What is Connectivism trying to be?

After a little reflection, and reading a little more from the forums/blogs etc., I thought I’d ask what Connectivism is trying to be rather than the more obvious ‘What is Connectivism’ – but there is a reason for this, and I think it might help me get to heart of my issues with Connectivism. Apologies for any incoherent rambling below…

First up, the usual caveat: I haven’t completed this weeks required reading. In fact, I can’t access Stephen Downes’ website at the moment (has it crashed?) – and so can’t view the video from last week’s chat between Stephen and George Siemens where they discuss “What is Connectivism” – should be here: http://connect.downes.ca/cgi-bin/page.cgi?post=44447

Here are two things that Connectivism could be:

  • A general theory of learning and knowledge in networks. This would not need to claim that all learning occurs in networks – but seek to understand and explain how learning does occur in networks, and use this understanding to develop improved pedagogies/andragogies.
  • A grand unified theory of learning in homo-sapiens (incorporating what McLuhan calls ‘the extensions of man’ – i.e. technologies from writing to smartphones and everything in between).

Other people have picked out some quotes from Stephen and George which indicate that they are indeed trying to develop a general theory of learning in networks – e.g.:

A lot of discussions about networks that occur in society fall under the heading of discussions of social networks, so you get people like Duncan Watts and others and you’re looking at stuff like scale-free networks and all of that. And you’re quite right, connectionist literature focuses on what they call neural networks or simulations of neural networks and it’s focused on things like the brain and perception and recognition by computers and so on. Part of my position is that the two phenomena are one and the same, that what we’re seeing at the micro level in the brain is the same kind of thing that we’re seeing in society, that we’re seeing in different ways in different places in society. The same principles that govern crickets interacting with each other govern bloggers citing and quoting each other, govern the development of river systems and trees – those principles are also the principles that govern things like human brains and computer networks set up in certain ways.

Stephen Downes, 2007 – Quote picked out by Roel on the CCK09 forums.

This seems quite reasonable in many ways – networks are indeed a distinct class of phenomenon that are open to study irrespective of what the networks are made up of. Which is why, to pick an example I’m familiar with, it is possible to model language change and evolution using mathematical and simulation models that were originally convieved to model physical or biological phenomena. My own experience of this feels like half a life-time ago, but is encapsulated in my PhD thesis, “Computer Models of the Evolution of Language and Languages”, defended way back in September 2003.

The ability to study networks independent of what the networks are of is important in complexity science, and has led to books like ‘Weak Links‘ by Peter Csermely, which has the subtitle: Stabilizers of Complex Systems from Proteins to Social Networks. Proteins, neurons, social networks – all very different, but at some level all the same at a very abstract level.

Where I have problems is (as identified in my last post) is the ‘bit in the middle’ – the conceptual layer of Connectivism. At the bottom layer we have neural networks, at the top we have social and technological networks. In the middle, we learn by developing networks of concepts. As George said in reply to my previous post:

I learned by encountering concepts and connecting them in different ways. In fact, I’d go so far as to state that our understanding is related to how we have connected concepts. An expert has a more nuanced conceptual network – she understands how the introduction of a new element influences what already exists. A psychologist has an easier time learning a new theory of motivation than does a farmer. Why? The existing state of understanding (patterns of connections between concepts/ideas) is more developed in the psychologist in relation to psychological concepts. The farmer, in contrast, will better understand new fertilizers or the impact of weather on particular crops (where to irrigate and when).

And concepts are non-symbolic patterns:

concepts are not words and that’s why it’s not going to be a rule-based system; they are patterns in a network and that like the human brain or a network like society as a whole. In these networks, there’s no specific place where the concept is located. The concept is distributed as a set of connections across the same network and other concepts are embedded in the same network; they form parts of each other and they affect each other. (Stephen Downes, 2007)

It’s here at the level of concepts that Connectivism extends itself from simply trying to describe how learning occurs in networks into a new grand-unified theory of learning – by asking us to adopt this theory in place of the like of Constructivism. Without this middle layer of concepts, Connectivism would perhaps sit more easily alongside a wide range of other learning theories and pedagogies. As it is, trying to explain human learning simply in terms of networks of concepts seems tricky – again in his comments George admits:

The conceptual level then refers to “learning”…and, unfortunately, this is an area that is still rather underdeveloped…

I think that at this middle level, when concepts and networks are invoked we are actually dealing with a metaphor for learning – based on connectionist principles, agreed, but a metaphor nonetheless. And a very vague one at that. Connections are made or are not:

Connections form naturally, through a process of association, and are not ‘constructed’ through some sort of intentional action (Downes 2007)

So how do I use that to develop a pedagogy, or improve my teaching? I model some concepts for students and the connections either form or they don’t? George Siemens’ prepared a What is Connectivism googledoc that emphasises the challenge in developing a practical pedagogy from Connectivism. In a table, a number of learning theories are compared according to a range of factors. From this we see the following:

  • Influencing factors in constructivism: Engagement, participation, social, cultural
  • Influencing factors in connectivism: Diversity of network, strength of ties, context of occurrence

If I see a student in my class who is playing web-games instead of attempting the current activity I might consider that the student is not engaging with the class and possibly chat with that student and try to develop some strategy to improve that student’s engagement. This clearly is a constructivist approach.

What practical pedagogy does connectivism offer here? The idea that a student might have some specific learning intentions appears to be rejected by connectivism, and the question of engagement does not appear to fit with connectivist theory.

And this is, I think, a symptom of attempting to use general theories about networks and turn them into a grand-unified theory of learning. For me, Connectivism’s biggest issue is scope.

Enough for now. Back to work…

Edit: The table in the GoogleDoc mentioned above has a final row titled “Types of learning best explained” – this implies that Connectivism (at least as seen by George) is not trying to be a grand-unifying learning theory – but one that sits easily alongside others. In which case, why does it need the bit in the middle? After all learning involving “Complex learning, rapid changing core, diverse knowledge sources” could be described without reference to the neural level at all, and without insistence on the networks-of-concepts metaphor.

6 thoughts on “What is Connectivism trying to be?

  1. Martin Aa

    Very interesting post. I completely agree that the scope is a major concern.

    Grand unified theories tend to show up as pseudoscience, I think Connectivism would be better off with a more narrow scope.

    Reply
  2. Stephen Downes

    Yes, the website is down, the victim of a very badly timed (for me) hack. The completely rebuilt site should be back soon (thank goodness for multiple backups).

    I think we need that bit in the middle, specifically “concepts are not words and that’s why it’s not going to be a rule-based system; they are patterns in a network and that like the human brain or a network like society as a whole.”

    Whether this means a “grand-unified theory of learning” is an open question. What it does mean is that I think that this is how language (and learning) in fact works. Whether this means you have to abandon constructivism depends on the constructivists – I won’t presume to theorize for them – but I will say the sort of things that are said in constructivism that are metaphors, and that they’re misleading metaphors in some important ways.

    Reply
  3. Daniel Livingstone

    Thanks Stephen for the response. Good luck getting your site back up – I had a site hacked a few months ago, luckily I was able to pass the recovery work to someone else…

    I will extend this in a future post I think – for a whole range of disciplines and areas emergent phenomenon exist at multiple layers or levels, while productive theories tend to concentrate on only a single (or perhaps two closely related) levels of phenomena. Medicine as applied/emergent biology, biology as emergent chemistry, chemistry as emergent physics… that kind of idea.

    Alternatively, socio-linguistics (and historical linguistics) as emergent systems arising from networks of individual speakers (psycho-linguistics) who learn language during their life – constrained by evolved neural mechanisms (evolutionary and cognitive neuro-science).

    It is possible to derive theories at one level with a large degree of independence to theories at lower/higher levels. And while we can see that socio-linguistics is clearly emergent, without reference to specific phenomena that only exist at the social level the ability to understand and explain language change in society becomes quite constrained.

    As a heads up, when I get the chance I’ll even propose three forms of Connectivism:
    exo-Connectivism: Connectivism theories principly concerned with the ‘out there’; Inter-networking people and technology
    endo-Connectivism: Connectivism theories principly concerned with theories of networks relating to concept formation and the internal processes of learning

    and of course:
    holo-Connectivism: Connectivism theories that attempt to combine exo- and endo- forms into a unified theory of learning.

    Why bother with this?
    Because I have a feeling that for many people exo-connectivism may be a useful way to consider learners and learning in todays networked world – even where the endo-connectivist ideas have failed to convince.

    Reply
  4. Bill Kerr

    You might like this snippet from Marvin Minsky which for me puts it into perspective in a way that has both humour and insight:
    “Mathematician: It is always best to express things with logic
    Connectionist: No, logic is far too inflexible to represent commonsense knowledge. Instead, you ought to use Connectionist Networks
    Linguist: No, because Connectionist Nets are even more rigid. They represent things in numerical ways that are hard to convert to useful abstractions. Instead, why not simply use everyday language – with its unrivaled expressiveness
    Conceptualist: No, language is much too ambiguous. You should use Semantic Networks instead – in which ideas get connected by definite concepts!
    Statistician: Those linkages are too definite and don’t express the uncertainties we face, so you need to use probabilities
    Mathematician: All such informal schemes are so unconstrained that they can be self contradictory. Only logic can ensure us against those circular inconsistencies”
    - The Emotion Machine, pp. 295-6

    Minsky concludes that “it makes no sense to seek a single best way to represent knowledge”

    Reply
  5. Bibiana

    hello Daniel,
    Thanks for verbalizing so clearly some of my own concerns about Connectivism. I do think scope might be one of the problems and, definitely for me, I feel uncomfortable with the idea that learners volition -me in this course- has nothing to do with what is happening.
    Good post and good reflections!

    Reply

Leave a Reply