An international team of researchers have uncovered a previously undetected structure in the birdsong of Bengalese Finches that has the same statistical patterns found in human languages.
The songbirds – famous for their complex, learned songs – organise their songs using the same mathematical formula found in human speech – a trait that has developed over time which makes complex communication easier to learn and pass from one generation to the next.
Systematic structure
The findings build on the team’s previous discovery that humpback whales also rely on the similar systematic structure.
Language has long been considered a uniquely human trait, with features that mark it out as distinct from the communication of all other species.
The findings suggests that some of language's most fundamental building blocks may not be uniquely human after all, but a universal feature of learned communication, the researchers say.
Communication processes
The research suggests that like human language the communication processes for songbirds and humpback whales are an evolutionary solution with them being transmitted culturally, meaning that they learn their communication systems from each other.
The team, led by researchers from the University of Edinburgh, the Hebrew University of Jerusalem and Teikyo University in Japan used methods inspired by how babies discover words in speech, and applied them to song data collected from Bengalese Finches in Japan.
The team observed the Finch songs contained statistically coherent sequences of notes, the avian equivalent of "words", and these sequences followed the same patterns observed in human language where a few words are used very frequently and others are used more rarely – known as a Zipfian distribution.
Data was collected across the various stage of development of the birds and the patterns were already present during the birds' earliest attempts at singing, suggesting they emerge naturally as part of the learning process.
Statistical process
The young song birds listen to the sounds around them to unconsciously learn the statistical patterns of the sounds and organise their songs – similar to how babies figure out where one word ends and the next begins by recognising which sounds frequently occur together, the research found.
This statistical process is thought to make language easier to learn, helping infants identify recurring "chunks" of speech within the continuous stream of sounds they hear every day.
Recently, the lead researchers of this study, together with Dr. Ellen Garland from the University of St. Andrews, found similar statistical patterns in the songs of humpback whales.
The teams says the new findings suggest that some of the fundamental building blocks that make language learnable can emerge wherever communication is learned and passed from one generation to the next – even in the songs of small birds.