Does Epistemology Drive Metaphysics?

I've wanted to post something on Deborah Modrak's interesting lecture in BACAP last Thursday, "Aristotelian Form, Function, and Definition". Modrak's distinctive approach to the Metaphysics was to see the central books of the treatise as motivated, at least in part, by epistemology.

The objects of perception are concrete particulars. They provide the basis for knowledge even though the objects of knowledge are universals. This sets up a tension within Aristotle’s epistemology. Its resolution depends upon his success in providing an ontology that grounds the universal in the particular. Not only is there a tension between universal and particular, there is also a tension between that which is intelligible and that which is observable and a tension between simplicity and complexity. Aristotle calls our attention to the relation between perception and knowledge in the opening lines of Metaphysics I. In the central books of the Metaphysics, he expends considerable effort to present and explicate an ontology of form, function and composite substance. In light of the many references Aristotle makes to knowledge and definition in these books, it seems likely that he intends to offer an ontology that will resolve the tensions inherent in his picture of knowledge.

My first reaction, however, was to doubt that this is correct. Why? Three reasons:

1. If Aristotle's metaphysics is supposed to respond to epistemological difficulties, then (unless we invoke developmentalism) he ought to draw explicitly upon that metaphysics, in passages outside the central books of the Metaphysics, which pertain to how we come to know, e.g. Post. An. II.19 or Physics I.1. And yet there are no signs of this. (Or are there?)
2. It doesn't look as though epistemological problems figure importantly among the perplexities (in Beta) which apparently motivate the Metaphysics.
3. Generally Aristotle seems to see no 'tensions' where we might think there are such. He seems dismissive, for instance, of the Heraclitean account of the sensible world which, Aristotle thinks, tempted Plato. If epistemology begins with the aporia, 'How is it possible that we have knowledge?', are there signs that Aristotle was ever troubled by this difficulty?

In brief: if, as we might agree, ancient scepticism differs from modern, then mustn't ancient epistemology differ as well? And yet I worry that Modrak is perhaps bringing in a modern notion of epistemology.