growth Advisors for more background on what adaptive learning really looks like – full disclosure: I had a hand in writing it.)
If the early results hold, we may soon be able to argue quite compellingly that these forms of computer-aided instruction actually produce better outcomes – in certain settings at least – than traditional forms of teaching and assessment do. In the future, as Darwin might have said were he still here, it won’t be the students who can withstand the brute force approach to higher education who survive, but those who prove themselves to be the most adaptive.
A recent poll of college and university presidents conducted by Inside Higher Ed and Gallup showed that a greater number of the survey’s respondents saw potential in adaptive learning to make a “positive impact on higher education” (66 percent) than they saw in MOOCs (42 percent). This is somewhat surprising given the vastly differing quantities of ink spilled on these respective topics, but it’s encouraging that adaptive learning is on the radar of so many college and university leaders. In some respects, adaptive learning has been one of higher education’s best-kept secrets.
For over a decade, Carnegie Mellon University’s Open Learning Initiative has been conducting research on how to develop technology-assisted course materials that provide real-time remediation and encourage deeper engagement among students en route to achieving improved outcomes. So adaptive learning is not necessarily new, and its origins go back even further to computer-based tutoring systems of various stripes.
But the interest in adaptive learning within the higher education community has increased significantly in the last year or two – particularly as software companies like Knewton have attracted tens of millions of dollars in venture capital and worked with high-visibility institutions like Arizona State University. (See Inside Higher Ed’s extensive profile of Knewton’s collaboration with ASU, from January of this year, here.)
Some of our biggest education companies have been paying attention, too. Pearson and Knewton are now working together to convert Pearson learning materials into adaptive courses and modules. Other big publishers have developed their own adaptive learning solutions – like McGraw-Hill’s LearnSmart division.
But a variety of early-stage companies are emerging, too. Not just in the U.S., but all around the world. Take CogBooks, based in Scotland, whose solution’s algorithms permit students to follow a nonlinear path through a web of learning content according to their particular areas of strength and weakness as captured by the CogBooks system. Or consider Smart Sparrow, based in Australia, whose system supports simulations and virtual laboratories and is currently being deployed in a variety of institutions both at home and here in the U.S., including ASU.
There is also Cerego, founded in Japan but now moving into the U.S., with a solution that focuses on memory optimization by delivering tailored content to students that is based not only on a recognition of which content they have mastered but also with an understanding of how memory degrades and how learning can be optimized by delivering remediation at just the right point in the arc of memory decay.
These adaptive learning companies, and many others working alongside them, share a common interest in bringing brain science and learning theory into play in designing learning experiences that achieve higher impact.
They differ in their points of emphasis – a consequence, in part, of their varying origin stories. Some companies emerged from the test prep field, while others began life as data analytics engines, and so on. But they are converging on a goal – drawing on big data to inform a more rigorous and scientific approach to curriculum development, delivery, and student assessment and remediation.
In the months ahead, you should expect to be seeing more and more coverage and other discussion of companies like these, as well as the institutions that are deploying their solutions in increasingly high-impact ways. Last month, the Bill & Melinda Gates Foundation issued an RFP inviting institutions to collaborate with companies such as these in seeking $100,000 grants to support new adaptive learning implementations. The grants are contingent, in part, on the winning proposals outlining how they’ll measure the impact of those implementations.
Before long, then, we may have much more we can say about just how far adaptive learning can take us in moving beyond a one-size-fits-all approach to teaching and learning – and in achieving better outcomes as a result. And for some students, their survival may depend upon it.