My unscientific take on this is it's about micro adjustments and just in time feedback. By focusing on where you ARE going, you will make that happen subconsciously through your motor skills. If you are going the wrong way (in micro amounts), you will self-correct.
I tried to dig out the references from my mess of bibliography (I worked on this topic long before Mendeley and alike appeared), unsuccessfully.
I remember that I ran into this technique independently first (in a context of encoding expression trees) and then found a number of other examples in the literature (e.g., encoding the neural network topology, etc.).
To play devil's advocate, there's certainly a limit to computer-based searches. If the computer's model (which was made by a person or persons) is under-specified, then it will produce suboptimal results. At least with a person, there is always the possibility of integrating new facts on the fly.
Another angle is a groupon type service. Do something fun together, get a coupon for food/drink/another fun activity that unlocks when all members on the group reunite in a month. Create mechanism for 'chaining' to encourage stickiness. People get slight push to maintain friendships they might let otherwise let slip away.
Haha, testing automation framework. Test engineers took our framework and wrote tests with it. Yes Perl professionally is not much fun but I fell in love with Larry Wall after reading the Perl book. Though Perl isn't the best language, a lot of his philosophy shines through and makes it much more than just another programming language.