Dopamine Index
Caffeine
In plain English
Caffeine does interact with dopamine, but not the way most people picture. Real caffeine changed how many dopamine receptors were available afterwards, rather than clearly flooding the brain with dopamine. Oddly, simply expecting caffeine, without actually taking any, produced a genuine dopamine release on its own. So the story is not a simple stimulant hit — belief and pharmacology both play a real, separately measured part.
The evidence
- Evidence
- BB — Good
- Human evidence
- Yes · Two independent human PET studies directly measured dopamine changes.
- Studies reviewed
- 3
- Verdict
- Explainer — no single claim judged
- How sure are we?
- Uncertain
- Last reviewed
- Grade status
- Provisional — awaiting editorial validation
Short answer
Caffeine is the stimulant compound found in coffee, tea and many soft drinks, and human PET studies confirm it interacts with the dopamine system — but not in the simple way most people assume. Real caffeine at normal doses appears to increase the number of available dopamine D2/D3 receptors rather than clearly flooding the brain with extra dopamine, while merely expecting caffeine (a placebo pill people believed might be caffeine) released measurable dopamine on its own. For the beverage itself, with its hundreds of other compounds, see the separate coffee entry.
- Evidence
- BEvidence grade B · Good human evidence with limitations
- Why this grade
- Two independent human PET studies directly measured dopamine-system changes with caffeine — one showing placebo caffeine (expectation alone) reduces raclopride binding, consistent with dopamine release 1, and one showing real caffeine increases D2/D3 receptor availability rather than clearly releasing dopamine 2. The findings are real, direct measurements, but they point in different directions and the literature is small, so grade B rather than A.
- Last reviewed
- 24 September 2026
The quick explanation
Caffeine blocks adenosine receptors, which indirectly changes how the dopamine system behaves, and PET scans have measured this in people rather than just guessing at it. One study gave people a placebo pill and told them it had a 50% chance of being caffeine: their brains released real dopamine in the thalamus just from that expectation. A separate study giving people actual caffeine found something different — more available dopamine D2/D3 receptors afterwards, which the researchers interpreted as receptor upregulation rather than a straightforward dopamine surge. Both are genuine PET findings; they simply show the dopamine story around caffeine is more layered than ‘caffeine equals a dopamine hit’.
What it is
Caffeine is a naturally occurring stimulant compound found in coffee, tea, cacao, guarana and many soft and energy drinks, and is the world’s most widely used psychoactive substance.
Does dopamine play a role?
Partly, and in an unexpected direction — human PET studies find real dopamine-system changes with caffeine, but the clearest direct dopamine release measured so far came from expecting caffeine, not from the drug itself.
Human evidence
Caffeine has one of the more interesting entries in this index because two well-designed human PET studies on it point in different directions, and both are worth taking seriously rather than picking whichever fits a simpler story.
The first is a genuine landmark. Eight habitual coffee drinkers were scanned twice with [11C]raclopride PET: once with no treatment, and once after being given a placebo pill and told they had a 50% chance it contained caffeine (it never did). The placebo condition produced a significant 15% drop in raclopride binding in the thalamus — evidence of real dopamine release triggered purely by the expectation of caffeine.1
The second study gave 20 healthy adults actual caffeine (300mg, roughly two to three cups of coffee) versus placebo, and scanned striatal D2/D3 receptors with PET. Real caffeine significantly increased receptor availability in the putamen and ventral striatum, and this increase tracked with how much more alert people felt.2
At first glance those two findings look contradictory, but they measure different things: a drop in binding (as with placebo) is usually read as more dopamine competing with the tracer, while a rise in binding (as with real caffeine) points the other way — towards more available receptors rather than more dopamine flooding the synapse. The study authors were explicit that this pattern argues against real caffeine simply releasing a wave of dopamine in the striatum, at least at the dose and receptor sites they measured.2
| Study | What was given | PET finding | Interpretation |
|---|---|---|---|
| Kaasinen et al.1 | Placebo pill, expectation of caffeine | 15% drop in thalamic raclopride binding | Dopamine release from expectation alone |
| Volkow et al.2 | Real caffeine, 300mg oral | Increased D2/D3 receptor availability (putamen, ventral striatum) | Receptor upregulation, not a clear dopamine surge |
Animal evidence
Older animal work reviewed in a 1999 paper on caffeine dependence found that, at doses comparable to typical human intake, caffeine did not clearly release dopamine in the nucleus accumbens shell — the region most tied to the reinforcing effects of classic drugs of abuse — though it did affect dopamine activity in the prefrontal cortex. Only much higher doses, well beyond ordinary consumption, produced broader, non-specific effects across brain structures.3
That review is dated and mostly describes rodent microdialysis data rather than modern human imaging, but it is consistent with the more recent human PET work above: caffeine’s relationship with dopamine looks different from, and generally weaker than, that of drugs specifically built to hijack reward circuitry.3
Mechanism
Caffeine’s primary, well-established action is blocking adenosine A1 and A2A receptors. Adenosine normally dampens neural activity, so blocking it increases arousal broadly across the brain, including in circuits that also carry dopamine signalling. Because A2A receptors physically sit alongside D2 receptors in the striatum and can change how they behave, caffeine’s dopamine-related effects are widely thought to be indirect, running through this adenosine-dopamine interaction rather than caffeine acting on dopamine neurons directly.2
Placebo caffeine vs real caffeine
It remains unresolved why an expectation of caffeine produced a classic ‘dopamine release’ PET signature in one study, while a real, larger dose of caffeine produced the opposite PET pattern in another. Different brain regions, doses, populations and scan timing are all plausible explanations; no study has yet directly compared placebo and real caffeine in the same design and dose to settle it.12
The practical takeaway is that ‘caffeine dopamine hit’ is a much less settled phrase than it sounds. The honest, evidence-based version is narrower: caffeine and the expectation of caffeine both measurably change the dopamine system in humans, in ways linked to alertness, but the mechanism is not the simple flood implied by comparisons to stimulant drugs of abuse.2
There is also a dose and delivery question worth flagging. Both PET studies used a defined oral dose given at one point in time — a placebo capsule in one case, 300mg of caffeine in the other — which is a controlled way to test pharmacology but does not resemble how most people actually take in caffeine, sipped gradually across a morning in coffee, tea or soft drinks alongside hundreds of other compounds.1
That gap between a single laboratory capsule and everyday drinking is exactly why this site keeps caffeine (the isolated chemical, tested this way) and coffee (the drink, with everything else it contains) as separate entries rather than treating them as interchangeable.2
Evidence strength
Two independent human PET studies directly measured dopamine-system changes with caffeine — one showing placebo caffeine (expectation alone) reduces raclopride binding, consistent with dopamine release 1, and one showing real caffeine increases D2/D3 receptor availability rather than clearly releasing dopamine 2. The findings are real, direct measurements, but they point in different directions and the literature is small, so grade B rather than A.
Common claims
| What people say | What the evidence says |
|---|---|
| Caffeine floods your brain with dopamine like a stimulant drug | misleading — the receptor pattern for real caffeine looks different from a classic dopamine surge 2 |
| Even a placebo coffee can give you a mental lift because you expect caffeine | true, and this has been measured directly with PET 1 |
| Caffeine works purely by blocking adenosine, dopamine is irrelevant | unproven as an absolute statement — both PET studies here found the dopamine system did change, just not in the way a simple ‘blockade’ story predicts 1 2 |
What the evidence supports
What the evidence does not support
- Real caffeine increased receptor availability rather than producing the receptor-binding drop typically read as a dopamine surge, which the study authors explicitly say argues against a simple ‘caffeine releases dopamine’ story 2.
- Animal studies reviewed alongside the human PET work found that caffeine at ordinary human doses does not clearly act on the nucleus accumbens shell the way classic drugs of abuse do 3.
What we know
- In habitual coffee drinkers given a placebo pill and told it might be caffeine, PET scans showed a significant 15% reduction in thalamic raclopride binding compared with no treatment, consistent with real dopamine release triggered by expectation alone 1.
- In that same study, self-reported arousal after the placebo correlated with tracer binding changes in the putamen 1.
- In a separate PET study, a single 300mg oral dose of real caffeine significantly increased D2/D3 receptor availability in the putamen and ventral striatum (but not the caudate) compared with placebo, in 20 healthy adults 2.
- That increase in receptor availability was associated with caffeine-induced increases in alertness, and the authors argued the pattern reflects more available receptors rather than a straightforward dopamine surge, because a surge would be expected to reduce, not increase, receptor availability 2.
- An earlier review of caffeine dependence noted that, at typical human doses, caffeine does not clearly release dopamine in the nucleus accumbens shell (the region most associated with drug reward) in animal studies, unlike classic drugs of abuse 3.
What we don’t know
- Why placebo caffeine appears to release dopamine directly 1 while real caffeine instead shows increased receptor availability 2 — whether this reflects different brain regions, different doses, or a real mechanistic difference between expectation and pharmacology has not been resolved.
- Whether regular, heavy coffee drinkers show a different pattern than the occasional users typically recruited into these small PET studies.
- How much of everyday caffeine’s alerting effect in habitual users is dopaminergic versus driven by adenosine receptor blockade acting through other systems.
How sure are we?
- Very confident
- Reasonably confident
- Uncertain
- Very uncertain
Uncertain. Both PET findings are real and direct 12, but they point in different directions and the literature is small, so the overall mechanism is not settled.
Confidence describes how settled the answer on this page is, not how important the topic is. It can change as new research is published.
Where scientists disagree
Placebo caffeine produced a classic dopamine-release signature (less receptor binding), while real caffeine produced the opposite pattern (more receptor availability) in a different study 12. One reading is that expectation and pharmacology act through different routes; another is that the difference reflects dose, brain region or timing rather than a true mechanistic split.
Why studies may disagree
No study has directly compared placebo and real caffeine in the same design and dose, so the two findings cannot yet be reconciled with certainty 12.
What would change the answer?
The evidence that is currently missing:
- A single study directly comparing placebo and real caffeine at the same dose and timing
- Larger PET samples across habitual and occasional caffeine users
- Studies using naturalistic, gradual dosing rather than one capsule at one timepoint
Common misconception
Caffeine gives you a dopamine hit the same way cocaine or nicotine does.
The clearest human PET evidence for caffeine shows changes in receptor availability, not the same pattern of direct dopamine release seen with stimulant drugs, and animal work suggests caffeine at normal doses does not clearly act on the brain’s core reward circuitry the way classic drugs of abuse do 2 3.
Key studies
- PET study, 8 habitual coffee drinkers: a placebo pill believed to have a 50% chance of being caffeine produced a significant 15% drop in thalamic raclopride binding, indicating real dopamine release from expectation alone.1
- PET study, 20 healthy adults: a single 300mg dose of real caffeine significantly increased striatal D2/D3 receptor availability compared with placebo, linked to increased alertness.2
- 1999 review of human and animal caffeine dependence data: at doses reflecting typical human consumption, caffeine did not clearly release dopamine in the nucleus accumbens shell in animal studies, unlike classic drugs of abuse.3
The bottom line
Caffeine genuinely changes the dopamine system, but the evidence does not show a simple dopamine hit like classic stimulant drugs. Expecting caffeine can itself release dopamine, which is a striking finding worth remembering on its own.
Related questions
The Brief
Understand your brain a little better each week.
Join The Brief from dopamine.co.uk. Free. No spam, no supplements, no selling your address. Unsubscribe in one click.
References
- 1. Kaasinen V, et al. Expectation of caffeine induces dopaminergic responses in humans. The European journal of neuroscience. 2004. PMID 15090062. doi:10.1111/j.1460-9568.2004.03310.x
- 2. Volkow ND, et al. Caffeine increases striatal dopamine D2/D3 receptor availability in the human brain. Translational psychiatry. 2015. PMID 25871974. doi:10.1038/tp.2015.46
- 3. Nehlig A Are we dependent upon coffee and caffeine? A review on human and animal data. Neuroscience and biobehavioral reviews. 1999. PMID 10073894. doi:10.1016/s0149-7634(98)00050-5