Peptides for focus and memory research divides along a line that most summaries miss: on one side, short regulatory peptides derived from endogenous sequences — ACTH fragments, tuftsin analogs, angiotensin IV derivatives — that modulate neurotrophin expression and neurotransmitter systems; on the other, neurotrophic mimetics and brain-derived peptide preparations aimed at synaptogenesis and neuronal survival. The two groups differ in mechanism, in evidence quality, and in the experimental models where they make sense. This overview separates them, compares the reference standards, and sets out the selection logic. Everything referenced is supplied for laboratory research use only.
What cognitive endpoints look like in this literature
Preclinical cognition work uses a small and well-established set of readouts. Behavioural endpoints in rodents include Morris water maze and Barnes maze for spatial memory, novel object recognition for recognition memory, passive avoidance for aversive learning, and elevated plus maze or open field for the anxiety dimension that several of these compounds also touch. Molecular endpoints are BDNF and NGF expression by ELISA or qPCR, hippocampal long-term potentiation in slice electrophysiology, dendritic spine density on Golgi staining, and neurogenesis markers such as doublecortin and BrdU incorporation in the dentate gyrus.
Human work, where it exists, uses standard cognitive batteries and, in the Russian clinical literature specifically, a set of instruments and endpoints that do not map cleanly onto Western trial conventions. That mismatch is a recurring problem when reading the Semax and Selank evidence base, and it is a limitation rather than a dismissal.
The classes studied
ACTH fragment analogs
Semax is a heptapeptide, Met-Glu-His-Phe-Pro-Gly-Pro at 813.93 Da, built from ACTH(4-7) with a Pro-Gly-Pro tail added to resist enzymatic cleavage. Crucially, the modification removes the corticotropic activity of the parent hormone, so it engages central effects without driving the adrenal axis. Published rodent work reports rapid and substantial increases in BDNF and its receptor TrkB in the hippocampus within hours of administration, along with effects on the dopaminergic and serotonergic systems. N-Acetyl Semax Amidate adds N-terminal acetylation and C-terminal amidation, both standard strategies for extending peptide survival against exopeptidases.
Tuftsin analogs
Selank is a heptapeptide at 751.89 Da based on the immunomodulatory tetrapeptide tuftsin, again with a Pro-Gly-Pro extension. Its published profile is anxiolytic rather than stimulant: reported mechanisms include inhibition of enkephalin degradation, modulation of GABA-A receptor subunit expression, and changes to monoamine turnover. It is the compound researchers reach for when the question concerns anxiety-related interference with cognition rather than cognition directly, and it is frequently paired with Semax for that reason.
Neurotrophic mimetics
Dihexa, 439.55 Da, is an N-hexanoic-Tyr-Ile-(6)-aminohexanoic amide derived from angiotensin IV. Published work from Washington State University reports that it potentiates hepatocyte growth factor signalling at the c-Met receptor and increases dendritic spine density and synaptogenesis in hippocampal cultures at picomolar concentrations. P-21 is a CNTF-derived peptidergic compound studied for adult hippocampal neurogenesis and BDNF elevation, with published work in transgenic models of neurodegeneration. PE-22-28 is a spadin analog derived from the sortilin propeptide, acting as a TREK-1 potassium channel blocker — an unusual and well-defined target studied mainly in depression rather than cognition models.
Brain-derived peptide preparations
Cerebrolysin is not a single molecule. It is a standardised enzymatic hydrolysate of porcine brain protein, containing low-molecular-weight peptides and free amino acids, and it has been studied in registered clinical trials for stroke and dementia in Europe and Asia. Its composition means it cannot be characterised by the same analytical methods as a synthetic peptide — batch consistency is a manufacturing question rather than an HPLC purity figure, and researchers should account for that accordingly.
What the published record shows, by study type
Cell and slice work
Dihexa has the cleanest in-vitro signal: dendritic spine formation in hippocampal culture at very low concentrations, with the c-Met dependence demonstrated by receptor blockade. Semax raises BDNF and TrkB expression in cultured neurons and in hippocampal tissue. P-21 increases neurogenesis markers in culture and in vivo. PE-22-28's TREK-1 blockade is confirmed by electrophysiology, which is a stronger form of target validation than expression data.
Rodent studies
The Russian literature on Semax and Selank is substantial and covers learning, ischaemia and stress models across several decades. Dihexa has published cognitive rescue data in scopolamine and lesion models. Cerebrolysin has extensive rodent stroke and dementia model data. The general weaknesses are inconsistent blinding, reliance on single behavioural tasks, and — for the Russian-origin compounds — limited replication by laboratories outside the originating network.
Human data
Semax and Selank are registered medicines in Russia with clinical trial data published largely in Russian-language journals; neither is approved in the United States or the European Union. Cerebrolysin has the most substantial international trial record of anything on this page, including meta-analysed stroke and vascular dementia data with mixed conclusions. Dihexa, P-21 and PE-22-28 have no human trial data at all.
Comparison table: cognitive research reference standards
| Compound | Origin | MW | Reported target | Model where best characterised | Research sizes |
|---|---|---|---|---|---|
| Semax | ACTH(4-7) analog | 813.93 Da | BDNF/TrkB expression, monoamines | Rodent learning and ischaemia | 5-30 mg |
| N-Acetyl Semax Amidate | Modified Semax | Semax + acetyl/amide | As Semax, extended stability | Comparative stability work | 10 mg |
| Selank | Tuftsin analog | 751.89 Da | Enkephalinase inhibition, GABA-A | Rodent anxiety models | 5 mg, 10 mg |
| Dihexa | Angiotensin IV derivative | 439.55 Da | HGF/c-Met potentiation | Hippocampal spine density | 10 mg, 20 mg |
| P-21 | CNTF-derived | 1017.24 Da | Neurogenesis, BDNF | Transgenic neurodegeneration | 5 mg |
| PE-22-28 | Spadin analog | 773.89 Da | TREK-1 channel blockade | Rodent depression models | 10 mg |
| Cerebrolysin | Porcine brain hydrolysate | Mixture | Multiple neurotrophic fractions | Stroke and dementia trials | 5 mL, 10 mL ampoules |
| Noopept | Prolyl dipeptide ester | 318.37 Da | Cycloprolylglycine metabolite, BDNF | Rodent learning models | 10-30 mg capsules |
Noopept is included as a small-molecule comparator rather than a peptide in the synthetic sense; it is a prolyl dipeptide ester whose active metabolite is the endogenous dipeptide cycloprolylglycine.
How researchers choose peptides for focus and memory
- Cognition or affect? Semax and Dihexa are studied against learning and synaptic endpoints; Selank and PE-22-28 against anxiety and mood endpoints. Using an anxiolytic in a spatial memory task and reporting improved performance risks measuring reduced stress interference rather than memory.
- Is the target defined? Dihexa's c-Met dependence and PE-22-28's TREK-1 blockade allow receptor-level controls. Semax and Selank act on expression and turnover, so their mechanistic claims close less tightly.
- Does the compound need to cross the blood-brain barrier? Intranasal delivery is standard in the Semax and Selank literature specifically because of it, and the pre-diluted nasal formats exist for that reason. Slice and culture work sidesteps the question entirely.
- Is the material a defined molecule? Cerebrolysin is a mixture and cannot be specified by sequence or mass. Protocols that handle it as a single compound will not be reproducible across batches.
The full catalogue for this goal is at peptides for focus and memory, with the narrower grouping under nootropic peptides. The pairing researchers ask about most is covered in our Semax versus Selank comparison.
Formats and handling
Lyophilised vials remain the reference format and are the only one that allows concentration to be set at reconstitution. Pre-diluted nasal solutions such as the NAD+ nasal format and the equivalent Semax and Selank sprays remove reconstitution error at the cost of fixed concentration, and they introduce a vehicle — preservative, tonicity agent, buffered pH — that must appear in the control condition. Dissolving strips and capsules exist for the orally available small molecules in this category and are not appropriate for the peptides, which are degraded in the gut.
Handling notes: Semax contains methionine at position 1 and is therefore susceptible to methionine oxidation, which shows up as a satellite peak on the chromatogram and is a real degradation route in poorly stored material. Dihexa is active at picomolar concentrations in culture, so serial dilution error dominates the assay and carrier protein in the diluent is often needed to limit adsorption losses to plasticware. All lyophilised material is stored sealed at -20 °C, protected from light, and aliquoted at reconstitution rather than repeatedly thawed.
Adjacent compounds in the same protocols
A number of materials sit alongside these peptides in cognitive research without sharing their chemistry. Cortexin, like Cerebrolysin, is a brain-tissue peptide fraction rather than a defined molecule and carries the same batch-characterisation caveat. Pinealon is a Khavinson-series tripeptide studied for neuronal endpoints, with the same pattern of a large internal literature and thin external replication that characterises that whole series. Methylene blue appears in cognitive protocols because of its bioenergetic action in neurons rather than any neurotrophic mechanism, and its biphasic concentration response applies there as elsewhere. Bromantane and phenylpiracetam are small-molecule stimulant and racetam-class compounds respectively, sharing an application area with these peptides and nothing else.
The practical consequence is that a literature review organised around the phrase rather than around mechanism will pull in compounds that cannot be meaningfully compared. Grouping by target — neurotrophin signalling, ion channel, monoamine, bioenergetic — produces a far more useful map, and it is the organisation reviewers expect in a methods section.
Common design errors
The most frequent is the single-task behavioural conclusion. A change in one maze does not establish a memory effect, and the standard remedy — a second, mechanistically different task plus a locomotor control — is cheap and rarely done. Related to it is failing to separate anxiolysis from cognition, which the anxiolytic compounds in this group make very easy to get wrong.
The second is reading BDNF elevation as a cognitive result. BDNF expression is a molecular intermediate, not an endpoint, and it moves in response to handling stress, exercise and circadian time as well as to the compound under test.
The third is adsorption loss with the picomolar-active compounds. Peptides at very low concentrations bind to polypropylene and glass, and without a carrier protein or low-binding plasticware the delivered concentration can be a fraction of the nominal one — which produces irreproducible concentration-response curves that look like biological variability.
A fourth error is ignoring circadian timing. Hippocampal BDNF expression, long-term potentiation thresholds and rodent performance in every standard maze all vary with time of day, and rodents tested during their inactive phase behave differently from those tested during the active phase. Studies that do not fix and report the testing window are introducing a variance term larger than most of the effects being sought.
Purity, identity and regulatory status
The synthetic peptides here are short and should arrive at 98% or better by HPLC with mass-spectrometric identity confirmation. For Semax specifically, ask whether the chromatogram shows the oxidised-methionine satellite; for the acetylated and amidated variants, the mass difference from the parent is small but unambiguous on a mass spectrum, and it is the only reliable way to confirm you have the modified molecule. Cerebrolysin cannot be assessed this way and is a manufacturing-consistency question instead.
Semax and Selank are registered medicines in Russia and are not approved in the United States or the European Union. Cerebrolysin is approved in a number of countries but not in the United States. Dihexa, P-21 and PE-22-28 have no approval anywhere and no human trial data. Everything referenced here is research-grade material supplied for in-vitro and laboratory research use only, and is not for human or veterinary administration.