Machine learning algorithms can predict emotional valence across ungulate vocalizations
Summary: Vocalizations can vary as a function of their context of production and provide an immediate measure of an animal’s affective states. If vocal expression of emotions has been conserved throughout evolution, direct between-species comparisons using the same set of acoustic indicators should...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-02-01
|
| Series: | iScience |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S258900422500094X |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832087454333534208 |
|---|---|
| author | Romain A. Lefèvre Ciara C.R. Sypherd Élodie F. Briefer |
| author_facet | Romain A. Lefèvre Ciara C.R. Sypherd Élodie F. Briefer |
| author_sort | Romain A. Lefèvre |
| collection | DOAJ |
| description | Summary: Vocalizations can vary as a function of their context of production and provide an immediate measure of an animal’s affective states. If vocal expression of emotions has been conserved throughout evolution, direct between-species comparisons using the same set of acoustic indicators should be possible. The present study used a machine learning algorithm (eXtreme Gradient Boosting [XGBoost]) to distinguish between contact calls indicating positive (pleasant) and negative (unpleasant) emotional valence, produced in various contexts by seven species of ungulates. With an accuracy of 89.49% (balanced accuracy: 83.90%), we found that the most important predictors of emotional valence were acoustic variables reflecting changes in duration, energy quartiles, fundamental frequency, and amplitude modulation. This approach is critical in the field of emotional communication, where more information is needed to reach a better understanding of the emotional origins of human language. In addition, these results can help toward the development of automated tools for animal well-being monitoring. |
| format | Article |
| id | doaj-art-115383339e51439282a54a74a2f2e722 |
| institution | Kabale University |
| issn | 2589-0042 |
| language | English |
| publishDate | 2025-02-01 |
| publisher | Elsevier |
| record_format | Article |
| series | iScience |
| spelling | doaj-art-115383339e51439282a54a74a2f2e7222025-02-06T05:12:39ZengElsevieriScience2589-00422025-02-01282111834Machine learning algorithms can predict emotional valence across ungulate vocalizationsRomain A. Lefèvre0Ciara C.R. Sypherd1Élodie F. Briefer2Behavioural Ecology Group, Section for Ecology & Evolution, Department of Biology, University of Copenhagen, 2100 Copenhagen Ø, Denmark; Corresponding authorBehavioural Ecology Group, Section for Ecology & Evolution, Department of Biology, University of Copenhagen, 2100 Copenhagen Ø, Denmark; School of Engineering and Applied Sciences, Harvard University, Cambridge, MA, USABehavioural Ecology Group, Section for Ecology & Evolution, Department of Biology, University of Copenhagen, 2100 Copenhagen Ø, Denmark; Corresponding authorSummary: Vocalizations can vary as a function of their context of production and provide an immediate measure of an animal’s affective states. If vocal expression of emotions has been conserved throughout evolution, direct between-species comparisons using the same set of acoustic indicators should be possible. The present study used a machine learning algorithm (eXtreme Gradient Boosting [XGBoost]) to distinguish between contact calls indicating positive (pleasant) and negative (unpleasant) emotional valence, produced in various contexts by seven species of ungulates. With an accuracy of 89.49% (balanced accuracy: 83.90%), we found that the most important predictors of emotional valence were acoustic variables reflecting changes in duration, energy quartiles, fundamental frequency, and amplitude modulation. This approach is critical in the field of emotional communication, where more information is needed to reach a better understanding of the emotional origins of human language. In addition, these results can help toward the development of automated tools for animal well-being monitoring.http://www.sciencedirect.com/science/article/pii/S258900422500094Xalgorithmsartificial intelligencebioacousticswildlife behaviorzoology |
| spellingShingle | Romain A. Lefèvre Ciara C.R. Sypherd Élodie F. Briefer Machine learning algorithms can predict emotional valence across ungulate vocalizations iScience algorithms artificial intelligence bioacoustics wildlife behavior zoology |
| title | Machine learning algorithms can predict emotional valence across ungulate vocalizations |
| title_full | Machine learning algorithms can predict emotional valence across ungulate vocalizations |
| title_fullStr | Machine learning algorithms can predict emotional valence across ungulate vocalizations |
| title_full_unstemmed | Machine learning algorithms can predict emotional valence across ungulate vocalizations |
| title_short | Machine learning algorithms can predict emotional valence across ungulate vocalizations |
| title_sort | machine learning algorithms can predict emotional valence across ungulate vocalizations |
| topic | algorithms artificial intelligence bioacoustics wildlife behavior zoology |
| url | http://www.sciencedirect.com/science/article/pii/S258900422500094X |
| work_keys_str_mv | AT romainalefevre machinelearningalgorithmscanpredictemotionalvalenceacrossungulatevocalizations AT ciaracrsypherd machinelearningalgorithmscanpredictemotionalvalenceacrossungulatevocalizations AT elodiefbriefer machinelearningalgorithmscanpredictemotionalvalenceacrossungulatevocalizations |