Behavioral responses for all participants. Perceptual anchoring relative to500 μs—more punctate, compact, or foreground-like sounds—are ranked with high values while more “wide,” “spacious,” or “diffuse” sounds, i.e., similar to a zero IAC noise, are ranked with low values. Gray points show individual responses. Boxplots show the median by the horizontal black line while 25% and 75% percentile are indicated by two hinges. The orange line shows the fitted model including both the exponential and sinusoidal terms (STAR Methods). Note that500 μs anchor was included in the fitting as having the highest rank. Mean and standard error are shown in dark red. Participants ranked ITDs at1,000 and2,000 μs as more compact, foreground-like, than1,500 μs, for which changes in sound location faded into background, and overall percept had a greater sense of spaciousness. This was also the case for ITDs at3,500 μs, which was significantly more compact than at2,500 and4,000 μs. The sinusoidal and exponential terms of the model are shown by the solid and dashed gray lines, respectively.
