Figure 1.

Distributions of the acoustic characteristics for multitalker, naturally uttered bisyllabic words, produced as questions or statements. Panels (a) through (c) show the amounts of F0 variation, peak intensity ratio, and duration ratio, respectively. The x-axis displays the utterance types (i.e., “question” vs. “statement”); the y-axis displays the overall amount of changes in each acoustic dimension. The mean and median are displayed by the dotted and solid lines across each box, respectively. The upper and lower bounds of each box represent the quartiles, the whisker away from the box bounds showed the ± 1.25 SD of the mean, and the filled circles represent the 5th and 95th percentiles bounds, if they are outside of the end of whisker. The p value is indicated on each panel.