Week 3: Reward Loop Architecture

We’re entering the halfway point of the course: this is where we step up the complexity and start connecting the dots. Strap yourself in; this is going to be a fun one.

Why we’re covering this

  • Simplistic habit loop diagrams are everywhere, but behavior isn't simple. You need  more than just ‘cue → action → reward’ to be successful tackling tough problems.
  • There are many different ways to create reward loops; the most effective and retentive approaches typically involve layering many different kinds.
  • The majority of important outcomes require a behavior to be repeated many times, so to create the greatest impact, you'll need to keep people engaged.


Our spin on it

  • While oversimplifying things is to be avoided, the lens of operant conditioning can be a useful one for analyzing or designing iterative behaviors—if you approach things with the appropriate level of nuance
  • Way more things are a loop that you might think. You’ll start to see reward loops (and their little sibling, microloops) everywhere and know how to repair them.
  • Negative reinforcement, reciprocal determinism, and cognitive misfires around reward probability and behavior cost are under-appreciated tools for understanding reward loops. (You'll have these down-pat by the end).


Finishing this week’s lesson might be like traveling through the Matrix: you might start seeing things very differently afterwards. And if you're able to systematically create rewarding experiences that drive behavior, there's no limit to what you'll be able to do.

🎉 Let's get this party started.

Start lesson →