Phosphatidylserine for Memory: Old Bovine vs. Modern Soy/Sunflower
Educational overview — not medical advice. No trial here is a dementia-treatment or dementia-prevention study; the Alzheimer's-population trial (Crook 1992) measured cognitive test scores over 12 weeks in a small, discontinued-ingredient trial, not disease progression or diagnosis-level outcomes. This statement has not been evaluated by the FDA. This product is not intended to diagnose, treat, cure, or prevent any disease.
The honest answer
Real evidence for memory exists, but it's mostly old and uses an ingredient nobody sells anymore. Crook 1991 and Crook 1992 found bovine-cortex PS improved memory-test scores vs. placebo — but that ingredient was discontinued in the 1990s. The one direct soy-PS replication at the same dose (Jorissen 2001) found no significant difference from placebo on any outcome. A soy-PS-plus-omega-3 trial (Vakhapova 2010) found a benefit only in a higher-baseline-performance subgroup. The newest sunflower trial (Friling 2025) was null on its primary outcome. Nothing sold today has cleanly replicated the old bovine result.
Five trials, in chronological and ingredient order
Reading these in order matters, because the story is the ingredient switch, not just a list of trials. The two positive results both used bovine-cortex PS; every trial since, on soy or sunflower PS, has been null on its primary outcome or positive only in a subgroup.
| Study | Population & ingredient | Dose & duration | Result |
|---|---|---|---|
| Crook 1991 PMID 2027477 | RCT, n=149, age-associated memory impairment. Bovine-cortex PS (BC-PS). | 100 mg 3x/day = 300 mg/day, 12wk | Improved vs. placebo on learning/memory tests and clinical global ratings; greatest response in lower-baseline-performance patients. Ingredient discontinued industry-wide (BSE-era) — not sold today. |
| Crook 1992 PMID 1609044 | RCT, n=51, probable Alzheimer's disease. Bovine-cortex PS. | 100 mg 3x/day, 12wk | Improved on several cognitive measures, most pronounced in milder-impairment patients. Same bovine-source caveat. Small, short, not a treatment/cure claim. |
| Jorissen 2001 PMID 11842880 | RCT, n=120, age-associated memory impairment (>57y). Soy-derived PS — the direct replication of the Crook design. | 300 or 600 mg/day, 12wk + 3wk washout | NULL — "no significant differences...between the treatment groups" on any outcome, including delayed recall/recognition primaries. Authors concluded soy-PS "does not affect memory or other cognitive functions." |
| Vakhapova 2010/2011 PMID 20523044 / 21711517 | RCT, n=157 randomized (122 analyzed), non-demented elderly with memory complaints. Soy-PS + omega-3 combination. | 300 mg PS/day, 15wk; extension to 100mg maintenance, 30wk total | Significant improvement in immediate verbal recall vs. placebo, concentrated in a higher-baseline-performance subgroup. Combination product — can't isolate PS alone. 30wk extension: well tolerated, reduced diastolic BP, slight weight gain (longest human PS safety data in this pack). |
| Friling 2025 PMID 41318468 | RCT, healthy children 8-12y. Sunflower-derived PS (Sharp-PS® Green) — the modern ingredient actually sold today. | 100 mg/day, 12wk | NULL on primary/secondary outcomes, full cohort. Benefit only in a pre-defined below-median-baseline-visuospatial-memory subgroup. Four co-authors are IFF Health employees (IFF commercializes Sharp-PS® Green) — disclosed industry funding. |
Why this is chosen over a cortisol/stress page for the site's one "best-for" spot
The cortisol-blunting signal is real across four small trials (see the dosage guide for the doses), but it has never been tied to a functional performance or well-being outcome the way the memory trials at least attempt to measure — one trial (Kingsley 2006, not covered on this page) even found an exercise-capacity benefit with no cortisol effect, and the cortisol-positive trials measured no functional outcome at all. Memory is the "least-unsupported" thread once bovine and soy/sunflower evidence are honestly separated — not because the memory evidence is strong, but because it's the thread with the most attempts at direct replication, and those replication attempts are exactly what expose the honest picture: the ingredient changed, and the result changed with it.
The net honest verdict Nothing sold today has replicated the old bovine result cleanly. The soy-PS direct replication at the classic dose was null. The one soy-PS combination trial with a positive result was subgroup-only and can't isolate PS's contribution. The newest sunflower trial was null on its primary outcome. Real memory evidence for phosphatidylserine, as actually sold today, is thin.
Frequently asked questions
Does phosphatidylserine improve memory in older adults?
Depends on the ingredient. Bovine-cortex PS trials were positive (now discontinued). The direct soy-PS replication was null. A combination trial was positive only in a subgroup. The newest sunflower trial was null on its primary outcome.
Why did the soy replication of the classic trial fail?
Jorissen 2001 replicated Crook's exact design and dose with soy PS instead of bovine PS and found no significant benefit on any outcome — consistent with EFSA's finding that the two are different substances.
What did the Vakhapova trials actually find?
A soy-PS-plus-omega-3 combination improved recall, but only in a higher-baseline-performance subgroup, and the combination design can't isolate PS's own contribution.
What did the Friling 2025 sunflower trial find?
Null on primary/secondary outcomes in the full cohort of children; a benefit only in a below-median-baseline subgroup. Industry-funded (IFF Health, Sharp-PS® Green's maker).
Related
- Phosphatidylserine: what it is & who it's for
- PS dosage guide — 300mg memory dose vs 600-800mg cortisol dose
- Best PS — ranked by disclosed source, then cost per mg
- The FDA "qualified" claim, explained
Sources
- Crook TH, Tinklenberg J, Yesavage J, et al. "Effects of phosphatidylserine in age-associated memory impairment." Neurology. 1991. PMID: 2027477
- Crook TH, Petrie W, Wells C, Massari DC. "Effects of phosphatidylserine in Alzheimer's disease." Psychopharmacol Bull. 1992. PMID: 1609044
- Jorissen BL, Brouns F, van Boxtel MP, et al. "The influence of soy-derived phosphatidylserine on cognition in age-associated memory impairment." Nutr Neurosci. 2001. PMID: 11842880
- Vakhapova V, Cohen T, Richter Y, Herzog Y, Korczyn AD. "Phosphatidylserine containing omega-3 fatty acids may improve memory abilities in non-demented elderly with memory complaints." Dement Geriatr Cogn Disord. 2010. PMID: 20523044
- Vakhapova V, Richter Y, Cohen T, Herzog Y, Korczyn AD. Safety and tolerability extension study. BMC Neurology. 2011. PMID: 21711517
- Friling R, Jackson G, Kennedy D, et al. Sunflower-derived phosphatidylserine (Sharp-PS® Green) in healthy children. Nutr J. 2025. PMID: 41318468