Anti-Voice Adaptation Suggests Prototype-Based Coding of Voice Identity - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Frontiers in Psychology Année : 2011

Anti-Voice Adaptation Suggests Prototype-Based Coding of Voice Identity

Résumé

the steady-state nature of vowels offers a good analogy with the static images of faces used in face processing research as a simplified model of facial recognition. This simplified model may not generalize to more naturalistic situations when several other cues such as speaking style, dialect, idiosyncratic expressions, etc., contribute to recognition but, as for face perception, this approach potentially can lead to important insights. In this simplified framework, we have previously shown that speaker identity perception in simple vowels can be adequately summarized (78 and 81% of the variance explained for male and female speakers, respectively) by representing voices as points in a two-dimensional "perceptual voice space" (Baumann and Belin, 2010). This voice identity space is largely independent of which vowel is uttered. The two axes of this space reflect contributions of the source, and filter aspects of phonation, respectively. In that space, voices located close to one another are perceived with similar identities, whereas voices located far apart are perceived with different identities (Baumann and Belin, 2010). This is analogous to the dominant framework of facial identity representation in which faces are represented in a multidimensional space and coded either as vectors referenced to the central, prototypical face [prototype-based model or axis model (Tsao and Freiwald, 2006)] or as prototype-independent exemplars (Valentine, 1991). Importantly, one characteristic distinguishes between the two models: the role of the average face (Valentine, 1991; Tsao and Freiwald, 2006). In the axis, or prototype-based model, all faces are encoded in terms of their deviation from the prototypical, average face (thought to be the center of mass of the multidimensional face space); on the contrary, the exemplar
Fichier principal
Vignette du fichier
Latinus & Belin_FrontiersPerceptionScience_2011.pdf (1.21 Mo) Télécharger le fichier
Origine : Fichiers éditeurs autorisés sur une archive ouverte

Dates et versions

hal-04323252 , version 1 (05-12-2023)

Identifiants

Citer

Marianne Latinus, Pascal Belin. Anti-Voice Adaptation Suggests Prototype-Based Coding of Voice Identity. Frontiers in Psychology, 2011, 2, ⟨10.3389/fpsyg.2011.00175⟩. ⟨hal-04323252⟩
2 Consultations
3 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More