Cognition · CREA-COG-01 · v2
The question is whether a plain creatine supplement makes healthy, rested adults think better day to day. The best direct tests here point mostly the wrong way for that broad claim: the largest clean trial found only a small effect that did not reach statistical significance on either of its two main measures [1], and two other trials found nothing [2][4]. Where creatine clearly helped, it was either paired with another compound [3][5], combined with exercise [11], or given to people under stress such as sleep deprivation [7] or low oxygen [10], where it restored impaired performance rather than lifting a normal baseline. The effects also varied by task within single studies, so a blanket 'improves cognitive performance' does not hold. The panel concluded a small, scattered, mostly conditional signal exists, but the corpus does not support the unqualified general claim.
The direct baseline evidence in healthy adults is mixed and leans null. The largest clean RCT [1] (N=123, 5 g/day for 6 weeks) reported Bayesian support for a small effect but missed significance on both primary outcomes (BDS p=0.064, RAPM p=0.327), with no benefit on exploratory tasks. A dose-response RCT [2] (N=30) found no significant effect at 10 or 20 g/day, and a conference-abstract RCT [4] (N=25) found no significant between-group changes. The clearest positives come from a combination [3], where the most consistent reaction-time gains were in the CrM+GAA arm, and from conditional states: single-dose creatine mitigated sleep-deprivation decline [7] and restored hypoxia-induced attention loss [10]. The contested point: The division is whether conditional and combination positives count as support for a baseline claim. [7] and [10] show clear benefit, but only as restoration of performance impaired by sleep deprivation or oxygen deprivation, not improvement in rested, normoxic adults; and [3][5][11] pair creatine with GAA, caffeine, or exercise. The panel treated these as indirect or confounded rather than clean support.
All 11 papers behind this claim with their DOIs and what each does to it, the panel’s arbitration, the evidence gap map and the open research questions. Free, and sent to your inbox.
We’ll also send you other claims from Creatine, one at a time, and tell you when a release or a pivotal trial changes a grade. Unsubscribe by replying.
One reviewer's only job is to attack the finding. The challenges below were weighed before the grade was set.
The two largest clean RCTs in healthy adults deliver a borderline-but-non-significant primary ([1], BDS p=0.064, RAPM p=0.327) and an outright null ([2]), with a third null ([4]). The direct baseline evidence contradicts an unqualified 'improves'.
The most robust positive effects [7][10] are observed only under an energetic insult (sleep deprivation, hypoxia) and describe restoration of an impaired state, not improvement of baseline cognition in rested, normoxic healthy adults — indirectness relative to the claim's wording.
Where it remains uncertain. The most convincing benefits appear only under sleep deprivation [7] or hypoxia [10], or alongside a co-agent such as GAA [3] or caffeine [5], none of which establishes baseline improvement.
Aleatory
Even the supportive signal in [1] is small and near the significance boundary (BDS p=0.064), so sampling noise cannot be separated from a true effect without larger trials.
Epistemic
Only [1] is large (N=123); [2] (N=30) and [4] (N=25) are small, and [4] is a conference abstract with no full text, limiting what the corpus can resolve.
Inferential
Clear benefits in [7] and [10] arise under sleep deprivation or hypoxia and describe restoration of impaired performance, which does not license an inference to rested, normoxic baseline cognition.
The load-bearing papers from a corpus of 11, each classified by what it does to the claim. Every DOI is taken from the corpus the claim was graded on.
Largest direct trial (N=123): at most a small, borderline, non-significant effect on one primary outcome; null on the other and on exploratory tasks.
“Bayesian evidence supported a small beneficial effect of creatine. The creatine effect bordered significance for BDS (p = 0.064, η2P = 0.029) but not RAPM (p = 0.327, η2P = 0.008). There was no indication that creatine improved the performance of our exploratory cognitive tasks.”
Panel auditInternal validity: high · scope of inference: Healthy adults (mixed vegetarians/omnivores) on reasoning and working-memory tasks after 6 weeks of 5 g/day.Key limitations: primary outcomes non-significant · small effect only · borderline p-value for one test
Direct null in healthy young adults across doses — contradicts the claim.
“Overall, six weeks of Cr supplementation at a moderate or high dose does not improve cognitive performance or change PFC activation in young adults.”
Panel auditInternal validity: medium · scope of inference: Healthy young adults on processing speed, episodic memory and attention after 6 weeks of moderate-to-high dose creatine.Key limitations: small sample · no effect at either dose · short duration
Positive signal in healthy adults, but most consistent effects driven by CrM+GAA combination, limiting attribution to creatine alone.
“Results provided evidence that CrM supplementation (10 g/d) with and without GAA (2 g/d) supplementation can enhance cognitive function in healthy adults and that the more consistent positive effects were observed from supplementation of CrM and GAA.”
Panel auditInternal validity: medium · scope of inference: Healthy recreationally active adults (mean ~41 years) on reaction time and memory tasks after 6 weeks.Key limitations: many comparisons tested · strongest effects from combination not creatine alone · some creatine-vs-placebo contrasts only trends
Cognitive gain required caffeine co-ingestion in trained athletes; creatine effect not isolable.
“Creatine nitrate and caffeine that were co-ingested significantly enhanced cognitive function, as indicated by improved scores in the Stroop Word-Color Interference test ( p = 0.04; effect size = 0.163).”
Panel auditInternal validity: medium · scope of inference: Resistance-trained male athletes on acute Stroop performance 45 min after a combined creatine-nitrate/caffeine dose.Key limitations: creatine effect confounded with caffeine · creatine nitrate not monohydrate · acute single-session design · n=12 males only
4 of 11 papers shown. The full review lists all of them, with the panel’s arbitration →
Claims are graded at each quarterly release, and re-graded between releases only when a pivotal trial lands. Every change is a new version of the claim.
A small, inconsistent, and largely conditional signal exists, but the corpus does not carry the unqualified general claim that creatine improves baseline cognition in healthy adults.
Release 2026-q4 re-graded every claim from its corpus under protocol v0.19. An earlier release's grade is shown for the record; it came from an earlier protocol and is not a movement.
Assumptions that carry weight but are not themselves established.
The panel records the gap that would most move the claim, and the study that would close it.
Adequately powered baseline RCT with a pre-specified domain. The cleanest large test [1] was underpowered to confirm a small effect and missed significance on both primaries, leaving the baseline claim unresolved. The panel rates this the highest-importance open gap for the claim, and names the study design that would close it: A large, preregistered, placebo-controlled RCT in healthy rested adults with a single pre-specified primary cognitive domain and creatine monohydrate alone.
Open question: Does creatine monohydrate alone improve a pre-specified cognitive domain in healthy rested adults at adequate power? Why it is still open: The one large trial [1] supported only a small effect and missed significance on both primaries (BDS p=0.064, RAPM p=0.327), while [2] and [4] were null.
Priors is a sceptical first reader, not peer review, and not medical advice.
Release 2026-q4 · graded 21 Oct 2026 · protocol v0.19