Nikhil Phaniraj
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Nikhil Phaniraj has a total of 24 co-authors. Here are the top co-authors by number of shared publications:
- Judith Maria Burkart
Department of Evolutionary Anthropology, University of Zurich, Zurich, Switzerland
9 shared publications - Kaja Wierucka
Behavioral Ecology and Sociobiology Unit, German Primate Center – Leibniz Institute for Primate Research, Göttingen, Germany
4 shared publications - Rahel K. Brügger
Department of Evolutionary Anthropology, University of Zurich, Zurich, Switzerland
2 shared publications Yvonne Zürcher
Institute of Evolutionary Anthropology (IEA), University of Zurich, Winterthurerstrasse 190, 8057 Zürich, Switzerland
2 shared publicationsAditya Pujari
Division of Biology, Indian Institute of Science Education and Research Pune
1 shared publication- Claudia Fichtel
Behavioral Ecology & Sociobiology Unit German Primate Center Göttingen Germany
1 shared publication
Latest Publications
- Open Access
Norm-like vocal behavior in a nonhuman primate
Human interlocutors have strong expectations that guide when, how, and what is communicated, but the evolutionary origins of such normative expectations remain poorly understood. Vocal accommodation in marmosets, where individuals modify their calls to match partners, offers a potential model. Within an interactive closed-loop playback paradigm, marmosets experienced partner calls that either converged toward or diverged away from their own call structure. Marmosets responded affiliatively to convergence, whereas diverging calls elicited protest vocalizations. Importantly, the protests stopped when the calls converged again. Effects were partner-specific, indicating that marmosets expect bond partners, but not strangers, to converge during a conversation. Furthermore, individuals modified their calls to restore similarity to dissimilar partner playbacks. These findings identify marmoset vocal accommodation as a socially regulated behavior with normative properties.
Open AccessNorm-like vocal behavior in a nonhuman primate
Human interlocutors have strong expectations that guide when, how, and what is communicated, but the evolutionary origins of such normative expectations remain poorly understood. Vocal accommodation in marmosets, where individuals modify their calls to match partners, offers a potential model. Within an interactive closed-loop playback paradigm, marmosets experienced partner calls that either converged toward or diverged away from their own call structure. Marmosets responded affiliatively to convergence, whereas diverging calls elicited protest vocalizations. Importantly, the protests stopped when the calls converged again. Effects were partner-specific, indicating that marmosets expect bond partners, but not strangers, to converge during a conversation. Furthermore, individuals modified their calls to restore similarity to dissimilar partner playbacks. These findings identify marmoset vocal accommodation as a socially regulated behavior with normative properties.
- Open Access
- 3 June 2026
- Published by The Royal Society in Proceedings Of The Royal Society B-Biological Sciences
- Vol. 293 (2072)
Dynamic vocal learning in adult marmoset monkeys
Socially mediated vocal plasticity, such as vocal accommodation, is an important mechanism for maintaining social cohesion and coordination in group-living animals. Yet, the cognitive mechanisms underlying vocal accommodation remain poorly understood. Here we scrutinize the temporal dynamics of vocal accommodation in common marmosets by tracking the vocalizations of pairs during the first two months after pair formation. We show that vocal accommodation in marmoset pairs displayed the following properties: (i) bidirectional learning (i.e. both males and females underwent equal amount of vocal change); (ii) an exponential decrease in vocal distance with time; (iii) sensitivity to initial vocal distance, that is, pairs converged to a higher extent when their vocalizations were initially more dissimilar; and (iv) synchronous dyadic movement through acoustic space. We find that mathematical models capturing these characteristics and accurately simulating the phenomena rely on dynamic template matching. This suggests that marmosets match their vocalizations to an auditory template continuously updated to track changes in partners’ vocalizations, a dynamic form of vocal learning. The model thus suggests that vocal convergence is a highly interactive process that may have provided a foundation for more advanced vocal learning in primates, resonating with recent claims about its role in language evolution.
- 3 June 2026
- Published by The Royal Society in Proceedings Of The Royal Society B-Biological Sciences
- Vol. 293 (2072)
Open AccessDynamic vocal learning in adult marmoset monkeys
Socially mediated vocal plasticity, such as vocal accommodation, is an important mechanism for maintaining social cohesion and coordination in group-living animals. Yet, the cognitive mechanisms underlying vocal accommodation remain poorly understood. Here we scrutinize the temporal dynamics of vocal accommodation in common marmosets by tracking the vocalizations of pairs during the first two months after pair formation. We show that vocal accommodation in marmoset pairs displayed the following properties: (i) bidirectional learning (i.e. both males and females underwent equal amount of vocal change); (ii) an exponential decrease in vocal distance with time; (iii) sensitivity to initial vocal distance, that is, pairs converged to a higher extent when their vocalizations were initially more dissimilar; and (iv) synchronous dyadic movement through acoustic space. We find that mathematical models capturing these characteristics and accurately simulating the phenomena rely on dynamic template matching. This suggests that marmosets match their vocalizations to an auditory template continuously updated to track changes in partners’ vocalizations, a dynamic form of vocal learning. The model thus suggests that vocal convergence is a highly interactive process that may have provided a foundation for more advanced vocal learning in primates, resonating with recent claims about its role in language evolution.
- Open Access
- 5 February 2026
- Published by The Royal Society in Philosophical Transactions of the Royal Society B: Biological Sciences
- Vol. 381 (1943)
Opportunities and mechanisms for learning through social interactions: lessons from marmosets
Social interactions are crucial for learning not only in humans but also in non-human animals. To date, comparative studies have typically focused on what is learned from others and on purely observational learning, while paying less attention on how social interactions facilitate learning. Here, we present how computational approaches can be leveraged to examine the role of contingency in learning through social interactions, focusing on callitrichid monkeys. They are the only primates besides humans in which group members other than the mother contribute significantly to infant care (i.e. cooperative breeding), resulting in more and richer opportunities for social interactions and learning, in both immatures and adults. We first review how contingency, turn-taking and social timing fundamentally shape learning and inter-individual coordination. We then zoom in on turn-taking in vigilance and vocal accommodation, illustrating how mathematical modelling can offer unique insights into the underlying mechanisms. We also highlight the potential of hybrid approaches that combine machine learning’s pattern detection strengths with the interpretability and explanatory power of mathematical models. Overall, we find that contingency is key to learning through social interactions not only in humans but also in non-human primates, and perhaps particularly so in the cooperatively breeding callitrichids. This article is part of the theme issue ‘Mechanisms of learning from social interaction’.
- 5 February 2026
- Published by The Royal Society in Philosophical Transactions of the Royal Society B: Biological Sciences
- Vol. 381 (1943)
Open AccessOpportunities and mechanisms for learning through social interactions: lessons from marmosets
Social interactions are crucial for learning not only in humans but also in non-human animals. To date, comparative studies have typically focused on what is learned from others and on purely observational learning, while paying less attention on how social interactions facilitate learning. Here, we present how computational approaches can be leveraged to examine the role of contingency in learning through social interactions, focusing on callitrichid monkeys. They are the only primates besides humans in which group members other than the mother contribute significantly to infant care (i.e. cooperative breeding), resulting in more and richer opportunities for social interactions and learning, in both immatures and adults. We first review how contingency, turn-taking and social timing fundamentally shape learning and inter-individual coordination. We then zoom in on turn-taking in vigilance and vocal accommodation, illustrating how mathematical modelling can offer unique insights into the underlying mechanisms. We also highlight the potential of hybrid approaches that combine machine learning’s pattern detection strengths with the interpretability and explanatory power of mathematical models. Overall, we find that contingency is key to learning through social interactions not only in humans but also in non-human primates, and perhaps particularly so in the cooperatively breeding callitrichids. This article is part of the theme issue ‘Mechanisms of learning from social interaction’.
- Open Access
- 9 July 2025
- Published by Wiley in Methods in Ecology and Evolution
- Vol. 16 (8) , 1574-1586
Same data, different results? Machine learning approaches in bioacoustics
Automated acoustic analysis is increasingly used in behavioural ecology, and determining caller identity is a key element for many investigations. However, variability in feature extraction and classification methods limits the comparability of results across species and studies, constraining conclusions we can draw about the ecology and evolution of the groups under study. We investigated the impact of using different feature extraction (spectro‐temporal measurements, linear and Mel‐frequency cepstral coefficients (MFCC), as well as highly comparative time‐series analysis) and classification methods (discriminant function analysis, neural networks, random forests (RF), and support vector machines) on the consistency of caller identity classification accuracy across 16 mammalian datasets. We found that MFCCs and RFs yield consistently reliable results across datasets, facilitating a standardised approach across species that generates directly comparable data. These findings remained consistent across vocalisation sample sizes and number of individuals considered. We offer guidelines for processing and analysing mammalian vocalisations, fostering greater comparability and advancing our understanding of the evolutionary significance of acoustic communication in diverse mammalian species.
- 9 July 2025
- Published by Wiley in Methods in Ecology and Evolution
- Vol. 16 (8) , 1574-1586
Open AccessSame data, different results? Machine learning approaches in bioacoustics
Automated acoustic analysis is increasingly used in behavioural ecology, and determining caller identity is a key element for many investigations. However, variability in feature extraction and classification methods limits the comparability of results across species and studies, constraining conclusions we can draw about the ecology and evolution of the groups under study. We investigated the impact of using different feature extraction (spectro‐temporal measurements, linear and Mel‐frequency cepstral coefficients (MFCC), as well as highly comparative time‐series analysis) and classification methods (discriminant function analysis, neural networks, random forests (RF), and support vector machines) on the consistency of caller identity classification accuracy across 16 mammalian datasets. We found that MFCCs and RFs yield consistently reliable results across datasets, facilitating a standardised approach across species that generates directly comparable data. These findings remained consistent across vocalisation sample sizes and number of individuals considered. We offer guidelines for processing and analysing mammalian vocalisations, fostering greater comparability and advancing our understanding of the evolutionary significance of acoustic communication in diverse mammalian species.
- Open Access
- 15 May 2024
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 20 (5) , e1012104
Marmosets mutually compensate for differences in rhythms when coordinating vigilance
Synchronization is widespread in animals, and studies have often emphasized how this seemingly complex phenomenon can emerge from very simple rules. However, the amount of flexibility and control that animals might have over synchronization properties, such as the strength of coupling, remains underexplored. Here, we studied how pairs of marmoset monkeys coordinated vigilance while feeding. By modeling them as coupled oscillators, we noted that (1) individual marmosets do not show perfect periodicity in vigilance behaviors, (2) nevertheless, marmoset pairs started to take turns being vigilant over time, a case of anti-phase synchrony, (3) marmosets could couple flexibly; the coupling strength varied with every new joint feeding bout, and (4) marmosets could control the coupling strength; dyads showed increased coupling if they began in a more desynchronized state. Such flexibility and control over synchronization require more than simple interaction rules. Minimally, animals must estimate the current degree of asynchrony and adjust their behavior accordingly. Moreover, the fact that each marmoset is inherently non-periodic adds to the cognitive demand. Overall, our study provides a mathematical framework to investigate the cognitive demands involved in coordinating behaviors in animals, regardless of whether individual behaviors are rhythmic or not.
- 15 May 2024
- Published by Public Library of Science (PLoS) in PLOS Computational Biology
- Vol. 20 (5) , e1012104
Open AccessMarmosets mutually compensate for differences in rhythms when coordinating vigilance
Synchronization is widespread in animals, and studies have often emphasized how this seemingly complex phenomenon can emerge from very simple rules. However, the amount of flexibility and control that animals might have over synchronization properties, such as the strength of coupling, remains underexplored. Here, we studied how pairs of marmoset monkeys coordinated vigilance while feeding. By modeling them as coupled oscillators, we noted that (1) individual marmosets do not show perfect periodicity in vigilance behaviors, (2) nevertheless, marmoset pairs started to take turns being vigilant over time, a case of anti-phase synchrony, (3) marmosets could couple flexibly; the coupling strength varied with every new joint feeding bout, and (4) marmosets could control the coupling strength; dyads showed increased coupling if they began in a more desynchronized state. Such flexibility and control over synchronization require more than simple interaction rules. Minimally, animals must estimate the current degree of asynchrony and adjust their behavior accordingly. Moreover, the fact that each marmoset is inherently non-periodic adds to the cognitive demand. Overall, our study provides a mathematical framework to investigate the cognitive demands involved in coordinating behaviors in animals, regardless of whether individual behaviors are rhythmic or not.
- Open Access
- 18 October 2023
- Published by The Royal Society in Journal of The Royal Society Interface
- Vol. 20 (207) , 20230399
Who is calling? Optimizing source identification from marmoset vocalizations with hierarchical machine learning classifiers
With their highly social nature and complex vocal communication system, marmosets are important models for comparative studies of vocal communication and, eventually, language evolution. However, our knowledge about marmoset vocalizations predominantly originates from playback studies or vocal interactions between dyads, and there is a need to move towards studying group-level communication dynamics. Efficient source identification from marmoset vocalizations is essential for this challenge, and machine learning algorithms (MLAs) can aid it. Here we built a pipeline capable of plentiful feature extraction, meaningful feature selection, and supervised classification of vocalizations of up to 18 marmosets. We optimized the classifier by building a hierarchical MLA that first learned to determine the sex of the source, narrowed down the possible source individuals based on their sex and then determined the source identity. We were able to correctly identify the source individual with high precisions (87.21%–94.42%, depending on call type, and up to 97.79% after the removal of twins from the dataset). We also examine the robustness of identification across varying sample sizes. Our pipeline is a promising tool not only for source identification from marmoset vocalizations but also for analysing vocalizations of other species.
- 18 October 2023
- Published by The Royal Society in Journal of The Royal Society Interface
- Vol. 20 (207) , 20230399
Open AccessWho is calling? Optimizing source identification from marmoset vocalizations with hierarchical machine learning classifiers
With their highly social nature and complex vocal communication system, marmosets are important models for comparative studies of vocal communication and, eventually, language evolution. However, our knowledge about marmoset vocalizations predominantly originates from playback studies or vocal interactions between dyads, and there is a need to move towards studying group-level communication dynamics. Efficient source identification from marmoset vocalizations is essential for this challenge, and machine learning algorithms (MLAs) can aid it. Here we built a pipeline capable of plentiful feature extraction, meaningful feature selection, and supervised classification of vocalizations of up to 18 marmosets. We optimized the classifier by building a hierarchical MLA that first learned to determine the sex of the source, narrowed down the possible source individuals based on their sex and then determined the source identity. We were able to correctly identify the source individual with high precisions (87.21%–94.42%, depending on call type, and up to 97.79% after the removal of twins from the dataset). We also examine the robustness of identification across varying sample sizes. Our pipeline is a promising tool not only for source identification from marmoset vocalizations but also for analysing vocalizations of other species.
- Open Access
Marmosets mutually compensate for differences in rhythms when coordinating vigilance
Synchronisation is widespread in animals, and studies have often emphasised how this seemingly complex phenomenon can emerge from very simple rules. However, the amount of flexibility and control that animals might have over synchronisation properties, such as the strength of coupling, remains underexplored. Here, we studied how pairs of marmoset monkeys coordinated vigilance while feeding. By modelling them as coupled oscillators, we noted that (1) individual marmosets do not show perfect periodicity in vigilance behaviours, (2) even then, pairs of marmosets developed a tendency to take turns being vigilant, a case of anti-phase synchrony, (3) marmosets could couple flexibly; the coupling strength varied with every new joint feeding bout, and (4) marmosets could control the coupling strength; dyads showed increased coupling if they began in a more desynchronised state. Such flexibility and control over synchronisation require more than simple interaction rules. Minimally, animals must estimate the current degree of asynchrony and adjust their behaviour accordingly. Moreover, the fact that each marmoset is inherently non-periodic adds to the cognitive demand. Overall, our study taps into the cognitive aspects of synchronisation and provides a mathematical framework to investigate the phenomenon more widely, where individuals may not display perfectly rhythmic behaviours.
Open AccessMarmosets mutually compensate for differences in rhythms when coordinating vigilance
Synchronisation is widespread in animals, and studies have often emphasised how this seemingly complex phenomenon can emerge from very simple rules. However, the amount of flexibility and control that animals might have over synchronisation properties, such as the strength of coupling, remains underexplored. Here, we studied how pairs of marmoset monkeys coordinated vigilance while feeding. By modelling them as coupled oscillators, we noted that (1) individual marmosets do not show perfect periodicity in vigilance behaviours, (2) even then, pairs of marmosets developed a tendency to take turns being vigilant, a case of anti-phase synchrony, (3) marmosets could couple flexibly; the coupling strength varied with every new joint feeding bout, and (4) marmosets could control the coupling strength; dyads showed increased coupling if they began in a more desynchronised state. Such flexibility and control over synchronisation require more than simple interaction rules. Minimally, animals must estimate the current degree of asynchrony and adjust their behaviour accordingly. Moreover, the fact that each marmoset is inherently non-periodic adds to the cognitive demand. Overall, our study taps into the cognitive aspects of synchronisation and provides a mathematical framework to investigate the phenomenon more widely, where individuals may not display perfectly rhythmic behaviours.
- Open Access
Dynamic vocal learning in adult marmoset monkeys
While vocal learning is vital to language acquisition in children, adults continue to adjust their speech while adapting to different social environments in the form of social vocal accommodation (SVA). Even though adult and infant vocal learning seemingly differ in their properties, whether the mechanisms underlying them differ remains unknown. The complex structure of language creates a challenge in quantifying vocal changes during SVA. Consequently, animals with simpler vocal communication systems are powerful tools for understanding the mechanisms underlying SVA. Here, we tracked acoustic changes in the vocalizations of adult common marmoset pairs, a highly vocal primate species known to show SVA, for up to 85 days after pairing with a new partner. We identified four properties of SVA in marmosets: (1) bidirectional learning, (2) exponential decrease in vocal distance with time, (3) sensitivity to initial vocal distance, and (4) dyadic acoustic feature synchrony. We developed a mathematical model that shows all four properties. The model suggests that marmosets continuously update the memory of their partners’ vocalizations and modify their own vocalizations to match them, a dynamic form of vocal learning. The model provides crucial insights into the mechanisms underlying SVA in adult animals and how they might differ from infant vocal learning.
Open AccessDynamic vocal learning in adult marmoset monkeys
While vocal learning is vital to language acquisition in children, adults continue to adjust their speech while adapting to different social environments in the form of social vocal accommodation (SVA). Even though adult and infant vocal learning seemingly differ in their properties, whether the mechanisms underlying them differ remains unknown. The complex structure of language creates a challenge in quantifying vocal changes during SVA. Consequently, animals with simpler vocal communication systems are powerful tools for understanding the mechanisms underlying SVA. Here, we tracked acoustic changes in the vocalizations of adult common marmoset pairs, a highly vocal primate species known to show SVA, for up to 85 days after pairing with a new partner. We identified four properties of SVA in marmosets: (1) bidirectional learning, (2) exponential decrease in vocal distance with time, (3) sensitivity to initial vocal distance, and (4) dyadic acoustic feature synchrony. We developed a mathematical model that shows all four properties. The model suggests that marmosets continuously update the memory of their partners’ vocalizations and modify their own vocalizations to match them, a dynamic form of vocal learning. The model provides crucial insights into the mechanisms underlying SVA in adult animals and how they might differ from infant vocal learning.
- Open Access
Optimising source identification from marmoset vocalisations with hierarchical machine learning classifiers
Marmosets, with their highly social nature and complex vocal communication system, are important models for comparative studies of vocal communication and, eventually, language evolution. However, our knowledge about marmoset vocalisations predominantly originates from playback studies or vocal interactions between dyads, and there is a need to move towards studying group-level communication dynamics. Efficient source identification from marmoset vocalisations is essential for this challenge, and machine learning algorithms (MLAs) can aid it. Here we built a pipeline capable of plentiful feature extraction, meaningful feature selection, and supervised classification of vocalisations of up to 18 marmosets. We optimised the classifier by building a hierarchical MLA that first learned to determine the sex of the source, narrowed down the possible source individuals based on their sex, and then determined the source identity. We were able to correctly identify the source individual with high precisions (87.21% – 94.42%, depending on call type, and up to 97.79% after the removal of twins from the dataset). We also examine the robustness of identification across varying sample sizes. Our pipeline is a promising tool not only for source identification from marmoset vocalisations but also for analysing vocalisations and tracking vocal learning trajectories of other species.
Open AccessOptimising source identification from marmoset vocalisations with hierarchical machine learning classifiers
Marmosets, with their highly social nature and complex vocal communication system, are important models for comparative studies of vocal communication and, eventually, language evolution. However, our knowledge about marmoset vocalisations predominantly originates from playback studies or vocal interactions between dyads, and there is a need to move towards studying group-level communication dynamics. Efficient source identification from marmoset vocalisations is essential for this challenge, and machine learning algorithms (MLAs) can aid it. Here we built a pipeline capable of plentiful feature extraction, meaningful feature selection, and supervised classification of vocalisations of up to 18 marmosets. We optimised the classifier by building a hierarchical MLA that first learned to determine the sex of the source, narrowed down the possible source individuals based on their sex, and then determined the source identity. We were able to correctly identify the source individual with high precisions (87.21% – 94.42%, depending on call type, and up to 97.79% after the removal of twins from the dataset). We also examine the robustness of identification across varying sample sizes. Our pipeline is a promising tool not only for source identification from marmoset vocalisations but also for analysing vocalisations and tracking vocal learning trajectories of other species.
- Open Access
CineFinch: An animated female zebra finch for studying courtship interactions
Dummies, videos and computer animations have been used extensively in animal behaviour to study simple social interactions. These methods allow complete control of one interacting animal, making it possible to test hypotheses about the significance and relevance of different elements of animal displays. Recent studies have demonstrated the potential of videos and interactive displays for studying more complex courtship interactions in the zebra finch, a well-studied songbird. Here, we extended these techniques by developing an animated female zebra finch and showed that ~40% of male zebra finches (n=5/12) sing to this animation. To study real-time social interactions, we developed two possible methods for closed loop control of animations; (1) an arduino based system to initiate videos/animations based on perch hops and (2) a video game engine based system to change animations. Overall, our results provide an important tool for understanding the dynamics of complex social interactions during courtship. SUMMARY STATEMENT We develop and test an animation of a female zebra finch to study song and courtship interactions in the male zebra finch.
Open AccessCineFinch: An animated female zebra finch for studying courtship interactions
Dummies, videos and computer animations have been used extensively in animal behaviour to study simple social interactions. These methods allow complete control of one interacting animal, making it possible to test hypotheses about the significance and relevance of different elements of animal displays. Recent studies have demonstrated the potential of videos and interactive displays for studying more complex courtship interactions in the zebra finch, a well-studied songbird. Here, we extended these techniques by developing an animated female zebra finch and showed that ~40% of male zebra finches (n=5/12) sing to this animation. To study real-time social interactions, we developed two possible methods for closed loop control of animations; (1) an arduino based system to initiate videos/animations based on perch hops and (2) a video game engine based system to change animations. Overall, our results provide an important tool for understanding the dynamics of complex social interactions during courtship. SUMMARY STATEMENT We develop and test an animation of a female zebra finch to study song and courtship interactions in the male zebra finch.