Learnlets
Clark Quinn's Learnings about Learning

Two recent events converged to spark some new thinking.

First, I had the pleasure of meeting up with Dave Gray, who I’d first met in Abu Dhabi where we both were presenting at a conference. Dave’s an interesting guy; he started XPlane as a firm to deliver meaningful graphics (which was recently bought by Dachis Group, and he’s recently been lead author on the book Gamestorming.

What Gamestorming is, I found out, is a really nice way to frame some common activities that help facilitate creative thinking.  Dave’s all over creativity, and took the intersection of game rules and structured activities to facilitate innovative thinking, and came up with a model that guides thinking about social interaction to optimize useful outcomes.  The approach incorporates, on a quick survey, a lot of techniques to overcome our cognitive limitations. I really like his approach to provide an underlying rationale about why activities that follow the structure implicitly address our cognitive limitations and are highly effective at getting individuals to contribute to some emergent outcomes.

I also happened to have a conversation with a lady who has been creating some local salons, particular get-togethers that have a structured approach to interaction (I’ve attended another such).  Hers was based upon biasing the conversation to the creative side, a very intriguing approach. Not only was she thinking of leveraging this for tech topics, but she was also thinking about leveraging new technologies, e.g., a Second Life Salon.

Which got me thinking that there were some relationships between Dave’s Gamestorming approach and the salons . I wouldn’t be surprised to find salons in Dave’s book!  Moreover, however, was that there are intriguing potentials from tapping into virtual worlds to remove the geographic constraints on such social interactions.

What was also interesting to me, reflecting on an early experience with the Active Worlds virtual world, your attention eventually focused on the chat stream, because that’s where all meaningful interaction really happened.  Which is really what #lrnchat is, a chat.   One of the nice properties of a chat is that you’re not limited to turn-taking.  A problem in the real world is that the more people you add, the less time each gets to contribute in a conversation. In a simultaneous medium like #lrnchat, everyone can contribute as fast as they can, and the only limitations are on the participants ability to process the stream and contribute (which are, admittedly, finite).  Still, it’s a richer medium for contribution, as I find I can process more chats in the same time only one person would talk (of course, the 140 char limit helps too).

The important thing to me is that social media have new capabilities to enable contribution, and achieve the innovation end that Dave’s excited about in ways that maximize the outcomes based upon new technology affordances that we are just beginning to appreciate.  Can we do better than we’ve done in the past, leveraging new technologies?  I think Dave’s model can serve for virtual as well as real events, and we may be able to improve upon the activities with some technology capabilities.  To do so, however, means we really have to look at our capabilities in conjunction with new technologies.  Yeah, I think we can have some fun with that ;).

On our weekly twitter learning fest, #lrnchat, I regularly identify myself as a learning experience design strategist.  I don’t always assume people know what that means, but for that audience I figure they can infer what it means.  However, I think the idea is worth exploring, because increasingly I think that not only is that what I do, but it also is important.

First, I think it is important to stop thinking about content, and start thinking about learning experience.  It’s too easy, when focusing on content, to focus on knowledge, not skills, yet skills are what will make the difference – the ability to do.  Also, it helps focus on the conative side of learning, the motivation for and anxiety about learning when you think about the learner experience. And, as always, I take a broad interpretation of learning, so this holds true beyond formal learning; it applies to thinking about performer experience when you consider the tools they’ll have, and even the way that access to communities and other informal learning components will be made available in situ.

When you think about creating learning experiences, you are talking about design.  How do you create effective and engaging learning experiences?  You need a design process, tools, and good concepts around learning and engagement.  Really, both my book on designing engaging learning experiences, and my forthcoming one on mobile learning, are at core about design.  And there are levels of design, from individual experiences to the architecture and infrastructure that can support the rich suite of experiences that characterize an organization’s full needs.

Which takes us to the last part, strategy.  By and large, I don’t do the design anymore, since I can add more value at a higher level.  Increasingly, what I’m doing is helping organizations look at their needs, current state, teams, processes, and more, and helping them develop a strategic approach to delivering learning experiences.  I help design pedagogies, processes, templates, and short-, medium-, and long-term steps.  And it is in this way that I accomplish what my first real client told me I did for them, I helped them take their solutions to the ‘next level’.

I think learning experience design is important, so important that I want to not just execute against a project at a time, but find ways to develop capability so a lot more good learning experience is created.  That means working with groups and systems. More organizations need this than might be imagined: I’ve done this for for-profit education, education publishing, those servicing corporate learning needs, and of course organizations (governmental and corporate)  wanting their external or internal learning solutions to be effective and engaging.  The sad fact is, too much ‘learning design’ is content design, still.  I’m always looking for ways to help spread a better way of creating learning.

For example, I ran a ‘deeper ID’ workshop this week for a team, and presented the concepts, modeled the application to samples of their learning objectives, gave them a practice opportunity, and wrapped up, across each of the learning elements. It was a way to address learning design in a bigger way. An extension would be to then submit sample content to me to have me comment, developing their abilities over time, as I did with another client working on integrating scenarios.

There are lots of ways this plays out, not just workshops but developing content models, spreading new metaphors for mobile learning, creating pedagogy templates, and more, but I reckon it is important work, and I have the background to do it.  I’ve found it hard to describe in the past, and I do question whether the ‘learning’ label is somewhat limiting, given my engagement in social learning with ITA and more, but I reckon it’s the right way to think about it. So I’ll keep describing it this way, and doing this work, until someone gives me a better idea!

In reflecting on the presentation I gave earlier this week, I realize that I didn’t make it clear that just making it social will make activities lead to better processing.  Of course, my goal was evangelizing, but I reckon I should followup with some clarity.  There are some design principles involved.

First, the assignment itself needs to be designed to involve valuable processing activities.  If it’s merely reviewing other’s comments (after you’ve had them either “restate the concepts in your own words” or “indicate how this explains something in your past or will influence your future behavior”), asking for a “contentful contribution” (where you’ve made clear that a contentful contribution addresses the substance of their post in an elaborative or constructively critical way) is fairly straightforward. If, however, you’re looking for discussion, you will need to strive for a topic that is likely to have different points of view, either from a base of values or from different conceptions.  Areas where misconceptions are rife are useful as they can be used for constructive feedback.

If you’re asking them to collaborate to apply the knowledge to a problem (which I encourage), then you’ll want to find an application exercises the core knowledge in ways that is as closely related as possible to how they’ll need to apply it in the world.  Choose appropriately challenging applications that will bring out differences of opinion that will need active interaction to resolve.  Having teams submit intermediate representations gives the instructor a chance to provide guidance, ala Laurillard.

However, there’s more than just the assignment.  For one, do not assume learners know how to interact well on a collaborative project.  When I first assigned such to online learning teams, they questioned how to work together. I’m glad they did, as I was able to develop a set of guidelines for them that subsequently smoothed the process.  Things like each coming up with their draft response, and then sharing before negotiating a shared approach are not necessarily obvious to learners.

Finally, you need to have an environment where learners understand the expectations about taking responsibility for learning and contributing sincerely on projects, as well as tolerating differences of opinion and tolerating diversity.  Don’t assume it, engineer it by stating at the outset what’s appropriate, and always welcome inquiries on process.

Social learning does provide richer processing (next to an individual Socratic tutor, but that’s not very scalable), but it takes careful design as well.  Design your learning experiences well, and generate powerful outcomes!

I’ve argued before that you need to emotionally hook learners even before you cognitively activate related knowledge.  I reckon that learners are more likely to be open to any manipulation you might provide if they understand viscerally why something’s important before they are informed cognitively.  Some new research might support this argument.

An article points to a theory proposed by two philosophers that interprets a broad range of cognitive phenomena in terms of human communication and argumentation. In particular, some reliable flaws exhibited in our thinking, such as confirmation bias, are hypothesized to exist because we’ve evolved to be able to argue for our beliefs. We argue, therefore we are.

This isn’t to say that we can’t evaluate arguments effectively under the right contexts (when we have no bias or when we’re searching for ‘the truth’), but that when we’re creating arguments we are likely to be suboptimal from a logical standpoint, but very good at trying to marshal the evidence in a particular direction when we care.  As the authors make clear.

My particular take on this, however, is that we should ensure to marshal a convincing case about why this learning is important or our learners may make a convincing case to the contrary.  Hook the learner’s interests and motivations, and the rest of the work will be easier.

And, of course, I’m making my case in the same way they argue we should, but that doesn’t undermine the quality of the reasoning ;).

In my previous post, I mentioned that we needed to start thinking about designing not just formal learning content, or formal learning experiences, but learning experiences in the context of the informal learning resources (job aids, social tools), and moreover, learning in the context of a workflow.  I’d sold myself on this, when I realized just where my ITA colleagues would draw me up short: it’s still the thinking that we can design solutions a priori!

Things are moving so fast, and increasingly the work will be solving new problems, designing new solutions/products/services, etc, that we won’t be able to anticipate the actual work needs.  What we will need to do, instead, is ensure that a full suite of tools are available, and provide individuals with the ability to work together to create worthwhile working/learning environments.

In short, tying back to my post on collaboratively designing job aids, I think we need to be collaboratively designing workflows. What I mean is that the learning function role will move to facilitating individuals tailoring content and tools to achieve their learning goals.  (And not, I should add, to ‘accreditation‘!)

And that’s where I tie back to Explorability and Incremental Advantage: we need easy to use tools that let us build not just pages, but environments.  The ‘pods’ that you can drag around and reconfigure interfaces are a part, but there’s a semantic level behind it as well. No one wants to get tied to a) learning a complex system that’s separate from their goals, or b) depending on some department to do it when and where convenient for the department.

Obviously, providing a good default is the starting point, but if people can invest as much as they want to get the power they want to configure the system to work the way they want, with minimal assistance, we’re making progress.

So that, to me, facilitating the development of personal (and group) learning environments is a valuable role for the learning function, and a necessary tool will be an easily configurable environment.

I’ve talked before about how our design task will need to accommodate both the formal learning and the informal job resources, but as I’ve been thinking about (and working on) this model, it occurs to me that there is another way to think about learning design that we have to consider.

The first notion is that we should not design our formal learning solutions without thinking about what the performance support aspects are as well.  We need to co-design our performance support solutions along with our preparation for performance so that they mutually reflect (and reference) each other. Our goal has to be to look at the total development and execution of the task.

The other way I’ve now been thinking of it, however, is to think about designing the workflow and the learning ‘flow’ together.  Visualize the formal and informal learning flows as components within an overall workflow.  You want the performer focusing on the task, and learning tools ‘to hand’ within the task flow.  Ideally, the person is able to find the answers, or even learn some new things, while still in the work context. (Context is so important in learning that we spend large amounts to recreate context away from our existing work context!)

The point being, not only is formal learning and informal learning co-designed, but they’re both co-designed in the context of understanding the flow of performance, so you’re designing the work/learning context.  Which means we’re incorporating user-interface and user-experience design, as well as resource design (e.g. technical communications) on top of our learning design.  And probably more.

Now, are you ready to buy this?  Because I’d talked myself to this point and then realized: “but wait, there’s more. If you call now, we’ll throw in” an obvious extension. To be covered in the next and last post of this series (tying it back to the context of explorability and incremental advantage I started with in my last post.

During a summer internship at NASA, many years ago, I met a researcher who was conceptualizing the interface property of ‘explorability’. I can’t claim that I accurately can communicate the nuances of Jean-Marc Robert‘s model, but I was intrigued with the notion. The idea that interfaces could differ on the extent that they supported experimentation and subsequent comprehension seemed valuable. The requisite property would be predictability, requiring consistency, and learnable interfaces would empower users.

A related concept is Andi diSessa‘s ‘incremental advantage’, where he proposed that interfaces should elegantly allow the investment by a user to learn more to provide more power. So his Boxer software environment supported gradual addition of concepts to yield more computational capability. The underlying notion of ‘the more you learn, the  more you can do’ again seems like a user-empowering concept.

Fast-forward a few years, and as a newly-minted academic using HyperCard for student interface design projects, I recognized that the notion of buttons, fields, and backgrounds provided a reasonable implementation of the ideas of explorability and incremental advantage. I proposed that the key idea was supporting correct inferences about how to make things happen. Interestingly, the English-like nature of HyperTalk supported both some correct and some incorrect inferences about making more complex logic.

As a side note, a combination between software design supporting a strong conceptual model, and software training that builds the model (not rote procedures), strikes me as a learning approach that is far more powerful but seldom seen.  Similarly for other learning outcomes, models are powerful thinking tools that we do not leverage sufficiently.

The reason I mention this is two-fold; I want to bring this concept to light, and to build on it. As I mentioned before, I think we need to make editable environments to support collaborative tool building. This will become more important, going forward, for reasons that I intend to elaborate across two subsequent posts. Stay tuned!

On twitter today a brief conversation ensued about best practices versus best principles.  I’ve gone off on this before ( I think Dilbert sums it up nicely), and my tweet today captures my belief:

“please, *not* best practices; abstract best principles and recontextualize!”

However, I want to go further.

Several times recently I’ve had people ask for research that justifies a particular position. And at a micro-level, that makes sense.  But there’s little ‘micro’ about the types of problems we solve.  So I hear it at a larger level: “why should we make learning more scenario-based”, or “what is the empirical evidence about social learning in the organization”.  And the problem is, you can’t really answer the question the way they think you should be able to. On principle (heh).

The problem is, most empirical research tends to be done around very small situations: these 3 classrooms were trialed in this state or province.  In many cases, there just hasn’t been the specific studies that are close enough to make a reasonable inference. And it’s hard to coordinate large studies that are really generalizable for pragmatic reasons that include logistics and funding.

What’s done instead, when sufficient cases arise, are meta-studies (as the recent one that said online learning was somewhat better than face to face), that tend to look across research, but you need a sufficient quantity of comparable studies (and someone capable and motivated).  Or, you can point to long programs of studies that are based around theoretical positions (e.g. John Sweller’s Cognitive Load Theory).  And expert practitioners typically have created  or procedures across long experience that can guide you.  In any case, you’re making inferences from a variety of studies and models.   One of my favorite models (Cognitive Apprenticeship) actually came from finding some synergy across several bodies of work.

So what’s a person to do? Sure, if you can find that specific relevant experiment, go for it.  Otherwise:

• look to what others do, but don’t try to immediately adopt their practices, look to find the underlying principles and adapt those,
• look to theories folks have proposed, and see how they might guide your approach,
• bring in someone who’s had experience doing this,
• or, think through it yourself, conceptualize the relationships, and determine what should be appropriate approaches.

(Note that the latter likely will take longer.) This is a ‘design-based research‘ approach, and to continue you need to trial, evaluate, and refine. Please do bring your reflections back to the conceptual domain.  We need more transparency!

The point I’m trying to make here is that, particularly in the learning sciences (e.g. when you’re working with the human brain), the properties aren’t as predictable as cement or steel; there is a bit of ‘uncertainty principle‘ going on (studying it changes the situation), and your intervention can very much affect how the individual perceives the task and possibilities.  You should expect to do some iteration and tuning.  And your bases for decision will not be individual research studies, by and large, but frameworks, models, and inferences.

Still, it’s systematic, based upon research and theory, and the best we can do.  So what are you waiting for?

The Learning Circuits Blog Big Question of the Month is “Does the discussion of “how the brain learns” impact your eLearning design?”  My answer is in several parts.

The short answer is “yes”, of course, because my PhD is in Cognitive Psychology (really, applied cognitive science), and I’ve looked at cognitive learning, behavioral learning, social constructivist learning, connectionist learning, even machine learning, looking for guidance about how to design better learning experiences.  And there is good guidance.  However, most of it comes from research on learning, not from neuroscience.

The longer answer has some caveats.  Some of the so-called brain science ranges from misguided to outright misleading.  Some of the ‘learning styles’ materials claim to be based in brain structure, but the evidence is suspect at best.  Similarly, some of the inferences from neural structures are taken inappropriately.  There’s quite a bit of excitement, and fortunately some light amidst the heat and smoke.  In short, there’s a lot of misinformation out there.

At the end of the day, the best guidance is still the combination of empirical results from research on how we learn, a ‘design research’ approach with iterative testing, and some inspiration in lieu of what still needs to be tested (e.g. engagement).  I think that we know a lot about designing effective learning, that is based in how our brains work, but few implications from the physiology of the brain.  As others have said, the implications at one layer of ‘architecture’ don’t necessarily imply higher levels of phenomena.  We’ve lots to learn yet about our brains.

As with so many other ‘snake oil’ issues, like multigenerational differences, learning styles, digital natives, etc, brain-based learning appears to be trying to sell you a program rather than a solution. Look for good research, not good marketing.  Caveat emptor!

Many years ago, I responded to a broad query by Jefferey Bonar asking what was the interface metaphor we really wanted.  I responded something to the effect of wanting ‘magic’.   This was in the early days of the desktop metaphor, and we were already looking to go beyond, and I was looking for the ultimate metaphor of control.

Now I didn’t mean magic in the ‘legerdemain’, sleight-of-hand type of thing, nor the magic I feel when sitting on the deck on a warm summer evening with my family, but instead the classic form with incantations, artifacts, etc. What I really wanted was to be empowered, and the best metaphor for total power I can imagine is having the ability to bring things into being, to have questions answered, to control the world with mere gestures and commands. And yet, even that has to have some structure.  As Clay Kallam wrote in a recent column comparing two recent fantasy books:

“The plot of both books relies heavily on the magic, but Coe is careful to explain how his works and its limitations and impact.  Drake seems to just call on some whenever it suits him, and nothing is explained.”

So, what I meant was that there was rigor underlying the metaphor of magic, rigor that roughly parallels the structures of programming languages.  For example, Rob Moser (my PhD student) prototyped a game for his thesis that taught programming via learning to cast magic spells in a fantasy world.  My vision was that in any place you wanted to, you could learn the underlying magic (language) to accomplish what you wanted, but if you didn’t, you’d be able to buy artifacts (e.g. wands, crystal balls, etc) that did specific things that you wanted without having to program.

The reason I mention this, before you think I’m going off with the fairies and unicorns, is that there are reasons to start thinking about magic.  As Arthur C. Clarke has said:

Any truly advanced technology is indistinguisable from magic.

And I really think we’re there. That is, our technology has advanced to the point that the technology is no longer a barrier.  We can truly bring any information, any person (at least virtually), anywhere we want.  We can augment our world with information to make us substantially more effective: we can talk through ‘mirrors’ (video portals) to others, actually seeing them; we can bring up ‘demons’ (agents) to go find information for us, we can send out commands to make things happen at a distance, we can unveil previously hidden information about the environment to start making conceptual links between there and our understanding to make us smarter.

There’s more required, such as Andi diSessa’s “incremental advantage”, and more accessible ways to specify our intentions, but with really powerful metaphors emerging (styles is something everyone should get their minds around), with gestural interfaces, and the ability to control games with our bodies, and with augmented reality aka Heads-Up Displays for civilians, we’ve got the tools.  What we need is the perspectives and the will.

This is important from the point of view of designing new solutions.  Years ago, when I taught interface design, I told my students that one of the pieces in their exploration of the design space should be to imagine what they would do when they had ‘magic’.  To be more specific, once you’ve gathered the requirements, before you see what others have done and start limiting yourself to pragmatics, imagine what you’d do with no limitations (ok, except mind-reading, I’m just not going there).  Given that among our cognitive architectural pre-dispositions is to prematurely converge on solutions, we need lateral input.  By exploring the possibilities space in a more unhampered way, we might come across a solution that’s inspired, not tired, and revolutionary, not evolutionary.

This, however, is not just interface design, but specifically learning and performance support design.  What would you do if you had magic to help meet your learning and performance needs?  Because you have it.  Really.

So think magically, not in the trivial sense, but in the sense that we have awesome powers at our command.  The limitations are no longer the technology, the limits are between our ears (and, occasionally, in our wallets or will).  Go forth and empower!