W.D. Hamilton thought humanity was roughly divided into people concerned with other people- the florid psychotic representing one extreme- and people concerned with things- with the autistic savant at the the other end of the spectrum.
Of course, any binary opposition can be used to construct a continuum- for e.g cat people vs. dog people- but the same thing can be done with a single
signifier, e.g. buggering a Brummie. Everybody falls along a continuum defined by their tropism, under circumstances of equal endowment, to bugger the said Brummie. However, this continuum describes a boustropheodonic time-path because the subject with least tropism to bugger the Brummie at time t is precisely the person most deeply up said Brummie's arse in the previous period. (Making the entirely reasonable assumption that Brummies feed exclusively on highly spiced Balti curries.)
Now rates of 'tunneling' through the continuum for any individual will differ considerably and thus the instrumental cognitive or heuristic benefit of retaining the continuum is likely to be low.
True, one can create an ideal of the continuum to exclude this cyclicity but such ideals are not robust to small changes in the specification of exclusion.
For this reason, the autism/schizophrenia dichotomy fails quite independently of the success or failure of scientific hypotheses re. imprinting.