I suspect this had a two-step teaching process for neophytes... First, they'd play with the cardboard machine and get a feel for assembly programming, instruction processing, memory, etc. Once then, after a bit more hacking on things like Star Trek or 4x4x4 tic-tac-toe they'd set out to write an electronic version (virtual machine!) of the cardiac "computer". Debugging that process taught all sorts of relevant things.

And it vaguely felt like a PDP-8, and I suspect it also felt like whatever very early minicomputer that was available.