Connectivism

I frequently emphasize the substrates at which learning and knowledge are connected or networked. As more attention is paid to learning networks and personal learning, it’s important to highlight that most of the discussion is focused on the social/external substrate, ignoring other dimensions of networkedness (I’m sure that’s a word).

By quick review:

1. Neuronal – brains don’t hold knowledge in chunks – it’s networked. A simple task, such as picking up a pencil, requires numerous areas of the brain to harmonize their distributed activity (sometimes referred to as the “binding problem”) in order to produce the intended action. Recognizing a human face is an astonishingly complex distributed neural activity – an image of a face doesn’t exist in our brains. Instead, different regions of the brain contributed to producing recognition. Olaf Sporns has explored the similarity between some network attributes of the neocortex and other scale-free networks.

2. Conceptual – connections generate meanings. When two or more concepts are brought into some type of relationship, they produce something different than their individual attributes would suggest. Conceptual blending attempts to describe what’s involved in the process of bringing concepts in relation to each other. Burkes’ Knowledge Web is similarly based on trying to find how knowledge is connected/related. As does Danny Hillis’ article Aristotle: The Knowledge Web. Or consider a tool like Brainscanr that attempts to detail relationships between concepts in psychology. We are constantly forming and blending concepts. When we are involved in formal learning, we are more conscious of the process as we’re bringing together our life experiences and current understanding of a topic with new information provided by a course or program of study.

3. Social and technological networks – we live these daily and tools like Facebook and Twitter have made these more explicit. Publications from mathematicians and physicians over the last decade have increased attention on networks (Barabási, for example). However, sociologists have been playing in the domain of networks long before the current hype drove networks into popular society. Barry Wellman, Mark Granovetter, and Paul Lazerfeld lay much of the foundation for what is now being “discovered” about social networks. Researchers are beginning to take a multi-disciplinary approach to networks, realizing that network attributes exists in food chains, transportation systems, etc. Basically, networks underpin life and human existence. The internet, web, and now social media raise the profile of networks because we now experience them daily. When directed toward learning, networks (web, citations, social, etc) are inescapable. As human knowledge becomes more explicit – i.e. stored in a database, waiting analysis – analytics becomes increasingly important in order to understand complexity. The discovery of the corona virus (SARS) was accomplished in a period of a month – an extremely short period of time considering the complexity involved. This was enabled by researchers connecting to each other and sharing information. Understanding how and why people and information connect is a key task of analytics (have I mentioned TEKRI is organizing a conference on Learning and Knowledge Analytics?). Knowledge in any moderately complex task or activity is networked (building a plane, designing a road system, printing a book).

I’m working on an article on the discussion around connectivism over the last six years. A key problem that arises with criticism about connectivism, and I think the efforts of proponents to explain it, centre on dramatically different views of ontology, epistemology, and language. In some areas – such as when people ask “how is this different from social constructivism”- it appears that some view differences as trivial. In other areas – such as when people begin to contrast distributed knowledge and social learning networks in relation to the existing education system – it appears that differences are enormous.

I’ve been grappling with a thought experiment that might help to clarify differences and provide a platform with which to think about learning and knowledge. Zombies and other planets are well explored thought-experiment models, likely because they allow the thinker to jettison some of the assumptions that are inherent when thinking about entities that have real world presence. By moving to other planets, or stripping the cognitive capacity of zombies, we are better able to isolate the phenomenon that we want to consider. Here is my current version of a connectivist thought experiment – I appreciate any feedback, questions, disagreements, withering critiques:

We travel to a different planet (Planet Connecton). The ecosystem is similar to what we experience on earth, so we are free to move about and explore. During our exploration, we encounter a human-like species. As we observe their interactions with others, we quickly notice a distinct difference: as each “person” communicates, a cloud appears above their heads. In this cloud we see explicitly their knowledge. The knowledge we observe is networked, so we see real-time changes to their knowledge patterns as they read, learn, and as they interact with others. When they express an idea to a fellow Connectite, we can observe how their thoughts begin to form, which areas of expertise they draw from, which contrary ideas they briefly entertain in an attempt to communicate…but then decide to dismiss. Even more fascinating, we are able to see how a new concept that they learn is broken down into a neural (biological) network. Different conceptual and neural networks are constantly activated and suppressed depending on the context or situation of learning/interaction. We also see how clouds between individuals connect. For example, if one Connectite tries to solve a problem, we observe cross-cloud connections as different levels of skill and knowledge are required for different tasks. Even the simplest task requires the activation of connections to the clouds of others and even objects. These objects can be seen as equivalent to our cognitive objects – books, papers, Google.

We observe one individual reading a book on biology (at roughly the equivalent of an earth-based under graduate degree). As she (sure, gender still exists here) reads about DNA, we begin to see isolated nodes – the fragments of knowledge – appear in the network within her thought cloud. Some of these nodes are quickly connected to existing conceptual patterns. Some nodes, those that don’t readily “cohere” or “resonate” with the existing knowledge of the learner, remain isolated or at best have simple, weak connections. Were we to observe this learner for a period of time, we could see various nodes cohering and strengthening in prominence and, in other cases, weakening and fading, as the learner moves between different subject areas or connects with other learners. Knowledge growth is constant in all domains of her personal and professional life.

As visitors to this planet, we are able to observe every aspect of knowledge and learning through the formation of connections – at the neural, social/object/external, and conceptual substrates. We see the interplay of a social interaction that influences new neural connections, which in turn update and readjust the conceptual understanding of individuals. Surprisingly, the rich, advanced, and varied knowledge of this species can be thoroughly explained through the connection clouds.

We’re a bit surprised because what we’ve learned about, well, learning and knowledge, on earth is so much more complex – theories of intricate details about motivation, images, emotions, and so on. On seeing the Connectites interact, the simplicity of connection-based learning and knowledge push many of our earth-based theories to the outer edges of relevance. Instead of starting with learning at different stages (institution, individual, organization) or seeing numerous views of learning (social, constructivist, cognitivist, situative), we break from our insistence of complicated explanations to complex phenomenon and collapse down to connections as the basic unit for understanding knowledge and the process of learning. The elements that impact connection-forming in the process of learning – such as emotions, pervious experience, and motivation – are not nodes within the connection clouds. Instead, they are enablers or influencing elements that impact whether or not a connection will form or the way in which that connection will resonate with the rest of the network.

Question:

What more do we need for a theory of learning and knowledge than what we observed in our interactions with people on Connecton? What can’t we explain with this model?

Secondly, what questions and reservations do you have about this model?