Archive for November, 2007

Only Connect…

Thursday, November 29th, 2007

Only connect! That was the whole of her sermon. Only connect the prose and the passion, and both will be exalted, and human love will be seen at its height. Live in fragments no longer.
E.M. Forster, 1910

Networks and connections are deceptive. It would not appear that the formation of a simple connection has the capacity to reverberate across a network, rewriting both form and function. And yet it does. Latent semantic analysis (.pdf) suggests that “people acquire much more knowledge than appears to be available in experience”, or put another way, the addition of a new element of information or knowledge yields a greater impact than what exists within the information or knowledge itself. It appears that the new node creates a ripple effect altering the meaning of other nodes within the network. The above cited paper suggests this is due to an inductive property of learning itself, but I’m more inclined to see it as a function of network formation. A new node of information results in new connections, which in turn results in new shape of knowledge, and thereby our own understanding. I tackled this briefly on my Knowing Knowledge blog. The core assertion – which was criticized (see comments in the post) – is that knowledge is a function of connections and understanding is the emergent shape of the network.
But what is the nature of the network? Lately I’ve been detailing network formation as occurring on numerous levels:

  1. Neural level – the formation of neural connections as new stimuli, input, and experiences shape the physical development of the brain (Bechtel & Abrahamsen, 1991). Research arising in the study of memory suggests a similar network-like function (Lynne Reder). Knowledge and learning are not held at any particular point in the human brain. Instead, they are distributed across numerous sections.
  2. Conceptual level – within a discipline of field of knowledge. Key concepts of a field – those which are foundational to the knowledge structure of the discipline – are networked (Joseph Novak) in structure. Novice learners seeking to develop advanced understanding of a discipline do so through the formation of conceptual maps reflective of those held by experts within the field.
  3. External. The formation of networks has been significantly aided through the development of participatory web technologies. Blogs, wikis, social bookmarking, and social networking sites, raise the capacity of individuals to connect with each other, with experts, and with content. Understanding, in a networked sense, is an emergent element related to the shape and structure of the learner’s personal network of information. The development of RSS as a means of aggregating information and mashups as a means of combining information in various contexts, contributes to the external formation of networks which in turn assist learners in forming accurate conceptual relationships within the field.

I suspect many of the attributes of network structure and behaviour (expressed rather nicely by Stephen Downes in various recent posts: Personal Network Effect and How the net works) in networks occurs in all three levels (neural, conceptual, external). What a node is, however, will differ in each instance. A node in a neural network is a neuron. In a conceptual network, a node is an idea or collection of ideas (networks can serve as nodes when connected to larger network structures). In an external network, a node is a person, an information source, or similar entity capable of accepting connections, and thereby participating in a network.

All the knowledge is in the connections
David Rumelhart

Applying this threefold structure to knowledge requires that we see distinctions between knowledge at different levels. For example, access to a data base is not knowledge. Searching Google or Wikipedia is not knowledge (that would be information). When, however, I am related or connected to the information in a certain way, that is, it is connected to and enriches my larger conceptual and neural network, then it is knowledge.
This raises a difficult question: how do we validate or determine something to be knowledge? Paul Boghossian, in his slim volume Fear of Knowledge, relies on Plato’s traditional knowledge definition – justified true belief – to argue for the need for “privileged ways of seeing the world” through approaches such as the scientific method. Not all knowledge is subjective or strictly determined by the observer. Some forms of knowledge (such as the number of moons surrounding Jupiter) can be correct or incorrect. I don’t want to get into a discussion here about the way we determine something to be a moon or how Pluto sometimes is/isn’t a planet. At this point, I’m simply suggesting that knowledge requires some type of relation (or embodiment according to Andy Clark) between an object, the environment in which it exists (context) and the agent (this is starting to sound eerily like Gibson’s affordance theory (.pdf)). In most instances, this knowledge is shaped, formed, and connected/created/grown by the agents doing the connecting. In some instances, the method of validation determines whether the knowledge is accurate (and thereby knowledge) or inaccurate. Put another way: not all knowledge is purely subjective.
Of what value is this discussion? My question rests in how learning occurs in a networked world. We’ve had many taxonomies (Fink, Wiggins, Bloom) that detail knowledge and learning in gradients, levels, and stages. Perhaps we have been conditioned to expect something as complex as learning to require a complex process or explanation. But what if forming a connection is enough? What if learning is as simple (for the purposes of most educators) of getting learners to form diverse networks representing divergent viewpoints and cultures? What if exposing learners to rich networks of content and conversation is sufficient? The learners will, after all, begin to “play”, make sense, interact, and grow in knowledge and understanding (I find the concept of wayfinding to be of value here).
A second component requires consideration: the depth and quality of learning in a network. Will Richardson and I recently discussed weak tie and strong tie learning. Sometimes our learning is about forming networks and connections at a basic Level (often with the intent of creating awareness of related fields which may impact our own are of expertise). This is weak tie learning. Learning in this instance is defined by creating connections to peripheral fields or simply interacting briefly with new information and then moving on. Strong tie learning, on the other hand, involves more time, effort, expertise, and sustained focus. Geetha Narayanan defines this as slow learning. And I think it’s an important concept. Sometimes we want our learners to gain an awareness of factors, other times we want them to interact with elements in order to understand deeply. Sometimes we want knowledge for foundation building, other times we want knowledge for action or reform. Different knowledge-network connections, defined by strength of the tie, result in different depth of learning. Perhaps “only connect” is still (almost 100 years later) a sufficient motto. Perhaps the elimination of barriers to connection is the greatest systemic challenges our institutions face. And the role of teaching is one of guiding, directing, and curating the quality of networks learners are forming.
It seems too simple. I desire a more complex explanation. But, perhaps, just connecting is sufficient.

More than anything else, being an educated person means being able to see connections so as to be able to make sense of the world and act within it in creative ways. All of the other qualities that I’ve just described—listening, reading, writing, talking, puzzle-solving, seeing the world through others’ eyes, empowering others, leading—every last one of these things is finally about connecting.
William Cronon

A few recent presentations

Thursday, November 15th, 2007

I’ve posted a few presentations on connectivism (delivered to University of Alaska Fairbanks). Slideshare version available here:
Connectivism 101
Practical Connectivism
Organizational Impact of Networked Learning