FIG. 4.

(Color online) Mean SRTs measured in experiment 3, obtained by musicians (top panels) and non-musicians (bottom panels), for a monotonized target voice against speech-modulated buzzes (left panels) or two interfering voices (right panels). The masker F0 was either fixed at 125 Hz or varied randomly across trials between 100 and 150 Hz. Relative to the masker F0, the target F0 was either identical, 2 semitones above, below, or swapped across trials within a block. Here, the asterisk indicates the musician advantage effect.