Cheat codes

Most things worth do­ing take se­ri­ous, sus­tained effort. If you want to be­come an ex­pert vi­o­linist, you’re go­ing to have to spend a lot of time prac­tic­ing. If you want to write a good book, there re­ally is no quick-and-dirty way to do it. But sus­tained effort is hard, and can be difficult to get rol­ling. Maybe there are some eas­ier gains to be had with sim­ple, lo­cal op­ti­miza­tions. Con­trary to oft-re­peated cached wis­dom, not ev­ery­thing worth do­ing is hard. Some lit­tle things you can do are like cheat codes for the real world.

Take habits, for ex­am­ple: your habits are not fixed. My diet got dra­mat­i­cally bet­ter once I figured out how to change my own habits, and ac­tu­ally ap­plied that knowl­edge. The gen­eral trick was to figure out a new, sta­ble state to change my habits to, then use willpower for a week or two un­til I set­tle into that sta­ble state. In the case of diet, a sta­ble state was one where junk food was re­placed with fruit, tea, or hav­ing a slightly more sub­stan­tial meal be­fore­hand so I wouldn’t feel hun­gry for snacks. That’s an equil­ibrium I can live with, long-term, with­out need­ing to worry about “fal­ling off the wagon.” Once I figured out the pat­tern—work out a sta­ble state, and force my­self into it over 1-2 weeks—I was able to im­prove sev­eral habits, per­ma­nently. It was amaz­ing. Why didn’t any­body tell me about this?

In ed­u­ca­tion, there are similar easy wins. If you’re try­ing to com­mit a lot of things to mem­ory, there’s solid ev­i­dence that spaced rep­e­ti­tion works. If you’re try­ing to learn from a difficult text­book, read­ing in mul­ti­ple over­lap­ping passes is of­ten more time-effi­cient than read­ing through lin­early. And I’ve per­son­ally wit­nessed sev­eral peo­ple aca­dem­i­cally un-crip­ple them­selves by learn­ing to re­flex­ively look ev­ery­thing up on Wikipe­dia. None of this stuff is par­tic­u­larly hard. The prob­lem is just that a lot of peo­ple don’t know about it.

What other easy things have a high marginal re­turn-on-effort? Feel free to in­clude spec­u­la­tive ones, if they’re testable.