💭 Thoughts on the article A primer on dopamine.

I really enjoyed reading about dopamine, desire and pleasure here! It’s worth a read, even if it slightly veers into the “how to abuse this knowledge for personal gain with your new product” space.

I’ve been (ha!) pleasantly surprised by how this article pulls lots of different concepts — half-understood by me — together. I’ll be using it to try and provide motivation for myself while doing things that won’t pay off for a while (hello 2023/24 tax return).


The more interesting part is that dopamine (and hence desire) is intensified when we get positively surprised.

our brain is constantly betting: which choice can lead to the highest gain for the minimum investment.

Games are problems people pay to solve.

You can imagine dopamine as signalling the value of work.

all classic high-dopamine activities are low-effort, high-reward situations

It’s easy to confuse “liking” and “wanting”

If you’re addicted to something, you have an intense desire which fails to resolve into an equally intense pleasure.

desire is generated after encountering something rewarding.

Because dopamine is typically generated after pleasure is felt, it is the “desire” molecule, not the “pleasure” one (as it is commonly mistaken).

even in the absence of perfect information, we have to take actions.

we tend to choose an action with the highest expected gain of our wealth

The gap in knowledge about the reward generating process is intensely motivating

the brain is a prediction machine for rewards (I do this -> I get this) and dopamine represents errors in predicting rewards.

Higher the surprise, higher the reward prediction error, higher the activity of dopamine neurons.

dopamine helps in learning by adjusting the expected reward to be closer to the actual reward, so that next time cost-benefit calculations have less error. This is why the music that sounded strange the first time starts sounding a little better the next time.

since decision-making is energetically expensive, it makes sense for rewarding behaviors to become habitual and automatic over time. Neuroplasticity literally ensures frequently rewarded behaviors become more and more efficient (and hence automatic over time). This is why, once spotted, you’re unable to stop yourself from opening that bag of chips.

Information about a reward is (almost) as valuable as the reward itself, so (safe) exploration of the environment producing intermittent rewards makes sense. This is why we can’t stop checking social media feeds, as it’s impossible to predict when and how we get socially valuable rewards in it, so the brain remains hooked.

it triggers many of the adaptive behaviors of learning, exploration and automaticity.

If you were expecting a reward and get exactly that, your brain doesn’t increase dopamine production. That’s because there’s no point in optimizing things even further, since you got what you were expecting. No further learning, exploration or automaticity is needed.

Desire (to explore/try/want) is generated when learning is incomplete.

we actively seek high-dopamine producing situations

we DON’T mind occasional lack of expected rewards as long as we get unexpected rewards.

For video games, dopamine’s behavior suggests that continuing to unseat expectations is the key job.

Fun is pleasure with surprises.

We love stories because they’re partial puzzles for our brain to solve.

From the associated tweet:

Stories have a 4 part structure that teaches us an essential lesson.

- Characters (are they like me)
- Conflict (what issue they faced)
- Struggle (what did they do)
- Resolution (did they solve it)

We _tell_ stories because it's the way we share how to solve problems.
We _listen_ to stories because they're the best way to learn how somebody else solved a problem.

I’m not sure I agree that this is all that stories are, but to the extent that it can be fun and engaging to connect with a story, I think these are the angles that make sense.

Plus, why everything looks the same now.

Hint: that’s because before beating expectations, you have to meet them.

I don’t agree with the base premise here, I think this is inverted. You can set yourself apart from expectations & reset them; if you end up looking like others then you have to meet all the expectations folks have of everyone else, but if you can find a niche and be different enough that people roll back their higher-level expectations, then you can carve out some particularly interesting reward.

we find it joyful to put in (some) effort to figure out a reward

Additionally, from my time at Tesco working on Price Drop, if someone doesn’t have to work to derive reward then it instantly becomes expectation, and pleasure can’t be derived from that source any more.