I got my thoughts in order to compose the serotonin lesson, realized that I didn't quite understand how it worked, and decided I needed more research.

...and of course the field has changed so much since I first learned it 10 years ago that my understandings are now completely wrong. Seriously, what I learned 10 years ago is now the complete opposite of the current consensus.

Isn't it always the case - you think you know something until you try to explain it.

OK, 2/3 of this particular lesson arc was mentioned by David Huberman in a recent podcast, he's a Stanford researcher, so I think I'm OK for those parts.

On balance, I've decided to keep the serotonin explanation as is, with a big caveat that I might be getting the names wrong. I think the effects and actions are correct, but I might have the names of the messenger chemicals wrong. There's no good consensus on what serotonin does anyway, so this might be right.

I can revisit this later to correct errors, but I think the described process is correct.

Generally: The brain has a differential reward detection circuit (officially, the "dopamine reward prediction error") that fires when there is a perceived change of value. This is an error feedback that detects *changes* in state, and acts to train our value predictor.

This feedback signal explains many features of extrinsic motivation: time extinction, initial excitement, and so on. Also some child behaviors.

You get an idea, and the predicted reward is suddenly high (high differential) so you're psyched to do the project. A couple of days later and the reward value hasn't changed, so the differential is zero and your initial excitement has died down. Dopamine from progress towards the goal should have taken over by that time to carry you through to the end. This has implications for projects that require ordering parts online - by the time the parts arrive, your initial excitement has died down and you're no longer motivated to do the project.

Explaining this in a simple, easy-to-understand way is a challenge, and simplifying it will make it wrong anyway, and I couldn't find a definitive correct explanation online, so I decided to just keep the original version that sounds simple and logical, but is wrong.