WAKEFULNESS IS NOT INTELLIGENCE // MODAFINIL
The executive drug of the last decade, and the meta-analytic answer is not the one the market believes. Across 14 randomized trials in healthy, non-sleep-deprived adults — 64 effect sizes — the pooled cognitive effect of modafinil is SMD 0.12. Conventionally, 0.2 is a small effect. Only one sub-domain reached significance: memory updating, at 0.28. The authors state it plainly: there is a user perception that these drugs are effective cognitive enhancers, and the evidence does not support it. What modafinil does do, and does well, is restore a deficit. Against 24 hours of sleep deprivation it returns working memory toward baseline. Against a rested brain it has almost nothing to add — because wakefulness was never the thing that was limiting you.
- ›Understand what you are buying before you buy it. Modafinil raises wakefulness. Wakefulness is not intelligence, and the pooled trial data say the second does not follow from the first in someone who is already rested.
- ›If you are sleep-deprived, the drug is not the intervention — the sleep is. Modafinil restores what deprivation took, which means it is a way to keep paying a debt rather than a way to settle it. The debt still compounds underneath.
- ›Treat the European withdrawal as evidence, not as bureaucracy. A regulator reviewed these same trials and removed every indication except narcolepsy, on the grounds that the risks outweighed the demonstrated benefit. That is a formal reading of the evidence by people with no product to sell.
- ›Prescription-only, everywhere that matters. There is no version of this that is a supplement decision, and sourcing it outside a prescription means unverified material with an SJS/TEN risk attached.
This is the dossier that tests whether the compendium is willing to grade against its own audience. Modafinil is the substance the executive market most wants validated, and the honest read is that for a rested adult it does close to nothing — an effect one-third the size of caffeine's reputation, in a molecule with a fatal-skin-reaction warning and a European regulator that struck out every indication but one. The value is not in the discouragement. It is in the redirection: the operator reaching for a wakefulness drug almost always has a sleep problem, a circadian problem, or a load problem — and all three are addressable, measurable, and covered elsewhere in this compendium.
⧗ WHERE CERTAINTY ENDS — The trials measure single doses against laboratory tasks in small samples. Nobody has run the study the executive actually cares about: sustained real-world cognitive output, over months, in a high-functioning adult. Its absence is not evidence of benefit; it is the reason the certainty bar reads 37 rather than higher or lower. Two further limits: the pooled SMD of 0.12 is an average, and an average can conceal responders — no trial has identified who they are, so nobody can tell you in advance whether you are one. And the safety signal that moved the EMA comes from clinical populations on chronic prescriptions; the risk profile of intermittent off-label use in healthy adults has not been characterised at all. Directional context only. Not a diagnosis, not a recommendation, and not a substitute for a physician who knows your history.
Peer-reviewed trials — sample size, effect size, stage.
Applies to healthy apex, not only to clinical deficit.
Survives a demanding calendar. Zero executive friction.
Signal is real but unproven. Watchlist, not protocol.
The score is a geometric mean — a single failed layer collapses it. Excellence in two cannot rescue a gap in the third. That is why hype scores low and proven, feasible, broadly-applicable work scores high. The restraint is the product.