Fig. 2

Absence of reliable sexual signature in human babies’ cries. a Principal component analysis (PCA). Each disk or triangle represents the centroid of one baby’s cries (red disks = boys; purple triangles = girls). The radar plot on the top left represents the loadings of the acoustic variables in the PCA circle of correlations. Spectrograms on the sides illustrate the similarity between girls’ and boys’ cries (Additional file 1). b The distribution of mean pitch (mean F0) does not significantly differ between the sexes (see Table 1). Boxplots display the distribution of cry pitch for each recorded baby, showing that both boys’ and girls’ cries are distributed widely on the pitch scale. c Frequency distribution of discriminant scores (Permuted Discriminant Function Analysis; mean discrimination success rate for the validating data set = 53.3 %, chance level = 50 %). n ₁  = 15 boys, n ₂  = 13 girls; boys in red, girls in purple (a-c)