I have a new unconfirmed hypothesis about this, which revolves around the darn parentheses.

The specific shape of the parentheses matters. Say we have:

  (let ((s nil))
    (dotimes (i 16)
      (setq s (rbt-insert (1+ i) s)))
    (pp s))

But say we replace the ( and ) glyphs with the ^ and $ shapes:

  ^let ^^s nil$$
    ^dotimes ^i 16$
      ^setq s ^rbt-insert ^1+ i$ s$$$
    ^pp s$$

There is a night-and-day difference between those two, for you and me. Now imagine you have the hypothetical cognitive disorder which I have christened "dyslispia". Maybe to someone with dyslispia, there isn't much difference between these two; their brain doesn't process and "auto-complete" the enclosure hint of the shape of the parentheses. All they see is noise. Those people are helped by dimming the color or using some indentation-only notation.

I also observe the curious effect that if we swap the shapes of the parentheses, I can still train myself to see the closure by rewiring my cognition to see the parentheses as pointing toward an interior in the convex direction:

  )let ))s nil((
    )dotimes )i 16(
      )setq s )rbt-insert )1+ i( s(((
    )pp s((

Whatever the reason, this is not as bad as ^ and $. It helps if I imagine this in 3D, and pretend that the ))s nil(( is pushed into a pillow, sorta thing.

If dyslispia is real, it's basically a form of dyslexias; it's fundamental brain wiring problem for which there is no cure. No amount of explanations about Lisp will fix it.

In general, to enjoy working with Lisp, the raw experience as such, it probably helps to be well-sighted (no visual impairment), and no dyslispia-like cognitive impairment. My remarks here are mainly about those who report persistent difficulties with Lisp, but who do not mention any visual impairment.