> Infix and symbols don't produce cache misses for the vast majority of humans

You tried to prove your theory by finding positive evidence. But the evidence is very weak and far fetched.

> which support the standpoint that symbols-and-infix arithmetics is good for humans, based on how their mind works.

Given that we largely don't know how the mind 'works', that's a weak argument.

> Show me a Lisp program with more than casual usage of arithmetics and tell my why you consider it readable.

Given that a lot math code is expressed in low-level Fortran, I'll take Lisp every day.

From a statistics system in Lisp:

    (defgeneric gaussian-probability-density (x params)
      (:documentation "general gaussian density method.")
      (:method ((x number)
                (params gaussian-probability-univariate-parameters))
        (\ (exp (* -1.0 (/ (- x (mean params))
                           (standard-deviation params)))
           (sqrt (* 2.0 pi (variance params))))))

I find that perfectly readable.