Does Decaf Coffee Have Caffeine? How Much It Contains and Why Some People Still Feel It
Author: Ukrainian Psychological Hub · Published: September 28, 2026 · Editorial Policy
Yes. Decaf coffee contains caffeine, but usually much less than regular coffee. The U.S. Food and Drug Administration says decaffeinated coffee typically contains about 2–15 milligrams of caffeine in an 8-fluid-ounce cup. “Decaffeinated” therefore describes a coffee from which most caffeine has been removed; it is not a guarantee that the finished drink contains zero caffeine.
That 2–15 mg range is the most useful mainstream answer, but it is not a specification for every cup. The amount in a particular decaf depends on the coffee, how thoroughly it was decaffeinated, how much ground coffee was used, how it was brewed, whether the drink is based on one or several decaf espresso shots, and the final serving size. Laboratory measurements of café decaf have found meaningful variation even among drinks sold under the same broad label. McCusker, Goldberger, and Cone measured 0–13.9 mg in 16-ounce decaf coffees from different establishments and 3.0–15.8 mg per decaf espresso shot in repeated samples from one outlet.
A second question is why someone can still feel decaf. There is no single explanation. Residual caffeine is pharmacologically real, and low doses can produce measurable effects under some experimental conditions. At the same time, coffee is a learned sensory cue: aroma, taste, the cup, the time of day, and the expectation of “having coffee” can alter alertness, craving, and subjective arousal. Those effects are documented, but they are variable and they do not mean that decaf secretly contains the caffeine dose of regular coffee.
How much caffeine is in decaf coffee?
For a standard consumer reference, the FDA’s current range is the clearest starting point: about 2–15 mg per 8 fluid ounces of decaf coffee. This is a range, not an average and not a legal promise about the cup in your hand. It also tells you why two familiar statements can both be true: decaf is a low-caffeine coffee, and decaf can still contain enough caffeine to matter to a person who is deliberately minimizing caffeine.
Health Canada provides a useful same-size comparison using average values. Its caffeine table lists about 3 mg for 8 ounces (237 mL) of roasted-and-ground decaffeinated coffee and about 135 mg for the same volume of brewed regular coffee; it lists about 5 mg for 8 ounces of instant decaf. These are population-level reference values, not measurements of every brand or brew.
The difference between the FDA range and Health Canada’s average is not a contradiction. A range answers “what amounts are commonly possible?” while an average answers “what representative amount can be used for comparison?” Real coffee is variable. The strongest conclusion supported by both authorities is that decaf generally contains a small fraction of the caffeine in regular coffee, while remaining measurably caffeinated in many products.
Real-world decaf can vary more than a single chart suggests
In the often-cited analytical study by McCusker and colleagues, researchers bought decaffeinated drinks from coffee establishments rather than relying on menu estimates. Ten 16-ounce decaf samples ranged from 0 to 13.9 mg of caffeine. When the researchers repeatedly sampled decaf beverages from the same outlet, decaf espresso ranged from 3.0 to 15.8 mg per shot, while 16-ounce brewed decaf samples ranged from 12.0 to 13.4 mg. The study was small and dates from 2006, so it should not be treated as a current brand directory. Its lasting value is methodological: “decaf” is a category, not a precise caffeine assay.
One detail is especially important: the study found one 16-ounce sample with no detected caffeine while other samples contained measurable amounts. That means the scientifically accurate statement is not “every decaf coffee contains caffeine.” It is that decaf coffee as a category is not guaranteed to be caffeine-free. If the exact milligram amount matters, product-specific testing or a manufacturer’s verified caffeine information is more informative than the word decaf alone.
Decaf vs regular coffee: how large is the caffeine difference?
Decaf usually creates a large reduction in caffeine exposure. For regular brewed non-specialty coffee, the FDA currently gives a typical range of 113–247 mg per 12-fluid-ounce serving. For decaf, the same agency gives 2–15 mg per 8 fluid ounces. Those serving sizes differ, so dividing one range by the other would create a false precision. The practical message is simpler: regular brewed coffee is usually measured in tens to hundreds of milligrams per serving, while decaf is usually measured in single digits to low tens of milligrams.
If you want a same-volume illustration, Health Canada’s table places an average 8-ounce brewed coffee at 135 mg and roasted-and-ground decaf at 3 mg. Again, these are reference values. A very large decaf, a drink built from multiple decaf espresso shots, and a small regular coffee are different serving systems. Beverage names do not replace dose measurement.
For a broader explanation of why caffeine changes with bean species, dry coffee dose, extraction, brew method, concentration, and serving size, see How Much Caffeine Is in Coffee? Bean, Roast, Brew, Serving Size, and Why Effects Vary. The present article owns the narrower question: what remains in decaf and how that residual dose relates to the experience of drinking it.
Why does decaf still contain caffeine?
Coffee beans naturally contain caffeine. Decaffeination removes caffeine from green coffee before the beans are roasted, but the process is designed to reduce caffeine dramatically while preserving as much of the rest of the coffee as practical. It is an extraction problem, not a magical switch that turns a caffeinated biological material into chemically pure zero-caffeine coffee.
The technical literature describes several major approaches, including water-based methods, solvent-based methods, and supercritical carbon dioxide. Ramalakshmi and Raghavan’s review summarizes these principal decaffeination routes and their tradeoffs. The exact process matters for manufacturing and can influence the chemical and sensory profile, but a process name by itself does not tell you the exact milligrams in your final brewed cup.
That last distinction prevents a common category error. Residual caffeine in the bean is only one stage. Brewing then determines how much material is extracted from the ground coffee into water, and serving design determines how much of that beverage you actually drink. A decaf espresso, a large decaf drip coffee, and a milk drink made with three decaf espresso shots can all be “decaf” while delivering different total caffeine doses.
For the coffee-production side of the topic—green-bean decaffeination, water, solvent and carbon-dioxide approaches, taste changes, and why decaf can still taste and feel recognizably like coffee—see Decaf Coffee: How It Is Made, Caffeine, Taste, and Why It Can Still Feel Like Coffee.
What determines the caffeine in your specific decaf cup?
The residual caffeine left after decaffeination
Different green coffees begin with different caffeine concentrations, and decaffeination does not leave every batch with the same residue. That residual amount becomes the raw material for the brewing step. This is one reason a universal statement such as “decaf always has 4 mg” is too precise.
How much coffee is used
Total dose depends partly on the mass of decaffeinated coffee used to make the beverage. If two coffees have similar residual caffeine per gram but one recipe uses much more ground coffee, the second recipe has more caffeine available for extraction. This is why a drink’s size alone does not determine its dose.
Extraction and beverage yield
Caffeine is water-soluble and is extracted during brewing. Grind size, contact time, temperature, water-to-coffee relationship, pressure or flow, and beverage yield can influence how much caffeine reaches the cup. These variables matter to regular and decaf coffee alike; decaf simply starts from a much lower caffeine baseline.
Espresso shot count
A decaf latte, flat white, cappuccino, or Americano usually gets its caffeine from the decaf espresso used in the recipe. If a drink contains two or three decaf shots, the total residual caffeine is the sum of those shots. The McCusker study’s 3.0–15.8 mg range per decaf espresso shot illustrates why shot count can matter, although those historical measurements should not be used as a current menu value for a particular chain.
Serving size
A larger cup can contain more total caffeine even when its caffeine concentration is low. This is the same distinction that matters throughout coffee science: concentration is caffeine per unit of beverage; total dose is the caffeine in everything you drink. A “weak-tasting” large decaf can therefore contain more total caffeine than a small, intense-tasting decaf.
Brand and product variation
Packaged instant decaf, home-brewed decaf, café drip decaf, and decaf espresso are different products. If a manufacturer publishes a verified caffeine amount for the exact product and serving, that is more useful than a generic internet chart. The FDA notes that caffeine content varies among products and serving sizes and encourages consumers to ask retailers about caffeine when that information matters.
Why can some people still feel decaf?
The phrase “I can still feel decaf” can describe several different experiences: mild alertness, less fatigue, a sense of activation, a familiar morning shift, relief of caffeine withdrawal, jitteriness, or simply the expectation that coffee should do something. Those experiences should not be collapsed into one mechanism.
1. Residual caffeine is still caffeine
A decaf dose of 2–15 mg is small compared with most regular coffee, but small is not identical to pharmacologically inert. In a controlled dose-response experiment after overnight caffeine abstinence, Smit and Rogers tested 0, 12.5, 25, 50, and 100 mg and found measurable performance effects at every caffeine dose studied. The study included only 23 adults and was conducted under specific abstinence conditions, so 12.5 mg should not be treated as a universal “minimum dose you can feel.” It does show that a dose in the upper part of the FDA decaf range can be behaviorally active under some conditions.
Subjective perception is even less suitable for a fixed threshold. Someone may notice a physiological or cognitive change without labeling it as caffeine, while another person may report feeling strongly stimulated by a context in which very little caffeine is present. The safest inference is therefore modest: residual caffeine can contribute to the experience of decaf, but the experience alone cannot reveal the dose.
2. People differ in caffeine sensitivity and clearance
Caffeine response is not uniform across adults. The FDA explicitly notes wide variation in both how sensitive people are to caffeine and how quickly they eliminate it. Medicines and some health conditions can also change individual response. This matters at the margins: a trace dose that is irrelevant to one person may be more noticeable to another, especially when the goal is near-total caffeine avoidance.
Individual sensitivity does not justify diagnosing yourself from a cup of decaf. Feeling anxious, wired, sleepy, focused, or calm after coffee is not a biomarker of caffeine metabolism, an anxiety disorder, ADHD, or any other diagnosis. The beverage dose, expectation, sleep state, food intake, prior caffeine use, and context can all influence the experience.
3. Multiple decaf drinks can add up
Caffeine totals accumulate across servings. Three decaf drinks containing 10 mg each would provide 30 mg in total. That arithmetic does not mean three decafs typically contain 30 mg; it shows why the number of servings matters when each serving is not zero. For people tracking caffeine closely, count total milligrams rather than counting the word “decaf.”
4. Coffee aroma and taste can become conditioned cues
Coffee is repeatedly paired with caffeine in the lives of habitual coffee drinkers. Over time, sensory features of the beverage can become learned signals for the state that usually follows coffee. Experimental work has tested this directly. Flaten and Blumenthal found that decaffeinated coffee used as a caffeine-associated stimulus increased some subjective and physiological arousal measures in habitual coffee users.
A later experiment also manipulated what participants were told about caffeine. Mikalsen, Bertelsen, and Flaten reported that caffeine-associated stimuli and caffeine-related information could alter measures of arousal. These findings support a learned-cue pathway: smelling and tasting coffee can change a person’s state because those cues have a history.
Conditioning is not the same as caffeine pharmacology. A cue-triggered increase in alertness does not mean the decaf contains a hidden regular-coffee dose, and it does not mean the response is imaginary. Learning can produce real psychological and physiological responses without reproducing the full drug effect.
5. Expectation can change what people notice and report
Expectancy effects are real but inconsistent. In one study, participants expected coffee to increase alertness and then received decaf or caffeine. Flaten, Aasli, and Blumenthal found only limited support for a simple expectancy theory: decaf changed some physiological measures, expected alertness predicted some reported alertness, but strong expectations did not reliably create a full placebo response.
This nuance is important because “placebo effect” is often used as if it meant “nothing happened.” That is the wrong model. Expectation can shape attention, interpretation, craving, and some behavioral or physiological measures, but it does not reliably duplicate a pharmacological caffeine dose. The best evidence treats pharmacology and expectancy as separable influences that can sometimes interact. For the focused placebo-and-expectancy question, see Decaf and the Placebo Effect: Can Coffee Feel Energizing Without Much Caffeine?.
6. Decaf can change caffeine withdrawal even when people know it is decaf
For habitual caffeine users who have recently stopped caffeine, the state before the cup matters. In a 2016 experiment, Mills, Boakes, and Colagiuri found that abstinent coffee drinkers who received decaf but were told it contained caffeine reported larger reductions in several withdrawal dimensions than participants told it was decaf. That result shows how belief can modify withdrawal experience.
More recently, an open-label study asked whether deception was necessary. Mills and colleagues gave 24-hour abstinent heavy coffee drinkers decaf that was truthfully described as decaf and still observed a larger reduction in withdrawal than in a water control group. The deceptive “told caffeine” condition produced a larger reduction than open-label decaf. This does not prove that ordinary decaf is a treatment for caffeine withdrawal; it shows that a familiar coffee experience can alter withdrawal-related experience through more than one pathway.
Balanced-placebo research also places limits on expectancy claims. Juliano and colleagues found expectancy effects on craving among 87 abstinent regular coffee users, but no expectancy effects on withdrawal symptoms or cognitive performance, while pharmacological caffeine affected all measured outcomes. Taken together, the literature argues for mixed mechanisms rather than a universal placebo explanation.
7. Coffee-related cues can alter craving and heart rate without reproducing caffeine
A 2022 balanced-placebo study separated caffeine administration, information about caffeine, and coffee-related sensory cues. Shephard and Barrett found that coffee cues influenced some craving and physiological measures while caffeine and expectancy produced distinct effects. This is a useful model for everyday decaf: the coffee experience can be psychologically active even when its caffeine content is low.
Feeling energized after decaf does not tell you how much caffeine was in it
Subjective effect is not a caffeine meter. If you feel alert after decaf, several explanations remain compatible with that observation: residual caffeine, learned coffee cues, expectation, relief from withdrawal-related fatigue, the behavioral act of taking a break, or ordinary changes in arousal that happened at the same time. The experience can be genuine while its cause remains uncertain.
The reverse is also true. If you feel nothing after a cup, that does not prove the coffee contained zero caffeine. Tolerance, current sleepiness, dose, timing, and what you were doing can all change whether a small effect reaches conscious awareness. Chemical presence and subjective salience are different questions.
This distinction protects against a common reasoning error: “I felt jittery, therefore the café served regular coffee.” A service mistake is possible in the real world, but symptoms alone cannot establish it. If an unexpectedly strong reaction matters, checking the product, recipe, shot count, or retailer information is more informative than trying to reverse-engineer milligrams from a feeling.
Can decaf coffee make you jittery or anxious?
It can be possible for a person to notice jitteriness or anxiety after decaf, but that observation does not identify a single cause. Residual caffeine is one candidate, particularly for someone who is very sensitive or who has consumed several decaf servings. Expectation and conditioned arousal are other candidates. Anxiety itself can also fluctuate independently of the drink.
The FDA lists anxiety, jitters, sleep disruption, increased heart rate, palpitations, and other effects among signs that caffeine intake may be too high for an individual. Those general caffeine effects are well established, but they should not be mechanically transferred to a typical decaf dose. Direct evidence about anxiety responses specifically to ordinary decaf-level caffeine is much thinner than evidence about caffeine at conventional doses.
For that reason, “decaf made me anxious” is best treated as a report of experience rather than a diagnosis or a precise pharmacological conclusion. The relevant questions are what was actually consumed, how much, what else contained caffeine that day, the person’s usual caffeine exposure, and whether the pattern repeats under comparable conditions.
Can decaf coffee keep you awake?
Decaf contains far less caffeine than regular coffee, so its caffeine-related sleep effect would generally be expected to be smaller. But “smaller” is not the same as “impossible.” The FDA’s 2–15 mg range means a late decaf can still contain caffeine, and the amount can accumulate across servings. People also differ in caffeine sensitivity and elimination.
There is a scientific limit here that matters: most sleep studies use caffeine doses far above the amounts typically found in decaf. It would be poor evidence practice to take a study using 100, 200, or 400 mg and claim that a 3 mg decaf cup has the same sleep effect scaled down in a simple linear way. Evidence specific to ordinary decaf doses and sleep is limited.
The coffee ritual itself can also be activating for some people because it is associated with work, conversation, late-night routines, or expected alertness. That psychological activation is conceptually separate from caffeine blocking sleep pressure. A person can therefore experience an evening decaf as stimulating without that proving a large pharmacological caffeine effect.
Is decaf a good choice when you are trying to reduce caffeine?
As a caffeine-reduction strategy, decaf generally makes a large quantitative difference because its caffeine content is usually far below regular coffee. The authoritative reference values make that clear. It also preserves much of the sensory and behavioral structure of drinking coffee: the aroma, warmth, taste, preparation, cup, pause, and social ritual.
For someone who simply wants less caffeine, that combination is the reason decaf exists. For someone who needs strict or medically directed caffeine avoidance, the standard changes: “low” may not be equivalent to “allowed.” The FDA specifically notes that people who react strongly to caffeine in a negative way may want to avoid decaffeinated beverages as well. Individual medical restrictions should follow the clinician or treatment plan that established them.
A practical way to track decaf caffeine
First, identify the serving rather than the beverage name. An 8-ounce cup, a 16-ounce café decaf, one espresso shot, and a three-shot milk drink are not interchangeable units. Second, use product-specific caffeine information when the manufacturer or retailer provides it. Third, count the number of servings or espresso shots. Fourth, remember other caffeine sources in the same day.
If no product-specific number is available, the FDA’s 2–15 mg per 8 ounces is a reasonable general range for decaf coffee, but it should remain a range. Avoid converting it into a fabricated exact number such as “my 12-ounce decaf contains 8.5 mg.” Brewing and product differences are too large for that precision without measurement.
Common myths about caffeine in decaf coffee
Myth: Decaf means caffeine-free
Decaf means substantially reduced caffeine. It does not guarantee zero. The FDA’s typical range of 2–15 mg per 8 ounces and direct laboratory measurements both establish that residual caffeine is common.
Myth: Every decaf cup has the same amount of caffeine
There is no single universal milligram value. Product, decaffeination, coffee dose, brewing, espresso shot count, and serving size all matter. The McCusker measurements show real-world variation even within the same general category.
Myth: Strong-tasting decaf must contain more caffeine
Flavor intensity and caffeine dose are different variables. A dark, bitter, concentrated-tasting decaf can still be low in caffeine, while a milder-tasting large decaf can contain more total caffeine simply because more coffee or more beverage was used. Taste cannot measure milligrams.
Myth: If decaf makes you feel alert, it was not really decaf
Residual caffeine can contribute, but coffee cues and expectations can also change arousal, and withdrawal relief can change how energetic a habitual coffee drinker feels. Controlled studies of conditioned coffee cues and expectancy demonstrate these effects without requiring a regular-coffee caffeine dose.
Myth: If you do not feel decaf, it contains no caffeine
A small dose can be chemically present without producing a noticeable subjective effect. Sensation is not an assay. The only way to know an exact amount is through reliable product information or chemical measurement.
Myth: A particular decaffeination method guarantees zero caffeine
Water, solvent, and carbon-dioxide methods are techniques for removing caffeine from coffee; the process name does not by itself prove that the finished brewed beverage contains zero milligrams. The relevant endpoint for the drinker is the measured residual caffeine in the product and serving.
What the evidence shows—and what it does not
Established evidence supports three core conclusions. First, decaf coffee usually contains much less caffeine than regular coffee, and authoritative agencies report residual caffeine in typical servings. Second, the exact amount varies across products and preparations. Third, coffee-related sensory cues and expectations can influence some measures of arousal, craving, and withdrawal.
The evidence does not support treating every sensation after decaf as a caffeine effect. It does not support using a person’s response to decaf to diagnose caffeine sensitivity, anxiety, ADHD, or any other condition. It does not support a single universal caffeine number for every decaf beverage. It also does not support the opposite extreme—that because expectancy matters, residual caffeine is irrelevant.
The strongest model is additive and conditional. Chemistry determines whether caffeine is present and how much. Brewing and serving determine exposure. Pharmacology influences the nervous system in a dose- and person-dependent way. Learning and expectation influence how coffee cues are processed. The final experience is the result of those layers interacting in a particular person at a particular time.
FAQ
Does decaf coffee have caffeine?
Yes, decaf usually contains some caffeine. The FDA gives a typical range of about 2–15 mg per 8-fluid-ounce cup. A particular product can fall outside a generic reference range, and some tested samples have had no detected caffeine, so decaf is best understood as low-caffeine rather than guaranteed caffeine-free.
How much caffeine is in an 8-ounce cup of decaf?
The FDA’s typical range is 2–15 mg per 8 ounces. Health Canada lists an average of about 3 mg for 8 ounces of roasted-and-ground decaf and 5 mg for instant decaf. Use a product-specific number when one is available.
How much caffeine is in decaf espresso?
There is no universal number. In a 2006 analytical study, repeated decaf espresso samples ranged from 3.0 to 15.8 mg per shot. That study demonstrates variability; it is not a current menu specification for every café.
Is decaf completely caffeine-free?
Not as a category. Some individual samples may test at zero or below a method’s detection threshold, but the label decaf does not guarantee that result. If you require zero caffeine, verify the specific product and follow any medical guidance that applies to you.
Can you feel 10 or 15 mg of caffeine?
It is possible for low caffeine doses to produce measurable effects under some conditions. A controlled study found performance effects at 12.5 mg after overnight caffeine abstinence. That does not establish 12.5 mg as a universal perception threshold, and subjective sensitivity varies.
Why does decaf sometimes make me feel energized?
Residual caffeine is one possibility. Learned coffee cues, expectation, relief of withdrawal-related fatigue, and the behavioral context of drinking coffee are others. Experiments show that decaf and coffee-related cues can alter some measures of alertness, arousal, craving, and withdrawal even when caffeine pharmacology is absent or minimal.
Can decaf cause jitters?
A person can report jitters after decaf, but the symptom does not prove a high caffeine dose. Residual caffeine, multiple servings, sensitivity, expectation, and unrelated fluctuations in arousal can all contribute. If the reaction is strong or medically important, verify the product and discuss persistent concerns with a qualified health professional rather than inferring a diagnosis from the drink.
Will decaf keep me awake?
It contains much less caffeine than regular coffee, so a caffeine-related effect on sleep would generally be expected to be smaller. However, the dose is not always zero, people differ in sensitivity, and direct sleep research at ordinary decaf doses is limited. Timing, serving size, other caffeine sources, and learned coffee routines also matter.
Does drinking several cups of decaf add up?
Yes. Total caffeine is additive across servings. If each cup contains some residual caffeine, several cups deliver more caffeine than one. The exact total depends on the caffeine content of each serving.
Is decaf better than regular coffee for reducing caffeine?
For reducing caffeine exposure, decaf generally provides a large reduction compared with regular coffee. Whether it fits a specific health restriction is an individual medical question because decaf is not guaranteed to be caffeine-free.
Does the decaffeination method tell me the final caffeine amount?
Not reliably. The method describes how caffeine was removed from green coffee, but the final cup also depends on residual caffeine, the coffee dose, brewing, and serving size. Product-specific measurements are more informative than process labels when exact milligrams matter.
Related Articles
References
Flaten, M. A., & Blumenthal, T. D. (1999). Caffeine-associated stimuli elicit conditioned responses: an experimental model of the placebo effect. Psychopharmacology, 145(1), 105–112. DOI: 10.1007/s002130051038. PubMed
Flaten, M. A., Aasli, O., & Blumenthal, T. D. (2003). Expectations and placebo responses to caffeine-associated stimuli. Psychopharmacology, 169(2), 198–204. DOI: 10.1007/s00213-003-1497-8. PubMed
Health Canada. Caffeine in Foods. Official guidance and caffeine table
Juliano, L. M., et al. (2019). Investigating the role of expectancy in caffeine withdrawal using the balanced placebo design. Human Psychopharmacology, 34(2), e2692. DOI: 10.1002/hup.2692. PubMed
McCusker, R. R., Goldberger, B. A., & Cone, E. J. (2006). Caffeine content of decaffeinated coffee. Journal of Analytical Toxicology, 30(8), 611–613. DOI: 10.1093/jat/30.8.611. PubMed
Mikalsen, A., Bertelsen, B., & Flaten, M. A. (2001). Effects of caffeine, caffeine-associated stimuli, and caffeine-related information on physiological and psychological arousal. Psychopharmacology, 157(4), 373–380. DOI: 10.1007/s002130100841. PubMed
Mills, L., Boakes, R. A., & Colagiuri, B. (2016). Placebo caffeine reduces withdrawal in abstinent coffee drinkers. Journal of Psychopharmacology, 30(4), 388–394. DOI: 10.1177/0269881116632374. PubMed
Mills, L., Lee, J. C., Boakes, R., & Colagiuri, B. (2023). Reduction in caffeine withdrawal after open-label decaffeinated coffee. Journal of Psychopharmacology, 37(2), 181–191. DOI: 10.1177/02698811221147152. PubMed
Ramalakshmi, K., & Raghavan, B. (1999). Caffeine in coffee: its removal. Why and how? Critical Reviews in Food Science and Nutrition, 39(5), 441–456. DOI: 10.1080/10408699991279231. PubMed
Shephard, A., & Barrett, S. P. (2022). The impacts of caffeine administration, expectancies, and related stimuli on coffee craving, withdrawal, and self-administration. Journal of Psychopharmacology, 36(3), 378–386. DOI: 10.1177/02698811211032447. PubMed
Smit, H. J., & Rogers, P. J. (2000). Effects of low doses of caffeine on cognitive performance, mood and thirst in low and higher caffeine consumers. Psychopharmacology, 152(2), 167–173. DOI: 10.1007/s002130000506. PubMed
U.S. Food and Drug Administration. Spilling the Beans: How Much Caffeine is Too Much? Official FDA consumer guidance