Dufour, V. | Search bioRxiv

bioRxiv2024

Can we trace the social affiliation of rooks (Corvusfrugilegus) through their vocal signature?

Inter-individual recognition is crucial for stable social relationships and it is frequently mediated through vocal signatures. In socially complex species, recognition may additionally require additional levels corresponding to other layers of social organisation such as the pair, family, social group or colony. Additional vocal signatures may encode these different levels of social organisations for recognition. We investigated this hypothesis in the calls of the rook (Corvus frugilegus), a highly social corvid. Rooks form large breeding colonies where multiple pairs nest in clusters. We recorded the calls of five colonies located in France and in Great Britain, including both wild and captive colonies. To exclude variations due to different call types, we focused on the loud nest call produced exclusively by nesting females during the breeding season. We compared the acoustic distance of calls from each individual and between individuals at various levels of nest proximity, i.e. from the same nest cluster, from different nest clusters, from colonies within the same country, and from colonies in different countries. The only vocal signatures we found were at the individual level, but not at the nest cluster or colony level. This suggests a lack of vocal convergence in this species, at least for the nest call, which may be important for pair recognition in large colonies. Further studies should now evaluate if types of calls other than the nest call better carry vocal signatures as markers of different layers of sociality in this species, or if vocal divergence is a more general vocal phenomenon. In that case, applying new methods of monitoring vocal signatures in wild individuals should help understand the cognitive, social and environmental mechanisms underlying this vocal singularisation. 1. Significance statementInter-individual recognition is crucial for social relationships in animals, and is often mediated by individual-specific acoustic characteristics in vocalisations, called a vocal signature. High levels of social organisations, such as a social group of familiar conspecifics or a breeding colony, may likewise be signalled by vocal signatures shared by multiple individuals. We used machine-learning techniques to investigate vocal signatures at multiple social levels in the nest call of brooding female rooks, a corvid species that breeds colonially but lives year-round in social groups. We find evidence of a strong individual vocal signature, but no common vocal signature even in females that nest close together, or in the same colony. A strong individual vocal signature may be a potent tool to monitor populations in this species with minimal disturbance and minimal material, especially as corvids are frequently targeted by human-fauna conflicts in continental Europe.

animal behavior and cognition↗

bioRxiv2022

Acoustic detection and identification of individual rooks in field recordings using multi-task neural networks

Individual-level monitoring is essential in many behavioural and bioacoustics studies. Collecting and annotating those data is costly in terms of human effort, but necessary prior to conducting analysis. In particular, many studies on bird vocalisations also involve manipulating the animals or human presence during observations, which may bias vocal production. Autonomous recording units can be used to collect large amounts of data without human supervision, largely removing those sources of bias. Deep learning can further facilitate the annotation of large amounts of data, for instance to detect vocalisations, identify the species, or recognise the vocalisation types in recordings. Acoustic individual identification, however, has so far largely remained limited to a single vocalisation type for a given species. This has limited the use of those techniques for automated data collection on raw recordings, where many individuals can produce vocalisations of varying complexity, potentially overlapping one another, with the additional presence of unknown and varying background noise. This paper aims at bridging this gap by developing a system to identify individual animals in those difficult conditions. Our system leverages a combination of multi-scale information integration, multi-channel audio and multi-task learning. The multi-task learning paradigm is based the overall task into four sub-tasks, three of which are auxiliary tasks: the detection and segmentation of vocalisations against other noises, the classification of individuals vocalising at any point during a sample, and the sexing of detected vocalisations. The fourth task is the overall identification of individuals. To test our approach, we recorded a captive group of rooks, a Eurasian social corvid with a diverse vocal repertoire. We used a multi-microphone array and collected a large scale dataset of time-stamped and identified vocalisations recorded, and found the system to work reliably for the defined tasks. To our knowledge, the system is the first to acoustically identify individuals regardless of the vocalisation produced. Our system can readily assist data collection and individual monitoring of groups of animals in both outdoor and indoor settings, even across long periods of time, and regardless of a species vocal complexity. All data and code used in this article is available online.

biophysics↗