L-theanine is an amino acid found in tea and almost nowhere else in the diet. Put it alongside caffeine in a controlled trial and the caffeine behaves differently, though not always in the direction the supplement industry suggests.
The combination is one of the most studied pairings in the nootropics world, and it is sold on the promise of alertness without the edge. The trials are more interesting than that, because two of the better-designed ones found the theanine cancelled the caffeine out. What separates the studies that worked from the ones that did not turns out to be the ratio.
What L-theanine actually is
L-theanine is a non-protein amino acid produced by the tea plant, Camellia sinensis. Green tea, black tea and white tea all carry it. Coffee carries none, which is the whole reason the comparison exists. Two drinks deliver caffeine, one of them arrives with a second compound attached.
The amount in a cup is small. Brewed tea gives you somewhere in the region of 5 to 25 mg depending on the leaf, the quantity and how long it sits in the water. Every trial described below used a supplemental dose, and the smallest of them was double what a strong cup of tea would give you. This matters when people describe drinking tea as getting the benefits of the combination, because the doses are not close.
In the brain, theanine crosses the blood-brain barrier and increases alpha-band oscillatory activity, the electrical pattern associated with relaxed, wakeful attention. That mechanism is well described. Whether it produces anything you would notice is the part the trials disagree on.
What happens when you take them together
Six controlled trials give a reasonable picture. All are randomised and placebo-controlled, most are crossover designs, and they span doses from a light cup of tea to a strong pre-workout.
| Study | Caffeine | L-theanine | Ratio | What the combination did |
|---|---|---|---|---|
| Kelly, 2008 | 50 mg | 100 mg | 1:2 | Raised hit rate and target discriminability. Caffeine alone improved discriminability but not hit rate. |
| Owen, 2008 | 50 mg | 100 mg | 1:2 | Improved speed and accuracy on attention switching, and reduced susceptibility to distraction. |
| Haskell, 2008 | 150 mg | 250 mg | 1:1.7 | Beat caffeine alone on simple reaction time, working memory speed and sentence verification accuracy. |
| Nawarathna, 2025 | 160 mg | 200 mg | 1:1.25 | In sleep-deprived adults, better accuracy and a 38 ms larger reaction time gain than placebo. |
| Rogers, 2008 | 250 mg | 200 mg | 1:0.8 | Cancelled caffeine's blood pressure rise. No effect on jitteriness, alertness or mood. |
| Dodd, 2015 | 75 mg | 50 mg | 1:0.67 | Eliminated caffeine's cognitive and mood benefits along with its vasoconstriction. |
Read the right-hand column from the top and the results are good. Read the last two rows and they are not. The table is sorted by ratio, which is the clearest way to see what is going on.
The four that worked
Simon Kelly's group at the Nathan Kline Institute ran 100 mg of theanine with 50 mg of caffeine while recording EEG during a cued attention task. The combination raised both hit rate and the ability to tell targets from distractors. Caffeine on its own improved discriminability and left hit rate alone, so the second effect belonged to the pairing (Journal of Nutrition, 2008).
Owen and colleagues used the same doses on 27 people, testing at 60 and 90 minutes with a week between treatments. Caffeine alone lifted subjective alertness and, eventually, accuracy. The combination improved speed and accuracy together on attention switching, and made people less susceptible to distracting information in a memory task at both time points (Nutritional Neuroscience, 2008).
Haskell's team went higher, 150 mg caffeine and 250 mg theanine. Caffeine alone did the expected work on vigilance and mental fatigue. Adding theanine produced faster simple reaction time, faster numeric working memory, better sentence verification accuracy, lower tiredness ratings and higher alertness ratings on top of that. Headache ratings also came down (Biological Psychology, 2008).
The most recent one is the most relevant to anyone running on short sleep. Thirty-seven adults were kept awake overnight, then given 200 mg L-theanine with 160 mg caffeine or a placebo before a traffic-hazard discrimination task, with EEG recording throughout. The combination improved hit rate and target-distractor discriminability, and the reaction time gain was 38 ms larger than placebo. The P3b brain response grew in amplitude and shortened in latency, which the authors attribute to faster recruitment of the brain's attentional networks (British Journal of Nutrition, 2025).
The two that did not
Peter Rogers ran 250 mg of caffeine against 200 mg of theanine in 48 people. Caffeine raised alertness, jitteriness and blood pressure. Theanine flattened the blood pressure response and did nothing measurable to the jitteriness, the alertness or any other part of mood. It also slowed overall reaction time on a visual probe task (Psychopharmacology, 2008). The relaxing reputation did not survive contact with the measurement.
Dodd's 2015 study is the awkward one. It deliberately used low doses chosen to reflect the ratios in actual tea, 75 mg caffeine and 50 mg theanine, across 12 habitual and 12 non-habitual caffeine consumers, with cerebral blood flow tracked by near-infrared spectroscopy. Caffeine alone improved attention performance and mood and produced the expected vasoconstriction. Adding theanine removed the vasoconstriction, which was the hypothesis. It also removed the improvement in attention and the improvement in mood (Psychopharmacology, 2015). The authors' conclusion is blunt about it: at those levels the interaction is real, and it did not lead to a positive impact on behaviour.
Worth sitting with that one. It is the study closest to what you actually get from drinking tea, and in it the theanine subtracted from the caffeine.
The pattern in the ratios
Sort the six trials by how much theanine sits against each milligram of caffeine and the split is clean. Every trial where theanine matched or exceeded the caffeine found something. Both trials where theanine came in below the caffeine found nothing useful, and one found the caffeine's own benefits erased.
Treat that as an observation rather than a finding. Six trials run by separate labs on separate tasks do not add up to a controlled comparison of ratios, and the two null results also used the lowest absolute theanine doses in the set, 200 mg and 50 mg. Dose and ratio move together here, so you cannot fully separate them.
What you can take from it is a practical floor. The evidence supporting this combination sits almost entirely at theanine doses of 100 mg and up, with theanine at least matching the caffeine. Product claims built on the tea ratio are quoting studies that did not work.
What theanine does without caffeine
Taken alone and acutely, not much that helps. In Haskell's trial, theanine on its own increased headache ratings and reduced correct answers on serial sevens, a mental arithmetic task. That is a mild negative, and it comes from the same study that found the combination working well.
Taken daily over weeks, the picture changes. A four-week randomised crossover trial gave 30 healthy adults 200 mg a day. Scores fell on the Self-rating Depression Scale, trait anxiety and the Pittsburgh Sleep Quality Index, with the sleep improvements concentrated in how long people took to fall asleep and how disturbed their sleep was. Verbal fluency and executive function scores improved (Nutrients, 2019).
Acute effects and sustained effects are separate questions, and the answers differ. L-theanine holds no authorised health claim in Great Britain, so everything above is a description of what the trials measured and nothing more.
What this means if you drink coffee
Adding theanine to a coffee habit will not fix the underlying arithmetic. If you are at 400 mg a day, your brain has adapted to it. When coffee feels like it has stopped working, tolerance is a likely reason, and no amino acid touches that. We covered how that builds in the caffeine withdrawal timeline. For the way back, our guide to a caffeine tolerance reset covers the timing.
The honest read of the evidence is narrower and more useful. At a caffeine dose in the 50 to 160 mg range, with theanine matched or higher, several trials show better accuracy, better attention switching and more resistance to distraction than the same caffeine on its own. Those are moderate doses. The combination has been tested as a way to make a small amount of caffeine work harder, and that is the use it has evidence behind.
If you do not know your current daily figure, start there. Most people are out by a wide margin once tea, cola and anything taken before the gym goes into the total. Our caffeine calculator works it out in about a minute and shows how much is still circulating at bedtime.
Where this sits in a taper
The pairing is the reason Reset is built the way it is. Each Reset capsule carries 40 mg of caffeine with 50 mg of L-theanine, and it steps down by capsule count over 20 days, so nothing depends on you judging a dose by eye.
On the opening phase that comes to 160 mg of caffeine against 300 mg of L-theanine across the day, counting the theanine in both capsules. About 1 to 1.9, which sits inside the band where the trials found effects, and the caffeine figure is the same 160 mg used in the 2025 sleep-deprivation study.
Flow runs alongside at a constant dose for all 30 days and carries no added caffeine, which is what allows the protocol to reach zero while the theanine stays constant. It also includes magnesium, which contributes to the reduction of tiredness and fatigue and to normal psychological function, and pantothenic acid, which contributes to normal mental performance.
Preorders are open for the 30-day reset kit. Kits ship by 30 October. The full reasoning behind the step sizes and the ingredient choices is on the science page.
The short version
L-theanine and caffeine together do something caffeine alone does not, and the effect shows up most reliably in accuracy and attention. Raw wakefulness is the part caffeine already handles by itself. It needs theanine at 100 mg or more, at or above the caffeine dose. Below that the evidence thins out fast, and at tea-like ratios one good study found it working against the caffeine entirely.
Anyone selling you the tea ratio is selling you the studies that failed.
