This is a really interesting direction, especially the move away from representational models of perception. But I think there’s a subtle shift happening here thats easy to miss.

When Korzybski says “the map is not the territory” the point is not really about representations being optional or eliminable. It’s about not confusing a model - any model - with what it describes.

so even if we replace maps with sensorimotor contingencies, or with “pathfinding,” the same constraint still applies: whatever structure we use to navigate the world is still a way of accessing it, not the thing itself.

In that sense, the issue might not be whether perception is representational or not, but how we move from a difference in access map, model, action, etc to claims about what exists.

That step seems to be doing most of the work in these debates.