But I'm seeing most those things in the viewfinder at the time. The framing I see from moving around right in the viewfinder. The depth of field I can see by stopping down the aperture before I shoot the shot. I'm not visualizing anything.
What you see in the viewfinder is significantly different than what you see on the TV. The most important difference is, of course, that you see a three dimensional object, while the screen can only present a two dimensional view.
But there are other really important differences - some of which you have probably internalized.
You have knowledge about how your screen displays highlights and shadows, and you understand how that differs from the real world, so you when you view the scene through the finder, you make adjustments that take that into account.
You exclude elements in the scene, based on the frame that you are working with - the 16:9 aspect ratio.
There may be lots of great photo opportunities in front of you, but because they are inherently of different shapes than your TV screen, you reject them - because when you visualize them on your screen, they don't work right.
When you make depth of field choices, you are doing so based on how the result will look on your screen. In the real world, our eyes and brains make depth of field adjustments automatically.
Visualization is about taking the scene and translating it into the presentation medium in your mind's eye. Saying to yourself: "I'd like this scene to look like X on the TV screen, so I will make these adjustments at time of exposure". The viewfinder image is always significantly different than the TV screen image - visualization involves learning and applying the tools that control that translation.
My sense Alan is that the photos you like best for showing on your screen are the ones that are close to "realistic" analogues to what you perceive as the real world. That means that the translation adjustments you require may be more subtle than some. That doesn't mean though that you aren't making them.
When you look at a scene and then reject is as unusable for a photograph, you are visualizing it first as the photograph that might result.