Cervical and ocular vestibular evoked myogenic potentials (cVEMPs and oVEMPs) to air-conducted tone bursts (250–2000 Hz) were recorded in 14 patients with superior canal dehiscence (SCD) and 32 healthy controls. For cVEMPs, the most common ‘optimal frequency’ in control ears (48.2%) was 500 Hz; for oVEMPs, it was 1000 Hz (51.8%). We found a significant interaction between age and frequency, with a shift towards higher-frequency tuning in older subjects. cVEMP and oVEMP tuning in SCD was characterised by a broadening of amplitude and threshold tuning curves. The tendency of cVEMPs to tune to lower frequencies compared to oVEMP was enhanced in SCD. Differences in cVEMP and oVEMP ‘optimal frequencies’, demonstrated in 57.1% intact ears and 81.3% dehiscent ears, imply differences in the recruitment of hair cells generating these two reflexes. Age-matched oVEMP amplitudes provided excellent separation between SCD and control ears. Although cVEMP amplitudes overlapped between SCD and control ears, better separation was achieved by using a 2-kHz stimulus.