Stake It: Make the Future Cost Something Today
Tokens pay you for showing up.
But some things you'll skip even with a token on offer, because the pull of not doing it right now is just too strong. For those, you need the other side of the ledger: instead of only a reward for doing it, a cost for not. That's a commitment device, a constraint or a stake you set in advance, while you're thinking clearly, that binds your future self when you're not. You're using the planning brain to box in the in-the-moment brain. The cleanest demonstration: students who were allowed to set their own binding deadlines with real penalties performed better and turned work in less late than students given total flexibility, they chose to tie their own hands, and it worked. The principle generalizes hard. A self-restricting savings account, money you couldn't easily touch, produced 81% higher savings balances after a year versus a normal account. A deposit contract for quitting smoking, where you put down money you'd forfeit if you failed a later test, increased the number of people who actually quit. The engine underneath all of it is loss aversion: a staked loss hurts more than an equal gain feels good, so a small amount of money (or pride, or a promise) on the line generates more pull now than the distant payoff ever could. That's the trick, you're manufacturing an immediate consequence for a behavior whose natural consequences are all far away. Now the honest, genuinely useful catch, because it'll determine whether yours works: even when people choose to commit, they systematically under-bind. In the deadline study, people set deadlines that helped, but not aggressively enough to be optimal. You will feel the same pull: to set a stake so soft it doesn't actually bite. Resist it. A commitment device that costs you nothing to break is just a wish with extra steps. Set one real enough to make skipping genuinely uncomfortable, and just one, for the target that actually matters.
This article is drawn from the Attention Deficit course and describes population-level research, not personal medical advice. Talk to a qualified practitioner before changing anything.