Take It Personally: Adaptive Learning

It’s definitely a term that is being bandied around the ELT-osphere like a banned elastic band at a Band Aid concert.

band copy

The Impatient Optimists site (of the Gates Foundation) declared that:

“If 2012 was the MOOCs’ year for capturing venture capital and unrelenting media coverage, 2013 looks to be a big year for the adaptive learning space. A number of organizations are already seeking to raise funding for market expansion and continued product development efforts.”

When it comes down to it, however, what is Adaptive Learning all about and (more importantly) what do publishers need to learn from it? After all, where does the adaptation really need to occur?

Adaptive Learning is a methodology that is geared towards creating a learning experience that is unique to each individual learner through the intervention of computer software. Rather than viewing learners as a homogenous collective with more or less identical preferences, abilities, contexts and objectives who are shepherded through a glossy textbook with static activities/topics, AL attempts to tap into the rich meta-data that is constantly being generated by learners (and disregarded by educators) during the learning process.

Rather than pushing a course book at a class full of learners and hoping that it will (somehow) miraculously appeal to them all in a compelling, salubrious way, AL demonstrates that the content of a particular course would be more beneficial if it were dynamic and interactive. When there are as many responses, ideas, personalities and abilities as there are learners in the room, why wouldn’t you ensure that the content was able to map itself to them, rather than the other way around?

Alongside repurposing the content so that it is no longer linear, an AL platform would be able to monitor how a learner is responding and interacting with the material and respond accordingly. As the platform watches the learner demonstrating what they are capable of doing now and how they demonstrate this, the system can then deliver content/activities that are best matched to their abilities and behaviour.

It’s worth pointing out that adaptivity is a different proposition to personalized learning in that it involves a more sophisticated approach based on the data generated throughout learner engagement. An AL system will behave differently depending on how the learner interacts with the content. It goes beyond suggesting materials that might help the learner following their recent performance but instead will start building a profile of the learner upon which it will begin making decisions. This seems to feed into the Quantified Self culture as exemplified by wearable exercise monitors; as an individual is able to learn more about their performance/activity they are able to identify ways of improving or optimising their efforts. If we substitute learning for physical exercise, we can see how the learner will become an informed collaborator in their own learning journey.

So, what sort of data are AL capable of capturing? On one level it’s possible to see what activity types a learner responds best to, or whether they respond well to audio/visual input of if they prefer to read straight from the screen. Such insight could dramatically shape how a course is structured and what the learner experiences. On another level, an AL platform would be able to see where mistakes were occurring and what types of mistakes they were. Content that might help offer further support could then be integrated into the course in order to provide further opportunities for improvement.

Even deeper still, the AL software of many online classes is able to track learners’ onscreen behaviour; keystrokes, click-streams, delays, speed. Not only will there be data to indicate what a learner is able to do, but there is also a layer of data that indicates how the learner interacts with the content; are they taking time to think? Are they quick to choose an answer? Do they frequently skip back to re-read the salient parts of the text whilst doing an exercise? The opportunities that this type of data might create are mind boggling.

But what’s the point of all this? How does AL ‘add value’ to the learning experience? According to a white paper by Education Growth Advisors the benefits of implementing an AL approach are plentiful; higher retention through promoting learner engagement, customised learner pathways, more precise/specific measurement of learning outcomes, as well as an on-going pedagogical development. In this way, it seems, AL enables those involved in education to start to work on education.

Furthermore, an AL solution may well – according to EGA – be the much needed key to the “Iron Triangle” that poses a conundrum to HE providers; cost, access and quality. Any attempt to improve any one of those conditions impacts negatively on the others. If you want to increase access to a course you run the risk of escalating costs and jeopardising quality, and so on.

Currently, the AL landscape is divided into platform providers and traditional publishers who are custodians of content. The platform providers include CogBooks, LoudCloud, Smart Sparrow and the subject of a previous post, Knewton. The solutions offered by these companies are “content agnostic” and have applications across a wide range of academic subjects, although individual firms may focus on certain subject areas over others.

In the publisher corner we see some of the long-established companies trying to move away from a total reliance upon the printed textbook model, as well as younger, more agile digital publishing operations. Currently, the strategies of the ‘establishment’ have been to bundle adaptive resources with more traditional print and digital products to cater for their HE institutions expectations whilst not necessarily hitting the AL jackpot. It’s not surprising that the publishing heavyweights are seeking out collaborative projects with the platform providers in order to appropriate the AL segment while the going is good.

Of interest, meanwhile, are the emerging publishers who don’t have to worry about an entrenched print publishing reputation. These set-ups are able to create media-rich online solutions that integrate gamification and instructional design elements to promote learner engagement. Players in this field include Adapt Courseware and Open Learning Initiative. Whereas these digi-pubs are in the perfect position to rapidly develop their expertise and range of offerings, they are still competing in a publishing industry in which Content Is King, and the ‘establishment’ of traditional publishers may well be the most effective route to reaching the key stakeholders in HE institutions.

As ever, with the necessary paradigm shift that AL represents, publishers who seek to get in on the act need to assume the responsibility that is inherent in pursuing the model. While the implications of creating/implementing AL platforms are  at least innovative (and at most transformational), the publisher needs to be thoroughly aware of the reasons behind making such platforms. What is the desired result and how will we know when it’s been achieved? What does an AL ‘graduate’ look like? Perhaps the first stage requires publishers to be adaptable enough to ask themselves these questions.

Featured Photo Credit: Wi2_Photography via Compfight cc

8 thoughts on “Take It Personally: Adaptive Learning”

  1. I like the vision of algorithmic learning creating individual learning pathways in our differentiated environments. Hard to let go of the classroom-based shared experience of a unifying text though. I’m excited by blended add-ons but see a great role still for course-books.

  2. Fascinating post but nightmares of 1984 spring to mind … it sounds to me like science-fiction and not something that major publishing companies will be seriously interested in. Major publishing companies depend on formal education systems and AL is anything but formal. The major buyers – corporate clients, big educational institutions and the like want proven ROI and to do that they use KPIs. These KPIs need to be measurable and that will be very difficult with AL. So, I can only see this seriously happening if it is adopted as a part of more structured, formal courses.

    • Hi Fiona, thanks for the comment.
      I completely agree about the Orwellian overtones that an AL platform represents. That is definitely one of the key considerations with the approach; when the system depends so much on the meta-data that a learner produces, at what point is that info gathering infringing on their privacy?
      You’ve also hit on the main point re: the big publishers – they are dependant on formal education systems not adapting their approaches or expectations. Publishers have a vested interest in not pursuing more effective, leading-edge educational solutions as it’s bad for business.
      I’m interested in your reference to KPIs in an educational context. What would the main consumer institutions regard as KPIs for learning?


Leave a comment