Dopamine isn't just about individual reward learning. In 2005, my colleagues and I found that a dopaminergic reward prediction error signal extends to social exchange. People use this signal to build a model of others as they develop trust and cooperation. Once formed, this

She acknowledged that this was an error - that contrary to what her figure shows, there is no phasic dip after the conditioned cue when the reward follows. This mistaken view has been seen now by millions of readers and viewers - Hopefully, those who advocated for it will clarify

But, there's no evidence for this - and in fact as the Self commentary states, the paper reveals that "naturally evoked dopamine pulses are involved in both triggering and pursuit of reward-seeking behaviour." This is one of the key studies showing the exact opposite of what

At around the 22:00 mark of his podcast, @hubermanlab repeats the theory that a drop below baseline triggers the desire to obtain rewards and calls it the 'key thing':

Here's the original, well-known figure. It's from a commentary on a landmark 2003 paper (PMID: 12687000). But notice there's no phasic dip below baseline (the study itself doesn't find one either). There's no evidence from these studies of any phasic dip following a conditioned

Here's the claim below from her book - a detailed theory about a mini dopamine deficit = craving. This fits her view that addiction is driven by anhedonia - an aversive state we seek to relieve. Anyone familiar with dopamine research will know where this figure comes from. But

Some of the key claims in Lembke’s Dopamine Nation are simply wrong. One important example: Lembke's theory of craving as a dip below baseline, a claim @hubermanlab also makes in his Leverage Dopamine podcast. A 🧵on why it's wrong:

More dopamine nonsense based on outdated understanding of dopamine and wild extrapolations from substance abuse.

Even if we grant Goff that phenomenal properties aren’t reducible to dispositions, that still doesn’t justify the epistemic contrast he’s pushing. In computational models, feelings play an analogous role to black holes in physics: they are latent variables inferred from publicly

“Cheap dopamine” is a meme built on bad neuroscience: the outdated pleasure theory that equates cheap dopamine with cheap pleasures. It's really just a modern version of 19th century Thomas Carlyle's critique of “low” pleasures. But we don’t need faux dopamine talk to follow an

Pursuing worthwhile goals is good advice, but this gets the neuroscience wrong. Dopamine doesn’t deplete like a bank account, the popular “wave pool” analogy ignores how tightly it’s regulated by synthesis and reuptake, and everyday experiences don’t empty some “dopamine gas

Gravel riding isn't just fun; it sits at the center of the 'longevity protocol Venn diagram:' 1. Cardiovascular fitness across varied intensities, 2. Navigation and route-finding keep spatial circuits plastic, 3. High-demand attentional, balance, and sensorimotor control on loose

Saying calories can’t cause obesity because a calorie is “just a unit of energy” makes as much sense as saying interest rates can’t affect inflation because they’re “just a percentage.” In causal modeling (e.g., Pearl’s do-calculus), a cause is a manipulable variable where

Just because a therapy is delivered as a dietary pattern doesn’t mean its mechanism is better fuel. In epilepsy β-hydroxybutyrate acts as a signaling metabolite, inhibiting HDAC1/2, upregulating SIRT4 and GAD1, and shifting the GABA/glutamate balance, rather than simply serving

Columbia didn’t “prove” that. The paper found that, in very old adults, broad psychosocial composites correlated moderately with certain mitochondrial proteins in postmortem prefrontal cortex in a small subsample. No functional measures, no causal direction. A modest correlation

Social media doesn’t “atrophy” your cortex; it engages the same reinforcement-learning machinery that optimizes behavior in any complex environment. And social media use is well-captured by average-reward reinforcement learning models. Call it cheap reward if you want, but that's

Classic grift: wildly extrapolating from a study on gut cells in a lab dish to claiming a real-world hazard. Obviously intended to sell something no one needs, but these guys need to get a grip on the difference between a real hazard and a negligible risk no one needs to worry

More dopamine nonsense: there's no such thing as a 'dopamine reset.' Doomscrolling in the morning might be a bad idea, but not because it's ruining an imaginary reset. Your dopamine system isn't a dishwasher with a reset button...

Nina wildly overstates what the data show. 50% “reversal in 10 weeks” comes from a keto program in selected patients, not the population; 79% “less depression/anxiety” comes from small, uncontrolled studies. Recent meta-analysis found keto no better than high-carb for depression

The explore-exploit program isn't just about individual food choices - it's why Apple rolls out a new iphone every few years, why there's buzz about a new restaurant, and why one day everyone's wearing an Ed Hardy t-shirt then they suddenly disappear. It’s the same adaptive

Why are restrictive diets like keto so hard to maintain? Your brain runs an explore–exploit program. Exploitation repeats the best‑known option; exploration pushes you to try something new. When meals get repetitive, their reward value drops. That feels like boredom and cranks