At  an event the other evening, showing various career technology tools, someone  said something that I thought was just wrong. I asked afterwards, and then  explained why I thought it was wrong. The response was “well, there can be different ways to go about it”. And frankly, there really can’t.  Think for yourself about why I might say so, and then let me show you why.

The trigger was a  design program talking about their design courses. And the representative was saying that once a learner had created a project, it was shown to everybody. Which sounds good, since ‘sharing is caring’, or at least it’s a good example of working out loud. And, in general, this is a good idea. But I think it’s not in learning.

In brainstorming (e.g. informal learning), we know that sharing  before others have had their  chance to think, it can color their output. This limits the exploration of the total possible space of opportunities that would come from a diverse team. Hearing another response likely will limit  that  spaces that might get explored. Instead, the goal is to diverge before converging.

And so, too, in learning. I’ve argued for assignment submission systems that only allow you to see the other submissions  once you’ve submitted your own. Until you’ve struggled yourself with the challenge, you won’t  get the most out of seeing how others have solved the situation.

If you immediately share the first submission, it may affect those who aren’t that far along yet.  Some may even end up holding off to see what others do! This undermines the integrity of the assignment. One explanation that was given was to provide guidance to others, but that, to me, is the role of the assignment specification.

There is, however, real value in seeing the other submissions once you’ve completed yours. Seeing other approaches helps broaden the understanding. Better yet is to have discussion  on them, as when  critiquing others (constructively) you internalize the monitoring. This discussion  also  provides the opportunity to experiment with working out loud that eventually develops good working habits.

(I’ve similarly argued, by the way, that ‘rollover’ questions  -where the answer is shown once you move your pointer over the question- don’t lead to any meaningful learning. If you haven’t made the mental effort to  commit to a response, it won’t stick as well.)

So I believe that, if you’re developing people’s ability to  do, you have a responsibility to do so in the most advantageous way. That includes making effort to use the best approach to sharing assignments. I was surprised (and dismayed) to see someone arguing to the contrary! I implore you to do the details on the approaches you work, for your learners’, and the learning’s, sake.

I’m going a bit out on a limb here, with a somewhat heretical statement: what you learn is more important than how you learn!  (You could say pedagogy supersedes curricula, but that’s just being pedantic. ;)  And I’m pushing the boundaries of the concept a bit, but I think it’s worth floating as an idea. It’s meta-learning, of course, learning how to learn!   The important point is to focus on what’s being developed.  And I mean this at two levels.

This was triggered by seeing two separate announcements of new learning opportunities.  Both are focused on current skills, so both are focusing on advanced curricula, things that are modern. While the pedagogy of one isn’t obvious (though claimed to be very practical), the other clearly touts the ways in which the learning happens. And it’s good.

So the pedagogy is very hands on. In fact, it’s an activity-based curricula (in my terms), in that you progress by completing assignments very closely tied to what you’ll do on the job. There are content resources available (e.g. expert videos) and instructor feedback, all set in a story.  And this is better than a content-based curricula, so this pedagogy is really very apt for preparing people to do jobs.  In fact, they are currently applying it across three different roles that they have determined  are necessary.

But if you listen to the longer version  (video) of my activity-based learning curricula story, you’ll see I carry the pedagogy forward. I talk about handing over responsibility to the learner, gradually, to take responsibility for the activities, content, product, and reflection.  This is important for learners to start becoming self-improving learners.  The point is to develop their ability to do  meta-learning.

To do so, by the way, requires that you make your pedagogy visible  for the choices that you made, and why.  Learners, to adopt their own pedagogy, need to see  a pedagogy. If you narrate your pedagogy, that is document your alternatives and rationales of choices, they can actually understand more about the learning process itself.

And this, to me, is the essence of the claim. If you start a learning process about  something, and then hand off responsibility for the learning, while making clear the choices that led there, learners become self-learners. The courses that are designed in the above two cases  will, of necessity, change. And graduates from those courses might be out of date before long,  unless they’ve learned  how to stay current. Unless they’ve learned meta-learning.  That can be added in, and it may be implicit, but I’ll suggest that learning to learn is a more valuable long-term outcome than the immediate employability.

So that’s my claim: in the long term, the learner (and society) will be better off if the learner can learn to self-improve.  It’s not an immediate claim or benefit, but it can be wrapped around something that  is of immediate benefit.  It’s the ‘secret sauce’ that organizations could be adding in, whether internally or in their offerings. What surprises me is how seldom I see this approach taken, or even discussed.

So yesterday I talked about the value of diagrams, but I thought I’d add a bit about the process of actually  creating  diagrams. Naturally,  I created a diagram about it.

I created this diagram for a session  I ran on diagramming a number of years ago.  In that session I talked about our cognitive architecture, why we need models, how diagrams work, properties and design issues, and more. At the end, I proposed a potential process for it.  It captures an ideal picture of how diagramming could work.

So, you need to know the elements of the model you want to relate, identify the relationships, and any dimensions that characterize differences between the elements. Then you have to choose how you’re going to represent them: shape, color, weight, font, and more.

With your elements, you can then place them, connect them, and add the visual coding.  Then, of course, you tune.

This is an ideal process, but in reality it’s much more flexible, at least when I’m creating a model as a way of understanding.  I typically iterate, creating placeholder elements, and moving them around until I think I have the right ones.  Then I go about connecting them to make sure I have the relationships right. Then I work on adding dimensions, and colors, and aligning them, and grouping them, and… Except that I might add some elements, then group, or connect, then add more, and…  it’s a very iterative  process.  It’s a creative process that involves lots of experimentation, revision, and more.

Sometimes, I even use a diagram and then realize it’s not working and revise it. So, for instance, I blogged about a  representation of social process.  I got some feedback that it wasn’t very clear.  So, I made a second stab at it, and I think it worked better. Certainly, I continue to use it without complaints.

And I’m the first to admit that my diagrams may not look as good as the ones that professional graphic designers could create, but they’re good enough (and OmniGraffle does a good job of making it easy for me to make them up to a standard I think is at least acceptable and useful; it’s probably overkill but I’ve stuck with it for years now).  And that’s the point.  If they help you think better, it’s good enough. If it helps you communicate effectively with someone else, even better.  Diagrams are cognitive tools, offloading conceptual complexity to graphic relationships and visual processing.  And with the complex problems we increasingly face, I reckon the more tools the better.

One of the things that I feel is a really useful tool in my ongoing learning, in my ‘making sense of the world’ is diagramming.  I find diagrams to be  a really powerful way to understand not just elements, but relationships.  And yet it doesn’t feel like diagramming gets enough respect.  So I want to make a  case for the diagram.

Language is good. Our brains have evolved to use it. But it has trouble communicating complex relationships.  For an example, once I wrote this:

They found that while subjects would rate the analogies, from best to worst, as literally similar, true analogy, mere appearance, and false analogy, their recall for stories, from best to worst, was literally similar, mere appearance, true analogy, and false analogy.

Try discerning the important difference!  My PhD advisor kindly pointed out that actually parsing this was hard, and recommended a diagram instead. Here’s a rendition of what resulted:

In this case it’s much easier to see how the two differed.  (If you want to find out what’s important  in the diagram, I’m happy to talk about analogical reasoning for as long as you can stand it! ;)

The point I’m making is that there are times when diagrams are very useful for communicating.  And, if you’ve followed this blog for a fair amount of time, you’ve seen I use diagrams a lot. I use them to think ‘out loud’, and I think it’s important.  As Larkin & Simon argued in their Cognitive Science article, Why a Diagram is (Sometimes) Worth Ten  Thousand Words,  diagrams let us map conceptual relationships to spatial ones. And so if I want to understand the conceptual relationships, I start laying out spatially, and adjust until they make sense to me.

And my concern is that we aren’t using this powerful visual tool enough.  Sketchnotes are really nice ways to capture presentations, and depending on the skill of the noter, they may communicate it all, or help recall if you’ve seen it. Similarly, my mindmaps of keynotes capture the flow of the discussion and the relationships (at least as I parsed it), but may only make sense if you heard the talk.

But representing things with diagrams is not only a personal thinking tool, it can be a powerful way to communicate concepts, and that’s an important component of a good learning experience design, providing a conceptual model to guide performance.

So I’m surprised we don’t talk about diagrams more. It may seem hard (certainly trying to create an infographic is harder than it seems, from my experience ;), but there’s some systematicity to it. There are principles, and types of diagrams, and more to explore.  And tools that make it easier (though even Powerpoint or Keynote can be used to make diagrams).  Diagrams aren’t the only visuals that help (c.f. graphs and tables), but they’re an important tool in your thinking toolbox.  I encourage you, as part of your meta-learning toolkit, to play around and get your mind around diagrams. Your thinking, and your learning design, can be better as a consequence.