There is a familiar figure in technology: the person who never finished the relevant degree, or never started it, and taught themselves the thing they are now good at by following their own curiosity. The claim attached to them is that they share a set of measurable traits, and that their edge in solving problems comes not from knowing more than everyone else but from learning differently.
Two honest caveats before the list. “Self-taught” is a blurry category, because almost everyone who is good at something has mixed formal instruction with long stretches of teaching themselves, so this is really about a way of learning rather than a type of person. And the traits below are observed tendencies, not the output of a validated personality test.
“Measurable” is generous.
What does hold up, and is worth the piece, is the second half of the claim. A good deal of learning research suggests that outcomes track how people manage and test their own learning more than how much they have passively taken in. That is the real kernel here, and several of the traits below are recognisable versions of it.
1. They run their own learning loop
The clearest marker is that they set their own goal, attempt the thing, see where it fell short, and adjust, without waiting for anyone to assign the next step. They are, in effect, their own teacher and their own marker.
This lines up with what researchers call self-regulated learning, a model developed by the psychologist Barry Zimmerman, in which capable learners cycle through planning, monitoring, and reflecting on their own progress. The self-taught tend to do this by necessity, because there is no syllabus doing it for them.
2. They know the edge of what they know
Ask a strong self-directed learner what they are bad at and they usually have a fast, specific answer. They have spent a lot of time at the boundary of their own competence, so they tend to know where it sits.
That awareness of one’s own understanding is what John Flavell named metacognition, thinking about your own thinking. It is useful precisely because it tells you where to point your effort next, and it is the opposite of the confident vagueness that comes from never having tested yourself.
3. They are pulled by interest, not by grades
The engine is usually curiosity. They learned the thing because they wanted to know it or needed it for something they cared about, not to clear an assessment.
That distinction is the heart of self-determination theory, the work of Edward Deci and Richard Ryan on intrinsic motivation. Interest-driven learning is more durable and more self-sustaining than the kind chased for an external reward, and the research on this is not really contested. The failure mode worth naming is that a person who only learns what interests them can end up with deep patches and surprising holes — but those holes are usually cheaper to fix than the absence of any genuine drive.
4. They learn towards a problem, not through a syllabus
Formal courses tend to build foundations first and applications later. Self-teachers often run it backwards, starting from a problem they want to solve and pulling in whatever they need to solve it, in the order the problem demands.
This makes them quick to get something working and good at learning just in time. The trade is that the foundations they skipped can stay missing until something breaks in a way only the foundations would have predicted.
5. They test against reality, not a marker
A self-taught coder finds out whether they understood something when the program runs or fails. The feedback is immediate, external, and hard to argue with, which is a different and often harsher examiner than a graded paper.
Leaning on that kind of active checking has a strong evidence base. The testing effect, documented by Henry Roediger and Jeffrey Karpicke, shows that retrieving and applying knowledge cements it far better than passively reviewing it. Building and breaking things is retrieval practice with the stakes turned up.
6. They are comfortable being a beginner
Teaching yourself means spending a great deal of time confused and visibly not good yet, often in public. People who keep doing it tend to have made some peace with that state rather than fleeing it.
This is more observation than established finding, but it is a consistent one. The tolerance for feeling stupid for a while is part of what lets the learning continue past the point where it stops being comfortable.
7. They practise at the edge, not in the easy middle
Given a free choice, it is tempting to practise what you are already good at, because it feels good. Strong self-directed learners more often push into the parts they are bad at, which is slower and less pleasant.
That is close to what Anders Ericsson called deliberate practice, the focused, slightly uncomfortable work of targeting weaknesses rather than rehearsing strengths. It is also the part of self-teaching that most people quietly avoid, which is why not everyone who studies alone gets good.
8. They treat being wrong as information
Because their feedback comes from results rather than from a person’s approval, the self-taught tend to read a failure as data about the problem rather than a verdict on themselves. The bug is not an insult. It is the next clue.
That framing keeps them iterating where someone more invested in looking competent might stop. It is probably the single habit that does the most work, and it is also the one least tied to whether a person went to school.
What the list leaves out
None of these habits are owned by the self-taught. Plenty of people acquired them inside formal education, which at its best is a structured way of building exactly these habits, with the added benefit of someone catching your errors early and a credential at the end.
The self-taught come with real gaps — shaky foundations, blind spots, the particular overconfidence of someone who has never had a hard examiner tell them they are wrong. Worth naming, but worth keeping in proportion.
Because when you line the habits up against the outcomes, the people who run their own loop, test against reality, and treat being wrong as information tend to out-solve the people who waited to be taught. The classroom can produce that kind of thinker. It mostly doesn’t. And until formal education stops optimising for compliance and grades over curiosity and retrieval, the edge is going to sit with the ones who learned to teach themselves — and its defenders will have to explain why the structure they keep pointing to is producing so few of them.
