As an able bodied person, I see potential in mapping some of these to specific regular actions, rather than individual key presses. Using this, along with dictation and a keyboard/mouse combo could save a lot of time.
These defaults seem like the exact opposite of how you use your eyes to look at real things. Cartoons are not good reference material for HUI (unless you want cartoonish outcomes).
As someone who has the bad habit of lying in bed at night with my laptop on my chest and my hands tucked into the sheets reading and watching things, I think this is great!
Doing it because you can aside, I feel like if you're a handicapped person with full control of your face voice dictation would probably be a better solution.
I knew someone who had to switch to voice dictation due to RSI problems during an especially intensive work period, who ended up temporarily losing their voice through overuse. Of course that could happen with anything, overuse of a tool like this could lead to facial tics or neuralgia-like symptoms. But it's definitely got potential as an accessibility tool.
Only a matter of time before someone uses this to auto-post memes/gifs of common emotional responses.