Introduction Risset tones can be described as compounds of sinusoidal chirps separated in frequency by the same interval (e.g., one octave). These tones are continuous versions of Shepard tones [1] and evoke illusions of ever rising/falling pitch depending on the gliding direction of the chirps [2]. We found in [3] that when asked to estimate the apparent origin and motion of Risset tones, listeners were more likely to associate them with horizontal than with vertical movements. These findings

more »

... ndicated that, for Risset tones, the association between approaching sound sources and increasing pitch, the so-called Doppler illusion [4] , is stronger than the Pratt effect [5] . The latter suggests that low and high frequency tones are perceived accordingly in the vertical plane. In the Doppler illusion (not to be confused with Doppler effect) listeners associate approaching sound sources with increases in their frequency which is contrary to the nature of this phenomenon: when a sound source approaches a stationary listener located at a distance of its trajectory, the relative frequency of the source measured at the location of the listener never rises as a consequence of its movement, on the contrary, the frequency is always decreasing. In our previous research, a limited set of possible answers were presented to participants. For instance, in one of the sessions, they were forced to select between 'front' or 'above' as possible origins of the sound, and between 'approaching' or 'receding' as its possible motion. Further, diotic Risset tones were presented among binaural versions, and no direct comparison with stimuli such as speech (of which perceived motion is presumably rare) was conducted. The present study was conducted to investigate whether these methodological aspects had a significant influence on the experiment results. To that end, a greater freedom of choice for answers was introduced in an experiment using exclusively Risset tones presented diotically. Concretely, this study aims at i) replicating the results found for diotic Risset tones with upward and downward glides, ii) finding whether Risset tones with different frequency component separation yield different outcomes, and iii) whether Risset tone opinions differ from those of speech.