Design for the Cliff: Refresh the Surface, Keep the Cue

The obvious move is "add novelty back", and that's right, but there's a trap in it that will quietly undo everything you built in 8.3, so we have to be precise. Here's the tension, stated honestly: Lesson 8.3 said stable context is the carrier, don't change the cue. Lesson 8.4 says novelty fades, add novelty back. Those sound like opposite instructions, and if you resolve them wrong you'll "freshen things up" by blowing up the very cue that was holding the habit. The resolution is a clean distinction: refresh the surface, never the cue. Keep the trigger, the time, the place, and the core behavior fixed, that's the carrier, that's load-bearing, that's what's slowly fusing into automaticity. Rotate novelty into everything around it: the reward (swap the token from 7.2 for a new one; make it variable, a surprise reward restores the prediction-error hit precisely because it's unexpected again), the flavor (new playlist for the walk, new location for the same desk-work block, a new route on the same run), the way you track it (a fresh log, a new visual). You're re-spiking the dopamine that carries you through the boring middle without touching the cue→behavior wire that's doing the structural work. Think of it as repainting the car, not rebuilding the engine. This is also where a popular bit of ADHD folk wisdom earns a careful mention. You'll hear the advice, often credited to the "interest-based nervous system" idea, to rotate through dozens of techniques because novelty wears off. Be honest about what that is: the "interest-based nervous system" is clinical folk theory, not a validated construct, there's no study testing it as such. But its core intuition, rotate novelty because the pull decays, maps directly onto the real prediction-error and habituation mechanisms we just covered, so the practice is sound even though the label is unverified. The trap inside that advice is the cue trap again: if "rotate techniques" means you start a brand-new system every three weeks, you never let any cue→behavior wire reach automaticity, you just live in a permanent, exhausting honeymoon-and-crash loop, which is the ADHD pattern, not the cure. So take the useful half and refuse the harmful half: rotate the reward and the surface to beat habituation; hold the cue and the behavior constant so the habit can actually form underneath. Novelty is the fuel that gets you across the months; the stable cue is the road.