Albeit it is hardly possible to account entirely for all stimuli any organism senses, it is probably safe to argue that all salient experience the fly receives in the flight simulator is controlled by the experimenter: 1) the fly is tethered and hence the visual field of the fly is either stationary or coupled to the recording device. 2) Most other stimuli (odorants, moisture, air-pressure, magnetic or electrostatic fields etc.) are to a large degree constant during the course of the experiment.
Consequently, most motor-output recorded is either initiated by the fly on its own (initiating activity [Heisenberg, 1983] or rhythmic activity) or induced by the experiment.