yeah thats why i need help from the machine interpretability crowd to make sure my hypothesized reframing of reasoning has sufficient empirical basis and isn’t adrift in lalaland.