We did end up talking about my teaching games post (among other things) in the February VGHVI Symposium; sadly, I had weird network problems which meant that I missed maybe a third of the conversation entirely and could listen but not speak in another third. Which is especially a pity because I think Roger and I spent time airing potential disagreements and not enough time figuring out where common ground was.
One point which the discussion brought home: to me, the term “motivator” (in the context of intrinsic/extrinsic) is more useful than the term “motivation”. And the reason for that is that motivation is internal: so it’s hard to know what really motivates somebody else, it’s too easy to even be unsure of your own motivation and/or to recast that motivation after the fact. Whereas motivator is external, and in particular suggests something that I have more direct control over as a teacher: I can choose from a set of possible behaviors towards my students, and I think it’s not as hard to label some of those as extrinsic motivators. (Though admittedly the boundary between extrinsic motivators and non-extrinsically-motivational feedback is extremely blurry.)
Even that isn’t why I like Punished by Rewards, though: I like that book because it steps back one further level from the concept of intrinsic / extrinsic motivation. It grounds itself instead with studies that divide test subjects into two groups, has the experimenters behave towards test subjects in two different ways, and measures how the test subjects act along dimensions that the experimenters are interested in. That’s much more concrete; if a specific such experiment is replicable, it’s valuable data, and if multiple experiments following similar protocols lead to similar results, then it starts to make sense to come up with a label for the common aspects of the behavior in those protocols. (In this specific case, the label is “extrinsic motivator”.)
And yes, you can try to use that (along with other ideas, e.g. psychological insights) to then come up with further concepts (e.g. “intrinsic motivation”, which is much harder to get at directly with an experiment), and you can (as I did) try to apply that in your own teaching. But still, if you can’t touch back to experiments periodically, then it’s not surprising if you get rather different results. So if Roger and his colleagues are having success (which, I repeat, I’m convinced he is) applying techniques that I would tend to label as extrinsic motivators (which I’m less convinced is the case, though that is still my tentative hypothesis), then there are several possible explanations for that:
- The original studies are not replicable after all.
- Roger’s practomimetic teaching techniques aren’t similar enough to techniques used in those studies for generalizations from the studies to be relevant.
- The state of students when practomimetic techniques aren’t applied isn’t similar enough to how students behave in the non-extrinsic side of the original studies to expect extrinsic motivators to cause problems.
- Practomimetic teaching has benefits that are significantly greater than the comparatively small drawbacks from extrinsic motivators.
- Practomimetic teaching has aspects that actively innoculate against the drawbacks of extrinsic motivators.
There are probably other explanations that I’m not thinking of, too. And certainly none of those five explanations would surprise me at all in practice; in fact, I’d be surprised if the bottom three weren’t all true.
Still, this is all not so relevant to me, given that I’m no longer actively teaching in a classroom; given how last time went, I doubt we’ll get any further if we continue that line of discussion, so I’d just as soon drop it for the next Symposium.
What I will propose instead is a different line of discussion: how do practomimetic teaching techniques change as teachers leave more space for students to explore? For example, if we draw a spectrum between a situation where students are told what to do every minute and a situation where students are given a vague goal with opportunities for feedback from teachers once a week or less, do practomimetic techniques lend themselves to one portion of the spectrum? I would think not so much the “every minute” portion, because the nature of role playing already carves out some amount of space for freedom of action. (Though, then again, I’ve played enough computerized RPGs that got in my face every minute…) But I would also expect that, as students get more familiar with the interplay that underlies the subject they’re learning (a language, an instrument, …), the benefits of external systems become less important. Certainly as I progress in learning guitar, I’m finding my actions somewhat more driven by what I’m seeing in the instrument and in the music I’m playing and less exclusively by the game, though the latter continues to be important to me. I’ll throw in a link to the concept of Shu-Ha-Ri here, too.
I’d also like to talk about this broadening of options in the context of games themselves, leaving learning (or rather: learning about something other than the game) aside. The fact that I love go so much is very much linked to my progression from its being a bit of a blur at the start to me learning about concrete techniques I can study to having those techniques link to more abstract ideas that I can play around with to another higher-level round of concrete techniques, repeating itself at several different scales: it’s great being able to go back and forth between concrete drills and conceptual experimentations. And I suspect that’s the way Patrick feels about Starcraft too. It isn’t, however, the way I feel about many games: e.g. adventure games by their nature fight against this, I think, and while the battling and leveling systems of role-playing games sometimes can take you a bit in the direction of exploring the games’ systems at different levels of depth, they don’t generally have enough layers to go really far. Which isn’t to say that role-playing games aren’t great: it’s just that the ways that they are great are, to me, not linked to their mechanics of their systems in the absence of any referents external to the games.
Hmm, actually, I’d be curious to hear what Roger thinks about that last sentence, too, given that he’s thought so much more about role-playing games, and in particular their non-combat systems, than I have.
This post has not been revised since publication.