Main Content

A1 - Neurobiology of the vowel space

PI: Prof. Dr. M. Scharinger und Prof. Dr. A. Jansen
Ph.D. student: Leah Niber

Research context

Assumptions about speech sound representations usually imply an abstraction from speaker-related variation. Results from the first cohort of the Research Training Group, however, indicate that aside from acoustic-phonetic information, so-called »meta«- or »paralinguistic« information (e.g., age and sex) is equally anchored in those representations and play a crucial role during speech sound- and vowel processing. In this vein, the current dissertation project will ask how »categorical« paralinguistic information should be conceived of. Which aspects of the signal allow for a categorical interpretation, where does a rather gradient interpretation prevail? 
The dissertation project in A1 seeks to attract students with an interdisciplinary interest in Phonetics, Sociophonetics, Phonology, and Neurophonetics. Individual projects in A1 enable the combination of phonetic and sociolinguistic research (based on e.g., speech corpora) with neuro-scientific methods (e.g., electroencephalography [EEG] or functional magnetic resonance imaging [fMRI]).

Current Dissertation Project

Working Title: Unlearning the Gender Binary of Voices? Processing of Gender Ambiguous Voices by Listeners in Contact with Gender Diverse Speakers

In voice gender perception, voice gender has been shown to be categorically represented in two binary voice categories (Charest et al., 2013), and not continuously regarding the acoustic gender markers. This categorical perception is challenged by gender-ambiguous voices. 

State of Research

Voice gender ambiguity has been shown to influence word recognition (Doyle et al., 2024), reaction times in a gender classification task (Charest et al., 2013), and processing effort in general (Sokhi et al., 2005). Throughout the time course of an experiment, listeners have shown learning or adaptation processes to ambiguous signals, leading to decreasing effects of ambiguity on processing (van den Brink et al., 2012). If listeners can adapt within the time course of an experiment, they should be able to do so over the life span as well, e.g., through regular contact with individuals whose voices are gender ambiguous. This contact could help listeners “unlearn” the categorical binary perception and processing of voices. This might apply to listeners who are in regular contact with gender diverse speakers, such as trans or nonbinary individuals. Compared to cisgender listeners, gender diverse listeners have been shown to differ in their rating behavior of gender ambiguous voices (Hope & Lilley, 2020, 2022). It is unclear whether this behavior would be reflected on an early and perhaps less conscious level of neural processing. Further, these previous behavioral findings have been based on listeners’ own gender diversity, rather than their contact with gender diverse speakers (and thereby possibly gender ambiguous voices). This poses the question whether listeners’ contact with gender diverse speakers affects their processing of gender ambiguous voices.

Aims

In my PhD project, I want to address the question, whether the categorical binary processing of voice gender can be unlearned. More precisely, I want to examine possible differences in the neural processing of gender ambiguous voices between listeners who are or are not experienced with listening to gender ambiguous voices in their day-to-day life. This applies most likely especially to subjects who are in regular contact with gender diverse speakers.

Methods

The project will combine both behavioral and neuroimaging data. First, behavioral speech perception data will be collected in an online listening experiment, in which stimuli of gender ambiguous voices are validated and behavioral results on the interplay of listeners’ perception of gender ambiguous stimuli and their contact with gender diverse speakers are evaluated. Then, the underlying neural processes of possible behavioral differences will be addressed in a subsequent EEG study addressing the question of the neural processing of gender ambiguous voices.

Prior Work

This work builds on the project of Paula Rinke in the first cohort, who demonstrated in her work that speaker characteristics such as gender or age are represented early on in language processing (Rinke et al., 2023). I can also build on my master's thesis, in which I conducted an acoustic analysis of a group of nonbinary and cisgender speakers. Here, I was already able to show that a purely binary construction of speaker gender does not sufficiently explain group differences at the acoustic level (Nieber et al., under review).

Links to other Projects

There are methodological links to projects A3, B1, and B3, which also use neurophysiological methods to investigate linguistic representations.

References

Charest, I., Pernet, C., Latinus, M., Crabbe, F., & Belin, P. (2013). Cerebral processing of voice gender studied using a continuous carryover FMRI design. Cerebral Cortex, 23(4), 958–966. https://doi.org/10.1093/cercor/bhs090.

Doyle, K., Harel D., Feeny, G. T., Novak, V. D., & McAllister, T. (2024). Word and Gender Identification in the Speech of Transgender Individuals. Journal of Voice: Official Journal of the Voice Foundation. https://doi.org/10.1016/j.jvoice.2024.06.007.

Hope, M., & Lilley, J. (2020). Cues for Perception of Gender in Synthetic Voices and the Role of Identity. In Interspeech 2020 (pp. 4143–4147). ISCA. https://doi.org/10.21437/Interspeech.2020-2657.

Hope, M., & Lilley, J. (2022). Gender expansive listeners utilize a non-binary, multidimensional conception of gender to inform voice gender perception. Brain and Language, 224, 105049. https://doi.org/10.1016/j.bandl.2021.105049.

Nieber, L., Kachel, S., & Scharinger, M. (under review). Acoustic Representations of Different Dimensions of the Gendered Self. An Analysis of Nonbinary and Cisgender Speakers. 

Rinke, P., Lavan, N., & Scharinger, M. (2023). Interactions of speech and speaker processing in early evoked components. Proceedings of the 20th International Congress of Phonetic Sciences, Prague 2023, 3997–4001.

Sokhi, D. S., Hunter, M. D., Wilkinson, I. D., & Woodruff, P. W. R. (2005). Male and female voices activate distinct regions in the male brain. NeuroImage, 27(3), 572–578. https://doi.org/10.1016/j.neuroimage.2005.04.023.

van den Brink, D., van Berkum, J. J. A., Bastiaansen, M. C. M., Tesink, Cathelijne M. J. Y., Kos, M., Buitelaar, J. K., & Hagoort, P. (2012). Empathy matters: Erp evidence for inter-individual differences in social language processing. Social Cognitive and Affective Neuroscience, 7(2), 173–183. https://doi.org/10.1093/scan/nsq094.