The acoustic features of sincere joyful, sad, and fearful human non-verbal vocalizations and its effect on the emotional valence of cat's meowing.

The perception of emotional non-verbal vocalizations relies on a complex interplay of specific acoustic features that facilitate the identification of valence and enable appropriate behavioral responses. Here, we analyzed over 3,000 videos containing laughter, screams, and cries, selecting 664 highly recognizable and sincere sounds for further study. We computed both linear and nonlinear acoustic parameters, including spectral and temporal features, and a panel of experts evaluated each sound on scales of joy, sadness, fear, and sincerity. Joyful vocalizations were characterized by higher fractal dimensions (FD) and envelope mean frequency (EMF), while sad sounds were distinguished by loudness and reduced acoustic variability. Fearful vocalizations were identified by their minimal and maximal loudness levels and elevated power spectral density (PSD) in the 1-2 kHz range. Sincerity in non-verbal sounds correlated with nonlinear features and PSD in the 0.5-1 kHz range. Utilizing these acoustic parameters, we modified neutral cat meows to incorporate features of joyful, fearful, and sad emotional sounds. These modifications influenced human emotional perception of the meows, revealing an anthropocentric bias in emotional interpretation. Our findings suggest that the emotional perception of cat vocalizations is shaped by human-specific acoustic cues and modulated by the listener's well-being and mood. Supplementary Information: The online version contains supplementary material available at 10.1007/s11571-025-10370-7.