'Prototypical autism' research is likely a dead end

Efforts to define “frank” or “classic” forms of the condition build on several assumptions that the science has not yet borne out.