Project

Back to overview

Are cortical oscillations a useful ingredient of speech processing?

Applicant Giraud Mamessier Anne-Lise
Number 149319
Funding scheme Project funding (Div. I-III)
Research institution Dépt des Neurosciences Fondamentales Faculté de Médecine Université de Genève
Institution of higher education University of Geneva - GE
Main discipline Neurophysiology and Brain Research
Start/End 01.10.2013 - 31.12.2015
Approved amount 453'261.00
Show all

Keywords (2)

cortical oscillations; speech

Lay Summary (French)

Lead
Les oscillations corticales sont-elles un ingrédient indispensable du traitement de la parole par le cerveau humain
Lay summary

La structure rythmique de la parole se reflète dans le comportement oscillatoire de l'activité des cortex auditifs et moteurs, qui la reçoivent et la produise. Dans ce projet, nous faisons l'hypothèse que les oscillations corticales interagissent avec les rythmes de parole afin d'exécuter des opérations importantes dans le traitement de la parole. Nous proposons de tester au niveau théorique et expérimental leur rôle mécanique dans le découpage de la parole, l'encodage phonémique et de contrôle du transfert d'information à-travers différents niveaux de la hiérarchie corticale auditive. En combinant neurophysiologie et modélisation computationnelle nous projetons de caractériser les opérations bottom-up (ascendantes) et top-down (descendantes) qui fonctionnent sur la base d'oscillations corticales. Nous utiliserons l'électrophysiologie invasive et des modèles inspirés par la biophysique pour évaluer si ces opérations ont un rôle fondamental. Les résultats devraient nous permettre d'initier un travail de rétro-ingénierie qui permettra d'évaluer l'intérêt d'inclure des routines basées sur des comportements oscillatoires dans les processus de reconnaissance ou synthèse automatique de la parole. Dans son ensemble ce projet devraient nous permettre d'évaluer si oui ou non les oscillations corticales sont un ingrédient nécessaire à la perception de la parole continue.

Direct link to Lay Summary Last update: 01.10.2013

Lay Summary (English)

Lead
Are cortical oscillations a useful ingredient of speech processing?
Lay summary
The acoustic structure of speech is reflected in oscillatory neuronal behaviour of both the motor system that produces it and the auditory system that receives it. In this project we hypothesize that cortical oscillations in auditory cortices serve important temporal integration functions in speech processing. We propose to experimentally and theoretically address their mechanical role in speech parsing, phonemic encoding and in the control of information transfer across auditory cortical stages. Using a combination of experimental and theoretical approaches we will characterize the bottom-up and top-down oscillation-based computations that take place during the first steps of speech decoding by the auditory cortex. We further expect to gain enough experimental and theoretical material to initiate a reverse engineering work in which we will evaluate the potential usefulness of incorporating oscillation-based neuromimetic routines in automatic speech recognition devices. Altogether, this project is expected to bring us in the position to critically assess whether periodic ensemble neural activity as cortical oscillations are a necessary ingredient of speech processing.
Direct link to Lay Summary Last update: 01.10.2013

Responsible applicant and co-applicants

Employees

Publications

Publication
Atypical coordination of cortical oscillations in response to speech in autism.
Jochaut Delphine, Lehongre Katia, Saitovitch Ana, Devauchelle Anne-Dominique, Olasagasti Itsaso, Chabane Nadia, Zilbovicius Monica, Giraud Anne-Lise (2015), Atypical coordination of cortical oscillations in response to speech in autism., in Frontiers in human neuroscience, 9, 171-171.
Human screams occupy a privileged niche in the communication soundscape.
Arnal Luc H, Flinker Adeen, Kleinschmidt Andreas, Giraud Anne-Lise, Poeppel David (2015), Human screams occupy a privileged niche in the communication soundscape., in Current biology : CB, 25(15), 2051-6.
Neural Cross-Frequency Coupling: Connecting Architectures, Mechanisms, and Functions.
Hyafil Alexandre, Giraud Anne-Lise, Fontolan Lorenzo, Gutkin Boris (2015), Neural Cross-Frequency Coupling: Connecting Architectures, Mechanisms, and Functions., in Trends in neurosciences, 38(11), 725-40.
Prediction across sensory modalities: A neurocomputational model of the McGurk effect.
Olasagasti Itsaso, Bouton Sophie, Giraud Anne-Lise (2015), Prediction across sensory modalities: A neurocomputational model of the McGurk effect., in Cortex; a journal devoted to the study of the nervous system and behavior, 68, 61-75.
Speech encoding by coupled cortical theta and gamma oscillations.
Hyafil Alexandre, Fontolan Lorenzo, Kabdebon Claire, Gutkin Boris, Giraud Anne-Lise (2015), Speech encoding by coupled cortical theta and gamma oscillations., in eLife, 4, 06213-06213.
Temporal coding in the auditory cortex.
Arnal Luc H, Poeppel David, Giraud Anne-Lise (2015), Temporal coding in the auditory cortex., in Handbook of clinical neurology, 129, 85-98.
The contribution of frequency-specific activity to hierarchical information processing in the human auditory cortex.
Fontolan L, Morillon B, Liegeois-Chauvel C, Giraud Anne-Lise (2014), The contribution of frequency-specific activity to hierarchical information processing in the human auditory cortex., in Nature communications, 5, 4694-4694.
Impaired auditory sampling in dyslexia: further evidence from combined fMRI and EEG.
Lehongre Katia, Morillon Benjamin, Giraud Anne-Lise, Ramus Franck (2013), Impaired auditory sampling in dyslexia: further evidence from combined fMRI and EEG., in Frontiers in human neuroscience, 7, 454-454.
Neuronal oscillations and speech perception: critical-band temporal envelopes are the essence.
Ghitza Oded, Giraud Anne-Lise, Poeppel David (2013), Neuronal oscillations and speech perception: critical-band temporal envelopes are the essence., in Frontiers in human neuroscience, 6, 340-340.

Collaboration

Group / person Country
Types of collaboration
Neurosurgery department HUG Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure
Neurology department HUG Switzerland (Europe)
- in-depth/constructive exchanges on approaches, methods or results
- Publication
- Research Infrastructure

Scientific events



Self-organised

Title Date Place
Predictive coding and cortical oscillations 06.02.2014 Biotech Campus - Genève, Switzerland

Communication with the public

Communication Title Media Place Year
Media relations: radio, television Emission CQFD RTS Western Switzerland 2014

Associated projects

Number Title Start Funding scheme
163040 Predictive coding and cortical oscillations in speech processing 01.01.2016 Project funding (Div. I-III)

Abstract

Background: Speech is a quasi-periodic signal whose acoustic structure reflects the oscillatory neuronal behaviour of both the motor system that produces it and the auditory system that receives it. Here we hypothesize that cortical oscillations in auditory cortices serve important temporal integration functions in speech processing. We propose to experimentally and theoretically address their mechanical role in speech parsing, phonemic encoding and in the control of information transfer across auditory cortical stages.Aims/hypotheses: In Part 1 of this project we will determine whether theta auditory cortical oscillations are able to 1) phase-lock to speech modulations and adapt to individual speech rates, 2) temporally organize gamma activity so that gamma-framed spiking can encode syllables. In Part 2 we will specify how cortical oscillations are involved in speech predictive coding. We will address whether syllable-level predictions made during speech perception involve specific frequency channels.Methods: We will record neural activity from human auditory cortices using temporal lobe electrocoticography (ECoG) in epileptic patients (8/year), and acute combined macro- (local fiels potentials) and micro- (spiking) electrode recordings in oncologic patients with a tumour in auditory cortex vicinity (8/year). This effort to gain neurophysiological information at precise spatio-temporal scales will be combined with computational modelling. We will use an existing biophysical model of coupled gamma and theta oscillators that has the property to parse and encode speech (Hyafil et al., submitted). Using this model we will explore how auditory cortex adapts to speech rates, and emulate the auditory oscillatory dysfunctions we observed in dyslexia and autism (Lehongre et al., 2011; Giraud and Ramus, 2012; Roussillon et al. submitted). By the end of this project, we should be able to expand our model to three hierarchically organized cortical stages (n, n-1 and n+1) to address the control of information transfer by gamma and beta cortical oscillations.Potential value of the project: Using a combination of experimental and theoretical approaches we expect to characterize the bottom-up and top-down oscillation-based computations that take place during the first steps of speech decoding by the auditory cortex. We further expect to gain enough experimental and theoretical data to initiate a reverse engineering work in which we will evaluate the potential usefulness of incorporating oscillation-based neuromimetic routines in automatic speech recognition devices. Altogether, this project is expected to bring us in the position to critically assess whether periodic ensemble neural activity as cortical oscillations are a necessary ingredient of speech processing.
-