How does a 4K Retina display 'double up' the pixels?

I did the following experiment on a 21.5 retina display:

Open Pages, and write some lines with Courier New at point size 4, 3, 2 and even 1.

Use default screen 2K High Resolution and then set view to Actual Size

Use extreme 4K Resolution and again set view to Actual Size.

The typefaces with 3 and 2 points are readable in both modes, but they seem better in 4K. How is this?

I mean, is not "2K Retina" supposed to present a default resolution of 2K but render objects at 4K? Just to be sure, "2K low resolution" seems to have a worse rendering of the 4, 3 and 2 typefaces. But 4K seems the winner. It is just my bias, or is it a fact that it renders better than 2K?

On older displays, the pixels were sufficiently large that you could see each one on the screen.

Newer displays have smaller pixels, which gives a sharper, less 'blocky' image. Eventually, displays' pixels became so small, and so densely packed that the human eye cannot distinguish individual pixels from a standard distance. (That's why the term 'Retina' is used.)

The problem with such high-density displays is that text is also reduced in size, appearing very small, and difficult to read.

Apple solved this with its 'Retina display' screen rendering. What it does is to take a high-density (HiDPI) display and magnify everything by a factor of 2. So an image that would have been displayed using 512 x 512 pixels will actually be displayed using 1024 x 1024 pixels. The image stays sharp because the 2x2 square that represents "1 pixel" can actually have different colors in each pixel.

So by pretending the display is half the size, you get 'normal'-sized images and text, but with twice the fine detail. I would recommend using the Default scaling for most purposes. You can of course zoom in and out in Safari, Pages, Preview, etc.