Is it largely genetics? Does everyone have an "upper limit" of the "smart" they can be, which is determined by their genes?
Or is it just hyper-focus and concentration that makes someone smart?
Let us assume, smart here means the rate at which someone lears, grasps or solves things.
I often wonder what it would have been like working with somebody like John von Neumann or how they really see the world.
Cognitive strategies vary so much they are difficult to compare. Good memory vs deduction skills, tenacious will vs opportunistic, dogmatic structure vs chaotic exploration. Quirky yet elegant tastes vs OCD who picks at problems.
Does not regard convention. Beyond aware that there is such a thing, preoccupation of expectations is a barrier. One may not realize they are a smarty pants until long after everyone stalling comments on how far along you've gotten.
Obsession and compulsion has a lot to say for circumstances. Normal people are off enjoying their mental serenity, smarty pants are often agonizing over incongruous abstractions. A smooth character simply does not expose their restlessness.
Not knowing something is enough to stop a normal in their tracks, a smarty will observe, explore, relate, and experiment. Playing around, having interests, and not knowing there were supposed to be limits may be common traits.
Normals are too busy doing something else, comfortable in their not knowing, everything explained away.
All that said, if you have the time and resources to pour into unabashed and uninterrupted learning, growing and exploring, it seems to help develop above average skill. Secondly, precise and guided attention and mentoring is very effective at reducing knowledge barriers.
There are literal geniuses, like Gauss for example, who can solve something on intuition without being taught, and there are maybe like 2-4 people who are actually like this alive today. The rest are seething and coping and pretending they’re genius, and taking the rage of their insecurities out on everyone else—a genius is the last person to talk about IQ tests, wordcels or shape rotators etc. Memorizing facts is not genius, it’s a memory trick.
Although training in certain disciplines can help you score higher on intelligence tests by making the situations they test for less novel (and increase your "intelligence" as measured by the tests due to their foundational assumptions), it will do nothing to increase your actual intelligence - merely skew the results of the imperfect test a little bit.
That's not to say that training for an IQ test is a waste of time! A piece of paper showing a higher score is always valuable for getting ahead in society. And training to make large swaths of situations less novel makes you better able to navigate because now you have relevant experience (and thus require less intelligence to navigate successfully on your "first" try).
Tests are often singularly-focused, have clear right or wrong answers, and measure smartness.
Real life is uncertain, many competing layers and priorities, and measures wisdom.
I did that too.
When it comes to genetics, you have to figure that underlying genetics set the baseline for the ability to respond to both culture & environment.
OTOH, genetics itself is quite fluid and adaptable on its own, capable of responding to both culture and environment over (sometimes unimaginable numbers of) generations not near the scale of a single individual's lifetime.
Making the breadth of what can be accomplished for any one person quite limited compared to what extremes may be accomplished over millennia under conditions that favor the desirable characteristics.
Whether the strongest influence is culture, environment, or both, has got to be an overwhelmingly complex set of variables.
Sometimes the culture is more stable over generations than the environment, or vice versa, sometimes neither one. Sometimes the genetics are more stable too, sometimes not.
Plus you can only imagine how bizarre some prehistoric cultures and/or environments were, and how some of the flourishers or surviviors deviated from others. Especially considering some environments and cultures were more widespread than others, or at the other extreme, relatively to completely isolated for more time than anything in recordable history. With a difference amounting to many orders of magnitude over all variables.
What are the odds that only one culture or environment selected for the hallmark learning, grasping, solving behavior that others are just not going to reach in a single lifetime? And from the variation that can be seen today, what are the odds that some other culture(s) did not deselect for the exact same characteristics?
If there was some lost culture of isolated super-intellectuals in the far prehistoric past, they were all just average people among themselves anyway.
But anything that far back can sure have an outsized outcome if the unconventional course is followed from such an early start, and the deviation from mainstream widens accordingly.
Regardless, I believe it is accepted that nobody is using but a small fraction of their brain constructively, and there are so many different kinds of talent. There might even be a common trend to using only a small portion of a single side of the brain more than anything else. So what.
So just about anybody should be able to mostly double their mental ability in a chosen area, and have plenty of room to spare, still utilizing only a small amount of what they were given to work with.
It's not going to happen unless you put your mind to it.
I don't have proofs though.