Background: Sleep plays a critical role in maintaining health and well-being. Yet hospitalized patients are frequently exposed to noise that can disrupt sleep. Efforts to attenuate hospital noise have been limited by incomplete information on the interaction between sounds and sleep physiology. Objective: We examined the cortical (electroencephalographic; EEG) arousal responses during sleep to typical hospital noises by sound type, sleep stage, and sound level in order to determine profiles of acoustic disruption of sleep. Design: Three-day polysomnographic study with a baseline (sham) night followed by two intervention nights. Settings: Soundattenuated sleep laboratory. Participants: Volunteer sample of 12 healthy subjects. Intervention: We systematically administered 14 recorded hospital sounds at calibrated rising decibel levels (total=3,147) during specific sleep stages. Measurements: EEG arousals, using established criteria, during Rapid Eye Movement (REM) sleep, and non-REM stages 2 (N2) and 3 (N3). Results: Sound presentations yielded arousal response curves that varied due to sound level, sound type, and sleep stage. Higher sound levels led to a greater probability of sleep disruption. Electronic sounds tested were more arousing than other sounds, including human voices. The effects of sound level and sound type were modified by sleep physiology, producing unique arousal probability profiles in each sleep stage. In REM sleep, the range of responses across sound types was narrowest, suggesting less differentiation of sounds. Limitations: Results in healthy people may underestimate effects of noise in hospitalized patients. Conclusions: This study systematically quantifies the disruptive capacity of sounds on sleep. These arousal-probability profiles can be expected to drive innovation in design, construction, engineering, building materials, equipment and care-giving protocols, with the goal of providing environments conductive for sleep among hospitalized patients.