When we look at a painting, say, a Rembrandt portrait, we do not wonder what is to the left or right of the sitter; we do not wonder what she is looking at; we presume that everything we need to know has been put into the picture-into the frame-by the artist.
Audiences of screen acting apply the same criteria.
We presume that everything we need to know is there, and that those things not on the screen are unimportant or irrelevant. This means that if an actor does wonderful things that are not seen, then it is as if she had not done them!
A frame condenses both time and space.
In real life, or on the stage, someone holding a cup of tea holds it about level with her navel. This is both for comfort and for convenience; it is easy to lift it up to the face or put it down on a table.
Putting this picture inside a frame means that the face is very small and difficult to see. To get a bigger face requires a tighter shot, but then it would look as if she had no cup at all.
So the screen actor holds a cup of tea (or a mug of coffee, or a bottle of beer, or a file, or a notebook, etc.) up close to her face, so that when the camera sees it, it looks real, we know what she is doing, and we still see all the expressions on her face.