The Chemistry of Learning: What Actually Works

The internet will sell you a stack of seven capsules that "switches on neuroplasticity." It will contain a nootropic with an unpronounceable name, a dosage precise to the milligram, and a promise that studying will get easier.

The problem isn't that this is harmful. The problem is that the most powerful set of substances affecting learning is already inside you, it's free, and it's controlled not by capsules but by everything the previous articles were about.

This article is about how, exactly — and about which supplements genuinely make sense and which don't.

The orchestra and the conductor

Popular writing usually puts it like this: dopamine is the motivation chemical, serotonin the happiness one, acetylcholine the learning one. That's convenient, and wrong to roughly the same degree as "the violin is the instrument of sadness."

The brain is closer to an orchestra, where a neurotransmitter isn't an emotion but a role in the overall sound. The same instrument does different things in different passages, and what matters isn't quantity but coordination and timing.

From which it follows immediately why the "pour in more of the needed substance" approach works badly. An orchestra isn't helped by the volume of one violin — it needs a conductor. And the conductor here is a set of thoroughly everyday things: sleep, movement, attention, tangible progress.

You don't pour chemistry in from outside; you conduct it from within.

Four roles worth understanding

Acetylcholine — the tag "this matters, remember it." It takes part in holding attention and in how the cortex tunes itself to new experience. Animal experiments showed something curious: if you stimulate the pathways associated with it at the moment a tone is played, the area of auditory cortex representing that tone expands — the system literally marks what to learn. The practical meaning is direct: the marking is done by attention. Material that passed through fragmented, broken attention doesn't get tagged as important — hence "read for an hour, remember nothing."

Dopamine — the signal "that went better than expected." The key discovery here belongs to Wolfram Schultz: dopamine neurons respond not to reward as such but to the discrepancy between expectation and outcome. An expected reward produces almost no signal, an unexpected one produces a strong signal, and one that was promised and not delivered drops activity below baseline.

A great deal of practice follows from this. Visible progress motivates precisely because it delivers a string of small positive discrepancies. Routine without feedback doesn't motivate, because there are no discrepancies in it. And cramming without self-testing is deprived of the signal entirely: you never find out whether you hit or missed. Here too is the root of what the article on flow was about: immediate feedback isn't a pleasant extra but fuel for a system that learns from discrepancies.

Noradrenaline — the level of mobilisation. Too little means sluggishness and distractibility, too much means panic and narrowed thinking, and the optimum is in between. Yerkes and Dodson described this regularity in general form more than a century ago: performance as a function of arousal has the shape of an inverted arc, and for difficult tasks the optimum is lower than for easy ones. A direct consequence for exams, which the article on the cockpit will cover: the aim isn't to calm down completely but to stay below the top edge of the arc.

BDNF and related factors — the material for growth. BDNF came up in the article on the aquarium: it supports the survival of neurons and the formation of connections, and is most reliably raised by aerobic exercise. NGF, discovered by Rita Levi-Montalcini, plays a similar role for particular types of neurons. An important distinction: these factors don't "improve memory" directly, they create the conditions in which learning is physically possible.

I deliberately leave serotonin off this list as a "learning chemical": it's tied to mood, sleep and a multitude of other processes, and blunt simplifications of the form "more serotonin, better studying" don't survive scrutiny.

What's actually doing the conducting

Now the main point. All four roles are most reliably managed by behaviour rather than by substances, and you already have every instrument.

Sleep is the strongest intervention that exists. It determines whether what you learned reaches long-term memory, and it restores your capacity to encode the following day. No supplement comes close in effect.

Aerobic exercise is the only reliable way to raise BDNF, and it affects hippocampal volume over the long run.

Undivided attention governs the marking of what's important. A phone in another room acts on learning more powerfully than any nootropic.

Immediate feedback and visible progress govern the dopamine signal. Self-testing, problems with answers, an hours counter, a collection of "couldn't before — can now."

The right difficulty holds mobilisation at the optimum of that arc: too easy means sluggishness, too hard means panic.

Regularity and intervals distribute all of it across time so that consolidation has room to happen.

Boring? Yes. But this is what real biohacking is: the same systems, only reliably and for free.

Now about supplements

Let's go through an honest hierarchy: from what's confirmed to what's being sold.

Works reliably, with conditions: caffeine. It improves alertness, attention and subjective capacity to work — one of the best-studied effects in psychopharmacology. The caveats are substantial. It doesn't replace sleep, it masks sleep deprivation: what restores the ability to remember is sleep, not caffeine. Its half-life is on the order of five hours, so coffee after lunch degrades night-time sleep even in people who "fall asleep fine," and spoiled sleep hits learning harder than the morning effect helped. On top of that, tolerance develops: over time a significant share of the effect goes on relieving withdrawal symptoms.

Works for a deficiency, not above the norm: vitamins and minerals. A shortage of vitamin D, iron, B12 or iodine really does impair cognitive function, and making up the shortage helps. But taking more than you need, in a person without a deficiency, produces no additional gain. The practical conclusion: it's worth checking for deficiencies with tests rather than taking everything prophylactically.

Seriously researched, effects modest: omega-3 and creatine come up more often than others in the context of cognition, and the data on them are mixed — a small benefit in some places, nothing in others, with much depending on your baseline and diet. These aren't snake oil, but nor are they what will noticeably change your studying.

Actively sold, weakly evidenced: most nootropics. The general picture is worth understanding: there is broadly no convincing evidence that popular "brain supplements" produce a durable improvement in learning in healthy people. Individual substances show effects under narrow conditions or in people with specific impairments, and this regularly gets turned into advertising aimed at everyone. A separate category is prescription stimulants: taking them without a prescription is illegal, they carry real risks, and — importantly — studies in healthy people mostly show a rise in confidence about one's own productivity alongside a small or absent rise in actual results.

It matters what the argument here is about. Not "is this harmful," but opportunity cost: the time, money and attention that went into assembling a stack would almost always have yielded more invested in sleep, movement and self-testing.

Before adding chemistry from outside, check that the conductor isn't asleep.

A disclaimer that doesn't need to be in small print here. This section is a survey of the state of the research, not a recommendation. We deliberately give no protocols and no dosages: any substance interacts with your health, your medications and your particular body, and decisions of this kind are made with a doctor, not from an article on the internet.

What to do instead of a stack

If you want to "upgrade your biochemistry," the order of operations goes from strongest to weakest:

Fix your sleep. Duration, regularity, darkness, no caffeine in the second half of the day. This will give you more than everything else on the list combined.
Add aerobic exercise. Twenty to thirty minutes on study days, ideally a few hours after the main block.
Remove the fragmenting of attention. Phone in another room, work in blocks.
Build in feedback. Self-testing instead of rereading, a visible progress counter.
Tune the difficulty. Work in the channel, not on a shoal.
Check for deficiencies with tests, if there's reason to, and close whatever you find.
And only then think about anything else — and then with a doctor, and without expecting a miracle.

The first five points are, in essence, the content of this whole sphere. That's the whole trick: learning techniques are the management of neurochemistry, simply described in a different language.

Check yourself

Close the article and answer in your own words:

Why does the phrase "dopamine is the motivation chemical" oversimplify, and what does the dopamine signal actually track?
How are undivided attention and the material the brain tags as important connected?
What does the inverted Yerkes–Dodson arc describe, and what practical conclusion follows from it for an exam?
Why doesn't caffeine solve the problem of too little sleep?
Why is the main argument against supplement stacks not harm but opportunity cost?

In short

Neurotransmitters are roles in an orchestra, not emotions in a bottle: what matters is coordination, not quantity.
Acetylcholine is tied to marking what's important — and the marking is done by attention. Dopamine responds to the discrepancy between expectation and outcome — hence the value of feedback and visible progress.
Noradrenaline: performance depends on mobilisation along an inverted arc, and for difficult tasks the optimum is lower. BDNF and NGF create the conditions for connections to grow, most reliably through aerobic exercise.
All four roles are managed by behaviour: sleep, movement, whole attention, feedback, the right difficulty, regularity.
Among supplements, caffeine works reliably — but it masks sleep deprivation rather than replacing it, and ruins sleep when taken in the second half of the day. Vitamins help with a deficiency, not above the norm. For most nootropics there's no convincing evidence in healthy people.
You don't pour chemistry in from outside; you conduct it from within.