Karl Zener was a perceptual psychologist, and the deck he designed for J. B. Rhine at Duke in the early 1930s solves a measurement problem rather than expressing anything. Twenty five cards carry five symbols, five of each: a circle, a cross, three wavy lines, a square and a five pointed star.
Every constraint on that design is statistical or perceptual. The symbols must be quickly distinguishable from one another at a glance and unambiguous when named aloud, so that a wrong call is a wrong call and not a misunderstanding. They must be simple enough to be visualised. They must be free of emotional loading, so that no card is more attractive to guess than another. And they must be equally represented, so that the expected number of hits by chance in a run of twenty five is exactly five, which makes the arithmetic of any excess trivial to compute.
The procedures were formalised alongside the deck and given abbreviations that still appear in the literature: calling the cards as they are turned, calling the whole pack down before any are turned, calling a pack in a distant room, and versions where a sender looks at each card. The card as an object is a laboratory instrument. Its later career as a divinatory or novelty deck came afterwards and has nothing to do with what it was made for.