by Stephen Downes
May 1, 2014
Visualizing my understanding of connectivism
x28’s new Blog,
May 1, 2014
Connectivism, as I describe it, suggests that there are different types of knowing networks. A human neural network is one. A society (or social network) is another. A neural net computer program is another. So far so good. But Geoff Cain asks the question: "how are these related? The recent criticisms show that it is difficult to reconcile these two." I would point out that there is no particular requirement that they be reconciled. It would be nice for us if they did, but it's not a priori necessary.
I think there are two answers, which I'll call 'the Downes answer' and 'the Siemens answer' because I've talked mostly about one and George has talked mostly about the other. These two are not mutually exclusive, they might both be true, and (probably) elements of both of them are, as they vary mostly from the perspective of point of view as opposed to postulation of a different causal mechanism.
The Siemens answer is multimodal extension. The networks reach out and integrate with each other. Thus, for example, a concept might be contained partially in a human brain but the full extension of the concept might extend beyond the network of neurons to include, and interact with (as part of the same network) extra-neural entities (like computers, other people, and the like).
The Downes answer is pattern recognition (yes yes I know William Gibson wrote a book of that name, and that the concept is widely discussed by others). One network perceives patterns in another network and interprets or recognizes these patterns as something. So, for example, a social network might recognize 'genius' in a person via the presentation of patterns of behaviour by that person that cause responses typical of recognition of genius in society.
May 1, 2014
I'll just pass along these three items from Spark, a Canadian radio program on technology (thanks to Danny for the links):
- HR tech: LinkedIn, job boards, and portfolio sites make it easier than ever to look for work. So why does it seem harder than ever to find a job? Communications professor Ilana Gershon discusses her research into technology-driven change in hiring practices. [FULL POST]
- Personal education: Today's students leave lots of data trails - from demographic information, to how they read and highlight ebooks, and interact online. Researcher George Siemens explains how analyzing data about the way students learn lets schools customize education. [FULL POST]
- We asked the question: If everyone's learning experience is customized, does that mean everyone gets an A? Spark producer Michelle Parise asks the broader Spark community: Should education be personalized so everyone can succeed, or should students be allowed to sink or swim? [FULL POST]
Why Connectivism is a Learning Theory
Brainstorm in Progress,
May 1, 2014
Yeah, this is kind of how I see it too: "Most of the criticisms I have read of Connectivism boil down to the new theory is not like the old theories." This is asserted in by Geoff Cain as he explains why the 'incompleteness' (David Wiley's concern) of connectivism doesn't mean it's not a theory. And this is useful: Cain writes, "for me, a theory must
- account for current theories (either through refutation or inclusion)? A theory shouldn't just account for a given phenomena, it should do so in some measurably better way (more complete, elegant, etc.).
- sufficiently explain where we are now.
- make predictions. Any theory that can't predict anything is basically a conjecture at best.
- be subject to testing. Here I would emphasize that the theory should change what we do based on experiment and empirical data.
In my experience, Connectivism has met those four conditions." Again - it doesn't matter to me whether or not connectivism is a theory, but these practical concerns - for example, whether ti explains, whether it predicts - are important.
This newsletter is sent only at the request of subscribers. If you would like to unsubscribe,
Know a friend who might enjoy this newsletter? Feel free to forward OLDaily to your colleagues. If you received this issue from a friend and would like a free subscription of your own,
you can join our mailing list. Click here to subscribe.