Bootstrap analyses revealed that the bandwidth and slope of vocal signals emitted by male and female mice were consistently different.

For each acoustic feature, a subsample of 1000 randomly selected vocal signals (500 male and 500 female) was used to calculate an index quantifying sex differences (see methods). This was repeated 1000 times to create distributions of sex differences for the measured acoustic features. Indices to the right of the vertical red bar (zero line) indicate that the median male value exceeded the median female value from that subsample, while indices to the left indicate the opposite. The total number of indices above zero (male-driven, above arrow pointing towards positive values) and below zero (female-driven, above arrow pointing towards negative values) was calculated. Histograms show bandwidth (A), duration (B), high frequency (C), low frequency (D), mean frequency (E), slope (F), and amplitude (G).