Learnlets

Just as I did for mobile, here’re 5 paragraphs on games:

Serious Games (or, to be Politically Correctâ„¢, Immersive Learning Simulations) have hit the corporate learning mainstream, so you should be asking yourself: “why are people excited” Quite simply, because games (I‘m not PCâ„¢) are probably the most pragmatically effective learning practice you can get. Sure, mentored real performance is the ideal, but there are two potential hiccups: scaling individual mentors has proven to be unrealistically expensive, and mistakes in live practice often are expensive, dangerous, or both. Why do you think we have flight simulators?

For principled reasons, the best learning practice is contextualized, motivating, and challenging. Interestingly, so are the most engaging experiences. It turns out that the elements that cause effective educational practice line up perfectly with those that create engaging experiences. Thus, we can safely say that learning should be ‘hard fun‘.

Then the issue becomes if we can do this reliably, repeatably, and on a cost-effective basis. It turns out that the answer to this question is also in the affirmative. While you can‘t just shove gamers and educators in a room and expect the result to work (all the bad examples that led to ‘edutainment‘ becoming a bad word are evidence), if they understand the alignment above, systematically follow a creative process (no, systematic creativity is not an oxymoron; why do we have brainstorming processes?), and are willing to take time to ‘tune‘ the result, we can do this reliably.

The question is really: when to use games. The answer for engine-driven (read: programmed, variable) games is when we have a need for deep practice: when there are complex relationships to explore, or making the change will be really hard. Branching scenarios are useful when we want to experience some contextualized practice but we don‘t need a lot of it. And the principles suggest that at minimum, we should write better multiple-choice questions that put learners into contexts where they must make decisions where they‘re applying the knowledge, not just reciting it.

And, yes, we can spend millions of dollars (I can help :), but for many needs we may not need to. While there isn‘t any one tool that lets us do this, there are a number of cost-effective ways to develop and deliver on the resulting design. As I like to say “if you get the design right, there are lots of ways to implement it; if you don‘t get the design right, it doesn‘t matter how you implement it”.

Further resources include:

• My book on designing games
• The eLearning Guild’s Research Report on ILS
• The Serious Games site
• Clark Aldrich’s blog on learning games
• My other game blog posts

We’re developing a scenario (code word for serious game :), and I’ve come into a situation where I can see a plausible case for either side, and don’t know of any research results.

The scenario is serving as a the major organizing focus for a course (this specifically is designed for the formal education system). We’ve got a contextualized task that requires applying the curriculum material, and want to make the curriculum material available for access during the scenario.

Here’re the two options I see:

• One is to have all the tutorial material available from within the scenario. The notion is that once there’s a need, having the information available will optimize the moment of learning. The fear is that taking time to access the information could break the flow of the scenario experience. So, if you couldn’t decide how to set up the quantum physics experiment, you’d access a tutorial on said topic from the lab library.

• The other is to have a digested down version of the information (in a ‘performance support’ model, that serves as a reference guide and you stay very much ‘in the moment’, but if you don’t know the material, you exit the scenario to get the concepts, and then you go back into the scenario experience, and use the guides for assistance but they’re not sufficient to actually learn from (unless, of course, you’ve already got some foundation). So, if you couldn’t set up the quantum physics experiment with the ‘checklist’, you’d leave the scenario motivated to read the tutorial and then restart/reenter the scenario.

The tradeoff is learning material available versus any effect on ‘breaking the wall’ between the scenario and the external learning environment.

We’re building the scenario, so we’ll actually have the second, and of course if that’s insufficient we’ll add in the first, but I wondered if there’s any ‘a priori’ information. Research solicited, opinions welcome.

In the Serious Games discussion list, Richard Wainess posted a thoughtful and eloquent reply to my request for research on the value (or not) of fantastic settings, in which he argued about the necessary learning design depth required in game design. I‘m primed for the discussion since I‘ve just been in the process of designing a learning game with a team. I thoroughly agree with him, and I’d highly recommend you find and read his response except for the fact that it appears there’s no archive. However, I had assumed the issues he‘s suggested, and penned this (slightly modified) response:

I think you’re missing the value of fantastic settings in effectively adding on top of what you say. We could set a task (e.g. negotiation) in several real-life environments, including with a car dealer, with the boss for a raise, with the kids about bed time (bad idea), etc. Or we could set it in space, for example, negotiating with suppliers for equipment, with civilizations for territory, with buyers for products, etc. Once we ensure we’ve put the necessary skills into the game, across differing contexts, and added the post-game reflection, is there a potential benefit for having a more compelling storyline? That trades off positively against the less direct transfer?

Yes, it takes different contexts to abstract and generalize, but let’s not neglect the value of motivation. So I agree it absolutely *has* to encompass the essential skills across contexts (broad enough to generalize to all relevant situations, and to no irrelevant ones). But there’s more than just that. My hypothesis is that embedding them into an exaggerated storyline may enhance the outcomes more than a real-world setting (and the more so the more general the skill).

If it’s not a storyline that the learner cares about, they’re not going to engage like they will when it really matters to them (e.g. the car *they* want to purchase). So we need that motivation, that emotional engagement as well. And that’s when we’re going to want to align the cognitive and game engagement. When people really have to perform, they have external motivation. Don’t we want to embed that in the experience as well?

I suggest that once we get the educational process down and vary the settings in context, that increasing the motivation through a compelling storyline that both is a meaningful application of the skill and is a storyline that the learners care about, will increase the outcome measure more than an more realistic, and dull, exercise. It’s testable, and I want the answer rather than just relying on my intuition (which will suffice for now; I too am trying to meet real needs, not just satisfy academic interests, but I’d feel far better knowing the answer one way or another).

My feeling is, rightly or wrongly, that not enough people get the depths he talks about, and on the other side, the argument I make above. I‘d like the answer, but in lieu of that, I‘m going to stick with my belief. (And later, Richard responded about how my response made him smile, as he’s starting just this research.)

A further claim from another respondent said that we just need to make the next Oregon Trail, which spurred this rejoinder:

If you don’t have the academic underpinning that Richard argues so eloquently for, all the cool window-dressing won’t lead to a thing. If you’ve infinite resources, you can iterate ’til you get the outcomes you suggest, but I’d prefer to draw upon principled bases and shorten the development process by systematically combining deep learning design with creative engagement design.

It almost appears that the few good edutainment titles were more a case of “even a blind pig finds a truffle once in a while” (a botched metaphor, to be sure, but personally relevant as how my friend described me finding my wife) than the result of a real understanding; there are too many bad titles out there. I don’t want to trust to chance that NASA’s MMO will be effective, nor burn through too much $$ to ensure it. I’d like to use what we know to help do it reliably, and repeatably. We owe it to ourselves and to society to demonstrate that serious games are a viable learning vehicle, not a hit or miss (or money sink) proposition.

Ok, so I‘m opinionated. What did you expect? I didn‘t spend, off and on, 25+ years doing learning game design to just throw up my hands. So, am I off my rocker?

I didn’t blog it, since I *was* it (3 times), but Brent did. Here’re his notes from the session.