Skip to main content
. Author manuscript; available in PMC: 2018 May 29.
Published in final edited form as: Anim Cogn. 2015 Nov 18;19(2):285–315. doi: 10.1007/s10071-015-0933-6

Figure 12. Discriminant Functions (A) and positions of vocalization types in Discriminant coordinates (B and C).

Figure 12

A. The first 5 discriminant functions (DF) obtained in the RFLDA applied to the spectrogram feature space. These discriminant functions are displayed in a spectrographic representation. Each vocalization was then represented in RFLDA coordinates by projecting its spectrogram onto these Discriminant functions (using a vector dot product). B and C. The average position of each vocalization type (centroid) is shown as a colored rectangle in coordinate-pair scatter plots. The DF have been scaled so that the within vocalization type variance along each discriminant dimension is equal to 1. In B the positions of the centroids obtained from the RFLDA applied on the spectrogram feature space are shown. In C the positions of the centroids obtained from the RFLDA applied on the Predefined Acoustical Features are shown. Vocalization abbreviations are defined in Figure 3.