Rosemarie,

Believe me, I know that using the mouse without being able to see it is hard. It is, however, within screen readers that support it a very easy way to give a quick listen to what's on the screen that you might never see if you're doing things like listing all links and traversing the list or all headers and traversing that list, etc. I've also found it interesting that on most sites there are hidden links specifically for those using accessibility software that the rest of the world never sees.

I'm not so much interested (though I am interested, don't get me wrong) in whether someone uses the mouse as a pointing and activating device so much as whether they use it to get a sense of what might be on a screen just by running it around.

It's also nice that at least some laptops allow you to configure the mouse such that it will not go outside the physical screen boundary. That's under the "Momentum" settings for my Synaptics touch pad under "Enable bounce off screen edge". While that would drive me crazy, as I often want the mouse to disappear, it would be incredibly useful for someone doing screen review via mouse movement. This may be possible with certain mice on desktop machines, too, but I've never had one with that configuration option in that hardware environment.

It was also interesting for me to see what I consider the weird difference in accessibility in PDF-XChange Viewer when using the conventional menu keyboard shortcuts and arrow key traversal, which gives you no feedback at all, but when you hover over the same menus with the mouse and draw the mouse pointer down the menu, each and every thing, including the menu title, is announced (this is all with NVDA 2016.1). Why that would be is a mystery, since if there were nothing that NVDA could access somehow for announcing things then there should be no accessibility feedback in either method.

Brian