Isn't there a rather fundamental problem with the study? It's not blinded. It's pretty obvious to both the subject and the experimenter if the bed is rocking ...