A minimum-impact, flexible tool to study vocal communication of small animals with precise individual-level resolution

To understand both proximate and ultimate factors shaping vocal communication, it is fundamental to obtain reliable information of participating individuals on different levels: First, it is necessary to separate and assign the individuals' vocalisations. Secondly, the precise timing of vocal events needs to be retained. Thirdly, vocal behaviour should be recorded from undisturbed animals in meaningful settings. A growing number of studies used animal-attached microphones to tackle these issues, but the implications for the study species and the research question often receded into the background. Here, we aim to initiate a discussion about the limitations, possible applications and the broader potential of such methods. Using lightweight wireless microphone backpacks (075g including customised leg-loop harness) combined with multi-channel recording equipment, we captured vocal behaviour of small songbirds. We evaluated the effect of the devices at various levels, including an assessment of how vocal and locomotor activities were affected by initial device attachment and battery exchange. We compared our approach to existing studies and identified suitable research examples. We acquired continuous vocalisation recordings of zebra finches, and unequivocally assigned them to interacting individuals, with system-based synchrony, irrespective of background noise. We found effects of initial backpack attachment and of battery replacement on vocal and locomotor activity, but they were minimised through the extended recording duration (ca. 16days) that outlasted habituation effects (ca. 3days). This method provides the tools to integrate individual vocal communications into a group setting, while enabling animals to behave freely in undisturbed, structured and acoustically complex environments. By minimising the effects on the animals, the behaviour under study, and ultimately on the research question, this approach will revolutionise the ability to capture individual-level vocalisations in a variety of communication contexts, opening up many new opportunities to address novel research questions.