Andrew Huberman· PhD
So what you are doing is you are applying reward prediction error to the entire sequence of things that's involved in getting into the habit execution, getting through the habit execution, and getting out of the habit of execution.