Tap an ingredient for detail, or use the Compare tab to see all five side by side. Ratings reflect the current human clinical trial evidence, not preclinical or mechanistic plausibility. Bar lengths are qualitative indicators of relative evidence weight, not scores derived from a validated scale. Comparisons are approximate and context-dependent; similar bar lengths do not imply equivalent clinical utility.
Citicoline
Moderate
Trial quality
Consistency
Effect size
Best evidence in adults 40+ with attentional decline
Phosphatidylserine
Moderate
Trial quality
Consistency
Effect size
Evidence mostly from bovine-derived form; soy-form trials smaller
Bacopa monnieri
Moderate
Trial quality
Consistency
Effect size
Repeated directional findings across small trials; requires 8–12 weeks; memory consolidation
Lion's Mane
Emerging
Trial quality
Consistency
Effect size
Promising preclinical biology; human trials small and unreplicated
Ginkgo biloba
Insufficient
Trial quality
Consistency
Effect size
Two large prevention trials (n=5,000+) found no benefit
Citicoline
Cytidine diphosphocholine — membrane support and cholinergic pathway
Meta-analysis (Jasielski et al., 2020; 14 trials, ~1,000 participants) found repeated directional findings for attention and memory across small trials, with substantial heterogeneity across populations.
Effect sizes in healthy younger adults are small and clinical significance is uncertain. Marketing to this group as a performance enhancer goes beyond the evidence.
Well tolerated in trials up to 12 months. No major safety concerns identified at studied doses.
Phosphatidylserine
Neuronal membrane phospholipid — cell signalling and membrane fluidity
Moderate evidence
Evidence strength
Moderate
Multiple RCTs; form-specific caveat
Evidence type
Clinical
Cognitive composite and memory outcomes
Studied dose
200–400 mg/day
Usually split into 2–3 doses
Critical caveat
Form matters
Bovine evidence stronger than soy-derived
Evidence dimensions
Trial quality
Replication
Effect size
Form consistency
Who the evidence applies to
Older adults with age-associated memory impairmentChildren with attention difficulties (limited evidence)Healthy younger adults (not established)
Key points
The only cognitive supplement with a qualified FDA health claim — though the qualification explicitly limits it to "very limited and preliminary" evidence for dementia risk in the elderly.
Most supplements now use soy-derived phosphatidylserine. The strongest evidence base used bovine-derived form, which has a different structural profile. This limits direct extrapolation.
No strong evidence for use in healthy adults without baseline cognitive concerns. Purchasing it as a general performance enhancer is not supported by current trials.
Lion's Mane
Hericium erinaceus — NGF-stimulating hericenones and erinacines
Emerging evidence
Evidence strength
Emerging
Small trials; no large replication
Evidence type
Mechanistic + early Clinical
Strong preclinical; limited human RCTs
Studied dose
1,000–3,000 mg/day
Variable across trials; standardisation unclear
Largest trial
n=41
Docherty et al., 2023 (crossover)
Evidence dimensions
Trial quality
Replication
Preclinical biology
Effect size
Who the evidence applies to
Mild cognitive impairment (small trial signal)Healthy older adults (single small trial)Healthy younger adults (early signal only)
Key points
Has strong preclinical NGF-related biology. The mechanistic rationale for investigation is well established. Human clinical translation has not yet been confirmed.
The most widely cited trial (Mori et al., 2009) had only 30 participants and used an outcome scale (Hasegawa Dementia Scale) not commonly used in contemporary cognitive research.
Marketing language for lion's mane has substantially outrun the evidence. The ingredient warrants further investigation; confident recommendations are premature.
Bacopa monnieri
Ayurvedic herb — bacosides modulating synaptic transmission and oxidative stress
Moderate evidence
Evidence strength
Moderate
Multiple RCTs; meta-analysis available
Evidence type
Clinical
Memory consolidation and verbal learning
Studied dose
300–450 mg/day
Standardised extract (CDRI 08 / KeenMind)
Time to effect
8–12 weeks minimum
Acute dosing does not produce the same benefit
Evidence dimensions
Trial quality
Replication
Effect size
Independence of trials
Who the evidence applies to
Adults with memory consolidation concernsOlder adults with age-related memory declineYounger adults (positive trials exist; effect size modest)
Key points
Meta-analysis (Pase et al., 2012; 9 RCTs) found generally similar directional findings for delayed recall memory across small trials. Effect most pronounced for memory consolidation rather than immediate or working memory.
Most positive trials used a standardised extract (CDRI 08). Evidence does not straightforwardly transfer to products of unknown bacosides content.
Publication bias is plausible and likely inflates the observed effect sizes: most positive trials have connections to commercial suppliers. Independent large-scale replication is limited.
Ginkgo biloba
Most extensively studied herbal cognitive intervention — and the most instructive failure
Insufficient evidence
Evidence strength
Insufficient
Large trials found no benefit
Evidence type
Clinical
Dementia incidence; cognitive composites
Largest trial
n=3,069
GEM trial (DeKosky et al., 2008); 6 year follow-up
Second large trial
n=2,854
GuidAge (Vellas et al., 2012); 5 year follow-up
Evidence dimensions
Trial quality
Replication
Effect on dementia
Effect (performance)
Who the evidence applies to
Dementia prevention (not supported by large trials)Cognitive performance (inconsistent; not established)
Key points
The GEM trial (n=3,069, 6 years, 240 mg/day EGb 761) found no reduction in dementia incidence. The GuidAge trial (n=2,854, 5 years) reached the same conclusion. These are well-designed, adequately powered trials.
Earlier smaller positive trials were likely inflated by small-study effects. This is the normal pattern when a literature moves from small exploratory trials to large definitive ones.
Some acute effects on attention have been reported in younger adults, but these are inconsistent across studies and have no meaningful large-trial confirmation.
Preclinical biology strong; clinical trials all small and unreplicated
Large-scale clinical trials dominant; consistently negative for primary outcomes
Key limitation
Effect size modest; heterogeneous populations
Form difference (bovine vs soy) limits extrapolation
Commercial ties in most trials; publication bias plausible
All trials small; no independent replication
Two large trials null; earlier positive trials likely inflated
PS = Phosphatidylserine. Ratings reflect human clinical trial evidence only. Preclinical and mechanistic evidence is not included in the rating. Where effect sizes are described as small to moderate, these typically correspond to standardised mean differences in the range of d=0.2 to 0.5. Effect sizes in this range are statistically detectable but may not translate into meaningful real-world cognitive improvement, and clinical significance remains uncertain. Bar lengths are qualitative and heuristic; they reflect overall evidence balance rather than a formal weighted model.