PROJECTS
PROJECTS
A. PHONOLOGICAL DEVELOPMENT
Summary: In this first line of research, we revealed that processing of spectro-temporal modulations conveying native phonetic information is functional at birth and at 6 months (Bertoncini, Nazzi, Cabrera & Lorenzi, 2011; Cabrera & Gervain, 2020; Cabrera, Bertoncini & Lorenzi, 2013; Cabrera Lorenzi & Bertoncini, 2015), thus, even when infants are not fully “tuned” to the native language. However, results also suggest that young infants may require faster amplitude modulations to “easily” discriminate syllables (Cabrera & Werner, 2017). We also showed that the perception of spectro-temporal modulations of speech differs between 6 and 10 months of age as a function of language exposure (Mandarin versus French; Cabrera, Tsao, Gnansia, Bertoncini & Lorenzi, 2014; Cabrera, Tsao, Liu, Li, Hu, Lorenzi & Bertoncini, 2015). Thus, language-specific processing of phonetic contrasts after 6 months of age seems to closely relate to changes in the reliance upon spectro-temporal modulations of speech.
Related funded projects:
2018 - The interaction of speech and auditory processing during development: the perceptual weight of acoustic information in speech perception
Funding: 34 740€ from the Fyssen Foundation. Research Assistant: Monica Hegde
Aim: This research project investigates the interaction between the auditory processing of the acoustic components of speech sounds and phonological development during infancy. Are the acoustic cues of the speech signal similarly used and organized -weighed- from birth to adulthood when perceiving speech sounds? During the first year of life, infants become better at discriminating speech contrasts of their native language but they do not show the same improvement for non-native speech contrasts. The present research hypothesizes that auditory exposure to a specific language input favors the processing and weight of certain acoustic cues.
Methods: This project takes advantage of recent auditory models of speech processing describing the acoustic components of the speech signal. This new approach is originally combined with electroencephalography to measure the neural mechanisms underlying the perception of these speech acoustic components during the early stages of human language development.
B. SPEECH IN NOISE PERCEPTION
Summary: In a second line of research, we explored how young listeners use temporal modulations when speech is presented in challenging listening conditions. We showed that in noise, adults depend more on frequency modulations than 3-month-old infants (Cabrera & Werner, 2017). Thus, the relative importance of temporal modulations for speech perception may change over development, influencing efficient speech tracking. Moreover, during childhood, we observed that children are able to correctly extract temporal modulations, in the same way as adults, and that, this ability correlates with speech in noise abilities (Cabrera, Varnet, Buss, Rosen & Lorenzi, 2019). However, children are overall less efficient in detecting amplitude modulations until 11 years of age in a behavioral task. Computational auditory models helped identifying the stages of processing that may develop during childhood (Cabrera, Lorenzini, Varnet, Rosen & Lorenzi, 2022). This line of work is currently completed by Charlotte Benoit (ENT and doctoral candidate). Moreover, Clémence Basire (doctoral candidate) is now using brain imaging tools to assess the development of the neural stages underlying modulation processing.
Related funded projects:
2022- BabySIN: Perception of speech in noise in infancy, the role of auditory development.
Funding: 301 215€ from the National French Research Agency (ANR, PRC). Doctoral student: Clémence Basire.
Aim: BabySIN seeks predictors of speech perception in noise by focusing on the development of auditory processing at early ages. First, this proposal aims to evaluate the processing of specific acoustic cues for infants with normal hearing, and whether its maturation contributes to the ability to perceive speech in noisy background. Second, this proposal explores how congenital sensorineural hearing loss and rehabilitation (cochlear implants) influence auditory and speech processing at early ages.
Methods: This project aims to tease apart the role of sensory mechanisms (i.e., auditory modulation filters) vs processing inefficiencies (i.e., neural noise) in the early perception of speech- in-noise by combining i) electrophysiological measures of modulation processing to objectively evaluate the perceptual temporal limit of the auditory system, ii) behavioral measures of phoneme detection in noise, while using a longitudinal design.
~~~~~~~~~~~~~~
2017-2022 - DESIN: DEvelopment of Speech perception In Noise: Role of sensory factors and central inefficiencies for normal-hearing and hearing-impaired children
Funding: 213 613€ from the National French Research Agency (ANR, grant for young researcher, No ANR-17-CE28-008). Post-doctoral researcher: Irene Lorenzini.
Aim: This project seeks predictors of poor speech perception in noise in childhood by focusing on the development of auditory processing. More specifically, the DESIN project aims to characterize the role of sensory (i.e., development of auditory filters) and non-sensory mechanisms (i.e, internal noise) involved in the development of speech perception in noise for normal-hearing and hearing-impaired children.
Methods: Using a combination of psychophysical and psycholinguistic paradigms, we look for the dominant factors limiting speech understanding in noise between 6 and 11 years of age for children with normal hearing and children with hearing loss. In combination with ideal-observer analysis and computational modelling, our set of behavioral tasks aim to tease apart the role of sensory factors such as auditory temporal resolution from non-sensory factors such as internal noise, sampling efficiency, or limited echoic-memory.
~~~~~~~~~~~~~~
2015-2017: SHAPED: SpeecH and Auditory Processing: Effects of Development
Funding: 183 455€ from European Union’s Horizon 2020 research and innovation programme under the grant agreement No 659204.
Aim: The overall questions addressed during the fellowship were: (i) Does the experience with a language influence auditory processing? (ii) How does the interaction between audition and speech perception develop over age? (iii) How do hearing impairments affect this relationship?
Methods: behavioral psychophysics, electroencephalography
C. EFFECTS OF CONGENITAL HEARING LOSS
Summary: Finally, we showed that congenital hearing impairments affect temporal modulation processing in childhood (Cabrera & Halliday, 2020), and that, children using cochlear implants do not exhibit similar difficulties when perceiving native speech contrasts compared to their normally hearing peers (Cabrera, Liu, Granjon, Kao & Tsao, 2019). Our results suggest that children with hearing loss may learn to rely differently on the spectro-temporal modulations of speech and may develop different perceptual strategies than children with normal hearing.
Related funded projects:
2018-2022 Diderot: Auditory development of modulation perception in cochlear-implant children : Sensory and cognitive predictors of the success of pediatric cochlear implantation
Funding: Advance Bionics with Ecole Normale Supérieure Paris (Pr Christian Lorenzi) & Hopital Robert Debré (Pr Thierry van den Abbeele)
Aim: The present research program focuses on the in-depth characterization of the auditory development of temporal- modulation perception for prelingually deafened children wearing a cochlear implant.
Methods: Several behavioral tasks assessing auditory sensitivity to amplitude-modulation (AM) cues and speech intelligibility in noise for prelingually deafened children aged between 5 and 12 years and showing various amounts of cochlear-implant experience.