[media]
#[/media]
A connection I think worth making here is that, in terms of game design, a lot of being a game designer is being a professional motivator. You want to motivate players to engage in some sort of activity (presumably a fun one -- but; I'd argue it doesn't have to be fun). If you're good they'll continue to do that task for a long period of time, and if not, they'll quickly move on to something else. I know that's a leaky abstraction, but it's a potentially useful way to think about it.
Now, the most interesting point of that video comes about two minutes in when he's talking about how rewards affect motivation, and the surprising conclusion that science has come to, which is: adding rewards to tasks only increases performance for activities that are cognitively rudimentary. Once tasks require more than basic cognition skills, rewards actually hurt performance.
So, if I were to give you a dollar every time you pressed a button, you would press that button a lot more, but if I were to give you a dollar every time you solved a puzzle, it's likely that would actually solve fewer puzzles, even though you'd want to solve more for because it would net you more money; and increasing the reward seems to magnify that effect.
When you think about casual and social games, which tend to lean very heavily on reward systems to motivate players, it becomes obvious why they tend to be so stupidly simple. It has to be something anyone can do without thinking, or else it will become frustrating.
Now, what I'm curious about (and I don't have an actual answer for this), is this: while we know that changing how we motivate players affects performance, it's unclear to me how it affects engagement.
I mean, Farmville/Cityville/Whateverville games are not an intrinsically interesting activities on their own merits, I'd say. Those games rely heavily on intermittent reward schedules and social obligations. Yet they're pretty engaging to the people who play them. On the other hand, games like Dwarf Fortress and Minecraft don't rely on extrinsic motivation at all (if anything they tend to arbitrarily punish players), yet they're also quite engaging to the people that play them.
So it seems to me that both intrinsic motivation and extrinsic motivation can be effective, but it's not clear to me if mixing the two motivation types can amplify the amount of engagement they provide, or if they're fundamentally incompatible. I'm guessing it could potentially amplify, but I couldn't say for sure. As a thought experiment, say you put a leader board into Minecraft for who mined the most gold. Would that detract from players enjoying the game as a creative sandbox? Would players then not view the game as a sandbox, and instead focus solely on the most efficient way to dig for/find gold? Would that be less fun? I suspect the answer might be yes to all of those, but it's something to think about.
