Who Uses Cognitive Enhancers? What the Largest Global Survey Actually Found
Written and fact-checked by the NootropicsJet editorial team Last updated: [27-08-2026] · Original publication date unchanged
Method: built from the full text of one peer-reviewed cross-sectional study — Maier LJ, Ferris JA, Winstock AR, International Journal of Drug Policy 2018;58:104–112 — plus the prevalence literature that paper cites. Figures we could not trace to a primary source were excluded rather than estimated.
How many people actually use cognitive enhancers? Every source gives a different answer, and the spread is not subtle: published estimates of non-medical prescription stimulant use in the United States have ranged from 5% to 35%, depending mostly on how the question was asked.
This article works through the largest survey ever run on the behavior — 15 countries, two waves, more than 100,000 respondents — and explains why its numbers are simultaneously real and unusable as national prevalence. Describing what a self-selected population reports doing is not a recommendation that anyone do it. By the end you should be able to read any prevalence headline and identify what it is measuring.
Key takeaways
- Among Global Drug Survey respondents, 12-month use of stimulants or modafinil for enhancement rose from 4.9% (2015) to 13.7% (2017). The widely quoted “180% increase” is substantially a measurement artifact — the two waves asked the question differently and drew different samples.
- The most commonly reported enhancement substance in the 2017 wave was not a stimulant. It was cannabis (22.9%), followed by alcohol (15.7%).
- Roughly half of respondents who used alcohol for enhancement reported that their performance got worse (49.9%). For benzodiazepines, decreases and increases were near-identical (31.5% vs 31.8%).
- Daily use was uncommon (under 5%). The modal pattern was one to two weeks around exams or a busy work period, once or twice a year. Risk concentrated in one subgroup: among those mainly using illegal stimulants for enhancement, 7.1% reported daily use.
- The supply chain was social, not medical: friends supplied 47.8% of stimulant enhancement users; physicians, 3.8%. Of those who did get them from a physician, a quarter said risks and side effects were never discussed.
Why every estimate of cognitive enhancer use is different
Before any number is useful, three things have to be pinned down: what counts as a cognitive enhancer, what counts as use, and over what window.
Change any one and the number moves more than the behavior does. Maier and colleagues open their paper by noting that US lifetime estimates of non-medical prescription stimulant use have ranged from 5% to 35% — without specifying the purpose of use. That range is not a disagreement about reality. It is a disagreement about definitions. German work cited in the same paper found rates starting from 0.8% in the general population and around 2.0% among university students. Twelve-month enhancement prevalence in earlier literature ran 1–20% among students and 3–7% among employees.
Recall window does the same work. Lifetime prevalence and past-month prevalence of the same behavior can differ by an order of magnitude in the same population. So does whether caffeine is counted — include it and enhancement prevalence approaches 100% in most adult populations, which is why serious surveys exclude it and why caffeine deserves its own evidence base.
This is the same problem that makes the category itself hard to bound, and it is the reason the line between “natural” and “synthetic” nootropics keeps collapsing under scrutiny. A survey that asks about “smart drugs” and a survey that asks about “any substance used to improve performance at work or while studying” are not measuring the same thing, and averaging them produces nothing.
What the largest survey measured — and who answered it
The Global Drug Survey is an anonymous online survey promoted through media partners — newspapers, magazines, websites, social platforms. It recruits people who have used legal or illegal drugs recreationally. That recruitment design is deliberate and it is the single most important fact about every number that follows.
The authors are explicit: the sample should not be considered representative of any country’s general population. They also disclose that Adam Winstock, one of the three authors, is the founder and owner of Global Drug Survey. Both disclosures are in the paper; both belong in any honest summary of it.
The enhancement module ran in two waves. GDS2015 covered 79,640 eligible respondents aged 16–65 with no self-reported ADHD history, across 15 countries with at least 1,000 responses each. GDS2017 covered 29,758 under the same criteria. The 2017 sample is far smaller for a specific reason: the module was included only for the first month of that survey, then cut to shorten the questionnaire.
The two samples also differ in composition. Men rose from 60.2% to 68.8%. Students fell from 43.4% to 35.3%. Past-year illegal drug use rose from 59.9% to 67.6%. Mean age rose from 29 to 31. Hold that thought.
The headline number, and why it is mostly an artifact
Across the two waves, 12-month use of prescription stimulants, modafinil or illegal stimulants for enhancement went from 4.9% to 13.7% among GDS respondents. That is the 180% increase reported in coverage of this paper, usually as evidence that enhancement drug use is surging.
The paper does not support that reading, and says so. Three mechanisms are stacked inside the number.
Question routing changed. In GDS2015, the 12-month substance questions were shown only to respondents who had already indicated ever using something for enhancement. In GDS2017, every respondent was asked directly. The 2015 figure is therefore a filtered estimate and the 2017 figure is an unfiltered one. Some of the “increase” is simply people who were never asked in 2015 being asked in 2017.
The sample drifted. The 2017 wave had proportionally more men, more recent illegal drug users, and fewer students than the 2015 wave. Recent illegal drug use is the strongest predictor of enhancement use in this dataset. A sample with more of those people will report more enhancement use even if nobody’s behavior changed.
The adjusted effect is much smaller than the crude one. In the paper’s logistic regression, which controls for age, student status, past-year use of nine specific drugs and country, responding to GDS2017 rather than GDS2015 carried an odds ratio of 2.22. That adjusted figure still cannot separate a real rise from the question-wording change, because both are bundled into the survey-year variable.
What the authors write is that the increase cannot be explained only by self-selection and question wording — which is a statement that those factors are part of it, not that they are absent. There may be a real underlying rise. This design cannot size it.
What people actually reported taking
The 2017 wave asked about sedatives as well as stimulants, on the reasoning that people use downers to sleep, relax or manage exam nerves in service of performing better afterward. The resulting hierarchy is not the one the smart-drug press describes.
Among GDS2017 respondents, past-year use explicitly to improve performance at work or while studying:
- Cannabis — 22.9%
- Alcohol — 15.7%
- Illegal stimulants (amphetamine, cocaine, methamphetamine) — 8.3%
- Prescription stimulants (methylphenidate, dexamphetamine) — 6.6%
- Benzodiazepines or prescription sleeping pills — 5.7%
- Modafinil — 2.4%
- Beta-blockers — 1.2%
The category that dominates public discussion — prescription stimulants — ranks fourth. The two substances at the top are central nervous system depressants. Keep that in mind when we get to whether any of it worked.
Who used them
The regression identified what predicted enhancement use with prescription stimulants, modafinil or illegal stimulants in the past year. The pattern is blunt: the strongest predictor of using drugs for enhancement is using drugs.
| Predictor | Adjusted odds ratio |
|---|---|
| Past-year amphetamine use | 3.09 |
| US residence (reference: Australia) | 3.06 |
| Past-year methamphetamine use | 2.71 |
| Past-year cocaine use | 2.48 |
| Current student status | 1.32 |
Adjusted odds ratios, all p < .001. Compiled from Maier et al. 2018, Table 4.
Student status — the variable the entire cultural narrative is built around — carries an odds ratio of 1.32. Recent amphetamine use carries 3.09. Being a student raises the odds by about a third; recently using illegal stimulants roughly triples them. Enhancement use in this dataset is, to a first approximation, a subset of drug use, not a distinct behavior with its own population.
This page treats the student case as data only. For the applied question — what a study-oriented approach looks like in practice — see our nootropic stack for studying; no guidance is offered here.
Why they said they used them
Motives came from the 2015 wave, among respondents reporting past-year stimulant or modafinil enhancement use. Concentration led at roughly 78% for both sexes. Study assistance followed at 60–65%, stamina at 49–55%, fatigue at 45–50%, work performance at about 41%.
Asked for a single primary motivation, more than half selected improving performance while studying; around one in five selected work; and roughly a quarter selected socializing — inside a module explicitly about performance enhancement.
Sex differences were significant on several motives. Women more often reported using as a study aid, for stamina, for fatigue and for coping; men more often reported getting high and sexual performance. One motive is worth stating for what it reveals about the category rather than about the substances: weight loss was reported by 20.5% of female and 7.9% of male enhancement users. Some of what gets counted as cognitive enhancement is not about cognition at all.
How often
Enhancement use in this sample was episodic. Under 5% of respondents reported daily or near-daily use. Two-thirds reported ten or fewer occasions in twelve months. The single most common pattern was a one-to-two-week block before and during exams or an intense work period, once or twice a year.
That does not describe an epidemic. The authors characterize the overall picture as relatively low-risk use patterns, limited for the most part to stressful periods at work and exam preparation.
One subgroup breaks the pattern. Among GDS2015 respondents who used mainly illegal stimulants for enhancement, 7.1% reported daily use in the past year and 30.3% reported using at least weekly. Frequency, dose and purity are all least controlled in exactly the group with the least access to a prescribing framework. The risk concentrates in the opposite place from where public concern usually points.
Where they got them
Among GDS2015 respondents using stimulants for enhancement, the main source was the circle of friends — 47.8%. A dealer accounted for 11.8%, the internet 9.1%, a family member with a prescription 6.1%, and a physician 3.8%.
This is a social distribution network, not a medical one. And the medical route, where it operated, underperformed: of those who obtained stimulants from a physician, a quarter reported that the physician did not discuss the risks and side effects with them.
That last figure is an epidemiological observation about a small subgroup, reported here as such. It is not a route, and nothing in this section should be read as one.
Do cognitive enhancers work in people without ADHD?
Two different questions hide inside that one, and they have different answers.
What users believed happened
GDS2017 respondents who used a substance for enhancement were asked whether their performance felt decreased, unchanged or increased. The results invert the marketing.
| Substance used for enhancement | Reported performance decreased | Reported performance increased |
|---|---|---|
| Alcohol | 49.9% | 15.1% |
| Benzodiazepines | 31.5% | 31.8% |
| Prescription stimulants | 4.5% | 77.5% |
| Modafinil | 2.1% | 75.8% |
Self-reported perceptions, not measured cognitive performance. Subset of substances; percentages do not sum to 100 because “unchanged” is omitted. Compiled from values reported in Maier et al. 2018.
Cannabis sat between these groups, with 19.9% reporting decreased performance. In the earlier 2015 wave, about a fifth of enhancement users (21.6%) said the effect fell short of expectations, half (50.4%) said it matched, and 28.2% said it exceeded them.
Read the table’s top two rows again. The two most commonly reported enhancement substances in this sample were alcohol and cannabis, and alcohol is the one users most often said made them worse. Roughly half of people using alcohol to improve performance reported that it did the opposite. That is not a marginal finding — it is the largest single perceived-effect signal in the dataset.
What was actually measured
Nothing above is measured cognition. It is what people believed about themselves, reported retrospectively, in an anonymous survey, about a behavior they had chosen. Every bias that framing implies is present.
The trial literature tells a more restrained story. Summarizing it, Maier and colleagues state that stimulant drugs have “weak positive and sometimes even impairing effects” on cognition in individuals without an underlying mental disorder, citing a systematic review of modafinil and methylphenidate for neuroenhancement in healthy individuals (Repantis et al., 2010) and a review of psychostimulants and cognition (Wood et al., 2014). A preliminary study of Adderall in healthy adults (Farah et al., 2009) raised the possibility that alertness gains come with creativity costs. Effects across this literature are strongly dose-dependent.
There is also a consistent signal on who responds: individuals without an ADHD diagnosis who show ADHD-like traits, particularly impulsiveness, appear most likely to perceive benefit. That is the deficit-versus-enhancement distinction that runs through this entire field and that we set out in our history of piracetam and the origins of the nootropic concept — restoring a compromised system is a different pharmacological problem from improving an intact one, and the prevalence data here extends that framing rather than contradicting it.
The honest summary: among people without a clinical indication, self-reported benefit is high and measured benefit is small, inconsistent, and sometimes negative.
Why the country numbers vary so much
Among GDS respondents, country-level 12-month enhancement prevalence ranged from 1.9% to 29.6%. The highest rates were among US, Dutch and Canadian respondents; the lowest among respondents from the three German-speaking countries, New Zealand and Portugal.
For prescription stimulants specifically, the 2015 wave ranged from 0.8% among Austrian respondents to 18.7% among US respondents; the 2017 wave from 1.6% among Hungarian respondents to 21.6% among US respondents.
The authors’ explanation is structural rather than cultural. Differences in ADHD diagnostic frameworks, national approval of specific stimulant medications, and physician prescribing behavior change how much prescription stimulant exists in circulation to be diverted — and therefore how much is available for non-medical use. Countries where stimulant treatment arrived late or in a narrower range of approved medications show lower prescription-stimulant enhancement rates. This article takes no position on whether any country’s diagnostic thresholds are correct; that question is out of its scope.
The relative-change figures are a useful lesson in reading percentages. French respondents showed the largest between-wave increase of any country — around 500% — while sitting near the bottom of the table in absolute terms and carrying the lowest country odds ratio in the regression (0.24, reference Australia). US respondents showed the smallest relative increase (about 47%) while reporting the highest absolute rates in both waves. Large percentage increases from small bases are arithmetic, not epidemiology.
Prevalence and rule-making in competitive settings — chess, esports, regulated sport — run on a separate evidence base and a separate governance structure, covered in our piece on doping in esports and mind sports. This page covers work and study settings only.
What this dataset cannot tell you
- It cannot give you national prevalence. Not for the US, not for Germany, not for anywhere. Every figure above describes Global Drug Survey respondents, a self-selected sample of people who use drugs.
- The 2017 wave is a first-month subsample. The module was cut after one month to shorten the survey. Respondents who complete a survey in its first weeks may differ systematically from those who complete it later; nothing in the paper rules that out.
- The paper contradicts itself twice. Its Results section reports that about a quarter of enhancement users would like to reduce their use; its Conclusion says one third. Table 3 gives 28.3% of men and 27.7% of women — a quarter. Separately, the Results text reports a model R² of 0.39 while Table 4 reports Cox & Snell 0.13 and Nagelkerke 0.32. We have used the table values throughout. Neither discrepancy undermines the descriptive findings, but both are worth knowing if you are citing this paper yourself.
- The comparison has not been repeated at this scale. These figures describe data collected in late 2016 and early 2017. Anyone citing them in the present tense is citing a decade-old snapshot of a fast-moving behavior.
- Self-report throughout. Recall bias and deliberate misreporting apply to every figure. The authors note that a small number of respondents may have completed the 2015 survey more than once, though under 1% of the sample produced identical response sets.
Risks, and the finding nobody quotes
Stated once, by drug class, with no treatment plan attached. Non-medical use of prescription stimulants and use of illegal stimulants carries cardiovascular strain, sleep disruption, appetite suppression and dependence potential, with the additional hazard of unknown dose and purity in the illegal supply. Benzodiazepines carry tolerance, dependence and withdrawal risk, and acute cognitive impairment — which the perceived-effect data above independently reflects. Alcohol’s acute effects on attention, working memory and consolidation are among the best-established findings in the field. Non-medical possession and use of controlled stimulants is a criminal offense in most of the 15 countries surveyed.
These substances are contraindicated in pregnancy and breastfeeding, are not appropriate for anyone under 18, and interact meaningfully with cardiovascular medication, antidepressants and other CNS-active drug classes. Anyone with a cardiac, psychiatric or seizure history should treat this entire category as a conversation with a physician rather than a decision to make alone.
Two figures from the survey belong together. About a quarter of past-year enhancement users said they would like to use less in the coming year. Fewer than 3% said they would like to seek help, and fewer still had a plan to do so. That gap — wanting to reduce, not intending to ask — is the most actionable thing in the dataset. If you recognize yourself in it, a clinician is the correct next step, and the conversation is more routine than it feels.
The bottom line
There is no single prevalence figure for cognitive enhancers, and anyone offering you one is either quoting a survey without its sampling frame or averaging incompatible definitions.
What the largest global dataset shows is a behavior that is episodic rather than habitual, socially supplied rather than medically supplied, concentrated among people who already use drugs, and dominated — in the most recent wave — by two depressants that a large share of users said made their performance worse.
The reader this article is not for: anyone looking for a number to quote as national prevalence, and anyone hoping the data will settle whether these compounds work. It won’t. Self-reported benefit runs high; measured benefit in people without a clinical indication runs small, inconsistent, and occasionally negative. Those two facts have coexisted in this literature for fifteen years, and this survey does not resolve them — it documents the gap between them at a scale nobody had reached before.
For the wider evidence base underneath this one behavior, start with our overview of nootropics and cognitive enhancement.
Frequently asked questions
How many people use cognitive enhancers? No reliable global figure exists. Among Global Drug Survey respondents in 2017, 13.7% reported past-year use of prescription stimulants, modafinil or illegal stimulants for performance. That sample is self-selected from people who use drugs and is not representative. Representative surveys cited in the same paper found rates as low as 0.8%, and earlier 12-month estimates ranged 1–20% among students.
What is the most commonly used cognitive enhancer? Among GDS2017 respondents, cannabis was the most frequently reported substance used to improve performance at work or while studying, at 22.9%, ahead of alcohol at 15.7%. Prescription stimulants were fourth at 6.6% and modafinil sixth at 2.4%. The ranking reflects a sample recruited among people who use drugs, not the general population.
Do smart drugs work if you don’t have ADHD? The evidence is weaker than the marketing. Reviewing the trial literature, Maier et al. (2018) state that stimulants have weak positive and sometimes impairing effects on cognition in people without an underlying mental disorder. Self-reported benefit is much higher: 77.5% of prescription stimulant users and 75.8% of modafinil users in GDS2017 reported improvement. Those are beliefs, not measurements.
Are people using these daily? Mostly no. Among GDS2015 enhancement users, under 5% reported daily or near-daily use, and two-thirds reported ten or fewer occasions across twelve months. The most common pattern was a one-to-two-week period around exams or a busy work period, once or twice a year. The exception is illegal stimulants: 7.1% of that subgroup reported daily use.
Why do study drug prevalence estimates vary so much? Because definition drives the number more than behavior does. US lifetime estimates for non-medical prescription stimulant use have ranged from 5% to 35% without specifying purpose. Different studies count different substances, use different recall windows, sample different populations, and disagree on whether caffeine counts. Two honest surveys can differ tenfold without either being wrong.
Where do people get prescription stimulants for non-medical use? Among GDS2015 stimulant enhancement users, 47.8% named friends as their main source, 11.8% a dealer, 9.1% the internet, 6.1% a family member with a prescription, and 3.8% a physician. This is reported here as epidemiology. It describes an observed distribution pattern and is not guidance of any kind.
Reference
Maier LJ, Ferris JA, Winstock AR. Pharmacological cognitive enhancement among non-ADHD individuals — a cross-sectional study in 15 countries. International Journal of Drug Policy. 2018;58:104–112. doi:10.1016/j.drugpo.2018.05.009
Secondary citations referenced within that paper and named above: Wilens et al. 2008; Franke et al. 2011; Mache et al. 2012; Repantis et al. 2010; Wood et al. 2014; Farah et al. 2009; Maier et al. 2015; Peterkin et al. 2011.
This article is for informational purposes only. It describes what a survey population reported doing and does not recommend, endorse or provide instructions for any substance use. It is not medical advice, and it does not replace consultation with a qualified healthcare professional. Nothing here should be used to start, stop or adjust any prescription medication.