ABSTRACT Perceptual adaptation is often studied within a single sense. However, our experience of the world is naturally multisensory. Here, we investigated cross-sensory (visual-vestibular) adaptation of self-motion perception. It was previously found that relatively long visual self-motion stimuli (≳ 15s) are required to adapt subsequent vestibular perception, and that shorter duration stimuli do not elicit cross-sensory (visual↔vestibular) adaptation. However, it is not known whether several discrete short-duration stimuli may lead to cross-sensory adaptation (even when their sum, if presented together, would be too short to elicit cross-sensory adaptation). This would suggest that the brain monitors and adapts to supra-modal statistics of events in the environment. Here we investigated whether cross-sensory (visual↔vestibular) adaptation occurs after experiencing several short (1s) self-motion stimuli. Forty-five participants discriminated the headings of a series of self-motion stimuli. To expose adaptation effects, the trials were grouped in 140 batches, each comprising three ‘prior’ trials, with headings biased to the right or left, followed by a single unbiased ‘test’ trial. Right, and left-biased batches were interleaved pseudo-randomly. We found significant adaptation in both cross-sensory conditions (visual prior and vestibular test trials, and vice versa), as well as both unisensory conditions (when prior and test trials were of the same modality – either visual or vestibular). Fitting the data with a logistic regression model revealed that adaptation was elicited by the prior stimuli (not prior choices). These results suggest that the brain monitors supra-modal statistics of events in the environment, even for short-duration stimuli, leading to functional (supra-modal) adaptation of perception.