The tone, intonation and tempo of your voice convey emotions to the person you are speaking to, who usually knows how to decode them – unless they have Parkinson’s disease or autism, for example. Science knows little about how the brain processes the emotional content of vocal messages, a gap that could be filled by studying… songbirds.

Indeed, songbirds are very good at perceiving the subtle musical variations between a courtship song and a song announcing the dawn. Songbirds are therefore excellent models for studying the music of language, or prosody, in humans. This has been shown by the work of Sarah Woolley, a professor in the Department of Biology at McGill University and a researcher at the Centre for Research on Brain, Language and Music.

She and her team studied the zebra finch, a small songbird native to Australia. The scientists’ methodology included a behavioural approach and electrophysiology to analyze the reaction of the brain cells of female birds exposed to different songs made by males and determine what chemical substances are involved.

They found that dopamine, known as the pleasure hormone, is strongly activated by courtship songs. This chemical messenger helps the birds to distinguish songs intended to woo them.

Thanks to dopamine, as well as oxytocin, the attachment hormone, birds can detect subtle variations in even very short sounds. To develop this capacity, chicks must be exposed to different types of songs from birth.

Taking the experiments a step further, the research team found that it was possible to compromise this ability to process prosody, and to modify the birds’ song preferences, by altering their dopamine levels.

Sarah Woolley and her colleagues eventually developed an algorithm and used machine learning to extract from the spectrogram (sound diagram) the specific acoustic features by which the birds recognize songs.

This approach could be replicated to study prosody in humans and better understand why, in the case of Parkinson’s disease in particular, patients have difficulty decoding emotions.