The Principle of Adaptive Learning
The Principle of Adaptive Learning
Of all the new big things in higher education, one of the most talked about is adaptive learning because it resonates with the principle that students achieve best when they control their own learning space. Just as artificial intelligence is experiencing a resurgence from its former doldrums, contemporary adaptive platforms make it possible to design courses and programs that personalize the learning experience by customizing content and assessing learning outcomes continually and individually. The big breakthrough, however, is that learning time is no longer constant—a constraint that causes students to achieve at substantially different levels, forcing a serious reevaluation of the higher education learning experience. The definition of a course, for instance, becomes more nuanced by providing students with unique, real-time adjusted learning paths with embedded assessment that can accelerate their learning or go beyond traditional boundaries such as semesters; remediation can be part of the learning path as well. Generally, adaptive “courses” organize themselves into a set of content modules enabled by a more specific set of learning activities, sometimes called nodes, that take on a variety of learning contexts: video, text, simulations, exercises, problem solving, games, or other faculty designed instructional devices. Rather quickly, good adaptive systems that incorporate some form of machine learning determine the most effective content presentation mode for individual students, enabling assessment to become much more authentic. For example, when asking aspiring nurses to evaluate patient telemetry, each case will receive different values that require a unique interpretation, thus gaining more than mere knowledge acquisition. This approach embeds mastery learning where students can review and repeat nodes and modules until they have acquired a specified level of achievement or develop competencies to their satisfaction. In addition, students have the opportunity to “test out” of portions of courses by demonstrating content proficiency. These pedagogical innovations, working in concert with and enabled by adaptive learning technologies, have the potential to make students more successful in gateway courses, thereby reducing dropout potential.
The Many Flavors of Adaptive Learning
Our last search found well over twenty adaptive learning platforms with providers offering a wide variety of approaches, ranging from Bayesian decision algorithms to rule-based recommendations with many other variations. Some come with vendor provided content that may or may not be modified, while others are content agnostic, requiring instructors to create their own course curriculum and material. Both approaches have advantages and challenges. Systems that include content necessitate much less front end work but lack the flexibility that some courses require whereas systems that require content input from the faculty member offer flexibility but require a great deal of course organization and preparation work.
“At the pedagogical level, adaptive learning offers the flexibility in educational programs, freeing students from the traditional lock step educational system”
Adaptive Learning Possibilities
In our increasingly diverse society, students—if they are to succeed in college—require more flexibility in their educational programs. At the pedagogical level, adaptive learning offers that possibility, freeing students from the traditional lock step educational system. Consistently, removing learning time as a constant has been shown to be a positive element in successful course completion. Further, students from underserved communities have much less financial and temporal slack in their lives to manage the many pressures that poverty creates; these students have too many needs and not enough resources, causing trade-off thinking where their studies become secondary to their life demands. Adaptive learning has the potential to relieve some of those pressures especially if combined with online learning.
These methods also maximize student retention and success with predictive analytics, identifying early on when intervention is most effective for students at higher risk of failure or dropping out.
Adaptive Learning Challenges
Adaptive learning presents a number of challenges to higher education in its current form. Adaptiveness is not consistent across platforms and mismatches in design and operational characteristics with the university culture can be formidable to overcome. Relaxing the learning time constraint causes instruction to exceed or shorten traditional education boundaries such as semesters and academic years. How do we accommodate students who need a few more weeks to achieve mastery in terms of transcripts and financial aid? What is to be done with students who complete intermediate algebra in half a semester and are ready to go on to college algebra in the middle of the term? Adaptive learning is much better suited for a continuous skill set than discrete components such as courses.
Adaptiveness also changes the teaching and learning paradigm. Although the instructor remains a critical part of the process when serving in an augmented mentor role, students’ primary interaction is with the adaptive platform. This constitutes a fundamental role transformation for both teacher and student. Finally, how do we incorporate a social learning component into adaptive systems where education is more than the student interacting with the machine? These are substantial challenges but excellent opportunities as well. We just have to adapt to them.