Fair point. I'm not really thinking of it as little squares, I'm more just thinking in terms of integer coordinates in a 2-d Cartesian space. But I do think I'm inherently abstracting the "screen" as a grid of square pixels because it's a serviceable abstraction that is substantially simpler than "here is the whole signal processing pipeline that sits between my <image> tag and some phone user's eyes."

I suppose you're right that what I'm describing is just the lack of conscious mental model for what the pixel represents, versus when I'm writing a shader, where I'm actively thinking through the mental model of a pixel.