> The autopilot, being engaged, pitched the plane downwards to return to 37,000 feet (11,278 m).

That was the strong negative G maneuver that broke the ceiling by crashing people into it. There was a second strong acceleration when it returned to the desired altitude. My reading of this is that most of the injuries were due to the autopilot’s reaction to the turbulence. Otherwise, it just would have been an unexpected increase to about 1.4 G followed by the cabin pitching around.

For the other one, why is the seat capable of interfering with controls? (Also, that doesn’t match the pilot’s story, which was that the screens cycled off and on, unless the seat can press the “reboot plane” button).