Creatine and Cognition, Memory and Mental Fatigue: What Research Shows

How the evidence is labeled

Promising but limited means a few small or short trials point the same way. Mixed means studies disagree. Insufficient evidence means too little to judge. Proper evidence not found means we looked and found nothing usable. Biomarker changes, such as brain scans, are kept apart from how well people actually performed on tests.

What the research found, by group

1. Healthy adults, all ages

   Mixed

   A 2018 review of six trials (281 people) found possible gains in short-term memory and reasoning, and conflicting results for other skills. [3] A 2024 review of 16 trials (492 people, healthy and ill) found no clear gain in overall cognitive function or executive function. [4]

2. Younger adults

   Mixed

   The 2018 review reported that performance on cognitive tasks stayed unchanged in young people. [3] A 2023 meta-analysis found almost no memory change at ages 11 to 31. [5]

3. Older adults

   Promising but limited

   The same 2023 meta-analysis found a memory gain at ages 66 to 76, with wide uncertainty and large differences between trials. [5]

4. Vegetarians

   Promising but limited

   A 2020 review of nine studies reported memory and intelligence gains, with moderate to high risk of bias. [6]

5. Sleep loss

   Promising but limited

   Small lab trials, described below.

6. Brain fog

   Insufficient evidence

   Proper evidence not found: our PubMed searches for creatine with "brain fog" and with long COVID fatigue found no trial of creatine for it.

Older adults

The 2023 meta-analysis pooled ten trials of memory in healthy people. Overall, creatine did better than placebo. In the older-age group the effect was larger, but the range of possible effects was wide (95% CI 0.22 to 1.55) and the trials differed a lot from each other (I2 = 83%). [5] A 2026 review found six studies in 1542 older adults, but only two were double-blind trials; four estimated creatine intake from food recall. [7] Quality was rated "good" for one study, "fair" for two and "poor" for three. [7] For muscle and bone in later life, see creatine for older adults, women and cognition.

Vegetarians

People who eat little or no meat tend to start with lower creatine stores, so trials looked at them separately. The 2020 review found supplementation had no effect on brain phosphocreatine, even though memory scores improved. [6] That mismatch keeps the result tentative. See creatine for vegetarians and vegans.

Sleep loss

Sleep loss is where the most specific signals appear, from small laboratory trials.

• 2006 trial. Ten people in the creatine group took creatine four times a day for seven days before 24 hours without sleep. The authors reported a positive effect on mood and on tasks that stress the prefrontal cortex. [8]
• 2024 trial. A single high dose (0.35 g per kg of body weight) was given during about 21 hours of sleep deprivation, and the authors reported that it partly reversed fatigue-related decline in cognitive performance. [9]
• 2026 trial. Twenty-nine healthy people received a single lower dose (0.2 g per kg) and showed less decline in cognitive performance during sleep deprivation. [10]

These are study amounts, not advice, and they measured lab performance, not everyday life. For sleep itself, see better sleep.

Why the picture is still unclear

• A systematic review noted that creatine can raise brain creatine only "to a limited extent", and asked future studies to measure brain content. [2]
• A 2026 review said the brain response may depend on dose and duration. [11]
• Trials used amounts from about 2.2 to 20 g a day and lasted from 5 days to 24 weeks, so results cannot simply be added up. [5] Reviewers say the best protocol for raising brain creatine is still to be determined. [12]
• Some authors of the reviews cited here disclose ties to creatine makers; each citation below notes this where we saw it.