I don't have time for a long post but I wanted to start this topic with a couple of general thoughts which are prompted by our recent discussions of images in book four of Lucretius, plus the articles from the Encyclopedia Brittanica (thanks Nate) as to the development of skepticism within the Academy, as well as the Stoic/Academic interplay involving how the Stoics attempted to remain dogmatic. Let me particularly highlight this paragraph:
(3) The next stage in the Academic succession was the moderate scepticism of Carneades, which owed its existence to his opposition to Chrysippus, the Stoic. To the Stoical theory of perception, the φαντασία καταληπτική, by which they expressed a conviction of certainty arising from impressions so strong as to amount to science, he opposed the doctrine of acatalepsia, which denied any necessary correspondence between perceptions and the objects perceived. He saved himself, however, from absolute scepticism by the doctrine of probability or verisimilitude, which may serve as a practical guide in life. Thus his criterion of imagination (φαντασία) is that it must be credible, irrefutable and attested by comparison with other impressions; it may be wrong, but for the person concerned it is valid. In ethics he was an avowed sceptic. During his official visit to Rome, he gave public lectures, in which he successively proved and disproved with equal ease the existence of justice.
Now as to the Epicureans, here is the primary source for the idea that the canon has a "fourth leg":
Thus in The Canon Epicurus says that the tests of truth are the sensations and concepts [preconceptions / anticipations] and the feelings; the Epicureans add to these the intuitive apprehensions of the mind. And this he says himself too in the summary addressed to Herodotus and in the Principal Doctrines. For, he says, all sensation is irrational and does not admit of memory; for it is not set in motion by itself, nor when it is set in motion by something else, can it add to it or take from it. Nor is there anything which can refute the sensations.
Note that this listing distinctly separates the preconceptions as one of the first three legs by saying that "the Epicureans add to these the intuitive apprehensions of the mind." Have I not seen this regularly translated as "present impressions of the mind"? (Need to check; perhaps that is the letter to Herodotus)
Regardless, both "intuitive apprehensions of the mind" and/or "present impressions of the mind" seem to me to be very good descriptions of the process described in book four of Lucretius where the mind receives images directly from outside itself, much as Cicero discusses also with Cassius in his "spectres" back-and-forth.
In addition to the separation of preconceptions from "intuitive apprehensions of the mind" in Laertius, we have the Velleius description of anticipations:
“Anyone pondering on the baseless and irrational character of these doctrines ought to regard Epicurus with reverence, and to rank him as one of the very gods about whom we are inquiring. For he alone perceived, first, that the gods exist, because nature herself has imprinted a conception of them on the minds of all mankind. For what nation or what tribe of men is there but possesses untaught some ‘preconception’ of the gods? Such notions Epicurus designates by the word prolepsis, that is, a sort of preconceived mental picture of a thing, without which nothing can be understood or investigated or discussed. The force and value of this argument we learn in that work of genius, Epicurus's Rule or Standard of Judgment."
The purpose of adding this post today is that it seems to me that a very significant case can be made for the argument that the intuitive apprehensions of the mind, which Epicurus denied the status of being a leg of the canon, is meant to refer to the mind as a suprasensory organism (as Dewitt refers to it) that is active all the time receiving images from outside. An obvious reason why Epicurus would deny to this function the status of canonical is that the mind is active in selecting the images that it pays attention to; it therefore does not function without opinion and without bias as do the canonical faculties, which do not inject opinion. Yet it is easy to see too why later Epicureans might start to make the mistake of considering it canonical, because it is a faculty in which the mind is making contact with some kind of reality outside itself.
DeWitt goes into this much further and I'll probably paste some of his commentary here, but I think the discussion of the "fourth leg" and why Epicurus only had three is a very important path to pursue to bring out the nuance of what makes a canonical faculty and what doesn't. I agree with Dewitt that it is a dramatic and even to the point of fatal error for the "later" or "other" Epicureans (whoever they were) to depart from Epicurus on this point.
INSERT: Here is the key DeWitt section in which he observes why Epicurus would not have ranked present impressions of the mind as a criterion of truth (and therefore why that is separate from anticipations) So what I am suggesting here is that this analysis is also relevant to the mind's direct receipt of images and how they are separate from anticipations:
Now linking back to my first statement as to the inspiration for this post, it sounds to me like the later/other Epicureans were adversely influenced by the Stoic and Academic debates as to how truth is determined through the mind, and that more modern proponents of the "fourth leg" probably have fallen for the same mistake by accepting those viewpoints as amounting to something canonical, which Epicurus rejected.