Pharmacology Investigation 001
Can You Buy 38 Milliseconds?
Caffeine, L-theanine and the tired brain. What controlled trials in FPS players and sleep-deprived adults actually measured, in milliseconds, and why the biggest numbers come with the biggest caveats.
At the top of competitive FPS, rounds turn on gaps that are hard to perceive. A peek that lands one frame earlier on a 240 Hz monitor is about 4 milliseconds ahead. Ten frames is roughly 42 ms. So when a controlled trial reports that a compound sped up reaction time by 38.1 milliseconds, it gets attention.
That number is real. It comes from a double-blind, placebo-controlled crossover trial of caffeine combined with L-theanine. It also comes from people who had been awake for 17 hours, doing a task that looks nothing like Counter-Strike.
This investigation follows the numbers from the cleanest esports data to the most dramatic lab result, and asks which parts reach the server.
The adenosine brake
From the moment a person wakes, adenosine builds up in the brain as a by-product of neural activity. It binds to A1 and A2A receptors and acts as a slowly tightening brake on arousal. Late in a session, much of what feels like “being off” is that brake.
Caffeine does not add energy. It blocks the adenosine receptors, releasing the brake. Downstream, dopamine and noradrenaline signalling increases, and attention and vigilance hold up better.
There is a catch that most supplement marketing ignores. Regular users adapt: the brain grows more adenosine receptors, and missing the usual morning dose produces a measurable dip in alertness and mood. Some of the boost a habitual user feels is simply withdrawal being reversed. This does not make caffeine useless. It means the real advantage over a non-user is smaller than the felt advantage over one’s own caffeine-deprived self.
What happens in actual FPS players
For years, the caffeine and reaction-time literature came almost entirely from students and soldiers. That has changed.
Rogers, Trotter, Johnson, Desbrow & King, 2024
Direct relevance to competitive play
Frontiers in Sports and Active Living doi:10.3389/fspor.2024.1437700
- Sample
- n = 24, FPS players currently playing ranked (22 men, 2 women), mean age 22.3, habitual intake about 209 mg/day
- Design
- Randomised, counterbalanced, single-blind, within-subject. Three familiarisation sessions.
- Intervention
- Water vs 1 mg/kg (≈83 mg) vs 3 mg/kg (≈250 mg) caffeine. Tested 60 to 100 minutes after intake.
- Result
- PVT reaction time: 290.8 ± 31.1 ms (water), 279.8 ± 29.1 ms (1 mg/kg, p = .008), 273.0 ± 28.3 ms (3 mg/kg, p < .001); partial η² = .42. Static-clicking score rose from 117.8 to 120.8 and 121.5. Clicking accuracy did not change (88.3%, 88.9%, 88.8%).
- Limitations
- Caffeine abstinence was self-reported. The control was water rather than a placebo capsule, so expectancy was not controlled. Aim tasks were KovaaK's drills, not matches.
Three details matter more than the headline:
- Even the low dose worked. About 83 mg, roughly a small coffee, produced an 11 ms PVT improvement. Going from 1 to 3 mg/kg added only about 7 ms more.
- Caffeine sped players up without making them more accurate. Hit rate rose (2.23 to 2.29 hits per second) and so did reactive tracking accuracy (35.7% to 37.8%), but static clicking accuracy did not move. Caffeine made players faster at the same precision.
- These were habitual users after abstaining. Part of the 17.8 ms may be withdrawal reversal.
Wu, Chen, Chen, Liu, Liu & Chiu, 2024
High relevance to competitive play
Scientific Reports doi:10.1038/s41598-024-52599-y
- Sample
- n = 9, Elite male players ranked in national competition, low habitual intake (44 mg/day)
- Design
- Single-blind, placebo-controlled crossover, 7 days apart
- Intervention
- 3 mg/kg caffeine vs flour placebo capsule, tested 1 hour later
- Result
- Aim-trainer kill ratio, hit accuracy and time to target all improved (d = 0.96 to 1.96). Visual search with 20 items: 728.0 → 637.0 ms (p = .004). The 5, 10 and 15 item conditions showed no significant change.
- Limitations
- Nine participants. Effect sizes this large in a sample this small are unstable. Inconsistent pattern across visual-search conditions.
The 91 ms figure is real, but it appeared in only one of four visual-search conditions, in nine people. It is a reason to run a bigger trial, not a number to quote as fact.
The L-theanine question
L-theanine is an amino acid found almost only in tea. It crosses the blood-brain barrier and is associated with increased alpha-band EEG activity, a signature often linked to relaxed but alert attention. The popular story says it “takes the edge off” caffeine. The more interesting finding is about attention.
Kahathuduwa et al., 2018
Moderate relevance to competitive play
Nutrition Research 49:67–78 doi:10.1016/j.nutres.2017.11.002
- Sample
- n = 9, Healthy adult men
- Design
- Randomised four-way crossover inside an fMRI scanner
- Intervention
- 200 mg L-theanine, 160 mg caffeine, both, or water
- Result
- Responses to targets were faster than placebo by 27.8 ms with L-theanine (p = .018) and 26.7 ms with the combination (p = .037). L-theanine reduced brain responses to distractors in attention-control regions.
- Limitations
- Nine men. One task. Millisecond values as reported in a 2021 systematic review of the paper.
The imaging result is the interesting part. L-theanine did not make the brain respond more strongly to targets. It made it respond less to distractors. In a game full of particle effects, flashes, killfeed updates and audio noise, filtering out irrelevant stimuli is arguably more valuable than raw speed. That link is a hypothesis, but it is a biologically coherent one.
The 38 milliseconds
Nawarathna, Ariyasinghe, Dassanayake et al., 2025
Moderate relevance to competitive play
British Journal of Nutrition 134(3):195–204 doi:10.1017/S0007114525104169
- Sample
- n = 37, Healthy adults aged 22 to 30 (21 men), after 17 hours awake
- Design
- Double-blind, placebo-controlled, counterbalanced crossover
- Intervention
- 200 mg L-theanine + 160 mg caffeine vs starch placebo; tested 50 minutes later on a visual oddball task
- Result
- Hit reaction time improved by 52.1 ms with the combination vs 14.0 ms with placebo (practice). Net difference 38.1 ms (95% CI 13.5 to 62.7), p = .003. P3b latency, an EEG marker of stimulus evaluation, fell by about 30 ms.
- Limitations
- Participants were sleep-deprived. The task was a traffic-scene oddball, not a game. The authors caution against extrapolating to habitual high-dose users.
The confidence interval runs from 13.5 to 62.7 ms. The true effect could be modest or very large, and the trial cannot tell which. What it shows convincingly is direction: in tired brains, the combination restored speed and accuracy, and the EEG data suggest it did so by speeding up stimulus evaluation, not just motor output.
What it does not show is that a rested player gets 38 ms. Across the caffeine literature, effects shrink when participants are well slept. Anyone quoting this number for a fresh morning session is quoting the wrong population.
Timing and the delivery problem
The pharmacology of the moment matters as much as the compound. Capsules are swallowed, dissolve in the stomach and are absorbed in the small intestine. The trials above tested 50 to 100 minutes after intake for that reason. A capsule is a pre-match decision.
Pouches change that. Placed against the gum, a pouch delivers its contents through the oral lining, the same route that makes nicotine pouches act within minutes. Players can take one mid-session, between maps or before a decider without swallowing anything or breaking focus. Nicotine has made the format familiar. L-theanine is now following, and brands such as Alyva have released pouches containing pure L-theanine. This kind of in-the-moment use is impossible with a capsule.
Whether L-theanine crosses the oral lining as efficiently as nicotine has not yet been measured. Its onset in pouch form is a strong rationale rather than a trial result. NeuroFPS covers the format in depth in Clutch Round Pharmacology.
What transfers to the server
| Finding | Strength | Transfers to matches? |
|---|---|---|
| Caffeine speeds PVT reaction time in FPS players (−11 to −18 ms) | Strong, replicated | Probably in part. Speed, not precision. |
| Caffeine improves aim-trainer scores | Moderate, small samples | Plausible. Untested in real matches. |
| L-theanine reduces distractor processing | Early, n = 9 | Plausible. Untested in games. |
| Caffeine + L-theanine restores speed when sleep-deprived (−38.1 ms) | Good, one strong RCT | Likely relevant to late-night sessions. Not to rested play. |
| Any of the above improves win rate | No data | Unknown |
The honest conclusion is less dramatic than the headline and more useful. The most reliable thing caffeine and L-theanine do is protect performance from fatigue. That is valuable in a sport where tournaments run late and ranked sessions run later. But a compound that defends against decline does not lift a player above their rested baseline. The rested baseline is built by sleep, which is the subject of the next investigation.
References
- Rogers EJ, Trotter MG, Johnson D, Desbrow B, King N. Caffeine improves the shooting performance and reaction time of first-person shooter esports players: a dose-response study. Front Sports Act Living. 2024;6:1437700. doi:10.3389/fspor.2024.1437700
- Wu SH, Chen YC, Chen CH, Liu HS, Liu ZX, Chiu CH. Caffeine supplementation improves the cognitive abilities and shooting performance of elite e-sports players: a crossover trial. Sci Rep. 2024;14:2074. doi:10.1038/s41598-024-52599-y
- Kahathuduwa CN, Dhanasekara CS, Chin SH, et al. L-theanine and caffeine improve target-specific attention to visual stimuli by decreasing mind wandering: a human functional magnetic resonance imaging study. Nutr Res. 2018;49:67–78. doi:10.1016/j.nutres.2017.11.002
- Nawarathna GS, Ariyasinghe DI, Dassanayake TL, et al. High-dose L-theanine–caffeine combination improves neurobehavioural and neurophysiological measures of selective attention in acutely sleep-deprived young adults. Br J Nutr. 2025;134(3):195–204. doi:10.1017/S0007114525104169
- Haskell CF, Kennedy DO, Milne AL, Wesnes KA, Scholey AB. The effects of L-theanine, caffeine and their combination on cognition and mood. Biol Psychol. 2008;77(2):113–122. doi:10.1016/j.biopsycho.2007.09.008
- Sohail AA, Ortiz F, Varghese T, et al. The cognitive-enhancing outcomes of caffeine and L-theanine: a systematic review. Cureus. 2021;13(12):e20828. doi:10.7759/cureus.20828
NeuroFPS publishes scientific analysis, not medical advice. Nothing here is a recommendation to take any substance or a dosing guide. Responses vary between individuals; speak to a doctor or pharmacist before changing anything you take.