Consensus checked against original sources: Creatine and Cognitive Performance

Cutoff: 12 October 2026. Consensus supports discovery and checks of specific source aspects; it is not a primary study, a full-paper audit or a causal probability meter.

Consensus describes Sandkuhler2023 as a controlled trial with varied multiweek dosing and negligible/inconsistent effects through secondary review links.

Original Methods specifies one5g/day dose for sixweeks per phase in a preregistered123-person randomized crossover. Original Results distinguishes a small uncertain BDS effect from no clearRaven/exploratory benefit; it does not prove literally zero possible effect. Original PMID37968687. Access: full_text_partial.

Consensus characterizes baseline dietary status as a major modifier and vegetarians as having intrinsically low tissuecreatine stores.

Original2023 Results directly tests the dietinteraction and finds no predicted difference; Rae2003 vegetarian recruitment does not supply an omnivore contrast or measured brain deficiency. Lower muscle stores cannot be assumed to mean lowerbrainstores. Original PMID37968687. Access: full_text_partial.

Consensus links the2024 cognitive andMRS changes but says elevated central phosphates cannot prove mediation.

OriginalMethods confirms a15-person randomized highdose sleep-loss crossover with spectroscopy/cognition and correlations. OriginalResults reports regionalcreatine/phosphate changes and performance shifts; no mediator perturbation occurs. Agreement on unproven mediation is retained, with multiple voxel/timepoint analysis limits stated. Original PMID38418482. Access: full_text_partial.

Consensus cites Gordji-Nejad2025 hemisphericATP balance alongside2024 cognitive effects.

Original2025 abstract identifies a new analysis of the previously published study. This is one15-person dataset with38418482, not independent replication or a second causal line. Original PMID40520504. Access: abstract_metadata.

Consensus says acute0.20–0.35g/kg protocols reverse sleep-loss deterioration in vigilance and verbal latency.

Original2026 selectedMethods uses a29-person crossover with participant-levelLMM plus pooled repeated-time contrasts. OriginalResults shows heterogeneous corrected outcomes and subgroup effects; a headline12percent or a single improved domain is not uniform reversal of impairment or a validated sex-specific dose rule. Original PMID42075005. Access: full_text_partial.

Consensus was asked to include CONCRET-MENOPA but does not describe its original findings in the rendered milestone table.

Original2026 abstract reports36 women randomized to three hydrochloride/ethylester formulations orplacebo and selected reactiontime/frontalcreatine gains. It is retained as a small recent positive with multiplecomparisons and unavailablebody, rather than erased by a provider coverage gap or accepted as broadmenopausecognition proof. Original PMID40854087. Access: abstract_metadata.

Consensus broadly acknowledges null geriatric executive findings without detailing the24week olderwomen factorial trial.

OriginalMethods/Results confirm56 healthy olderwomen, codedcreatine/placebo with andwithouttraining, and no cognitive differences; training mood/strength benefit was not an additivecreatine cognitive effect. Original PMID24098469. Access: full_text_partial.

Consensus gap text says simultaneous pre/post brain spectroscopy and cognition in restedomnivores is absent.

The examined2023 trial pairs cognition withfNIRS, notMRS, and is null. Other records pairMRS/cognition inchildren andmenopause, while2023adulttrial has noMRS. This bounded set supports a need for wellpowered pairedstudies but cannot verify an absolute absence claim about every restedomnivoretrial; providergrid counts are not a systematic census. Original PMID37759877. Access: full_text_partial.

Consensus correctly describes45 youngvegetarians,5g/day sixweeks and positiveworkingmemory/Raven scores, withnoMRSintr ial.

Originalabstract confirms the population/exposure/positive cognitive findings. Largeeffects remain a real result, but no samepersonbrainmeasurement establishes energeticmediation and subsequentlargertrial narrows expected magnitude. Original PMID14561278. Access: abstract_metadata.

Methods

Consensus was used to identify literature and cross-check specific claims about design, endpoints, scope and research gaps. Those aspects were checked against original primary abstracts/status records and selected original Methods/Results passages. Consensus output was not treated as original full text. Provider search omissions are not evidence that a study does not exist. Source access and examination receipts are retained. Scite use or monthly unavailability is recorded per case; provider citation counts do not change causal grades.