against “unity of experience over space”

Recently I have heard whispers arguing in favor of something called “unity of experience over space” and (perhaps) against something called “unity of experience over time”. I think this whole concept is probably underspecified and perhaps even confused. If we’re all roughly some sort of naive physicalists here, and I certainly haven’t figured out any metaphysical stance which substantially improves upon that so far, then we’re going to need to be pretty careful to separate phenomenological facts from (low-level) physical or neuroscientific ones, and to not make any unfounded assumptions about how these sets of facts intertwine.

Phenomenologically, I sure do seem to experience my whole visual field at the same time. There are some quibbles one could make, here! I’m not totally sure what’s up with my peripheral vision. And I would presume that many details of my visual field are essentially cached, or perhaps partially inferred / reconstructed, rather than every portion of my visual field directly corresponding to particular rays of light the same way that a portion of photo would. But it’s definitely possible for me to see two things at once.

My impression is that some skilled meditators report that, in fact, the stream of consciousness contains only one thing at a time. I’m not sure I’m convinced. Certainly my “meditated for thirty to sixty minutes a day for like a month once” credentials are outdone by many and do not license me to make any particularly confident claims. But my guess would be that it’s more like, if you pay attention you can notice your locus of attention shifting between various phenomenological objects, roughly one at a time. Or at least something like that is much closer to things I’ve experienced than “You Only Get To See One Qualium At A Time, They Pass Through Your Consciousness One After Another In A Linear Order” or whatever. So my guess is that surely the latter is not what anyone is actually trying to claim happens?

However, I think this has very little to do with the question of whether physics is like, applying some sort of magic rule that takes a spatially extended region containing some matter and turns it into qualia. Indeed, almost all of the physical space my brain takes up is unnecessary in principle for my subjective experience. Consider a particular moment of my consciousness, where I (for example) can see both my left hand and my right hand, and then (after introspecting and observing this) I choose to state out loud that I can see both of my hands. The computation my brain is performing there, if fully modeled, could be implemented in any number of ways, and still result in the decision to report the qualia of seeing both hands. So we could take every step that my brain performs in parallel¹¹ Probably a lot of them? and instead run them serially, using a computation device that only takes up a small amount of space. The main constraint, as far as I can tell, is memory storage.

I suppose I am forced to concede that any computation is bounded in how little physical space it can take up by its space complexity, including the space required for input and output. I don’t think this is really relevant. It still seems likely to me that really rather a lot of the computation involved in my experience of qualia could simply be done in serial rather than through spatial parallelization without impacting any properties I have introspective access to.

On this basis, as well as a more general skepticism, I am unconvinced that “unity of experience over space” is a very productive conceptual framework, that the concept strikes at the heart of how qualia work or any such thing, though I concede that any useful notion of “one” “moment” of “experience” probably involves a region of space larger than a single point.

But like, since when does literally anything only involve a single point of space or of time. This whole notion is ridiculous. This is the sort of thinking that causes you to start getting confused about paradoces involving sampling from intervals of real numbers instead of realizing that you can simply only sample a finite number of random bits, which are sufficient to specify a random real number only to some finite amount of precision. Simply don’t deal with infinities, you’ll get confused.