Interesting idea. Somewhere I have an old wireless keyboard that also worked on 27 MHz. These signals sounded too slow and had too long of a preamble to be keyboard presses, I think. I suspect 45.45 or 50 baud. They also were the same, relatively short, length, which wouldn't be likely to be a mouse.
I tried to make an IQ recording but the signals kept getting weaker and falling into the noise (or the stronger stations, assuming more than one station, were listening at those times).