Here’s a simplistic model of what I’ll call “preparatory” learning: you’d like to become a doctor, so you attend medical school and take classes for your first two years, and then in your third year you’re ready to see some patients (as a clerkship, with supervision). This style of learning is typical in formal educational contexts. The primary activity in those first two years is skill-building itself; you’re preparing yourself to participate in some future activity where the actual intrinsic value resides.
By contrast, in a “just-in-time” learning context, you learn things when you find you need them to do some other primary activity. For example, maybe you’re a medical resident, and you found yourself unable to follow along with the discussion of a patient who seems to have a movement disorder. You might then go home and review a book on the topic, hoping both to help your patient and to become more capable for future patients like them.
I think these “just-in-time” contexts are likely to be better for developing the Mnemonic medium, since they’ll more clearly demonstrate the connection between the medium and meaningful enablement.
“Preparatory” learning contexts are abstracted away from the activity we’re actually trying to enable. A course syllabus (or other structure) dictates the details you should learn. Students will tend to use and value the medium if the course’s feedback loop (problem sets / tests) rewards them for it. But we want the medium’s evolution to be driven by enabling people to do meaningful things, not by helping them get better grades. (The mnemonic medium should be developed in a context where learning is in service of some meaningful use) Grades will distort our work. For example, if cramming produces a decent result on their tests, that might seem like a better choice than the mnemonic medium. And we could easily find ourselves helping students get good test without expanding their capacity to do anything meaningful.
Practically speaking, this might mean:
One challenge here is that just-in-time learning usually can’t happen when you’re a total novice, since you often don’t know enough to begin a meaningful project. But that’s where the medium really shines: Spaced repetition memory systems can accelerate the unpleasant early stages of learning a subject. It’d be great if we could find some situation where people often find they need to learn something in the course of something meaningful, but they really do need to start with the basics.