Study Objectives: Children with congenital or late-onset central hypoventilation syndrome (CHS) hypoventilate during sleep, although they generally breathe adequately during wakefulness. Previous studies suggest that they have compromised central integration of afferent stimuli, rather than abnormal sensors or receptors. Cortical integration of afferent mechanical stimuli caused by respiratory loading or upper airway occlusion can be tested by measuring respiratory-related evoked potentials (RREP). We hypothesized that patients with CHS would have blunted RREP during both wakefulness and sleep. Design: RREP were produced with multiple upper airway occlusions, and were obtained from EEG electrodes Fz, Cz, Pz during wakefulness, stage 2, slow wave and rapid eye movement (REM) sleep. Setting: Sleep laboratory Participants: 10 patients with CHS and 20 controls. Each patient was age- and gender- matched to 2 controls. Wakefulness data were collected from 9 patients and 18 controls. Measurements and Results: During wakefulness, patients with CHS demonstrated reduced Nf and P300 responses compared to controls. During non-REM (NREM) sleep, patients with CHS demonstrated a reduced N350 response. In REM sleep, patients with CHS had a later P2 response. Conclusions: Results indicate that CHS patients are able to produce cortical responses to mechanical load stimulation during both wakefulness and sleep, but that central integration of the afferent signal is disrupted during wakefulness, and responses during NREM are damped relative to controls. The finding of differences between CHS andcontrols during NREM sleep, with minimal differences during REM, may be due to increased intrinsic excitatory inputs to the respiratory system during REM.