Decoding the structure of marmoset call sequences
According to scientists from the University of Zurich and the NCCR Evolving Language, some aspects of language may not be as unique as previously thought. In a new paper published in the Royal Society Open Science, a team of linguists and biologists worked together to shed light on the complexity of call combinations in a South American monkey species, the common marmoset. The researchers demonstrate this species is capable of combining more than two calls, with hierarchical rules – something previously believed to be impossible for animals.
Marmoset monkeys. © Judith M. Burkart, UZH.
More than two calls?
Common marmosets are very social monkeys, living in the dense tropical forests of Brazil. To communicate with one another, the species boasts a complex communication system, composed of eight different types of calls (ek, twitter, tsk, phee, chirp, whistle, trill, and chatter). Each call is produced by the monkeys in a different behavioural situation. Simon Townsend, one of the senior authors of the paper explains: “Contact calls aim at staying in contact with others in the group, while mobbing and alarm calls serve as instruments to fend off other groups of monkeys or inform others about a predator nearby. Lastly, food calls are intended to communicate with other sources of food.”
For humans, the complexity of language does not only come from the number of sounds we can produce. We are also able to combine these sounds into words, and these words into longer strings – sentences – where the relative position of each word impacts their meaning. Whilst some animals have the ability to combine their calls into strings, the current conclusion in animal behaviour research is that animals at most combine two calls. “We wanted to find out whether that is true in marmosets, and if the gap between animals and humans is as decisively large here as it seems,” explains Alexandra Bosshard, lead author of the study.
In their study, published in Royal Society Open Science, the researchers analysed marmoset calls showing that from 8’000 recorded vocalisations, more than 10% were part of sequences of three or more calls. “We show that marmosets combine calls on a regular basis. Interestingly, these combinations often consist of more than just two calls,” explains Bosshard.
Rules in call sequences
The researchers we also interested in how these calls within the sequences were combined. Through modelling the probabilities of calls occurring together, the researchers were able to uncover rules to determine what comes after a given call. Balthasar Bickel, senior author of the paper explains: “We found that more second-level Markovian modelling approaches best predicted the marmoset combinations, meaning that the next call in a call sequence is not only dependent on the current call but also on the previous two calls.”
According to the authors’ results, there are rules that determine what can come after a combination of two calls. For example, an ek call is most likely to follow a combination of calls containing tsk. Its occurrence can not be well predicted on the basis tsk alone.
The researchers also noticed that the combinations depend on behavioural states: it is for example frequent for a food call to be followed by a contact call and then switch back to a food call.
“These findings suggest that like in human language, the calls are not arranged like beads on a string but that there is some amount of hierarchical structure,” comments Bickel. “What we do not know yet is whether the marmoset structures are produced in similar ways as in humans and whether they involve differences in meaning,” the researcher concludes.
Reference
About the NCCR Evolving Language
The Swiss National Centre of Competence in Research (NCCR) Evolving Language is a nationwide interdisciplinary research consortium bringing together research groups from the humanities, from language and computer science, the social sciences, and the natural sciences at an unprecedented level. Together, we aim at solving one of humanity’s great mysteries: What is language? How did our species develop the capacity for linguistic expression, for processing language in the brain, and for consistently passing down new variations to the next generation? How will our capacity for language change in the face of digital communication and neuroengineering?
