Each sheet corresponds to a different dataset, and to dictionaries related to different vocal features of the produced syllables (pitch, amplitude, spectral entropy).
dictionaries_songbird_50: features above/below median
1st column: relevant codeword expressed in terms of the indexes of the interacting time bins t_j (from 1 to 20) that code for the vocal features.
2nd column: observed frequency
3rd column: expected frequency
4th column: 1-> coding for high value of feature, 0-> for low value, 10-> for both.
5th column: marginal magnetization, which determines significance in uBIA.
6th column: other observed frequency (when 4th column is 10)
dictionaries_songbird_p33: features in 33% percentile interval
1st column: relevant codeword expressed in terms of the indexes of the interacting time bins t_j (from 1 to 20) that code for the vocal features.
2nd column: observed frequency
3rd column: expected frequency
4th column: marginal magnetization, which determines significance in uBIA.
dictionaries_songbird_p20: features in 20% percentile interval
1st column: relevant codeword expressed in terms of the indexes of the interacting time bins t_j (from 1 to 20) that code for the vocal features.
2nd column: observed frequency
3rd column: expected frequency
4th column: marginal magnetization, which determines significance in uBIA.