Most of what we publish here is about separating real evidence from marketing hype. L-theanine is a good example of both a strong evidence base and an honest gap in how we currently sell it: it's already in two products in our range, but not yet as a standalone, individually-dosed capsule you can buy on its own. Here's what the evidence actually supports, and where dose transparency still needs work.
What the evidence actually shows
L-theanine is an amino acid found almost exclusively in tea leaves (Camellia sinensis), which is why "a cup of green tea to calm down" shows up so often in relaxation folklore. The clinical evidence is narrower and more specific than that folklore suggests — and in some places, more interesting.
A 2026 systematic review and meta-analysis pooled 31 randomized controlled trials (1,168 participants) on L-theanine's cognitive and affective effects.1 One finding is genuinely solid by supplement standards: a single 200mg dose, taken 30–60 minutes before cognitive testing, significantly improved choice reaction time, SMD 0.51 (95% CI, 0.25–0.77) — a real, measurable attention effect, and the one the reviewers themselves call robust.
Now the rest of the outcome list, which is where an ingredient's evidence usually gets quietly tidied up. Acute stress: a modest SMD of 0.31, which the reviewers state was largely influenced by studies with a high risk of bias — so the honest reading is that the stress signal rests disproportionately on the weakest studies in the set. Anxiety: the review's own words are "inconsistent and non-significant," with one exception, a psychotic-anxiety study using 400mg/day for eight weeks (SMD 0.54). Not "mixed." Non-significant. Depressive symptoms: SMD 0.69 (95% CI, 0.13–1.25) and statistically significant — but reached after excluding an outlier study, in healthy individuals on a single dose, which is a long way from an antidepressant claim. Fatigue: no significant effect at all. No serious adverse events were reported anywhere in the pooled data.
So the scorecard is one robust result, one that leans on high-bias studies, one that depends on dropping an outlier, one non-significant, and one flat null. The reviewers' own summary is worth borrowing verbatim: L-theanine "is safe and shows a robust short-term benefit on attention in healthy adults, with a potential antidepressant effect warranting confirmation in high-quality trials." Attention is the claim. Everything else is a hypothesis.
That is still a real, multi-trial evidence base for a specific dose and a specific outcome — which is the bar we use to decide what earns a place in the range. But "real" and "large" aren't the same thing, and the inconsistency across outcomes (robust on attention, modest and bias-prone on stress, non-significant on anxiety, null on fatigue) is itself informative about what this ingredient actually does.
It's also worth being honest about the limits of a pooled analysis like this one. Thirty-one trials sounds like a lot, but the individual studies varied in dose, timing, testing method, and how "stress" or "anxiety" was measured — some used validated psychometric scales, others relied on single self-report items or salivary cortisol. That heterogeneity is exactly why the review's authors flagged higher risk of bias on the stress and anxiety outcomes specifically, while the attention finding held up better across more consistently designed trials. When we say the evidence is "modest," this is the kind of detail we mean: a real signal, in a body of research that isn't yet uniform enough to call it settled.
The mechanism, and why the caffeine combination is a different story
Part of why the anxiety and stress data stays inconsistent is that L-theanine's mechanism doesn't map cleanly onto a single symptom. It's structurally similar to glutamate and is thought to modestly increase brain GABA activity while damping excitatory glutamate signaling, and EEG studies consistently show it increases alpha-wave activity — a brain-wave pattern linked to relaxed alertness rather than sedation. That's a real, replicated electrophysiological effect. It just isn't the same thing as "reduces anxiety," which is a much harder clinical outcome to move, and the gap between the two is exactly where a lot of L-theanine marketing overreaches.
Where the newest evidence gets more interesting is L-theanine paired with caffeine, rather than taken alone. A 2025 double-blind, placebo-controlled crossover trial in the British Journal of Nutrition tested 200mg L-theanine plus 160mg caffeine against a starch placebo in 37 healthy young adults (aged 22–30) who had been kept awake for 17 hours — a reasonable stand-in for a bad night before an exam or a late shift, though not for chronic sleep debt.2 On a computerised traffic-hazard task, the combination improved both accuracy and reaction time. Its limitations are worth naming: a lab task rather than real driving, a single night of sleep loss rather than sustained deprivation, and the open question of whether habitual caffeine users would respond the same way. If the real-world version of this problem is yours, the countermeasures for working nights and focus under pressure in exam season are the practical companions to it.
EEG recordings backed the behavioral result up: the P3b brain-wave component tied to attention and decision-making was larger and appeared faster after the active dose, consistent with the caffeine sharpening alertness while L-theanine's alpha-wave effect kept that alertness from tipping into jitteriness. This lines up with a pattern seen in earlier caffeine-plus-theanine cognition research: the pairing tends to outperform either ingredient taken alone on attention-switching tasks — the fuller version of that is in L-theanine plus caffeine: the calm-focus combo, explained. It's a genuinely useful synergy — but it's a caffeine story as much as a theanine story, which is a different claim than "L-theanine alone reliably calms you down," the one most often made in marketing copy.
None of this changes the safety picture, which is genuinely reassuring: across the pooled meta-analysis and the newer combination trial, no serious adverse events were reported at these doses, and L-theanine has long-standing GRAS (Generally Recognized As Safe) status in the US at typical supplemental amounts. The open questions here are about how strong and how specific the benefits are — not about whether the ingredient is safe to take.
| Protocol | Dose tested | What the trial data shows |
|---|---|---|
| Standalone L-theanine, attention | 200mg, single dose | Significant improvement in choice reaction time (SMD 0.51; 95% CI 0.25–0.77) |
| Standalone L-theanine, stress/anxiety | 200mg, single dose | Modest stress effect (SMD 0.31), driven largely by high-risk-of-bias studies; anxiety inconsistent and non-significant |
| Standalone L-theanine, depression symptoms | Single dose, healthy adults | SMD 0.69 (95% CI 0.13–1.25), significant only after excluding an outlier; unconfirmed in clinical populations |
| Standalone L-theanine, fatigue | Pooled across trials | No significant effect |
| L-theanine + caffeine, attention after 17 hours awake | 200mg L-theanine + 160mg caffeine | ~40ms faster reaction time, higher accuracy, larger/faster P3b EEG response vs placebo |
A real, replicated brain-wave effect is not the same claim as "reduces anxiety" — and that gap is where a lot of L-theanine marketing overreaches.
Where this sits in the range today
L-theanine is in two Proco products — Cognitive Support and Sleep Support — and in both cases it sits inside that product's blend without an individually disclosed dose. In Cognitive Support it is one of eight actives inside a 255mg proprietary blend; in Sleep Support it is one of fourteen ingredients inside a botanical blend carrying a single combined weight. So we cannot tell you whether you are getting the 200mg the attention trials used or a fraction of it, and we are not going to guess at a number.
Here is the part we have to state plainly rather than as a promise. Our manufacturer has confirmed that the per-ingredient amounts inside Cognitive Support's proprietary blend cannot be published. An earlier version of this article said getting that disclosure onto the label was "in progress." That was not something we could commit to, and saying so was wrong. Sleep Formula remains the one product in our range that discloses an exact per-serving dose for every ingredient it contains — it doesn't contain L-theanine, and we're not going to imply range-wide disclosure on the strength of one product that has it. (For the record, 5-HTP is in Sleep Support only; we previously listed it alongside Cognitive Support's ingredients, which was also wrong.)
Once an ingredient sits inside an undisclosed blend, you can't tell whether you're getting the dose the trials actually tested or a token amount added mainly for the label — the general version of that problem is in how to read a supplement label. L-theanine clearing the evidence bar on attention is precisely why it deserves a disclosed dose. That we can't currently give you one for either product it's in is a limitation of our range, stated as such. If knowing the exact milligrams matters more to you than the convenience of a blend, buying single ingredients separately is the honest alternative, and we've compared the two approaches in Cognitive Support versus stacking your own.
This is a familiar shape for us. When a product doesn't meet our standard, the reason is almost never "the ingredient doesn't work" — it's usually a dosing, sourcing, or transparency problem sitting on top of an ingredient that's actually fine. L-theanine is an unusually clean example, because the clinical case is stronger than most of what makes it onto shelves labeled "focus" or "calm" — which is exactly why the missing dose disclosure bothers us. Where we can fix it, we will; where the manufacturer has told us we can't, we'd rather say so than keep promising it.
Bottom line
L-theanine's evidence is solid and specific: a well-supported attention effect at 200mg, a genuinely useful pairing with caffeine, and more modest, less consistent results on stress and anxiety alone. It's in Cognitive Support and Sleep Support, in both cases inside a blend whose per-ingredient amounts are not disclosed — and for Cognitive Support, cannot be. That is the honest state of it, and it's a real limitation of buying L-theanine from us rather than as a single ingredient.
This article is for informational purposes only and has not been evaluated by the FDA. It is not intended to diagnose, treat, cure, or prevent any disease. Speak with a healthcare provider before starting any new supplement, especially if you take medication or have an existing health condition.
References
- Cognitive and affective effects of L-theanine: a systematic review and meta-analysis of 31 randomized trials. Molecular Psychiatry, 2026. nature.com
- High-dose L-theanine-caffeine combination improves neurobehavioural and neurophysiological measures of selective attention in acutely sleep-deprived young adults: a double-blind, placebo-controlled crossover study. British Journal of Nutrition, 2025. pmc.ncbi.nlm.nih.gov
The general standard this article measures L-theanine against — and how often a well-marketed ingredient fails it — is in why we don't sell "brain blends".
Want to know when this changes — including if a supplier arrangement ever makes per-ingredient disclosure possible? Join the list — no spam, just the range and the evidence behind it.