Skip to content

What is the theory that underpins our moocs?

If you’re even casually aware of what is happening in higher education, you’ve likely heard of massive open online courses (MOOCs). They have been covered by NY Times, Chronicle of Higher Education, TV programs, newspapers, and a mess or blogs. While MOOCs have been around since at least 2008, the landscape has changed dramatically over the past 10 months. In this timeframe, close to $100 million has been invested in corporate (Udacity) and university (EDx and Coursera) MOOCs . And hundreds of thousands of students have signed up and taken these online course offerings.

Personally, I’m very pleased to see the development of Coursera and EDx. The learning potential for society (globally) are wonderful. All of the critiques that I’ve read so far ring hollow compared with the tremendous learning opportunities that these MOOCs provide. This hit home for me when I was in India a few months ago. I met with numerous university students and the message was clear: we simply can’t get the quality of instruction from some of our colleges that we get from Coursera. While we debate pedagogical models and the ideologies informing different MOOCs and the corporate interests of open courses, the lives of students in different parts of the world are being changed with these projects. And that should be our real focus.

A secondary focus, for me (and far lower on the scale than the primary one mentioned above), is around the learning theory and pedagogical models that influence different types of MOOCs.

In 2008, Stephen Downes and I offered an open online course, Connectivism and Connective Knowledge (CCK08). As our registration numbers increased to about 2300 students, Dave Cormier and Bryan Alexander dubbed the course offering a “massive open online course” or MOOC. The term has stuck and both Dave and Bryan will eventually be inducted into the edtech hall of fame for great word inventage. Since that first course, Stephen, Dave, and I have offered a whack of different courses: CCK09, CCK11, CCK12, Future of Education, PLENK, LAK11, LAK12, Change11, Critical Literacies, and so on. All told, we are likely approaching about 20,000 registrants for our MOOCs (there is overlap from different courses, so the unique registrants would be less). 20,000 is an impressive number, but it’s hardly a blip on the Coursera scale (where student numbers in excess of 100,000 seems to be the norm). If Thrun, Page, Koller, and Ng are A-listers, we are “adding new letters to the back of the alphabetsters”.

The MOOCs that we’ve done – and the ones that Jim Groom and Alan Levine run (DS106) and Alec Couros runs (EC&i831) – are defined by a participative pedagogical model. They are unique and different from the emerging elite university MOOC model. There are many points of overlap, obviously, as both our MOOCs and the Coursera/EDx MOOCs taken advantage of distributed networks to reflect changing educational practice. In the presentation below, I detail some of the distinctions between the different open course formats:

What is the theory that underpins our MOOCs?

Aside from the surface level distinctions between our MOOCs and the Coursera/EDx model (yes, I’m “othering” them), some important differences exist in the underlying views of knowledge and learning that inform the different MOOC models. Here’s an 8-pack of MOOCy distinctions and what makes our approach distinct.

1. Connectivist. Our MOOCs are informed by connectivist views of learning, namely, that knowledge is distributed and learning is the process of navigating, growing, and pruning connections. Some theories of learning, notably the many shades of constructivism, have some similarities with this view. However, as the world becomes more digital and networked individualism becomes more prominent, the distinctions between connectivism and constructivism are becoming more clear. Eventually, even the constructivist will describe learning and knowledge through the lens of connectedness. This post needs to be updated, but it captures some of the unique attributes of connectivism: What the unique idea in connectivism?

2. Knowledge is generative. The Coursera/EDx MOOCs adopt a traditional view of knowledge and learning. Instead of distributed knowledge networks, their MOOCs are based on a hub and spoke model: the faculty/knowledge at the centre and the learners are replicators or duplicators of knowledge. That statement is a bit unfair (if you took the course with Scott E. Page at Coursera, you’ll recognize that the content is not always about duplication). Nor do our MOOCs rely only on generative knowledge. In all of the MOOCs I’ve run, readings and resources have been used that reflect the current understanding of experts in the field. We ask learners, however, to go beyond the declarations of knowledge and to reflect on how different contexts impact the structure (even relevance) of that knowledge. Broadly, however, generative vs. declarative knowledge captures the epistemological distinctions between our MOOCs and the Coursera/EDx MOOCs. Learners need to create and share stuff – blogs, articles, images, videos, artifacts, etc. The maker movement embodies the “create stuff” attitude. Wendy Drexler’s video, produced during CCK08, is a great illustration of creation and generation in open courses:

3. Coherence is learner formed & instructor guided. This attribute is closely related to the point above. In a traditional course, the instructor creates knowledge coherence by bounding the domain of knowledge that the learners will explore: i.e. this is the course text, here are the readings, quizzes will validate that you’ve learned what I think is important, etc. In our MOOCs, we have a sloppier relationship with coherence. I communicate my views of how different elements are related, but then ask learners to explore, deepen, and extend the ideas I/we express with additional narratives/opinions/views. Coherence then is something that the learner forms as she makes sense of and finds her way through the messy knowledge elements that comprise the many dimensions of a field.

4. Interactions are distributed, multi-spaced. Our first MOOC – CCK08 – started by being primarily centered in a Moodle discussion forum. As the course progressed, interactions were scattered over many tools and technologies. We ended up with many spaces of interactions: Second Life, PageFlakes, Google Groups, Twitter, Facebook, Plurk, blogs, wikis, YouTube, among dozens of others. Coursera/xED learning is heavily centered in their platform. DS106 is a similar multi-spaced course: blogs and personally owned spaces define much of the interaction and learning.

5. Synchronization. This is an important idea that I’ve long struggled to communicate. I still can’t get it quite right. However, this video partly gets at synchronization in learning: we act and react to those around us in ways that creates some structure or part uniformity of understanding. However, we are not robots (unlike the agents in the video), so our interactions don’t harmonize to the same level. Still, synchronization is an important element in our learning activities in MOOCs. We don’t online align ourselves to the course content and the instructor; we align ourselves to other learners and their knowledge.

6. Resonance. Like synchronization, resonance is a term that feels like it’s important but one that I can’t quite clearly articulate. The notion of “idea collision as innovation” is an example of resonance – two partial ideas combine to create something new This video reflects this concept. In a MOOC, learners possess different levels of knowledge. We fill each others knowledge gaps in MOOCs. Another aspect of resonance is concerned with why certain ideas have uptake in a network in contrast with others. In order for me to be able to interact with an idea, I need to have developed a suitably broad knowledge structure within which to situate new ideas. Certain concepts are inaccessible until I have formed a base network of knowledge. Resonance is concerned with why some concepts readily integrate with what I already know while others float by and are only “integratable” later on.

7. Innovation & impact focused. It is now a cliche to state that the world is complex and that knowledge is continually evolving (It was a cliche when I wrote Knowing Knowledge but less so than today). However, just because it’s a cliche doesn’t mean that it’s not true. We face complex challenges as a society. The solutions will be found in distributed/networked approaches. The challenges are too big to be addressed in traditional sub-clustered empirical knowledge models. Integrative and holistic knowledge approaches, distributed across global networks (for example, how the virus that causes SARS in 2003 was identified). With the MOOCs that we run, we attempt to emulate connective and integrative knowledge – a tug on what part of the knowledge network impacts other parts.

8. Fostering autonomous and self-regulated learners. At the core of the MOOCs that I’ve been involved with is a power question: what can learners do for themselves with digital tools and networks? MOOCs foster not only a particular type of knowledge in a particular area of inquiry; they also foster a self-regulated, motivated, and autonomous learner. When an instructor does for learners what learners should do for themselves, the learning experience is incomplete. Developing capacity for learning and the mindsets needed to be successful learners is a central attribute of our MOOCs. We are not only concerned with the epistemological development of learners (knowing stuff) – we target ontological development (being a certain type of person) as well.

These are a few of the more prominent attributes of MOOCs that I’ve been involved with. As stated above, there is overlap between our model at that of Coursera/EDx. However, Coursera/EDx emulates the existing education system, choosing instead to transfer it online rather than transform it online. No doubt they will continue to evolve their model and we’ll continue to evolve ours. Unfortunately, we will be spending slightly less than $100 million on ours!


  1. Dawn Worley wrote:

    You present a strong case for transforming MOOCs rather than transferring them online. The interactive content coupled with the development of information literacy skills are a definite draw for students and educators alike. As an educator and doctoral student, I have found that the greatest learning impact comes from discussions with others. Therefore, I appreciate the incorporation of shared knowledge in MOOCs. As you move forward with your project, have you considered offering credentials for your courses? If so, how might a system such as this work?

    Monday, June 4, 2012 at 9:37 am | Permalink
  2. Paul wrote:

    Interesting. So these neo-MOOCs essentially give traditional education a digital facelift. I haven’t tried any of them, but from what I’ve read in CHE and InsideHigherEd especially they sound very foreign to what I understand as a MOOC. The distributed knowledge network vs. hub and spoke model example clarifies what I had been thinking – that your MOOCs use a many-to-many relationship to develop massive interconnectedness while these new things use a one-to-many relationship to reach massive numbers.
    I like that Dawn listed “the development of information literacy skills” as something that attracts people to MOOCs. I don’t think I would have phrased it that way. It gives me food for thought.

    Monday, June 4, 2012 at 12:21 pm | Permalink
  3. Jon Dron wrote:

    This all synchronises and resonates very strongly with my thinking, not just on MOOCs but on LOOCs (large open online courses) and SOOCs (small open online courses) and TOOCs (tiny open online courses). I think the word ‘open’ is the big key here, though this is not to suggest that ‘more’ is not different and highly significant too. The word ‘open’ in neo-MOOCs is down to the fact that they are focused primarily on open *access*. Free as in beer, not free as in speech. What characterises the MOOC approach that you and others have developed is that ‘open’ relates much more strongly to process as well as product. It is about recognition that the ‘instructor’ is a co-traveller on the learning journey who might have a particular sense of direction, but is not a traditional guide on the side. Interestingly, though, the ‘instructor’ may well be a sage on the stage, albeit only one among many competing sages – a sage-bush if you like. I think both you and Stephen have spoken of ‘role modelling’ in this regard. It’s more to do with being a co-learner. I describe myself as a ‘professional learner’ for that reason. The openness is not so much due to the fact that anyone can join in as in the openness of the learning trajectory to perturbation and diverse directions.

    On the subject of resonance and synchronisation…

    I’ve described the synchronisation process in TOOCs, using my own software I wrote in the late 90s that was meant to support self-organisation of groups of learners, as ‘riding the wave’. There are shared values and intentions, but no one knows quite where it is all going and a self-organising collective can move along an interesting trajectory that no *single* person or group of people particularly planned but that definitely goes somewhere. The result of many plans leads to emergent paths that cannot be known in advance. I used to think of this as a bad thing as it is often useful to know the big picture when one is learning something, otherwise it may well lead more circuitously than you might wish to the destination. I’m not so sure that’s a bad thing any more. At least, it can actually be a good thing for ‘higher’ forms of learning, but it might not be the ideal way to learn, say, how to exercise effectively or to fix your drains. For that, the neo-MOOC seems well suited. The structure imposed by MOOCs you have run so far has provided mostly weekly limits on how far you can surf the wave with everyone else before the next wave hits and, I suspect, that granularity is well-suited to more expansive learning, less wonderful for figuring out how to operate your PVR. It might be more interesting though, rather than fitting things within arbitrary time-slots, to deliberately enable the crowd to decide when to move on. Some topics resonate more than others and deserve more attention. It would be hell on earth to organise that though!

    Your thoughts on resonance, as you describe it, seem to resonate with related concepts such as the ZPD, SDT’s notion of competence, the logic behind Bruner’s scaffolding, various similar ideas found in Knowles’s variant of andragogy and Dewey’s thoughts on informal learning, as well as many others. It strikes me that the ‘connect’ part of ‘connectivism’ relies on there being something old to connect with something new, or to connect ideas in new ways. If there’s nothing to connect with, then no connection happens.

    But somewhere, out there, there is something that is going to help you to make that connection.

    It’s a big Internet and there are few learning needs that are not catered for by something or someone if you could but find it. The big problem that started my PhD some 15 years ago and that continues to intrigue me, is how to find the right thing amongst that cornucopia of stuff (Walt Crawford called it a ‘stuff swamp’) to resonate at the right time. It’s not (just)about intentional guidance to the right solution but the creation of opportunities for serendipitous encounters that lead to ‘aha!’ moments. ‘Massive’ can help with that as many eyes make all bugs (knowledge gaps/errors/mismatches and alternative conceptions) shallow. But, without an ecology that supports evolution of structure, when it gets too big the problems get too diffuse and there are too many solutions to select from, which leaves the learner with no more (and possibly less) control over the process than if left alone. Which is where either a guided process (thinking of open source organisations of things like Linux, Apache, etc) or collectives (thinking of Google Search, Amazon Recommends, Slashdot, tag clouds, etc), or some combination of the two (thinking of Wikipedia, GitHub, etc) come in. I could ramble about that for hours. Time to stop!


    Monday, June 4, 2012 at 3:41 pm | Permalink
  4. Frances Bell wrote:

    Interesting article – it’s good to differentiate the flavours of MOOCs and the economic aspects
    However I am curious about your emphasis on the theories that underpin your MOOCs. Clearly, theories did interact with your MOOCs – connectivism, connective knowledge formed a trigger – and tbeories

    Wednesday, June 6, 2012 at 2:17 pm | Permalink
  5. Frances Bell wrote:

    (continued) and explications of theory such as Snowden’s conceptualisation of complexity and Cormier’s of rhizomatic learning. However MOOCs may also be regarded as emergent forms from which theories may be enriched (as part of the connectivist activitues)and rdfined

    Wednesday, June 6, 2012 at 2:24 pm | Permalink
  6. Frances Bell wrote:

    (continued) and refined via these activities and subsequent research projects.So not only are MOOCs emergent forms, but also is the theory that informs them. Apologies for punctuated comments #smartphonechallenges

    Wednesday, June 6, 2012 at 2:40 pm | Permalink
  7. I have spent some time in India, too – and it is pretty amazing the enthusiasm they have for learning. But I also saw how ultimately damaging it was when it all came from a traditional model – so many people (at least that I talked to) were held back in life because they only learned from the traditional model where knowledge was handed to them – they couldn’t think for themselves on many levels or initiate their own learning. So in the short term, many people in India might be happy for access to free knowledge that they might not have had access to before, but in the long run they will end up like so many college graduates there: stuck with a menial job, without any tools to break out the academic rut they are in.

    I also think of this in another way – when I meet people and they find out I work in online education, I always get to hear their opinion on education in general. There are many people that tell me that they prefer to sit in a lecture than take a class online. I hear stories about how traditional “sit and soak” courses are changing people lives all the time. Its not that these people’s stories make the model all of a sudden valid. It is really just a case of them not knowing how stunted their education really was. They are happy with a Big Mac even though they have access to a fancy steak.

    We don’t need to tell people that their criticism of the Big Mac rings hollow just because there are people out there that are happy with it. We need to get people to know that a much better option exists and they need to quit settling for less.

    Tuesday, June 12, 2012 at 6:56 am | Permalink
  8. One of the things that I find so interesting about this post (and indeed in this entire area of discussion and practice) is that it shows how MOOCS, still a relatively new modality / theory of learning, have continued to develop in several different ways while continuing to offer numerous opportunities for future directions. I think this is helpful, as it shows that MOOCS and connectivism may (or may not) be linked in ways that may (or may not) privilege any specific learning theory (at least now). With development and consideration on many different fronts, I find now a time of great opportunity for moving learning options forward, and having seen some of this move since the first CCK08 (something I still struggle to articulate and implement in learning practice), I think it has come a long way since now the discussion and players have expanded. Of course, this may be somewhat problematic from a theoretical perspective, as diversity in learning so often leads to diversity of the frames upon which the learning is designed and developed and supported. I wonder what the next 5 years will bring, especially whose voices (and the unspoken notions of cost and revenue that cannot be far away) will help to continue shaping this space.

    Sunday, June 17, 2012 at 5:24 am | Permalink
  9. gsiemens wrote:

    Hi Jon – thanks for a very thoughtful post. I imagine it’s a bit frustrating sometimes to see others walking paths you’ve walked a decade or two ago :) . I’m quite happy to have you ramble on for hours on this topic :) .

    There is something, I think, to be explored around resonance/synchronicity in relation to MOOCs. Your interest in how people find the right thing in the mess of stuff is quite important. I’ve used the terms sensemaking and wayfinding to describe the process, but I’m not convinced those capture the breadth of what we are looking at.

    Tuesday, June 19, 2012 at 8:21 am | Permalink
  10. gsiemens wrote:

    @Jeffrey – the mooc conversation has taken interesting turns over the past 4 years. It was largely dormant until last fall when things absolutely exploded with Stanford’s AI course. Now that we have universities such as MIT working on improving the software of open online courses, we should see a few exciting years of innovation…

    Tuesday, June 19, 2012 at 8:23 am | Permalink
  11. gsiemens wrote:

    @Matt – good points re: diversity of learning preferences. Personally, I find that my preferences change from topic to topic. I personally don’t mind a good lecture. Often, I prefer it over collaboration. Context, learner confidence, and familiarity with subject matter make a difference. As you noted, at minimum, learners should be familiar with the range of options available.

    Tuesday, June 19, 2012 at 8:26 am | Permalink
  12. gsiemens wrote:

    @Frances – I fully agree. moocs and the theory that informs them are emergent. Coursera has gone so far as to articulate the research that guides their courses…but after a few offerings of their courses, I’m sure they develop new views and theories on how to better design and teach. Both practice and theory are very much in flux.

    Tuesday, June 19, 2012 at 8:27 am | Permalink
  13. gsiemens wrote:

    @Dawn – we haven’t given much thought to credentialing. In the last open course we did Change11, students received credit in two possible graduate programs – Georgia Tech and Athbasca U. This wasn’t open credentialing, however. Students that received credit were already enrolled in those programs.

    Tuesday, June 19, 2012 at 8:29 am | Permalink
  14. gsiemens wrote:

    @Paul – you capture it perfectly! With our moocs, the model has been many to many in terms of interconnectedness. This presents its own problems, particularly when we have clear educational outcomes that we want to achieve and learners always feel that they are missing some part of the conversation somewhere (which is true!)

    Tuesday, June 19, 2012 at 8:31 am | Permalink