Supplements for studying: what helps students learn (and what doesn't)

Studying draws on attention, working memory (holding information in mind while working a problem) and getting new material into long-term memory. Researchers measure these with lab tests of attention, reaction time and recall, and less often with real quiz and exam scores.

One caveat up front. As of November 12, 2026 we found no randomized trials in high school or college students showing that the supplements on this page improved exam scores, course grades or grade point average (PubMed and Europe PMC searches, November 11, 2026 and November 12, 2026). The supplement evidence below comes mostly from lab tests in healthy adults and is not proof that a product raises grades.

Sleep and study methods have the student-specific evidence, so they come first. Each compound links to its full monograph, and every grade comes from our evidence matrix. Our page on how we grade evidence explains the rubric.

Sleep before and after learning

Short-term sleep loss affects the skills studying depends on. A meta-analysis of 70 articles covering 147 cognitive tests measured the effects of short-term total sleep deprivation (under 48 hours) on simple attention, complex attention, working memory, processing speed, short-term memory and reasoning [1]. The effects ranged from small and non-significant for reasoning accuracy to large for lapses in simple attention (g = -0.776) [1].

Sleep matters on both sides of a study session. Two meta-analyses of experiments involving at least one full night without sleep found worse memory for new material when the sleep loss came before learning (Hedges' g = 0.621) and when it came after learning (Hedges' g = 0.277) [2]. The authors also found signs of publication bias, and most of the studies had very low statistical power, so the true sizes are uncertain [2]. In a brain imaging study, a single night of sleep deprivation reduced activity in the hippocampus, a memory region, while people learned new material, and they remembered less of it later [3].

What about real grades? In an introductory college chemistry class, 88 students completed a study using wearable sleep trackers [4]. Better quality, longer and more consistent sleep was associated with better grades, and sleep measures accounted for nearly 25% of the variance in academic performance [4]. Sleep on the single night before a test was not related to the score; sleep over the month and the week before was [4]. These are correlations from an observational study, which cannot show that sleep caused the grades.

Larger reviews are more mixed. In a meta-analysis of studies in medical students, academic grades correlated with sleep quality scores (r = 0.15) but not with sleep duration (r = 0.03) [5]. In US secondary-school students, a meta-analysis of 11 studies found negligible correlations with academic performance, not significant for sleep duration and small but significant for sleep quality, and the authors pointed to inconsistent definitions and measures across studies [6]. Evidence on sleep and grades is mixed, possibly in part because studies defined and measured sleep and performance differently [6]; the experimental evidence on sleep loss and memory is more direct [2]. Our sleep hub covers the outcomes we grade for sleep.

Study methods with evidence

How students study has its own research base. A review of 10 common learning techniques rated practice testing and distributed practice (spacing study sessions out) as high utility, because they helped learners of different ages and abilities across many kinds of tests, including in real classrooms [7]. The same review rated summarization, highlighting, the keyword mnemonic, imagery for text and rereading as low utility, and noted that rereading and highlighting, though most students report using them, did not consistently improve performance [7].

Classroom studies point the same way for testing. A meta-analysis of 222 classroom studies with 48,478 students found that quizzing raised academic achievement to a medium extent (g = 0.499), with the size of the effect depending on factors such as corrective feedback and the number of test repetitions [8]. For spacing, a meta-analysis of 317 experiments found that the gap between study sessions that gave the best retention grew longer as the time until the final test grew longer [9].

A short bout of exercise

Exercise is the other free option studied. A 2025 meta-review pooled 30 systematic reviews with meta-analyses, covering 383 studies and 18,347 participants from children to older adults, and found that a single bout of exercise improved cognitive function with a small-to-medium effect (mean SMD = 0.33) [10]. Benefits appeared for attention (mean SMD = 0.37), executive function (mean SMD = 0.36) and memory (mean SMD = 0.23), and neither age, cognitive status nor the intensity, type and duration of exercise changed the result [10]. Benefits were largest when cognition was tested after exercise [10].

For caffeine, the main evidence here is on attention. A 2025 meta-analysis of 31 randomized, double-blind, placebo-controlled trials in 1,455 rested healthy adults found caffeine improved both accuracy (g = 0.27) and reaction time (g = 0.28) on attention tests [11]. The effect did not depend on habitual caffeine intake, task complexity or which kind of attention was tested [11]. Reaction time kept improving at higher doses, but accuracy improved only up to a point and then declined [11].

Two limits matter for students. First, part of the benefit in daily users may be reversal of withdrawal: a 2005 focused review found little evidence that long-term caffeine use improves performance or mood compared with long-term abstinence [14], although the 2025 meta-analysis found that the effect did not depend on habitual intake [11]. Withdrawal itself includes difficulty concentrating, irritability and feeling foggy, with symptoms starting 12-24 hours after the last dose [15]. Second, caffeine can shorten sleep [12], and in experiments a full night without sleep was linked to worse memory [2]; none of these studies tested whether caffeine's effect on sleep changes learning.

Caffeine and sleep

A meta-analysis of 24 studies found caffeine reduced total sleep time by 45 minutes and sleep efficiency by 7%, and lengthened the time to fall asleep by 9 minutes [12]. Its authors estimated that total sleep time was not reduced when a coffee containing 107 mg was finished at least 8.8 hours before bedtime, or a 217.5 mg pre-workout serving at least 13.2 hours before [12]. In a placebo-controlled home study, 400 mg given 6 hours before bedtime still disrupted sleep [13]. EFSA reports an average caffeine half-life of about four hours in healthy adults, with a range of two to eight hours [36], and FDA notes wide variation in how sensitive people are and how fast they clear it [37]. Our guide to caffeine half-life covers timing studies in more depth.

Doses used in studies and safety limits

In the attention meta-analysis, doses of 200 mg or more improved reaction time and accuracy more than lower doses [11], and the home sleep study used 400 mg [13]. These are doses used in studies, not advice for any one person. Regulators have set reference points for total intake. EFSA found that single doses up to 200 mg raise no safety concerns for healthy adults, and that daily intakes up to 400 mg raise none for healthy adults other than pregnant women [36]. FDA has cited 400 mg a day for most adults [37]. For children and adolescents, EFSA proposed a safety level of 3 mg/kg of body weight per day [36]. FDA estimates that toxic effects, such as seizures, can occur with rapid consumption of around 1,200 mg, or less than 1/2 teaspoon of pure caffeine [37].

See the caffeine monograph for interactions and legal status by market. Anyone taking medicines should check with a doctor or pharmacist before adding caffeine or any supplement.

L-theanine is an amino acid found naturally in tea [20]. A 2014 meta-analysis of 11 randomized placebo-controlled studies found moderate effects in favor of combined caffeine and L-theanine on self-rated alertness and on accuracy when switching between tasks, in the first 2 hours after a dose [16]. Effect sizes tended to change more with the caffeine dose than with the L-theanine dose [16], so it is hard to tell how much L-theanine adds.

In a crossover trial of 27 healthy volunteers, 50 mg of caffeine with 100 mg of L-theanine improved speed and accuracy on an attention-switching task at 60 minutes [18]. In a trial of 44 young adults, 97 mg of L-theanine with 40 mg of caffeine improved task-switching accuracy and self-rated alertness, but had no effect on visual search, choice reaction time or mental rotation [19]. Both trials were run by researchers at Unilever R&D [18][19], a company that produces and markets tea [17]. A 2025 meta-analysis, co-funded by Lipton Teas and Infusions, with one author employed by Lipton and one by Unilever, found a small-to-moderate difference favoring the combination on attention-switching accuracy (SMD 0.33), and its authors noted the confidence intervals often left the direction and size of effects uncertain [17]. In a 2008 trial, L-theanine alone (250 mg) increased headache ratings and lowered scores on one arithmetic test [20]. The 2025 meta-analysis found a small-to-moderate difference favoring L-theanine alone over placebo on choice reaction time in the first hour (SMD -0.35), with the same uncertainty [17]. See the L-theanine monograph.

For students, the age of the people studied matters. A meta-analysis of randomized trials in healthy people found an overall memory improvement compared with placebo (SMD = 0.29), but the benefit came mainly from adults aged 66-76 (SMD = 0.88); in people aged 11-31 there was no meaningful effect (SMD = 0.03) [21]. Doses in those trials ranged from about 2.2 to 20 g a day, for 5 days to 24 weeks [21]. In a 6-week placebo-controlled trial of 22 young adults (average age 21) who were not sleep deprived, creatine did not improve any of the cognitive tests [23].

A 2024 meta-analysis of 16 trials (492 participants) found benefits for memory but no improvement in overall cognitive function or executive function, and it included patients with specific diseases as well as healthy people [22]. In a preregistered crossover trial of 123 adults taking 5 g a day for 6 weeks, Bayesian analysis supported a small benefit, but the effect on a working memory test only bordered on significance and there was no statistically significant effect on a reasoning test; side effects were reported more often on creatine than on placebo [24]. Evidence is mixed, and the results in younger adults lean null [21][23]. See the creatine monograph.

Bacopa monnieri is a herb with a long history in Ayurvedic medicine [25], though traditional use is not evidence of an effect. A meta-analysis of 9 randomized placebo-controlled trials (518 participants, healthy and impaired) of standardized extracts taken for 12 weeks or more concluded that bacopa may improve cognition, particularly speed of attention [25]. A 2026 network meta-analysis of 29 trials (2,107 participants) in healthy adults found that low-dose bacopa (300 to under 600 mg a day) did better than placebo for delayed memory, with no differences in sustained attention, selective attention or processing speed; its authors noted that the lack of direct comparisons limits the strength of this evidence [26]. Evidence is mixed: the two analyses disagree on attention, and the 2014 abstract reports no result for healthy adults alone [25][26]. Effects were measured after weeks of daily use, so these trials say nothing about a single dose before an exam [25]. See the bacopa monograph.

A meta-analysis of randomized trials in adults and school-aged children found no evidence of an effect of omega-3 on cognition in the general population, and marginal evidence of improved short-term memory in people with low omega-3 status [27]. In 159 healthy young adults aged 18-35, 12 weeks of 1 g a day of fish oil had minimal effects compared with placebo [28]. In a 6-month trial of 176 adults aged 18-45 with low dietary DHA intake, 1.16 g of DHA a day improved episodic memory in women and working memory reaction times in men [29]. Those findings apply to people with low intake, and results differed by sex [29].

The one randomized trial we found that measured school grades gave adolescents with a low Omega-3 Index krill oil (400 to 800 mg of EPA plus DHA a day) or a placebo for a year, and found no effect on subject grades or a standardized math test [35]. Very few reached the target omega-3 level, many dropped out or did not stick to the regimen, and the trial was co-funded by Aker BioMarine [35]. See the omega-3 monograph.

Ginkgo. N, no benefit shown A meta-analysis of randomized trials in healthy people (1,132 participants for memory) found no ascertainable benefit of ginkgo on memory, executive function or attention [30]. In university students, a single 120 mg dose improved a sustained-attention task and a pattern-recognition memory task, but after 6 weeks of 120 mg a day there were no effects on any cognitive test [31]. See the ginkgo monograph.

Rhodiola. C, limited A systematic review found that three of five randomized trials on mental fatigue reported a benefit, but every included study had a high risk of bias or reporting flaws, and the authors described the research as contradictory [32]. One small pilot trial tested the SHR-5 extract in students for 20 days during an examination period and reported less mental fatigue, with no difference on a text correction test [33]. See the rhodiola monograph.

Energy drinks. FDA reports that energy drinks generally have 54 to 328 mg of caffeine per 16 fluid ounces [37], and that medical experts advise against them for children and teens because of their sugar and caffeine [37]. In a systematic review of 57 studies in young people up to age 21, energy drink use was associated with short sleep, poor sleep quality and low academic performance; the authors noted that longitudinal studies are needed to show cause [34]. In the UK, drinks with more than 150 mg/l of caffeine must carry the label "High caffeine content. Not recommended for children or pregnant or breast-feeding women" [40]. In Australia, energy drinks may contain at most 320 mg of caffeine per liter and must be labeled as not suitable for young children, pregnant or lactating women and people sensitive to caffeine [39].

Anything promising better grades. As of November 12, 2026 we found no randomized trials showing that these supplements improved exam scores, course grades or grade point average in students (PubMed and Europe PMC searches, November 11, 2026 and November 12, 2026). The one trial that measured school grades, of krill oil, found no effect [35]. A claim that one of these supplements raises grades is not supported by any trial we found.

Students are often teenagers, and the safety reference points above are set for healthy adults. For children and adolescents, EFSA proposed a lower caffeine safety level based on body weight [36], and FDA reports that too much caffeine in children and teens can cause increased heart rate, palpitations, high blood pressure, anxiety, and sleep and digestive problems [37]. EFSA's daily limit for adults excludes pregnant women [36].

Under US law, supplement firms are responsible for evaluating the safety and labeling of their products before marketing, and FDA can act against adulterated or misbranded products after they reach the market [38]. Our guide are nootropics safe covers product quality and how to read a label. Each compound's monograph lists its own safety notes and known interactions. Anyone under 18, pregnant, or taking medicines should talk to a doctor or pharmacist before trying any supplement on this page.

Each grade badge links to the studies behind it. Grades apply to one compound and one outcome, so caffeine's grade for attention says nothing about memory, and the L-theanine grade applies to L-theanine taken with caffeine. Explore every graded compound on our work and study, focus and memory hubs.

What is the best supplement for studying?

As of November 12, 2026 we found no randomized trials showing that the supplements covered here improved students' grades; one krill oil trial in adolescents found no effect [35]. On lab tests, caffeine improved attention accuracy and reaction time in a meta-analysis of 31 placebo-controlled trials in rested healthy adults [11]. Sleep and practice testing have their own evidence, covered above [2][8].

Does pulling an all-nighter before an exam work?

In a meta-analysis of experiments, at least one full night without sleep before learning was linked to worse memory for the new material [2]. In a college chemistry class, sleep over the month and week before a test tracked grades, while sleep on the single night before did not [4].

Does creatine help students remember more?

Evidence is mixed. A meta-analysis of randomized trials found memory benefits mainly in adults aged 66-76 and no meaningful effect in people aged 11-31 [21], and a 6-week trial in 22 young adults found no improvement in cognitive tests [23].