Learnlets

Secondary

Clark Quinn’s Learnings about Learning

You are here: Home / Archives for design

Translational research?

by 2 Comments

I came across the phrase “Translational Behavior-Analysis”. I had no idea what that was, so I looked it up.  And I found the answer interesting. The premise is that this is an intermediary between academic and applied work.  Which I think of as a good thing, but is it really a thing? Does it make sense?  I have mixed feelings, so I thought I’d lay them out.

So, one of the things that a few people do is translate research to practice. I’m thinking of folks who are quite explicit about it like Will Thalheimer, Patti Schank, Julie Dirksen, Ruth Clark, and Mirjam Neelen, amongst others.  They’re also practitioners, designing or consulting on solutions, but they can read research in untranslated academese and make sense of it. So is this that?

One definition I found said: “the process of applying ideas, insights, and discoveries generated through basic scientific inquiry to” <applied discipline>.  This is big in medical circles, apparently.  And that’s a good thing, I hope you’d agree.  However, they also say “occupies a conceptual space between basic research and applied research”.  Wait, I thought that was applied research!

Ok, so further research found this gem: “Applied research is any research that may possibly be useful for enhancing health or well-being. It does not necessarily have to have any effort connected with it to take the research to a practical level.”  Ah, so we can do things in applied research that we think might be good, even if it isn’t connected to basic research. Well, then.  When I think of applied cognition, which has showed up in interface design (and I try to push in learning experience design), I think of that as doing what they call translational, but perhaps it’s not that way in other fields.

Ultimately, this was about fast-tracking medical research into changing people’s lives. And that’s a good thing. And I think our ‘interpreters’ are indeed serving to help take academic research and fast-track it into our learning designs. Will has called himself a ‘translator’ and that’s a good thing.

We also need a way for our own innovations, for instance taking agile software development and applying it to learning design, to filter back to academia and get perhaps a rigorous test. There are people experimenting with VR and other technologies, for instance, and some of the experimentation is “why not this” instead of “theory suggests that”. And both are good.  We may need translators both ways, and I think the channel back to academia is a bit weak, at least in learning and technology. Happy to be wrong about that, by the way!

I’m mindful that we have to be careful about bandwagons. There’s a lot of smoke and hype that makes it easy to distract from fundamentals that we’re still not getting right.  And I’m not sure whether applied or transformational is the right label, but it is an important role.  I guess I still think that a tight coupling between basic and applied implies translational (I like Reeves’ Design Research as a bridge, myself), but I’m happy to accept more nuanced views.  How about you?

Are Decisions the Key?

by Leave a Comment

A number of years ago, now, Brenda Sugrue posited that Bloom’s Taxonomy was wrong. And, she proposed a simpler framework. I’ve never been a fan of Bloom’s; folks have trouble applying it systematically (reliably discriminating between various levels). And, while it pushed for higher levels, it left people off the hook if they decided a lower level would do it. Sugrue first proposed a simpler taxonomy, and also an alternative that was just performance. In her later version, she’s aligned the former to the work of the Science of Learning Center’s KLI (knowledge-learning-instruction) framework. But I want to go back to her ‘pure performance’ model, and make a the case that decisions are key, that they are necessary but also sufficient.

So her lates model discriminates between concept, process, fact, principle, etc.  And, I would agree, there are likely different pedagogies applied to each.  Is that a basis enough?  Let me suggest a different approach, because I don’t see how they differ in one meaningful way. For each, you need to take some action, whether it’s to:

  • classify as a fact (is it a this or a that)
  • perform the steps (which action to take now)
  • trouble shoot the process (what experiment now)
  • predict an outcome (what will happen)

Note, however, that for each, there’s an associated decision. And that, to me, is core.  Now, I’m not claiming that they all require the same approach.  For instance, to help people deal with ambiguous decisions, I suggested a collaborative approach to discuss the parameters and unpack the thinking. To teach trouble-shooting, I would give some practice making conceptual decisions about the systems that could cause the observed symptoms. In internal combustion engines (read: cars), if it’s not running, is the air/fuel system or the electricity? How could you narrow that down?  In a diesel, you could eliminate the electrical ;).

Van Merriënboer, in his Four Component Instructional Design, talks about the knowledge you need and the complex decisions you apply that to. I agree, and so it’s not just about decisions. However, even the knowledge needs to be applied to stick.  To test that learners have acquired the underpinning knowledge, you can hav them exercising the models in decisions.

Ok, so you might want to short-circuit the mapping from decision to practice. I think a good heuristic (ala Cathy Moore’s Action Mapping) is just to have them do what they need to do, and give them the necessary information. However, if you want to create a ‘cheat sheet’ to accelerate performance and success, with learning goals and associated pedagogies, I won’t quibble.

Now, you can’t provide all the situations, you you need to choose the right ones that will help facilitate abstraction and transfer. You may need to also ensure that they know the requisite information, so you may need to determine that, but I think exercising the models in simpler situations helps develop them more than just a presentation.

I’m suggesting that focusing radically on decisions is the best way to work with SMEs, and is the best guide for designing practice (e.g. put learners in situations to make decisions). Everything else revolves around that. Now, are these categories reliable types of decisions?  Will ponder. Your thoughts?

User-experienced stories

by Leave a Comment

Yesterday I wrote about examples as stories. And I received a comment that prompted some reflection. The comment suggested that scenarios were stories too. And I agree!  They’re not examples, but they are stories. With a twist.

So, as I’ve said many times, simulations are just a manipulable model of the world. And a motivated, self-capable learner can learn from them. But motivated and self-capable isn’t always a safe bet. So, instead, we put the simulation in an initial state, and ask the learner to take it to a goal state, and we choose those such that they can’t get there until they learn the relationships we want them to understand. That’s what I call a scenario.  And we can tune those into a game. (Yes, we turn them into games by tuning; making the setting compelling, adjusting the challenge, etc.)

Now, a scenario needs a number of things. It needs a context, a setting. It needs a goal, a situation to be achieved. And, I’ll suggest, it should also have a reason for that goal to make sense. If you see the alignment that says why games should be hard fun, you’ll see that making it meaningful is one of the elements. And that, I say, is a story. Or, at least, the beginning of one.

In short, a story has a setting, a goal, and a path to get there. We remove boring details, highlight the tension, etc.  We flesh out a setting that the learner cares about, provide a sense of urgency, and enable the goal achievement.  But it’s not all done.

The reason this isn’t a complete story is we don’t know the path the protagonist uses to accomplish the goal, or ultimately doesn’t.  We’ve provided tools for that to happen, but we, as designers, don’t control the protagonist. The learner, really, is the the protagonist!

What I’m talking about is that the story, certainly for the learner, is co-created between the world we’ve developed, and their use of the options or choices we provide. Together, a story is written for them by us and them.  And, their decisions and the feedback are the story and the learning!  It’s, voilà, a learning experience.

Learning is powerful. Creating experiences that facilitate learning are creative hard fun for the designer, and valuable hard fun for the learner. Learning is about stories, some told, some c0-created, but all valuable.

Telling stories

by Leave a Comment

I’ve talked about examples before, and have long suggested that examples for learning should be in the form of stories. Today I thought a little deeper about how those stories can be told. And I thought of a particular entailment about video. But let me go through the whole story.

So, first, an example needs to do several things.  Foremost, examples show how a concept is used in context to guide performance and achieve a solution. To put it another way, examples illustrate how a model is used to make decisions in a specific situation.  It could be situational coaching, a sales process, how to run a brainstorming session, or the basis of a good presentation. Importantly, examples need to show the underlying thinking behind what aspects of the situation made this model relevant, how it was mapped to the situation, how  it stipulated as actions to take, and what the outcomes were. It can also show mistakes, backtracking, and repair.

So how does this map to media?  There are (at least) three good ways to tell such a story. One, obviously, is prose. It’s easy to show the underlying thinking or discussions as “dialog” (internal or otherwise).  You can even illustrate model with a diagram, and convey context with a picture.  And this works.  However, for variety, you could use more visual treatments.

I’ve argued in the past that graphic novel formats also make sense. They strike a balance between conceptual and contextual, and tell a story nicely. You can use thought bubbles to show the underlying thinking, include diagrams as a separate panel. And you can include necessary context but simplify the unnecessary context.

Finally, there’s video. This can be an animated cartoon (e.g. dynamic graphic novel), a narrated slideshow (think: Ken Burns), or a documentary style.  The question: how do you represent the model? Is it a separate narrated animation, or an overlay on the video (or both)?  I don’t think it matters as long as it’s explicit, and the underlying thinking is shown linking the concept to the context.

The point is that the nuances matter. It’s not just ‘content’; the elements that we provide have specific roles and we need to understand those.

Social Cognition

by Leave a Comment

In the two preceding posts, I discussed situated and distributed cognition. In this closing post of the series, I want to talk about social cognition. They’re related, and yet each needs explicit consideration. If we don’t know how we think, work, and learn, we can’t optimally support both performance in the moment and continual innovation over time.

The traditional definition of social cognition is how we think about social interactions.  But here I’m emphasizing instead the fact that our thinking isn’t just in our heads or our tools, but also across our partners. That’s partly distributed cognition, but I want to emphasize it.  And this is true for formal and informal learning as well as performing.

There are two ways to think about this. For one, we benefit from formal social interactions as ways to get richer interpretations. It works the same way when we are problem-solving: working together (under constraints) increases the likelihood of the best outcome. As I like to say, the room is smarter than the smartest person in the room if you manage the process right. The implications of this are several.

First, we need to make sure we have the right constraints. When we have people working together, it helps if it’s the right people and the right environment.  We know that diversity helps, as long as there is overlap in values. Similarly, it needs to be psychologically safe to contribute, the environment helps to be open, and there needs to be time for reflection.

There’re also benefits to mentoring and coaching, helping people in the moment. We want to succeed, and we like to be challenged, and we learn when we are, so having scaffolding helps. Developing coaching and mentoring skills is a good investment in the workplace.

There’re also times when we want help, or someone else does and we can help. That is, we need to support serendipitous inquiry. It helps, by the way, to assist people in learning how to ask questions or answer them in useful ways. There also needs to be the channels to accomplish these goals.

Recognize that there are times when the answer can come from the network, not our own efforts. Particularly if things are changing fast, or the situation’s unique or hard to anticipate. In fact, it frees us up to do more if we take advantage of that as often as possible!  It takes nurturing the networks to become a community so that the answer’s likely to be right.

The point being, there are lots of considerations to making the ecosystem sociable as well as effectively distributed and situated.  If you want to optimize the environment, it helps to have the latest understanding of the users of that environment. Hope this makes sense, and in the spirit of social, I welcome your thoughts and comments!

Situated Cognition

by 5 Comments

In a recent article, I wrote about three types of cognition that are changing how we think about how we think (how meta!).  All are interesting, but they also have implications for understanding for supporting us in doing things.  I think it’s important to understand these cognitions, and their implications. First, I want to talk about situated cognition.

The psychological models of thinking really started with the behavioral models. The core argument was that we couldn’t look ‘inside the box’, and had to study inputs and outputs. Cognitive psychology was a rebellion from this perspective. The new frameworks started showing that we could posit quite a bit about what went on ‘in the box’. We got concepts like sensory, working, and long-term memory, and processes like attention, rehearsal, encoding, and retrieval. With most of our learning prescriptions. However, both were about the ‘the box’.

However, the observed behavior didn’t match the formal logical reasoning that underpinned the model. We needed new explanations. The computational model fell apart. And, despite rigorous attempts to create logical models that described human behavior, they were awkward at best. The shift came when Rumelhart & McClelland, in their PDP book, described what became known as neural networks. Associated with this was a new model of cognition.

What gets activated in the brain is not a reliably pure representation, and is strongly affected by the context. Thinking is ‘situated’ in the context it arises in. If our thinking is the emergent behavior of patterns across neurons, and those patterns are the result of both internal and external stimuli, then we’re very strongly influenced by what’s happening ‘in the moment’.  And that means that we can be captured (and fooled) by elements that may not even be consciously processed.

What this means in practice is that it’s harder than we think to get reliable performance across a range of conditions.  That we should ensure that patterns are generated across ‘noise’ so that they’re reliable in the face of the appropriate triggers, despite any accompanying contextual patterns. and recognize that decisions can be biased, and design scaffolding to prevent in appropriate outcomes. Developing mental models that provide reasoning abilities about causes and outcomes are useful here. This flexibility is advantageous (and why machine learning struggles outside it’s range of training), but we want to tap into it in helpful ways.

Our approaches should reflect what’s known, and therefore we need to keep up.  Situated cognition is a perspective that’s relevant to more effectively supporting individual and organizational performance and learning. So, what is your thinking about this?

Designing with science

by 1 Comment

How should we design? It’s all well and good to spout principles, but putting them into practice is another thing. While we always would like to follow learning science, there’re not always all the answers we need. I was thinking about this with a project I’m working on, and it occurred to me that there might be some confusion. So I thought I’d share how I like to think and go about it, and see what you think.

So, first of all, you should go with the science. There are good principles around in a variety of forms.  Some good guidance comes in books such as:

  • eLearning & the Science of Instruction (Clark & Mayer)
  • Design for How People Learn (Dirksen)
  • the Make it Learnable series (Shank)
  • and less directly but no less applicably, Michael Allen’s Guide to eLearning

There’s also ATD’s Science of Learning topic (with some good and some less good stuff).  And the 3 Star Learning site. Both of these, of course, aren’t as comprehensive as a book.   And, of course, you can also go right to the pure journals, like Instructional Science, and Learning Sciences, and the like, if you are fluent in academese.  For that matter, I’ve a video course that is about Deeper Instructional Design, e.g. a design approach with learning science ‘baked in’.

But what I was thinking of what happens when they don’t address the specific concern you are wondering about. The second approach I recommend is theory. In particular, Cognitive Apprenticeship (my favorite model; Collins & Brown), or other theories like Elaboration Theory (Reigeluth), Pebble in a Pond (Merrill), or 4 Component ID (Van Merriënboer). Or, arguably more modern, something from Jonassen on problem-based learning or other more social constructivist approaches.  They’re based on empirical data, but pulled together, and you can often make inferences in between the principles.  While the next step is arguably better, in the real world you want a scrutable approach but one that gets you moving forward the fastest.

Finally, you test. If science and theory can’t provide the answer, you either wing it, but it’s better if you set up an experiment. Ideally, with your sample population.  So, for instance, you don’t know whether to place the learner’s role in the simulation game as a consultant to many orgs or as a role in one org with many situations. There’re tradeoffs: in the former it’s easier to provide multiple contexts for practice, but the latter may be more closely aligned with job performance.   You can test it, and see what learners think about the experience. Of course, it may be that in the process of just designing both that you have some insight. And that’s ok.

And, if you’re a reflective practitioner (and we should be), you might share your findings.  What did you learn?  Learning science advances to the extent that we continue to explore and test.  Speaking of which, how does this approach match with what you do?

Silly Design

by Leave a Comment

Time for a brief rant on interface designs.  Here we’re talking about two different situations, one device, and and one interface.  And, hopefully, we can extract some lessons, because these are just silly design decisions.

OXO timerFirst up is our timer.  And it’s a good timer, and gets lots of use. Tea, rice, lots of things. And, sometimes, a few things at a time.  As you can see, there’re 3 timers. And, as far as I know, we’ve only used at most two at a time.  So what’s the problem?

Well, there’re different beeps signaling the end of different timers. And that’s a good thing. Mostly.  But there’s one very very silly design decision here. Let me tell you that one has one beep, one has two beeps, and one has three. So, guess which number of beeps goes to which timer?  You can see they’re numbered…

Got your guess? It’d be sensible, of course, if the one beep went with the first timer, and two beeps went with the second. But you know we’re not going there!  Nope, the first timer has two beeps. The second timer has 3 beeps. And the 3rd timer, of course, has one.

It’s a principle called ‘mapping’ (see Don Norman’s essential reading for anyone who designs for people: The Design of Everyday Things). In it, you make the mapping logical, so for instance between the number of the timer and the number of beeps.  How could you get this wrong? (Cliche cue: you had one job…)

iTunes way 1On to our second of today’s contestants, the iTunes interface.  Now, everyone likes to bash iTunes, and either it’s a bad design for what it’s doing, or it shouldn’t be trying to do too many things. I’m not going there today, I’m going off on something else.

I’ve always managed the files on the qPad through iTunes. It used to be straightforward, but they changed it. Of course.  There’re also more ways to do it: AirDrop & iFiles being two. And, frankly, they’re both somewhat confusing.  But that’s not my concern today.  The new way I use is only a slight modification on the old way, which is why I use it. And it works. But there’s a funny little hiccup…

So, there are two ways to bring up a list of things on your iPad.  For one, you select it from the device picture at the top (to the right of the forward/back arrows), and you see a list of things you can access/adjust: music, movies, etc. As you see to the left.

other way to access iPadOn the other hand (to the right), you select it from a list of devices, and you get the drop down you see to the right.  Note that the lists aren’t the same.

Wait, they’re not the same?  No, only one has “File Sharing”!  So, you have to remember which way to access the device before you can choose to add a file.  This is just silly!  Only recently have I started remembering which way works (bad design, BTW, trusting to memory), and before that I had to explore. It’s not much, just an extra click, but it’s unnecessary memory load.

The overhead isn’t much, to be clear, but it’s still irritating. Why, why would you have two different ways to access the device, and not have the same information come up?  It’s just silly!  Moreover, it violates a principle. Here, the principle is consistency (and, arguably, affordances). When you access a device, you expect to be able to manipulate the device. And you don’t expect that two different ways to get to what should be the same place would yield two different suites of information. (And don’t even get me started about the stupid inconsistencies between the mobile and web app versions of LinkedIn!)

At least if you haven’t communicated a clear model about why the one way is different than the other. But it’s not there.  It’s a seemingly arbitrary list. We operate on models, but there’s no obvious way to discriminate between these two, so the models are random. Choosing the device, either way, is supposed to access the device.  That’s the affordance.  Unless you convey clearly why these are different.

This holds for learning too. Interface folks argued that Gloria Gery’s Electronic Performance Support Systems were really making up for bad design. And so, too, is much training. Don argued in his The Invisible Computer that UI should be up front in product design, because they could catch the design decisions that would make it more difficult to use. I want to argue that it’s the same with the training folks: they should be up front in product or service design to catch decisions that will confuse the audience and require extra support costs.

Design, learning or product/service, works best when it aligns with how our brains work.  If we own that knowledge, we can then lobby to apply it, and help make our organizations more successful.  If we can make happier users, and less support costs, we should. And as Kathy Sierra suggests, really it’s all about learning.

 

A good publisher

by Leave a Comment

I shot this short little video because, well, I have to say that my experience with ATD has been excellent. They’ve done what I’ve needed: listening when they should, arguing with me when they should, responding to my questions, and executing on their responsibilities professionally. They’ve gone above and beyond, and I’m pleased to have them as my publisher on my most recent tome.

If you’re going to complain about the bad things (or, at least, make fun of them :), I reckon you should highlight the good ones too.  They showed their capabilities while serving as co-publisher on my last book, and now they’ve demonstrated the whole deal. Thanks, team!

Game Results

by Leave a Comment

The game we designed (as I talked about yesterday), Quest for Independence, was actually a fair bit of a hit. While we couldn’t talk to kids still, anecdotally we heard that the kids were playing it. And, as the design intended, it led them to talk to the Care counselors. That was good enough, but there was more.

First, the Aussie science program Quantum had a bit on it. They even interviewed me (with a big production about bringing stuff to our house), but never used the footage.  They also couldn’t talk to the kids ‘in care’, but it turns out Quest was being used by kids not in care!  High schools were using it to explore life after school as well. That was a nice outcome.

Another occurrence brought new action. Sometime after the game launched, I became aware of the Common Gateway Interface (CGI) standard for the web. What this did was allowed web pages to do backend processing based upon user actions, and then programmatically change the front end.  In short, web pages could react based upon what had happened before.

For Quest, this meant that we could port it to the web!  That is, you started the game, the player’s actions were sent to the web, the program could calculate the outcome and render an appropriate new page, with the graphics assembled to represent the game variables, the current location, and more.  This was exciting.

Splash screenAnd, again, I had a student wanting to do a project. So the project was to take the game graphics, and the programming, and make the game web-playable. And lo, it was done; the game could be played over the web.  Most wonderfully, it still can be!  (Yay, standards!)

Naturally, I wrote it up (with the student; a principle I always stuck to: even if I usually ended up writing it all, they got credit for their work and ideas).  And, as far as I know, it was the very first web-delivered serious game. At the least, without Flash.

The underlying principles in the game also became part of a couple of chapters, and ultimately the alignment between effective education and engaging experiences formed the core of my book on serious game design, Engaging Learning.  

One final reflection is that working on this, on a project that really helped real kids, was still one of the most rewarding projects I’ve ever worked on.  It’s nice to help clients deliver outcomes, but saving lives that were at risk?  That’s just too good.  Anyone up for some more ‘hard fun’?