The Cognitive Stack: Focus, Memory and Clarity
A short stack with three tiers, and an honest statement of the ceiling: in someone already sleeping well, the compounds do very little.
The Short Answer
A cognitive stack is worth assembling only after the things that dominate cognitive performance are in place, because otherwise it is working on the smallest term. Sleep, attention environment and the reversible physiological contributors, iron, B12, thyroid function, mood, account for most of the variance in how sharp someone feels. With those handled, a small stack has a modest and real role, and the honest ceiling is worth stating in advance.
Prerequisites, Which Are Not Optional
Three things come before any compound, and each has a larger effect than the whole stack.
Sleep. Restriction produces measurable deficits in attention, working memory and processing speed, and memory consolidation depends on sleep specifically. No compound compensates. Sleep-disordered breathing is a common and often unrecognised contributor.
Attention environment. Task-switching carries a real resumption cost, and notification-driven interruption is the dominant modern source. Removing it is free and produces a change most people notice within days.
The reversible physiological contributors. Low iron stores, low B12, thyroid dysfunction, depression, anticholinergic medication burden, alcohol and hearing loss. Each is common, each produces cognitive complaints, and each is measurable.
If any of these is unaddressed, the stack below will disappoint, and the disappointment will be attributed to the wrong thing.
Tier One: Foundational
Creatine monohydrate, 3 to 5 g daily. The best-evidenced compound in this stack, with cognitive effects demonstrated under sleep deprivation and metabolic stress, and larger effects in vegetarians with lower baseline stores. Inexpensive, well tolerated, no loading required.
Omega-3, where oily fish intake is low. DHA is a structural component of neuronal membranes. Trials in already replete populations are neutral, which fits a shortfall-correction rather than an enhancement effect. 2 g or more of combined EPA and DHA, and the omega-3 index can verify it.
Vitamin B12, where status is low. Low B12 causes genuine and reversible cognitive symptoms and is common in older adults, vegans, and anyone on long-term metformin or acid suppression.
Iron, where ferritin is low. Particularly in menstruating women. Fatigue and poor concentration appear before anaemia.
Vitamin D, where status is low.
Notice that four of five are corrections rather than enhancements. That is the honest shape of this stack.
Tier Two: Situational
| Compound | Use | Notes |
|---|---|---|
| Caffeine, 50 to 200 mg | Acute alertness and vigilance | Best-evidenced cognitive compound; timing is the whole question |
| L-theanine, 100 to 200 mg | With caffeine, for subjectively smoother alertness | Modest trial support |
| Bacopa monnieri, 300 mg standardised | Memory, over 8 to 12 weeks | Not acute; gastrointestinal effects common |
| Citicoline, 250 to 500 mg | Attention | Better supported than most lesser-known options |
| Rhodiola, 200 to 400 mg | Fatigue-related performance | Effects largest in fatigued states |
| Tyrosine, 500 to 2000 mg | Performance under acute stress or sleep loss | Situational rather than daily |
Caffeine deserves the emphasis. Its acute effects are better documented than anything else here, and its cost is on sleep, which harms cognition more than the acute benefit helps. A half-life around 5 hours, longer in slow metabolisers, means an afternoon dose leaves a meaningful fraction circulating at bedtime, and it reduces slow-wave sleep even when it does not delay onset.
Bacopa is the one compound here with a delayed rather than acute effect, which means it needs a defined 12-week trial rather than a subjective daily judgement.
What to Leave Out
Ginkgo biloba. Tested properly in large prevention trials and it did not reduce cognitive decline or dementia incidence. A negative result rather than absent evidence.
High-dose vitamin E. No prevention benefit and harm signals at high doses in other endpoints.
Proprietary nootropic blends. Undisclosed doses make efficacy and interaction assessment impossible, and they frequently combine several compounds acting on the same pathway.
Racetams and prescription stimulants outside a clinical indication. These are not supplements, and modafinil and amphetamine-class agents carry dependence, cardiovascular and sleep consequences.
Nicotine. Genuine acute cognitive effects, and addictive, which makes it a poor trade regardless of delivery form.
Anything sold on a mechanism without human data at an achievable dose. Much of the nootropic category is cell-culture pharmacology at concentrations oral dosing does not reach.
Stacking five new things at once, which makes it impossible to know what did anything.
Assembling and Testing It
Weeks 1 to 4. Prerequisites only. Sleep regularity, notification discipline, and a blood panel covering ferritin, B12, thyroid function, vitamin D and HbA1c. No compounds.
Week 4. Add tier one: creatine, plus corrections for whatever the panel showed low.
Week 8. Assess. Many people stop here because the prerequisites and corrections did the work.
Week 8 onward, if a specific complaint remains: add one tier two compound, matched to the complaint. Attention problems point to caffeine timing and citicoline. Memory complaints point to bacopa with a 12-week window. Fatigue-related performance points to rhodiola.
How to judge it. Output on work you do repeatedly is the most valid measure available. Specific observations beat general ones: "I lose the thread after 40 minutes" is trackable and points at attention; "brain fog" is not measurable. Track sleep and mood alongside, since both explain most variation.
Do not use a consumer cognitive test at short intervals, since practice effects and recent sleep dominate the result.
The Ceiling, Stated Plainly
In a person sleeping adequately, exercising, without a nutrient shortfall, with a managed attention environment and no mood or thyroid problem, cognitive supplements produce small effects at best.
That is not a reason to take none of them. Creatine is cheap and broadly evidenced, correcting a genuine shortfall is worthwhile, and caffeine works. It is a reason to hold expectations proportionate, and to recognise that large reported effects usually come from correcting something in the prerequisites rather than from the compound credited.
The other honest point concerns long-term protection. Nothing in this stack has been shown to reduce cognitive decline over decades. The interventions with that evidence are cardiovascular risk control, physical activity, addressing hearing loss, sleep and social engagement, none of which is a compound. A cognitive stack is for how you function this month; neuroprotection is a different and mostly cardiovascular project.
And if cognition is genuinely declining rather than suboptimal, this stack is the wrong response. New or progressive change, particularly noticed by others, warrants assessment, because the reversible causes are specific and time matters.
The AEONNN Perspective
AEONNN builds this stack only after the prerequisites, because the prerequisites carry most of the effect. Sleep, attention environment and the reversible contributors, ferritin, B12, thyroid function, mood, anticholinergic burden and hearing, dominate how sharp a member feels, and a stack assembled around an unaddressed one will disappoint for the wrong reason.
Four of the five tier-one items are corrections rather than enhancements, which is the honest shape of Pillar 5 supplementation. The Evidence layer keeps ginkgo out on the strength of large negative prevention trials, which is a finding rather than a gap, and keeps proprietary blends out because undisclosed doses defeat both efficacy and interaction assessment.
The platform states the ceiling in advance: in a member sleeping well without a shortfall, these compounds produce small effects. And it separates two goals the market merges. This stack addresses how someone functions this month. Long-term neuroprotection runs through Pillar 4 cardiovascular control, activity, hearing and sleep, and nothing in this stack has evidence for it.
Pillar Matrix mapping
Database Matrix layers
- Evidence Layer (PubMed, Cochrane, ClinicalTrials.gov)
- Safety Layer (DrugBank, FAERS)
- Pharmacokinetics Layer (HMDB, PubChem)
- Real-Time User Layer (wearable and adherence signals)
Frequently Asked
What should a cognitive stack contain?
Creatine at 3 to 5 g daily, plus corrections for any low ferritin, B12 or vitamin D, and omega-3 where oily fish intake is low. Optional situational additions are caffeine with theanine, citicoline or bacopa.
What comes before the supplements?
Sleep regularity, removing notification-driven interruption, and measuring ferritin, B12, thyroid function, vitamin D and HbA1c. Each has a larger effect than the whole stack.
Is caffeine part of a cognitive stack?
It is the best-evidenced cognitive compound available, and its cost is on sleep, which harms cognition more than the acute benefit helps. Timing is the whole question, given a half-life around five hours.
Should I take ginkgo?
No. It was tested in large properly designed prevention trials and did not reduce cognitive decline or dementia incidence.
How long before bacopa works?
Eight to twelve weeks. It is not an acute compound, which means it needs a defined trial window rather than a daily subjective judgement.
How do I judge whether the stack works?
Output on work you do repeatedly, plus specific observations rather than general ones. Track sleep and mood alongside, since both explain most of the variation.
Will this protect me from cognitive decline?
No. Nothing in this stack has evidence for reducing decline over decades. That evidence sits with cardiovascular risk control, physical activity, addressing hearing loss, sleep and social engagement.
Evidence and review
Any dosage ranges cited here reflect the ranges used in published human trials, not personal recommendations. Evidence in this field moves, so this article is reviewed quarterly and carries its last-updated date above. Nothing here is intended as medical advice, and supplementation should be discussed with a qualified clinician, particularly alongside prescribed medication or an existing condition.