This paper is an attempt to make sense of Aristotle's polemic -and, if taken at face value, quite evidently wrong-contention that the ancient thinkers globally failed to distinguish thought and knowledge from sensation. As I will try to show, what these thinkers missed, in Aristotle's view, was his own concept of intellect (nous) as a distinct mental faculty correlative to forms (eide). Lacking a concept of "forms" (formal and final causes), pre-Platonic thinkers, as Aristotle understands them, could conceive cognitive processes only as "alterations," i.e. at a merely physical level of description, which made them unable to account for the difference between truth and error: by this failure, Aristotle argues in Met. G 5, they unwittingly prepared the way for Protagorean relativism. This reconstruction will permit us also to have a more coherent understanding of some of Aristotle's criticisms of Parmenides and Melissus; finally, I will shortly point out some analogies and differences between Aristotle's theory of cognition and functionalist theories in contemporary philosophy of mind.