Figure 1. Stimuli and methods. (A) Auditory motion was produced by crossfading a sound signal between a series of horizontally aligned speaker locations to produce smooth continuous movement over space. Crossfading in this way varies inter-aural time and level differences as well as spectral cues to produce a strong motion signal. The signal was an amplitude-modulated white noise, with a modulation frequency of 20 Hz and modulation depth of 0.2. (B) Speaker locations were spaced to have a constant spatial separation of 5° of lateral angle from the participant’s position. (C) The visual stimulus was a high-contrast vertically oriented grating (0.8 cyc/deg) in a Gaussian window which translated horizontally over a background of low-contrast dynamic noise that was spatially filtered into a passband centred on the grating’s frequency. The stimulus was projected onto an acoustically transparent screen placed in front of the speakers so that they were not visible. (D) Seven speeds were randomly interleaved (20, 40, 50, 60, 70, 80, 100 deg/sec) and participants judged whether a given speed was faster or slower than the mean of the set^37,38, as represented in the left panel where the red arrow illustrates the mean. Over trials, leftward and rightward motions were interleaved randomly (right panel). To encourage reliance on velocity, path length and duration were randomly jittered over trials.