Given that similar claims have been around for a long time and it still lacks published results, I feel people have reason to be skeptical.

I wonder if this "training programme" could be implemented or tested using a VR workspace and varying the (virtual) distance of the working field while adjusting size so that it remains clear. The virtual environment could be finely and dynamically adjustable (as opposed to having to change lenses) and progress could be measured in a consistent environment.

On second thought, if that worked, the Apple Vision Pro might sell some additional units.