Why Thoughts Shrink, and How They Grow Back

A friend of mine stopped coding for four months. When he came back to a problem he used to solve without thinking, he stared at it for twenty minutes and couldn't remember where to start. He hadn't gotten stupider. Something more specific and more mechanical had happened to his brain, and once I understood what it was, I couldn't stop noticing it everywhere — in students who go blank in exams they were ready for, in why habits get harder to break the longer they run, in why some people seem to get sharper with age while others plateau by thirty.

What a thought is actually made of

Your brain has about 86 billion neurons in it. A neuron by itself doesn't do much — it's a single cell that passes an electrical and chemical signal to its neighbours. What it's connected to, and how strongly, is where everything interesting happens. Every time you learn something, a group of neurons fires together, and the connection between them — the synapse — gets a little more efficient. Do it again and again, and that pathway becomes fast enough that it stops feeling like thinking at all. It becomes a reflex.

Neuroscientists sum this up in one line: neurons that fire together, wire together. Walking, talking, driving, reading code, recognising that a problem wants a hash map instead of a nested loop — none of that was innate. Each one is a pathway your brain built because you used it enough times that building it became worth the cost.

The part people miss is the second half of that sentence. The brain doesn't just build pathways. It also removes them. And which one is happening to you, right now, in whatever you're doing today, is not random.

Why thinking shrinks

Nobody's thinking shrinks overnight, and it isn't because they got less intelligent. It shrinks because of three ordinary, mechanical things the brain does — none of which feel like anything is wrong while they're happening.

1. Pathways you stop using get taken apart

The brain runs on a strict energy budget, so it treats an unused connection as a cost with no return and prunes it — the same process, incidentally, that makes children's brains so plastic in the first place. You've felt this if you've ever taken a long break from something you used to be good at. Pick Java back up after four months away and the syntax that used to be automatic now needs conscious effort. Nothing was deleted from your memory in some dramatic sense. The pathway just isn't maintained anymore, so it's slower to reach.

2. Fear narrows what you can reach

When your brain detects a threat — and an exam or a hard deadline counts, biologically, as a threat — the amygdala takes over ahead of anything else. Under that kind of activation, logical reasoning slows down, creativity drops, and retrieval gets harder across the board. This is the actual mechanism behind blanking on an answer you knew perfectly well the night before. The knowledge hasn't gone anywhere. Stress just narrowed the channel it has to travel through to reach you.

3. Comfort doesn't ask the brain to build anything

Routine is efficient, and efficiency is exactly what the brain optimises for — which is precisely the problem. A pathway you've already built gets cheaper and cheaper to run the more you repeat it, but repeating it gives the brain no reason to build anything new. Stay inside familiar territory long enough and your thinking doesn't get worse, exactly. It just stops being asked to expand, and unasked things don't grow.

Why thinking grows

The reassuring part is that this is the same machinery running in the other direction. Nothing about a brain that prunes connections stops it from building new ones — building is, if anything, the default it reaches for whenever it's given a real reason to.

1. New pathways form the moment something is unfamiliar

A new concept activates neurons that haven't worked together before, and every additional exposure makes that group fire faster and more reliably as one unit. This is the entire difference between a beginner and someone experienced — not raw intelligence, but a denser, faster set of pathways built from repetition. An experienced programmer doesn't see a problem and think harder than a beginner does. They recognise the shape of it almost before they've finished reading, because that pattern-matching pathway has already been built.

2. Curiosity changes what the brain does with information

Curiosity isn't just a pleasant feeling attached to interesting things. It's a signal that shifts the brain into a state primed to encode whatever comes next — attention sharpens and memory formation improves measurably. Asking "why does this actually work?" before you're told the answer does more for retention than being handed the answer ever will, because the asking is what switches the encoding on.

3. Doing beats watching, every time

Reading a solution creates a thin, passive trace. Producing one yourself — wrong, corrected, wrong again, eventually right — creates a pathway built through actual use, which is the only kind that holds up under pressure later. This is the specific reason a hundred hours of tutorials can leave you less prepared than twenty hours of problems you actually struggled through.

4. Sleep is where it actually gets kept

Learning doesn't finish when you close the laptop. Sleep is when the brain replays the day's activity, reinforces the pathways worth keeping, and clears out the noise — which is also why cramming through the night before a test is such a bad trade. You're skipping the exact process that would have converted the studying into something that stays.

The five-move sequence for actually growing a thought

Biology explains how a pathway gets built. It doesn't tell you what to point your attention at. That part is a procedure, not a mystery, and it's the same five moves whether the problem in front of you is a bug, a bad decision, or a data structures question.

1. Observe before you touch anything

Most mistakes don't come from a lack of intelligence. They come from solving a problem before finishing the sentence describing it. Read it twice. Restate it in your own words before you write a single line — if you can't, you don't understand it yet, and anything you produce from here is a guess wearing the shape of an answer.

2. Analyze — break it into pieces you can actually hold

What's actually given? What's actually being asked for? What are the constraints, and which assumptions are you quietly making without having checked them? A problem that feels impossible as one large block is usually just three or four small, ordinary problems that haven't been separated yet.

3. Generate more than one route

The first approach that occurs to you is rarely the only one, and treating it as the only one is how thinking stays narrow. For a data structures problem that might mean weighing brute force against sorting, hashing, two pointers, binary search, or recursion before picking one. The point of generating options isn't that you'll use all of them — it's that comparing them is what teaches you why the winner actually wins.

4. Test the reasoning, not just the output

Don't assume a solution works because it looks right. Run it against an example built specifically to break it, not the friendly one from the top of the page. Strong thinkers go looking for the flaw in their own logic before anyone else finds it for them — it's a much better way to find it.

5. Learn from what actually happened

A wrong answer is information, not a verdict. Every failed attempt narrows down what the real shape of the problem is, provided you actually look at why it failed instead of just patching it and moving on. This step is the one people skip most often, and it's the one that turns one attempt into a pathway you'll still have next time.

Why DSA is quietly cognitive training

Most students treat data structures and algorithms as a hoop to jump through before a job offer, which undersells what's actually happening. Every real DSA problem forces you through the exact five-move sequence above: understand a requirement precisely, spot the pattern underneath it, weigh several approaches against each other, verify the reasoning against a case designed to break it, and extract something from wherever it went wrong. The code is the visible output. The thinking procedure is the actual thing being trained, and it transfers to essentially everything else you'll ever have to reason through.

What compounds

None of this requires a dramatic overhaul. A few unglamorous habits, repeated, do almost all of the work: read something outside your comfort zone regularly, sit with one genuinely hard problem instead of ten easy ones, ask why before accepting an answer, produce answers instead of only watching someone else produce them, sleep enough that the day's work actually gets consolidated, and look directly at your mistakes instead of patching around them. Each one is small. What compounds isn't any single habit — it's doing them on a day your brain would rather default to the pathway that's already there.

Thinking shrinks when the brain is never asked to build anything new. It grows under the same conditions every time: curiosity, real practice, honest correction, and enough rest to let the pathway set. Biology supplies the mechanism. The five-move process supplies the direction. Neither one is something you're issued at birth — both are things you build, one deliberately-faced problem at a time.

Working through DSA or trying to think more clearly under pressure? The archive has free PDF notes on data structures, algorithms, and computer science fundamentals — no account needed.