Monday, October 3, 2011

Complexity, Cognitive Load and MOOCs

I've just read two posts by Michael Gallagher Complexity, self-organization, and #Change11: reactions to Siemen's presentation(1) and Multiple interfaces, cognitive load and learning design: My appartment in Seoul (2). In the former Gallagher discusses the pattern making process we use to make sense from the chaotic stuff of life (and learning in a MOOC).  In the latter he takes us on a fanciful tour of his high tech Korean apartment,  a device to explore interface design and cognitive load.

Complexity and cognitive load - two of my favorite topics.

I love complexity. I love to slosh about in it.  But it isn't easy.  My training compels me to approach problems with a linear logic.  I learned to create a lovely 3x3x3 outline (three points on each of  three levels I, 1, a. b. c. ) in 6th grade.  I loved the rules of grammar.  How shocked I was to learn that poetry was not always composed in iambic pentameter. I learned well.

When I fight the compulsion I am thrilled by the meandering journey from hyperlink to hyperlink, the "collision of actors, agents, feedback, waste, noise, and then, ideally, pattern, understanding." (Gallagher, 1) 

I cannot even set goals until I muck about in the complexity of a topic.  I absorb a shallow Gestalt of a problem.  I see the problem through lenses I didn't know existed.  My interests are assaulted and piqued. Then I am able to dig deeply into an aspect of the problem, to which I think I have a chance of contributing.

When I am able the resist the temptation to take a prescribed set of steps toward a goal,  I remember that I don't fully understand the learning, sense making process. One does not get there from here simply by following these steps.  The collision of ideas and patterns is currently unpredictable.

Cognitive load.
Cognitive load: This term, coined by John Sweller, explains the ability or inability of a novice to process information - based on the mental demands required by germane, intrinsic and extraneous load. It attempts to reduce complexity so that learners do not apply all their cognitive resources to achieving a goal or making a "means-end analysis."

I have long balked at the idea of cognitive load. Even though I recognize it all around me and employ techniques both for coping with it and designing instruction to control for it.

Cognitive load theory is like iambic pentameter (well maybe not, but let's see how far I can go with the analogy).  It is a perfectly reasonable approach to instructional design. (Ok, so far. Iambic pentameter is a reasonable approach to poetry.)  Cognitive load theory works. You want someone to be able to tell you the process involved in solving a three step math problem with a known solution.  Providing a worked example reduces the load involved in solving the problem and enables the learner to see and repeat the solution.  (If you want to produce a catchy verse that can be quickly learned and repeated to spread a bawdy joke or the news of the kings heroism in battle, or to get a rapping gig, iambic pentameter is for you.)  But it isn't the only approach teaching or learning (or poetry).

I'm not sure about this yet but I think the strategies that Sweller offers have led instructional designers and educational policy makers to focus on a means-end analysis of the "problem" of education.  Means-end analyses are not bad analyses for experts, but for novices (if I understand Sweller) they lead to a narrow expedient approach to achieving a goal that misses the important aspects of learning. Cognitive load reduction leads to complexity reduction which leads to the ability to move quickly from one topic to the next, which leads to efficient memorization of factual information, with leads to high achievement on standardized test.  It does not lead to deep understanding or the ability to handle complexity in life.

Novices can develop without narrowly defining their responsibilities, without controlling for "extraneous" cognitive load (that's where the sparks of clashing ideas happen).  Watch any pre-schooler (and I mean any person who has not been to school).  It is messy.  It is hard to assess, because the assessor's goals and accomplishments the learner makes are often not aligned.   Deep understanding and understanding within complexity takes time, it requires the opportunity to observe, to see the decisions and outcomes of experts and developing novices.  It requires the opportunity to act and fail.

My experience being a novice in the complex learning environment of a MOOC has been messy.  In my first MOOC, PLENK2010, I often felt lost.  If only they would tell me what to do (whine, whine, whine).  They were not providing me the scaffolding I needed. (Sorry George and Stephen, they is you.)  I loved it.  I learned a lot.  But I wasn't sure what I was learning because the goal was not clear, and that was unnerving.  This time around, in my second MOOC, #Change11, I feel like my limitations have fallen away.  I don't feel compelled to do it all.  As I've said elsewhere, I don't feel compelled to do instructor promoted reading.  I'm still a novice MOOCer, but a developing one.  Complexity, collisions among actors, agents, feedback, waste, noise. Bring it on. My 3D pattern-finding glasses are on.

1 comment:

Michael Sean Gallagher said...

Good post! I think complexity and cognitive load go hand in hand, and in fact we are trying to stimulate both in a MOOC (or any advanced learning structure), but the trick (and you mention this) is to reduce the extraneous cognitive load, all that brian power necessary to learn a new system and then learn the learning objectives at hand. Good design will reduce that extraneous load (and MOOCs toy with extraneous cognitive load a lot by making the learning system the sole property of the learner) and focus on germaine cognitive load. That is the cognitive activity that attempts to make sense of complexity and that is the real power of this format. Great post and keep them coming!