EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Perceptual adaptation to human-likeness in voices

Christine Nussbaum, Sebastian Goerke, Ibrahim Vefa Arslan, Stefan R. Schweinberger

Computers in Human Behavior Reports · 2026-11 · 101447

DOI: 10.1016/j.chbr.2026.101447 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Computers in Human Behavior Reports, 2026-11. Verfasst von Christine Nussbaum und 3 weiteren Autor:innen. Die Arbeit nennt 78 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Friedrich Schiller University Jena.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

With synthetic voices now being a part of everyday life, it is important to understand how the perception of synthetic voices is shaped and modified through exposure. Therefore, we employed an adaptation paradigm to test whether perception of human-likeness is amenable to short-term perceptual encounters. In three experiments, we presented morphed target voices on a human-synthetic-continuum in three conditions: without prior adaptation, and after repeated exposure to adaptor voices that all were either synthetic or human. Experiment 1 (N = 26) was conducted online, Experiment 2 (N = 31) was a full replication in the lab, and Experiment 3 (N = 36) employed an adjusted procedure with blocked and reduced adaptor exposure to test the robustness of effects. In all experiments, we observed consistent contrastive aftereffects: After adaptation to synthetic voices, ambiguous voices were more often classified as human. After adaptation to human voices, the same ambiguous voices were more often classified as synthetic. Intriguingly, in Experiment 1 and 2, contrastive aftereffects were even larger in trials which employed digital caricatures (i.e., 25%-exaggerated human or synthetic voices) as adaptors, even though caricatures themselves were not consistently rated as more human (or synthetic, respectively) compared to original voices. While the adaptation effect survived the adjusted procedure (Experiment 3), substantial modulations by caricatured adaptors were no longer observed. Together, these findings provide robust evidence that impressions of vocal human-likeness are not only determined by the acoustic features but are the result of a complex interplay with listeners and their recent perceptual experiences.

Förderung

  • Friedrich Schiller University Jena

Zitieren

APA

Nussbaum, C., Goerke, S., Arslan, I. V., & Schweinberger, S. R. (2026). Perceptual adaptation to human-likeness in voices. Computers in Human Behavior Reports, 101447. https://doi.org/10.1016/j.chbr.2026.101447

Vancouver

Nussbaum C, Goerke S, Arslan IV, Schweinberger SR. Perceptual adaptation to human-likeness in voices. Computers in Human Behavior Reports. 2026:101447. doi:10.1016/j.chbr.2026.101447

BibTeX
@article{Nussbaum2026,
  author = {Nussbaum, Christine and Goerke, Sebastian and Arslan, Ibrahim Vefa and Schweinberger, Stefan R.},
  title = {Perceptual adaptation to human-likeness in voices},
  journal = {Computers in Human Behavior Reports},
  year = {2026},
  pages = {101447},
  doi = {10.1016/j.chbr.2026.101447}
}
Kostenlos starten →