I have no idea if I'm in the minority of humans in this, but when I picture something in my head, it has colors, shapes, and I guess I'm using the same brain hardware as when I use my eyes to process the "phantasm" of it. My mental model is that this facility is another "source", in addition to my eyes, that can be used to pipe sense data into the visual processing "sink".

When I was young I was into ray-tracing with POV-Ray. The exercise of positioning a camera in a scene using a text file got me thinking about what it would look like if my eyes were up in _that corner_ of my bedroom. I would sit at my desk and imagine what it would look like and I would be "seeing" a picture in my head of what it would look like (given what information I had about things like whether the top of the half-open door was painted or not, whether the top of the ceiling fan was dusty, that kind of thing).

Right now I can mentally picture the inside of my refrigerator and pantry, and that's pretty much how I keep track of whether I'm running out of various things (which might explain why I'm so terrible at doing so, it's only as good as my mental image). It's not like I have one of those eidetic memories; these mental images are flawed and only somewhat accurately represent reality, with more familiar things being represented more accurately.

It kind of doesn't matter whether my eyes are open or not, but it's a bit easier if they're closed. If they're open, my eyes naturally just go unfocused while I'm doing it (my parents called it "staring off into space").

I'll second the first paragraph here. The demonstration I've come up with involves holding a hand closely in front of only my left eye. My brain lets me see both things, but my hand also has a 'transparent' characteristic when I'm focusing on my right vision. Back to visualizing, the 'other source' isn't anywhere near as crisp as the demonstration, and often is a mix of memory / imagination / caricature / labeling it with the concept.

It's still not clear to me if there's a difference between imagining I'm seeing something and actually seeing something in my head. When I imagine what it's like to see something, of course it has colors, shapes, etc. I can also imagine what a room would look like from a particular vantage point (although I may not be as skillful as it as someone who has practiced doing it and experimented with computer graphics). I guess I'm still curious if there's truly any difference in our experiences. I wonder if optical illusions work when we're picturing things in our head.

That's so interesting! I did the test linked to in the article, and I noticed that I am able to "visualize" object, but have decidedly a hard time to think about how I would perceive them if I looked at them - so I guess just opposite from you.

I see them more as 3d objects, but not from the perspective of a camera, but just the model all at once. And it's always a simplification, only the concept of a tree, no tree with actual detail of a real tree. Many things only have shapes, no visual colors, no visual anything, only the shapes, as though it were a different sense.

> I wonder if optical illusions work when we're picturing things in our head.

I want to say that some won't work, since I know there are some that rely on design flaws in our eyeballs (placement of the optic nerve). I'm thinking specifically about the ones where you are directed to look at some part of the figure and notice that some other part has vanished.

I feel I am the same as this. I often build things from wood and steel. In order to plan parts to be made, where to do cuts and make joins I need to think and "see" how best to do that. I believe I do this quite well, and my thoughts work out well once I put my hand to my tools and make what I foresaw. But I don't believe I ever had a vivid hallucination of what I was making, more a hazy wireframe that gets the job done

Just by way of providing another data point: I was going down this exact same rabbit hole and my experience matches your description exactly.

Aphantasia probably has a single digit percentage incidence rate in the population but it is hard to know.