Adaptive Learning: Five Common Misconceptions

Adaptive learning brings together the latest in learning science, data, technology, and workplace principles. The concept is based on the realization that people develop in their own unique ways and, therefore, require a more customized experience than an academic, one-to-many approach offers. By using multidimensional data and cutting-edge technology, L&D can deliver the right information to the right person at the right time, thereby promoting value to the individual as well as the overall organization.

Adaptive learning is a bleeding-edge topic for corporate L&D. As such, there is plenty of room for misinterpretation and misunderstanding about what it takes to deliver an adaptive experience. In order to shift your L&D team from a one-size-fits-all approach to a right-size-fits-one mentality, you must overcome these misconceptions and clearly articulate the value of adaptive learning within your organization. After all, your stakeholders don’t want to hear about learning strategy. They are focused on getting their employees to the desired level of capability as quickly as possible in order to drive business value.

Read more


Pearson vs iTutorSoft about science-based learning design

This post is about the methodology of Adaptive Personalized Learning. It represents my comments on Michael Feldstein’s post that Pearson Releases a Significant Learning Design Aid.

It is not a secret, but rather a commonly forgotten or ignored fact, that scientific knowledge has a hierarchical structure in contrast to amorphous/messy heuristics, practices, pragmatic principles, rules, etc. A few of general interdisciplinary sciences, such as Systems, Control, Activity Theories, are on the top of this hierarchy. They generalize objects, models, and methods of more specific exact sciences and soft humanities, both located below in the hierarchy. In turn,  humanities include the learning sciences. In contrast to exact sciences (with well-defined objects, models, and methods), some humanities and all the learning sciences have an ill-defined, ill-observable, and ill-controllable object, the learning process. As a result, models and methods of learning sciences are often represented with amorphous/messy heuristics, best practices, principles, rules, etc. This is what Michael’s post is about. It is pretty challenging to analyze, understand and explain this mess to others who has the different mess in their minds. Michael’s talent is required.

The common problem is that we do not see a forest behind the trees. We are constructors without an architect. But according to the science hierarchy that forest, big picture, general models, methods, and methodology exist on the top level and can be found in such multidisciplinary theories as Systems, Control, Activity Theories.

It is not just an idea. In my R&D, I successfully used many interdisciplinary theories as a basis for developing general models and methods of cost-effective Intelligent Tutoring, Adaptive Learning Systems, and Platforms. According to the general-specific hierarchy,  all Pearson’s specific principles can be implemented within our general content-independent platform.


4 critical personalized learning questions, answered

Among the most popular buzz phrases in education over the last several years, “personalized learning” is also one of the K-12’s most promising trends as the sector works to move away from the “factory model” of the past century and toward what some have called “School 2.0.”

Continued Progress: Promising Evidence on Personalized Learning,” a 2015 RAND Corporation study funded by the Gates Foundation, spent two years measuring the academic progress of 11,000 students in 62 public charter and traditional schools utilizing a variety of personalized solutions, finding greater gains among those using personalized approaches than a similar comparison group.

Read more


Higher Education’s Struggle to Leverage Digital Teaching and Learning

According to“Evidence is mounting that the institution of higher education, as it is currently designed, is largely ill-suited to developing and leveraging more advanced uses of technology for teaching and learning. And given the institution’s near monopoly on widely recognised adult education in much of the West – higher education is likely inhibiting the development of more advanced forms of instructional technology and media, as well as new ways to bring these new forms to people at lower costs.”

For more details on this very sad subject see the full text of this excellent essay.


CLARITY shapes the Next Generation of eLearning, eLearning 3.0

What Is eLearning 3.0?

“Fundamentally it’s the end of courses that consist solely of video lectures and multiple-choice assessments. It’s an evolution to a student-centered world. It leverages advances in learning science and tools to make learning faster, more effective, more enjoyable, and applicable to a larger set of learning areas. It’s the end of systems where course production considerations trump learner experience. Hopefully it’s the end of wading through hours of boring videos and questions to learn something without really being able to apply it in the real world”.

For more on evolution of eLearning and details of eLearning 3.0, see a full paper.

For more on CLARITY platform click here.


Education Strategic Model First

This post is my feedback on the article

There is nothing wrong with research to test and verify what works, including research of blended and personalized learning.

There is nothing wrong with the research process, which involves identifying a problem to solve, studying academic literature related to that problem, developing detailed and context-specific descriptions of the problem, prototyping solutions, seeking focused feedback, revising solutions, testing solutions, and then scaling and automating solutions that prove successful. Becides a fact that is pretty obvious.

What I cannot accept here is a strategy of innovation: a problem by problem. It looks like a patching of existing obsolete system, not its holistic structural redesign.

The strategy has to include an innovative holistic vision first, then a holistic education model, and only after that its developing: part by part as it described above.
Absence of such a strategy is a root of most educational problems, patching, cycling, etc.

We have that strategic model of education and we use it to derive and define all parts and interrelations of our CLARITY platform level by level, step by step.


CLARITY platform to support the Universal Design for Learning

Universal Design for Learning (UDL) recognizes that every learner is unique and processes information differently. UDL is based on CAST’s research related to three primary brain networks [recognition network (What), strategic network (How), and affective network (Why)] and the roles they play in understanding these differences. UDL provides a conceptual framework to create and “manually” implement lessons with flexible goals, methods, materials, and assessments that support learning for all students.

Our CLARITY authoring tool (developed independently) includes three panes for systems specification of a Learning Activity: Objectives, Tasks/Performances, and Resources.

The Objectives pane reflects WHY aspect of an Activity and generally allows authors to define Learning Objectives, which can include affective Objectives of UDL, such as “engage”, “motivate”, …

The Task pane reflects HOW aspect of an Activity and generally allows defining Learning/Testing Tasks and their Expected Performances. It also allows defining UDL Tasks solving and planning.

The Resource pane reflects WHAT aspect of an Activity and generally allows authors to define/develop media/physical objects for presenting the Tasks to a learner, to be recognized and used during Task solving.

All three panes provide an author with functionality necessary for developing and associating sets of Objectives, Tasks, Performances and Resources within each Lesson, as well as their Alternatives necessary for adaptation to each particular learner.

To exclude (labor-consuming and error-prone) “manual” planning in UDL, CLARITY AI Tutoring Engine automatically plans, selects and presents each next Task/Resource to each particular learner based on her current profile updated after each interaction of the Engine with a learner.

Thus, CLARITY enforces UDL conceptual framework, clarifies and formalizes it for computerization, provides an Authoring Tool, excludes manual planning of personalized lessons, and automatically generates adaptive/intelligent tutoring of each particular learner. It opens new horizons for UDL scale-able online implementation.

_______________ _ ________________

Permalink Reply by Richard Jackson on February 3, 2011 at 12:29pm

Without digital technology (media and tools), UDL is just a good idea. With technology, UDL becomes palpable, visible and REAL!



Adaptive Learning Fails to Make the Grade. Or Does It?!

As the author of this post said “If you’re looking for evidence that adaptive learning is going to deliver on the promise of a robot tutor in the sky, you won’t find it there. But it’s easy to flatten that result into “adaptive learning doesn’t work.” I don’t believe that the SRI study shows any such thing.”

Me too actually. As one of the participants has mentioned during experiments they used “minimalistic technology”. So, all of it is not really about Adaptive Learning Technology and its implementation. It is just about more testing with traditional manual analysis and interpretation of results by teachers. Please do not generalize those experiments’ outcomes to modern science-loaded high-tech Adaptive Learning as CLARITY provides.


Personalized Learning Explainer: Teaching to the Back Row

Authors of this article frame personalized learning as “a set of ideas for solving [the] problem” of teachers being overloaded with work and not having enough time to give their students individual attention. That is a feature, not a bug. By framing it this way, we hope to open the door to questions like the following:

  • Why are faculty so overworked that “being the kind of teacher that [they] want to be gets harder and harder”?
  • How does the increasing diversity of our student population make good teaching more challenging and what is the best approach to meeting that challenge?
  • Are we providing faculty with the right support, incentives, and training to promote good teaching?
  • When is personalized learning good teaching practice and when is it cover for bad labor policy?
  • How can technology help us increase access to education without hurting quality and what are the limits of that capacity?

Reed the full article