Share this Post
In the last article, I criticised the LMS for being a gilded cage, locking people and content into a frictious, top down, narrow, reductionist and mostly rational experience. I asked how we might create a low friction flow of more natural learning, anchored through each person’s intrinsic motivations. In the next three articles, I’ll discuss the ideas and technologies that might provide answers to this question, starting with the LMS as a coach, not a manager.
Before we start, it’s worth noting that most of the ideas and examples i’ll be sharing exist right now. Many of them also have APIs. So what I am proposing, while expansive and complex, is not fanciful or way over the horizon. But integrating these ideas and systems would require shared approaches and standards to enable a unified and agile user experience.
Let’s start with adaptive learning. This interesting trend epitomised by Knewton, Smartsparrow, Zosmat, Snapp and others, uses algorithms to detect and respond in real time to what you do, and do not learn, determining why and then automatically optimising your learning experience, in some cases through mnemonic tactics like spaced repetition. Predictive analytics approaches such as eAdvisor, Course signals, Degree compass and learnalytics extend this idea to what you should or can learn in the future and why.
These systems have the potential to enable learning that aligns with your current and future needs within your moment to moment flow, invisibly doing the heavy lifting of working out what, how and when you should learn. This active support approach is already pervasive with digital assistants like Siri, Google Now and Cortana surfacing information you’ll be interested in or accurately anticipating your next action and providing actionable information.
But these adaptive learning platforms fail to address our emotions. This is where affective systems come in. Building on advances in hardware and software that can accurately detect emotional states through facial mapping, biofeedback and gaze tracking, platforms and games like Gaze Tutor and Nevermind detect and respond to your emotional state by adapting the narrative to suit and/or manipulate your emotional state. Add to this, quantified self hardware that detects and responds to your EEG patterns, your state of arousal and so on, like Dash, Muse, Hemavu and you have the beginnings of a system that can detect your physical, emotional and cognitive state and then influence your capacity and propensity to learn.
I’ve always been a big fan of Johari windows, as many people have low self-awareness. The emerging field of psychometric analytics epitomised by companies like silentale, eLoyalty and talentanalytics applies predictive analytics to behavioural profiling. I believe, this idea could be taken even further to analyse your personality and suggest learning activities to strengthen areas you wish to work on like empathy, discipline extraversion or leadership traits.
For example, over breakfast, my phone quizzes me on the key aspects of a marketing course I plan to attend, to build my background knowledge and playing tones at key points. It knows I have on average 3 minutes spare in the mornings and that I’m a morning person. It also adjusts the question difficulty based on the stressors in my voice and when it detects i’m on the move, it provides some verbal formative feedback and metacognitive insights for me to ponder on my way to work.
At work, while wearing a pair of AR glasses and reading the course materials, I touch an unfamiliar word and a brief definition pops over the page, then in response to a verbal query and knowing that one of the subjects in my upcoming course is related to the term, my glasses show me a set of popular learning resources related to the term I queried.
When my head hits the pillow, it decides to run an experiment. It tries to improve my recall of the knowledge I was learning that morning, by replaying the tones I heard that morning, while I sleep. The next morning , it adjusts my morning quiz to test the effectiveness of its experiment and benchmarks the results against global averages. On balance it decides I’m a good candidate for sleep learning and earmarks the activity for future use and experimentation.
Also overnight my learning system has decides it’s probable that for my next career move I need to be highly extraverted, so it looks at my offline and online social activity and once again based on benchmarking, decides I need to work on my social skills, adjusting my learning stack accordingly.
Obviously this example is a little fallacious, but you get the idea, which is this – We have the platforms to create an eCoach engine that could detect and adapt to our momentary learning needs, emotional state and physical environment, with very little user input. Moreover, this engine could help us identify and close gaps we were not even aware of, improving not just our skills and knowledge, but even personality traits we wish to strengthen.
While many LMSs have deployed personalised learning paths, it’s often driven by remedial content sets tied to keywords and failure points, so if you fail quiz X, it recommends you do resource Y. This is a shallow and prescriptive approach to personalisation.
The approach discussed in this article uses deep adaptive learning (what is the true cause of the failure and what do we do about it), affective analysis (how did we feel during the test and would we have perform better if the circumstances were different), predictive analytics (what does this tell us about what we need to learn in future and how we like to learn) and psychometric analysis (does this failure indicate a weakness in a foundation competency or personality trait that should be addressed?) It could do all of this through a dialectic or collaboration, offering me reflections and insights about how, why and when I could learn. But I’ll talk more about this in a later article.
This idea is probably closer to coaching than managing, which is why it is a fundamental departure from the prescriptive manager type approach taken by the LMS. In the next article, I’ll examine how we might realise this learning journey through human, contextual and algorithmic generative learning experiences. I’ll end by describing a value adding approach to generative learning and show how this is profoundly different to the philosophy underpinning most LMSs.
David is an Edtech Innovator, with over 20 years experience in the digital and adult learning spaces. Based in Melbourne Australia he runs the “Kill the LMS” workshop, designed to disrupt your thinking about how humans learn, reflect on the limitations your LMS imposes upon the performance of your people and look at ideas and architectures to remove those limits.
Click here to learn more about this workshop, or to book a free one hour phone conversation with David. You might also like tojoin David’s closed Linkedin group, exploring these issues in greater depth or follow our company page on linkedin for more great posts.
Share this Post