Biology subjects

Giraud, A.-L.

Publications and source records attributed to Giraud, A.-L..

At least 19 recordsLinked to original sources

bioRxiv2026

Cortico-hippocampal dynamics of hierarchical syntactic planning in natural speech production

The human brain must rapidly construct hierarchical structures to organize complex sequential behavior, yet the neural dynamics supporting this process during natural behavior remain poorly understood. Spoken language provides a powerful model system for investigating this computation, requiring rapid transformation of conceptual intent into structured sequential output. Using rare intracranial stereo-electroencephalography (SEEG) recordings from patients producing extended spontaneous speech, we examined how syntactic planning unfolds over time using measures of constituency, dependency structure, and probabilistic syntactic categories. We identified a hierarchical planning architecture in which global sentence structure and core syntactic categories (nouns and verbs) were specified before more local planning operations. Neural representations of these categories emerged up to 1 s before articulation and persisted throughout the planning period, whereas optional modifiers, including adjectives and adverbs, were recruited only closer to speech onset. These observations support a model of hierarchical incremental planning in which abstract sentence structure precedes the incremental specification of individual sentence elements. While core syntactic categories engaged a broader fronto-temporo-parietal network than other word classes, syntactic-depth-related activity emerged in parallel across cortical regions and the hippocampus, suggesting that hippocampal relational representations contribute to sentence structure building. Together, these findings support a cortico-hippocampal model of speech production in which hierarchical sentence structure and core syntactic categories are planned before secondary syntactic elements are incrementally incorporated into the evolving sentence plan. These results provide a neural account of how abstract linguistic structure is transformed into fluent speech.

neuroscience↗

bioRxiv2026

Contextual Prediction Tunes the Tempo of Speech Segmentation

Speech comprehension draws on both temporal structure and contextual prediction, yet how these mechanisms coordinate is poorly understood. Time-compressed speech provides a controlled probe: by degrading temporal structure, it reveals the architecture of ordinary speech comprehension. Using 3x compression with silence insertion, we varied delivery rate, temporal regularity, and boundary alignment (syllabic vs. time-defined) across two behavioural experiments. Comprehension peaked near the upper theta boundary and declined at slower and faster rates. Temporal regularity helped only when boundaries coincided with syllabic onsets, while periodic pacing alone was insufficient. Contextual predictability (word-level entropy) facilitated comprehension when temporal cues were least effective, but only under syllabic segmentation. Computational modeling confirmed that {beta}-mediated contextual prediction selectively benefited syllabic-aligned conditions, was detrimental under time-based segmentation, and better reproduced human pattern overall. Together, these results suggest that contextual prediction is continuously active but behaviorally visible only when temporal scaffolding is insufficient and syllabic structure is preserved.

neuroscience↗

bioRxiv2025

Coupled rhythms in early auditory cortex mirror speech acoustics

Theta and gamma neural dynamics dominate the human auditory cortex during speech perception and have been proposed to track syllable boundaries and encode phonemic information, respectively. To what extent these rhythms are intrinsically generated or imposed by speech acoustics remains unsolved. Applying analytic methods from neuroscience to speech audio corpora from 17 languages, we found that canonical brain features --theta, gamma, and their phase-amplitude coupling-- are a robust and specific acoustic signature of speech envelope across languages. They represent syllabic rate (2-6 Hz), vocalic features (30-50 Hz), and fundamental frequency (100-150 Hz). Intracerebral (sEEG) recordings from the auditory cortex of 18 epilepsy patients revealed that theta-gamma dynamics and their coupling are absent at rest. They emerge during speech perception and are linearly driven by the acoustic envelope, consistent with an evoked origin. Nevertheless, these responses originate from distinct yet functionally interconnected neural populations, indicating that the early auditory cortex demultiplexes speech timescales. Thus, early auditory cortex mirrors theta-gamma speech rhythms across segregated neural populations. IN BRIEFEarly auditory cortex demultiplexes speech: stimulus-locked gamma activity aligns theta phase with the nested acoustic structure, functionally organising speech timescales

neuroscience↗

bioRxiv2025

Light propofol anaesthesia for non-invasive auditory EEG recording in unrestrained non-human primates.

Non-invasive electroencephalographic (EEG) experiments have been instrumental in advancing our understanding of the brain mechanisms involved in the production and perception of sounds and human speech. Performing similar experiments in non-human primates (NHPs) would help further deepen our knowledge by allowing us to investigate the evolutionary roots of these processes. However, performing EEG on NHPs is a challenge, given its sensitivity to motion artefacts, device cost and durability, and animal training requirements. For these reasons, most attempts have used invasive intracranial recordings, which led us to develop an alternative that minimises stress and prioritises animal welfare. By using mild propofol sedation, neurophysiological experimentation can easily be integrated into the routine sanitary checks of captive animals and allows the optimisation of both EEG quality and animal welfare. To assess the influence of propofol on brain activity in NHPs, we sedated three olive baboons (Papio anubis), scored their sleep stages under different doses, and recorded auditory event-related potentials (ERP)in response to grunts. Analyses of the EEG recordings with regards to sleep stage and ERP components indicate that at low dose (< 0.1mg/kg/h), propofol induces a light sleep state conducive to recording stimulus-elicited auditory activity. Overall, this experiment confirms the use of propofol sedation as an appropriate technique to study auditory processes through unrestrained, non-invasive EEG in NHPs.

neuroscience↗

bioRxiv2024

From spiking neuronal networks to interpretable dynamics: a diffusion-approximation framework

Modeling and interpreting the complex recurrent dynamics of neuronal spiking activity is essential to understanding how networks implement behavior and cognition. Nonlinear Hawkes process models can capture a large range of spiking dynamics, but remain difficult to interpret, due to their discontinuous and stochastic nature. To address this challenge, we introduce a novel framework based on a piecewise deterministic Markov process representation of the nonlinear Hawkes process (NH-PDMP) followed by a diffusion approximation. We analytically derive stability conditions and dynamical properties of the obtained diffusion processes for single-neuron and network models. We established the accuracy of the diffusion approximation framework by comparing it with exact continuous-time simulations of the original neuronal NH-PDMP models. Our framework offers an analytical and geometric account of the neuronal dynamics repertoire captured by nonlinear Hawkes process models, both for the canonical responses of single-neurons and neuronal-network dynamics, such as winner-take-all and traveling wave phenomena. Applied to human and nonhuman primate recordings of neuronal spiking activity during speech processing and motor tasks, respectively, our approach revealed that task features can be retrieved from the dynamical landscape of the fitted models. The combination of NH-PDMP representations and diffusion approximations thus provides a novel dynamical analysis framework to reveal single-neuron and neuronal-population dynamics directly from models fitted to spiking data.

neuroscience↗

bioRxiv2024

Animal acoustic communication maintains a universal optimum rhythm

Most animals interact with conspecifics through acoustic signals that are modulated in frequency and rhythm. While small animals vocalize at higher pitch than large ones due to the smaller size of their vocal apparatus, the rules governing vocalization rhythms throughout the animal kingdom remain unknown. Vocal rhythms serve as a natural information parser, and one possibility is that they are constrained by the neural rhythms of transmitter and receiver, known to be relatively conserved across species and independent of their size. In this study, we quantified acoustic rhythms across taxa and investigated their evolutionary history with regard to phylogeny and selective pressure. In 98 species from six classes, we tested the main factors likely to influence their communication rhythms: morphology, physiology, social complexity, mastication and detectability. Phylogenetic modeling did not confirm the influence of these species-specific factors, but rather point to a scenario where acoustic communication rhythms have been maintained around an optimum at around 3Hz in the biological (neuronal) delta range (1-4Hz) well before the mammals split. These results suggest that the rhythm of acoustic communication signals, unlike their pitch, has a universal neural determinant that has been conserved throughout evolution, allowing for intra- and cross-species signaling.

animal behavior and cognition↗

bioRxiv2024

Consistency of resting-state correlations between fMRI networks and EEG band-power

Several simultaneous EEG-fMRI studies have aimed to identify the relationship between EEG band power and fMRI resting state networks (RSNs) to elucidate their neurobiological significance. Although common patterns have emerged, inconsistent results have also been reported. This study examines the consistency of these correlations across subjects and to understand how factors such as the hemodynamic response delay and the use of different EEG data spaces (source/scalp) influence them. Using three distinct EEG-fMRI datasets, acquired independently on 1.5T, 3T and 7T MRI scanners (comprising 42 subjects in total), we evaluate the generalizability of our findings across different acquisition conditions. We found consistent correlations between fMRI RSN and EEG band-power time-series across subjects in the three datasets studied, with systematic variations with RSN, EEG frequency-band, and HRF delay, but not with EEG space. Qualitatively, the majority of these correlations were similar across the three datasets, despite important differences in field strength, number of subjects and resting-state conditions. Our findings support consistent correlations across specific fMRI RSNs and EEG bands and highlight the importance of methodological considerations in interpreting them that may explain conflicting reports in existing literature.

neuroscience↗

bioRxiv2024

Atypical Audio-Visual Neural Synchrony and Speech Processing in children with Autism Spectrum Disorder

BackgroundChildren with Autism Spectrum Disorders (ASD) often exhibit communication difficulties that may stem from basic auditory temporal integration impairment but also be aggravated by an audio-visual integration deficit, resulting in a lack of interest in face-to-face communication. This study addresses whether speech processing anomalies in young (mean age 3.09-year-old) children with ASD are associated with alterations of audio-visual temporal integration. MethodsWe used high-density electroencephalography (HD-EEG) and eye tracking to record brain activity and gaze patterns in 31 children (6 females) with ASD and 33 typically developing (TD) children (11 females), while they watched cartoon videos. Neural responses to temporal audio-visual stimuli were analyzed using Temporal Response Functions model and phase analyses for audiovisual temporal coordination. ResultsThe reconstructability of speech signals from auditory responses was reduced in children with ASD compared to controls, but despite more restricted gaze patterns in ASD it was similar for visual responses in both groups. Speech reception was most strongly affected when visual speech information was also present, an interference that was not seen in TD children. These differences were associated with a broader phase angle distribution (exceeding pi/2) in the EEG theta range in autistic children, signaling reduced reliability of audio-visual temporal alignment. ConclusionThese findings show that speech processing anomalies in ASD do not stand alone and that they are associated already at a very early development stage with audio-visual imbalance with lousier auditory response encoding and disrupted audio-visual temporal coordination.

neuroscience↗

bioRxiv2024

Deciphering the Rhythmic Symphony of Speech: A Neural Framework for Robust and Time-Invariant Speech Comprehension

Unraveling the mysteries of how humans effortlessly grasp speech despite diverse environmental challenges has long intrigued researchers in systems and cognitive neuroscience. This study explores the neural intricacies underpinning robust speech comprehension, giving computational mechanistic proof for the hypothesis proposing a pivotal role for rhythmic, predictive top-down contextualization facilitated by the delta rhythm in achieving time-invariant speech processing. Our Brain-Rhythm-based Inference model, BRyBI, integrates three key rhythmic processes - theta-gamma interactions for parsing phoneme sequences, dynamic delta rhythm for inferred prosodic-phrase context, and resilient speech representations. Demonstrating mechanistic proof-of-principle, BRyBI replicates human behavioral experiments, showcasing its ability to handle pitch variations, time-warped speech, interruptions, and silences in non-comprehensible contexts. Intriguingly, the model aligns with human experiments, revealing optimal silence time scales in the theta- and delta-frequency ranges. Comparative analysis with deep neural network language models highlights distinctive performance patterns, emphasizing the unique capabilities of a rhythmic framework. In essence, our study sheds light on the neural underpinnings of speech processing, emphasizing the role of rhythmic brain mechanisms in structured temporal signal processing - an insight that challenges prevailing artificial intelligence paradigms and hints at potential advancements in compact and robust computing architectures.

neuroscience↗

bioRxiv2023

Adaptive LDA classifier enhances real-time control of an EEG Brain-computer interface for imagined-speech decoding

Brain-Computer Interfaces (BCI) aim to establish a pathway between the brain and an external device without the involvement of the motor system, relying exclusively on neural signals. Such systems have the potential to provide a means of communication for patients who have lost the ability to speak due to a neurological disorder. Traditional methodologies for decoding imagined speech directly from brain signals often deploy static classifiers, that is, decoders that are computed once at the beginning of the experiment and that remain unchanged throughout the BCI use. However, this approach might be inadequate in effectively handling the non-stationary nature of electroencephalography (EEG) signals and the learning that accompanies BCI use as parameters are expected to change, all the more in a real-time setting. To address this limitation, we have developed an adaptive classifier that updates its parameters based on the incoming data in real time. We first identified optimal parameters (the update coefficient, UC) to be used in an adaptive Linear Discriminant Analysis (LDA) classifier, using a previously recorded EEG dataset, acquired while healthy participants controlled a binary BCI based on imagined syllable decoding. We subsequently tested the effectiveness of this optimization in a real-time BCI-control setting. Twenty healthy participants performed two BCI-control sessions based on the imagery of two syllables, using a static LDA and the other the adaptive LDA classifier, in randomized order. In this real-time BCI-control task, the adaptive classifier led to better performances than the static one. Furthermore, the optimal parameters for the adaptive classifier were closely aligned in both datasets acquired with the same syllable imagery task. These findings highlight the effectiveness and re-liability of adaptive LDA classifiers for real-time imagined speech decoding, and its interest for non-invasive EEG-based BCI notably characterized by low decoding accuracies.

bioengineering↗

bioRxiv2023

Low-dimensional neuronal population dynamics in anterior superior temporal gyrus reactivate phonetic representations during semantic processing

AbstractSpeech processing involves a hierarchy of cognitive layers from low-level phonetic to high-level seman-tic representations. However, interactions are not only feed-forward; Feedback mechanisms are crucial for real-time, accurate speech recognition. Yet how distant levels interface during speech processing remains unclear. Here, we analyzed intracortical recordings from 624 neurons across three human participants implanted with microelectrode arrays in the anterior superior temporal gyrus during an auditory semantic categorization task and natural speech perception. We identified distinct neural subspaces, or manifolds, for semantic and phonetic features, with a functional separation of the corresponding low-dimensional dynamics. We contrasted a bottom-up cumulative and a top-down predictive hypothesis on phonetic-semantic temporal alignment, and found phonetic alignment to word-level semantic representations, signaling top-down prediction. These effects were consistent across participants at the spiking level, and remained undetected in adjacent ECoG recordings. These findings demonstrate the reorganization of neuronal population dynamics supporting phonetic representations during semantic prediction.

neuroscience↗

bioRxiv2023

Dogs' sensory-motor tuning shapes dog-human vocal interactions

Within species, vocal and auditory systems co-evolve to converge on a critical temporal acoustic structure that can be best produced and perceived. While dogs cannot produce articulated sounds, they respond to speech, raising the question as to whether this heterospecific receptive ability is shaped by exposure to speech or bounded by their own sensorimotor capacity. Acoustic analyses of vocalisations show that dogs main production rhythm is slower than the dominant (syllabic) speech rate, and that human dog-directed speech falls halfway in between. Comparative exploration of neural (electroencephalography) and behavioural responses to speech reveals that comprehension in dogs relies on a slower speech rhythm tracking (delta) than humans (theta), even though dogs are equally sensitive to human speech content and prosody. Thus, the dog audio-motor tuning differs from humans, who vocally adjust their speech rate to this shared temporal channel.

neuroscience↗

bioRxiv2023

Learning to operate an imagined speech Brain-Computer Interface involves the spatial and frequency tuning of neural activity

Brain-Computer Interfaces (BCI) will revolutionize the way people with impaired speech production can communicate. While recent studies confirm the possibility of decoding imagined speech based on pre-recorded intracranial neurophysiological signals, current efforts focus on collecting vast amounts of data to train classifiers, rather than exploring how the individuals brain adapts to improve BCI control, an important aspect given the known problem of "BCI illiteracy", the inability of some individuals to operate a BCI. This issue can be investigated by providing real-time feedback to allow users to identify the best control strategy. In this study, we trained 15 healthy participants to operate a simple binary BCI system based on electroencephalography (EEG) signals through syllable imagery for five consecutive days. We explored whether BCI-control improves with training and characterized the underlying neural dynamics, both in terms of EEG power changes and of the neural features contributing to real-time classification. Despite considerable interindividual variability in performance and learning, a significant improvement in BCI control was observed from day 1 to 5. Performance improvement was associated with a global EEG power increase in frontal theta and a focal increase in temporal low-gamma, showing that learning to operate an imagined-speech BCI involves global and local dynamical changes involving low- and high-frequency neural features, respectively. These findings indicate that both machine and human learning must be considered to reach optimal controllability of imagined-speech BCI, and that non-invasive BCI-learning can help predict the individual benefit from an invasive speech BCI and guide both the electrode implantation and decoding strategies.

neuroscience↗

bioRxiv2022

A deep hierarchy of predictions enables assignment of semantic roles in real-time speech comprehension

Understanding speech requires mapping fleeting and often ambiguous soundwaves to meaning. While humans are known to exploit their capacity to contextualize to facilitate this process, how internal knowledge is deployed on-line remains an open question. Here, we present a model that extracts multiple levels of information from continuous speech online. The model applies linguistic and nonlinguistic knowledge to speech processing, by periodically generating top-down predictions and incorporating bottom-up incoming evidence in a nested temporal hierarchy. We show that a nonlinguistic context level provides semantic predictions informed by sensory inputs, which are crucial for disambiguating among multiple meanings of the same word. The explicit knowledge hierarchy of the model enables a more holistic account of the neurophysiological responses to speech compared to using lexical predictions generated by a neural-network language model (GPT-2). We also show that hierarchical predictions reduce peripheral processing via minimizing uncertainty and prediction error. With this proof-of-concept model we demonstrate that the deployment of hierarchical predictions is a possible strategy for the brain to dynamically utilize structured knowledge and make sense of the speech input.

neuroscience↗

bioRxiv2022

Rhythmic modulation of prediction errors: a possible role for the beta-range in speech processing

Natural speech perception requires processing the current acoustic input while keeping in mind the preceding one and predicting the next. This complex computational problem could be handled by a multi timescale hierarchical inferential process that coordinates information flow up and down the language hierarchy. While theta and low-gamma neural frequency scales are convincingly involved in bottom-up syllable-tracking and phoneme-level speech encoding, the beta rhythm is more loosely associated with top-down processes without being assigned yet a specific computational function. Here we tested the hypothesis that the beta rhythm drives the precision of states during the speech recognition hierarchical inference process. We used a predictive coding model that recognizes syllables on-line in natural sentences, in which the precision of prediction errors is rhythmically modulated, resulting in alternating bottom-up vs. top-down processing regimes. We show that recognition performance increases with the rate of precision updates, with an optimal efficacy in the beta range (around 20 Hz). The model further performs when prediction errors pertaining respectively to syllable timing and syllable identity oscillate in antiphase. These results suggest that online syllable recognition globally benefits from the alternation of bottom-up and top-down dominant regime at beta rate, and that the gain is stronger when different features are also analyzed in alternation. These results speak to a discontinuous account of inferential operations in speech processing.

neuroscience↗

bioRxiv2022

Salient 40 Hz sounds probe affective aversion and neural excitability

The human auditory system is not equally reactive to all frequencies of the audible spectrum. Emotional and behavioral reactions to loud or aversive acoustic features can vary from one individual to another, to the point that some exhibit exaggerated or even pathological responses to certain sounds. The neural mechanisms underlying these interindividual differences remain unclear. Whether distinct aversion profiles map onto neural excitability at the individual level needs to be tested. Here, we measured behavioral and EEG responses to click trains (from 10 to 250 Hz, spanning the roughness and pitch perceptual ranges) to test the hypothesis that interindividual variability in aversion to rough sounds is reflected in neural response differences between participants. Linking subjective aversion to 40 Hz steady-state EEG responses, we demonstrate that participants experiencing enhanced aversion to roughness also show stronger neural responses to this attribute. Interestingly, this pattern also correlates with inter-individual anxiety levels, suggesting that this personality trait might interact with subjective sensitivity and neural excitability to these sounds. These results support the idea that 40 Hz sounds can probe the excitability of non-canonical auditory systems involved in exogenous salience processing and aversive responses at the individual level. By linking subjective aversion to neural excitability, 40 Hz sounds provide neuromarkers relevant to a variety of pathological conditions, such as those featuring enhanced emotional sensitivity (hyperacusis, anxiety) or aberrant neural responses at 40 Hz (autism, schizophrenia).

neuroscience↗

bioRxiv2021

Neural oscillation coupling selectively predicts speech reception in young children with Autism Spectrum Disorder

Communication difficulties in autism spectrum disorder (ASD) involve a speech reception deficit, whose biological causes are not yet identified. This deficit could denote atypical neuronal ensemble activity, as reflected by neural oscillations. Atypical cross-frequency oscillation coupling in particular could disrupt the possibility to jointly track and predict dynamic acoustic stimuli, a dual process that is essential for speech comprehension. Whether such oscillation anomalies can already be found in very young children with ASD, and with what specificity they relate to individual language reception capacity is unknown. In this study, neural activity was collected using EEG in 64 very young children with and without ASD (mean age 3) while they were exposed to naturalistic-continuous speech via an age-appropriate cartoon. EEG power typically associated with phrase-level chunking (delta, 1-3Hz), phonemic encoding (low-gamma, 25-35Hz) and top-down control (beta, 12-20Hz) was markedly reduced in ASD relative to typically developing (TD) children. Speech neural-tracking by delta and theta oscillations was also weaker in ASD than TD children. Critically, children with ASD exhibited slightly atypical theta/gamma coupling (PAC) involving a higher-than-normal gamma frequency, and markedly atypical beta/gamma PAC. Even though many oscillation features were atypical in our sample of 31 very young children with ASD, the beta/gamma coupling anomaly was the single best predictor of individual speech reception difficulties. These findings suggest that early interventions targeting the normalization of low-gamma and low-beta activity, might help young children with ASD to engage more in oral interactions. HighlightsVery young children diagnosed with autism spectrum disorder already present marked alterations of neural oscillatory activity in response to natural speech. The hierarchical processing of phonemic- and syllabic-range information (theta/gamma coupling) is atypical in young ASD children. Speech reception deficit in children with ASD is indexed by abnormal bottom-up (low-gamma) and top-down (low-beta) coordination.

neuroscience↗

bioRxiv2021

Imagined speech can be decoded from low- and cross-frequency features in perceptual space

Reconstructing intended speech from neural activity using brain-computer interfaces (BCIs) holds great promises for people with severe speech production deficits. While decoding overt speech has progressed, decoding imagined speech have met limited success, mainly because the associated neural signals are weak and variable hence difficult to decode by learning algorithms. Using three electrocorticography datasets totalizing 1444 electrodes from 13 patients who performed overt and imagined speech production tasks, and based on recent theories of speech neural processing, we extracted consistent and specific neural features usable for future BCIs, and assessed their performance to discriminate speech items in articulatory, phonetic, vocalic, and semantic representation spaces. While high-frequency activity provided the best signal for overt speech, both low- and higher-frequency power and local cross-frequency contributed to successful imagined speech decoding, in particular in phonetic and vocalic, i.e. perceptual, spaces. These findings demonstrate that low-frequency power and cross-frequency dynamics contain key information for imagined speech decoding, and that exploring perceptual spaces offers a promising avenue for future imagined speech BCIs.

neuroscience↗