Les visages fournissent à l'observateur un ensemble d'informations physiques, émotionnelles et sociales qui déterminent la manière dont les gens interagissent entre eux. Grâce aux cette informations, un humain peut se faire rapidement une première impression. La capacité de former des jugements de nature sociale est au centre de ce travail de thèse ainsi qu'à la manière dont la fiabilité d'autrui peut-être détectée spontanément à partir d'un visage. J'ai employé des techniques de suivi du mouvement oculaire, d'électrophysiologie (EEG) et comportementales. Le but de l'étude 1 visait à déterminer si la capacité d'évaluer la confiance est universelle. J'ai teste si les singes peuvent montrer une préférence spontanée envers des visages humains inspirant confiance, comme il l'a été observé chez les humains. Chez les deux espèces le temps de regard étais supérieur pour les visages inspirant confiance par rapport à ceux n'inspirant pas confiance. Un autre ensemble d'études s'intéressait au syndrome de Williams-Beuren (WS). La pathologie dont une des caractéristiques est un comportement d'appétence sociale a été utilisée comme modèle neurobiologique humain afin d'étudier la capacité à détecter les informations sociales du visage. Les patients WS sont-ils capables de détecter la confiance à partir d'un visage? Comment les patients WS se représentent un visage qui inspire confiance? J'ai observé que les patients WS regardent moins longtemps les visages qui inspirent confiance, suggérant qu'ils ont une tendance à davantage faire confiance à tout le monde. Nos résultats démontrent aussi qu'en comparaison à un groupe sain, ils ne présentent pas une image stéréotypique d'un visage qui inspire confiance. Dans une dernière étude, j'ai cherché à savoir si les sources neuronales éléctrophysiologiques, en particulier dans les régions du sulcus temporal supérieur (240ms), pouvaient expliquer leur comportement. J'ai observé que l'activité de la source était modulée de manière significative par rapport à la proximité des yeux, comme dans le groupe control. Les résultats suggèrent la présence d'une voie rapide dans le cerveau qui joue le rôle fondamental de moduler les comportements d'approche et d'évitement et que cette voie peut être altérée chez des patients caractérisés par un comportement d'appétence sociale / Faces provide a complex set of physical, emotional and social information to the observer that determines how people will interact with others. From facial information, human subjects can form rapid, first impression judgments. The ability to create social judgments from faces is the core topic of this work. This thesis will focus on how social information and trust is spontaneously detected from faces. In my studies I used eye tracking procedure, electrophysiology (EEG) and behavioral measures. In a first experiment, I investigated the evolutionary origin of trustworthiness detection testing whether monkeys (Macaca Mulatta and Fascicularis) have a spontaneous preference towards trustworthy human faces, thus suggesting a capacity to detect facial cues similar to those used by humans. Using a preference visual paradigm we observed that both species spent more time looking at trustworthy faces than untrustworthy ones. I further conducted three studies with patients affected by Williams-Beuren syndrome (WS). This pathology can be considered a neurobiological human model for the overexpressed social behavior. Are Williams-syndrome patients able to detect trustworthiness from faces? How WS patients form the representation of trustable faces ? Using a preference visual paradigm I observed that WS patients looked less the trustworthy faces compared to control group. This implicit behavior supports patients’ tendency to trust everybody. In a second experiment using reverse correlation paradigm - the procedure pushes subjects to select from noise the facial features that they believe are important for a specific judgment – I found that at group level patients did not show a stereotypical image of trustworthy faces compared to healthy controls. In a final study I investigate whether electrophysiological brain sources, with particular attention to the source localized in the superior temporal sulcus, could explain patients’ behaviour. I found that the activity of a source localized in the STS at 240ms was significantly modulated by eye proximity as in the control group. Overall the results of this work suggests the presence of a fast route in the brain that plays the fundamental role of modulating approach/avoidance behavior. This route may be altered in patients characterized by an overexpressed social behavior

La capacité à évaluer les résultats nos actions est fondamentale pour adapter et optimiser notre comportement et dépend d’un système superviseur chargé d’évaluer l’action, détecter les erreurs, déclencher des corrections.Le réseau neuronal sous-jacent la supervision de l’action n’a pas été complètement caractérisé chez l’homme.Dans une première étude nous avons enregistré dans l’Aire Motrice Supplémentaire (AMS) des LFP évoqués par les réponses et modulés par la performance. Des LFP évoqués exclusivement par les erreurs ont été enregistrés plus tardivement dans le cortex préfrontal médian.Dans la deuxième étude, nous avons observé que les activités de hautes-fréquences gamma sont, elles aussi, modulées par la performance des sujets, mais dans un vaste réseau frontal et extra-frontal.Dans une troisième étude, utilisant des enregistrements simultanés électroencéphalographiques (EEG) et magnétoencéphalographiques (MEG), nous observé une activité évoquée par un feedback interne sur l’EEG (mais pas en MEG), alors qu'une activité évoquée par le feedback externe était bien visible sur les enregistrements MEG, indiquant que les générateurs de ces deux activités cérébrales, sont différents. Nos résultats montrent une implication de l’AMSp dans la supervision de l’action chez l’homme, bien plus importante que ce que l’on soupçonnait auparavant. L’AMS évalue précocement, et de façon continue, l’action en cours et elle engage vraisemblablement des structures préfrontales en cas d’erreur seulement. Le traitement de l’erreur d’action, selon qu'il se fonde sur des informations internes ou externes est certainement sous-tendu par des réseaux corticaux différents. / The capacity to evaluate the outcome of our actions is fundamental for adapting and optimizing behaviour. This capability depends on an action monitoring system in charge of assessing ongoing actions, detecting errors, and evaluating outcomes.Electrical brain activity evoked by negative outcomes is thought to originate within the medial part of the frontal cortex. Nonetheless, the underlying neuronal network is incompletely characterised in humans.In the two first studies, we investigated the anatomical substrates of action monitoring in humans using intracerebral local field potential (LFP) recordings of cerebral cortex from epileptic patients. Response evoked LFPs sensitive to outcome were recorded from the Supplementary Motor Area proper (SMA), while LFPs evoked exclusively by errors were recorded later in the medial prefrontal cortex. High-gamma-frequency activity (60-180 Hz) was modulated as a function of action outcome in a vast frontal and extra-frontal network.In a third study using simultaneous recording of electroencephalography (EEG) and magnetoencephalography (MEG), we found that error related activity was detected by EEG (but not by MEG), while feedback-related activity was detected by MEG, indicating that the sources of these two forms of outcome-modulated brain activity are different.To conclude the SMA is much more involved in action monitoring than previously thought. SMA rapidly and continuously assesses ongoing actions and likely engages more rostral prefrontal structures in the case of error. Processing of action errors and of negative externally delivered feedback therefore appears to be supported by distinct cortical networks.

Cette thèse s'est intéressée à l'impact des interactions sociales sur l'acitivité motrice, à la fois au niveau comportempentale et cérébral et de caractériser cette modulation au cours d'actions simultanées et d'actions conjointes. Dans un premier temps, nous avons étudié les effets de l'observation d'une activation sur l'exécution motrice. En couplant analyse cinématique et enregistrement EEG, nous avons mis en évidence que l'activation du système des neurones miroirs facilitait l'action seulement lorsque les événements d'exécution et d'observation étaient temporellement coordonnées. Dans un deuxième temps, nous avons cherché à meiux caractériser l'influence du contexte interactif sur l'activité cérébrale de deux participants d'une interaction grâce à une étude en double-EEG et cinématique. Grâce à la comparaison des activités cérébrales motrices intra- et inter-individuelles chez deux participants qui effectuaient ou observaient des actions isolées ou des actions conjointes, nous avons pu mettre en évidence des modulations de l'activité motrice similaires chez les deux particiapnts ainsi que certaines modulations spécifiques à la fois du contexte social (humain vs robot) et du rôle dans l'interaction. L'analyse inter-individuelle a par ailleurs mis en évidence un couplage entre les aires visuelles et motrices des deux particiapnts pendant la condition d'action conjointe. L'ensemble des résultats suggère que le système moteur permet un couplage fin entre les participants d'une interaction, à la fois au travers d'un couplage temporel qui pourrait faciliter l'interaction et également au travers d'une boucle d'action-perception formant une représentation commune de l'action conjointe. / This research investigated the impact of social interaction on motor activity, both ona behavioural and cerebral level. More precisely, the aim of the thesis was to characterise this modulation during simultaneous actions, when two participants interact, jointly or not, and to determine its functions. First , we studied the effect of action observation on motor execution. Coupling kinematics analysis and EEG recording, we gave evidence that the temporal time course of motor system activation was crucial for motor interference. We found that motor performances were facilitated by the observation of an action, only when execution and observation events were temporally coordinated. Second, we sought to characterise the influence of an interactive context on brain activity of two interacting participants through the coupling of dual-EEG and kinematic recordings. Thanks to the comparison of intra- and inter-individual brain activity of two participants that observed or performed individual or joint actions, we were able to highlight modulations of cerebral motor activity common in both participants as well as modulations of brain activity specific to the social context (human vs robot) and the role in the interaction (actor vs observer). Moreover, the inter-individual analysis revealed a coupling between the two participants' visual and motor brain areas during the joint action condition. Taken toghther, these results suggest that the motor system allows a tight coupling between interacting participants, mediated by temporal coupling that could facilitate the interaction and through an action-perception loop forming a common motor representation of the joint action.

Cette thèse explore l'analyse de l'activité électrique du cerveau. Un défi important de ces signaux est leur grande variabilité à travers différents essais et/ou différents sujets. Nous proposons une nouvelle méthode appelée "adaptive waveform learning" (AWL). Cette méthode est suffisamment générale pour permettre la prise en compte de la variabilité empiriquement rencontrée dans les signaux neuroélectriques, mais peut être spécialisée afin de prévenir l'overfitting du bruit. La première partie de ce travail donne une introduction sur l'électrophysiologie du cerveau, présente les modalités d'enregistrement fréquemment utilisées et décrit l'état de l'art du traitement de signal neuroélectrique. La principale contribution de cette thèse consiste en 3 chapitres introduisant et évaluant la méthode AWL. Nous proposons d'abord un modèle de décomposition de signal général qui inclut explicitement différentes formes de variabilité entre les composantes de signal. Ce modèle est ensuite spécialisé pour deux applications concrètes: le traitement d'une série d'essais expérimentaux segmentés et l'apprentissage de structures répétées dans un seul signal. Deux algorithmes sont développés pour résoudre ces problèmes de décomposition. Leur implémentation efficace basée sur des techniques de minimisation alternée et de codage parcimonieux permet le traitement de grands jeux de données.Les algorithmes proposés sont évalués sur des données synthétiques et réelles contenant des pointes épileptiformes. Leurs performances sont comparées à celles de la PCA, l'ICA, et du template-matching pour la détection de pointe. / This thesis investigates the analysis of brain electrical activity. An important challenge is the presence of large variability in neuroelectrical recordings, both across different subjects and within a single subject, for example, across experimental trials. We propose a new method called adaptive waveform learning (AWL). It is general enough to include all types of relevant variability empirically found in neuroelectric recordings, but can be specialized for different concrete settings to prevent from overfitting irrelevant structures in the data. The first part of this work gives an introduction into the electrophysiology of the brain, presents frequently used recording modalities, and describes state-of-the-art methods for neuroelectrical signal processing. The main contribution of this thesis consists in three chapters introducing and evaluating the AWL method. We first provide a general signal decomposition model that explicitly includes different forms of variability across signal components. This model is then specialized for two concrete applications: processing a set of segmented experimental trials and learning repeating structures across a single recorded signal. Two algorithms are developed to solve these models. Their efficient implementation based on alternate minimization and sparse coding techniques allows the processing of large datasets. The proposed algorithms are evaluated on both synthetic data and real data containing epileptiform spikes. Their performances are compared to those of PCA, ICA, and template matching for spike detection.

Em Biomedicina, a detecção e a quanticação de anormalidades presentes num sinal são desejáveis. Uma estratégia de codicação baseada em extração de características, tais como picos ou frequências, pode não capturar todas as irregularidades. Assim, uma representação baseada em funções de base denidas com conhecimento a priori do sinal pode ser mais precisa para aplicações biomédicas. A escolha das funções base depende da natureza siológica do sinal e de suas peculiaridades. Sinais de eletrocardiograma (ECG) e eletroencefalograma (EEG) exibem características bem denidas. ECG, por exemplo, é um sinal elétrico composto de uma forma de onda especíca (P, QRS e T). Se as características de um sinal a ser sintetizado são bem compreendidas, é possível derivar uma assinatura para o sinal. Uma codicação apropriada permite a extração de parâmetros relevantes para sua análise, tais como anormalidades num ciclo cardíaco representadas por uma alteração no sinal de ECG, ou então uma excitação das ondas cerebrais representada por uma modicação no sinal de EEG. O objetivo deste projeto é introduzir uma nova técnica de codicação de sinais, que representa um sinal pela soma de funções sigmoides para aproximar iterativamente o sinal medido, com foco em aplicações biomédicas. Funções sigmoides tendem a reproduzir bem as grandes variações presentes em sinais biomédicos, daí a escolha de usá-las na codicação deste tipo de sinal. Serão explorados o nível de compressão dos dados, bem como a taxa de convergência. A técnica desenvolvida será comparada com técnicas convencionais de codicação e sua robustez será avaliada. Uma estratégia de codicação ótima pode trazer benefícios não só para a compressão, mas também na criação de assinaturas de sinais representando tanto condições siológicas normais como patológicas. / In Biomedicine, detection and quantication of abnormalities present in a signal are desired. An encoding strategy based on feature extraction, such as peaks or frequencies, may not capture all irregularities. Thus, a function-based representation, constructed using a priori knowledge of signal characteristics, may be more accurate for biomedical applications. The choice of the basis function depends on the physiological nature of the signal and its specic features. Electrocardiogram (ECG) and electroencephalogram (EEG) signals exhibit well-dened characteristics. ECG, for instance, is an electrical signal composed of specic waveform (P, QRS, and T). If the characteristics of a signal to be synthesized are well understood, its possible to derive a signal signature. An appropriate encoding allows the extraction of parameters relevant for its analysis, such as, abnormalities in a cardiac cycle represented by an alteration in the ECG signal, or an excitation of the brain waves represented by a modication of the EEG. The objective of this project is to introduce a novel signal encoding technique that represents a signal by a sum of sigmoidal functions to iteratively approximate the measured signal, targeted at biomedical applications. Sigmoidal functions tend to reproduce well large variations in biomedical signals, hence their use for coding this type of signal. We explore the data compression level as well as the convergence rate. We also compare it to conventional encoding techniques and assess the robustness of this model. An optimal encoding strategy may bring not only benets in compression, but also in the creation of signatures for signals representing both physiological and pathological conditions.

No description available.

Syftet med uppsatsen är att ge en uppdaterad översikt av området BCI (Brain Computer Interface) och undersöka vad som hänt sedan begreppet introducerades i forskningssammanhang; vilka praktiska resultat forskningen lett till och vilka tillämpningar som tillkommit. Metoden som företrädesvis används är litteraturstudie som tecknar bakgrund och enkät. Därefter följer en diskussion där utmaningar för framtiden, potential och tillämpningar i BCI-tekniken behandlas utifrån ett neuropsykologiskt perspektiv. Kommer BCI-tekniken att implementeras på samma sätt som radio, TV och telekommunikationer i samhället och vilka etiska och tekniska problem finns idag. För att skildra allmänhetens uppfattning om BCI genomfördes en webbaserad enkätundersökning (survey) i form av pilotstudie (n=32) som syftar till att ge en indikation på attityder och hur allmänhetens opinion med avseende på tillämpningar i samtiden och jämförelser med avseende på teknisk bakgrund.

Background. In everyday life it is often required to integrate multisensory input to successfully conduct response inhibition (RI) and thus major executive control processes. Both RI and multisensory processes have been suggested to be altered in autism spectrum disorder (ASD). It is, however, unclear which neurophysiological processes relate to changes in RI in ASD and in how far these processes are affected by possible multisensory integration deficits in ASD. Method. Combining high-density EEG recordings with source localization analyses, we examined a group of adolescent ASD patients (n = 20) and healthy controls (n = 20) using a novel RI task. Results. Compared to controls, RI processes are generally compromised in adolescent ASD. This aggravation of RI processes is modulated by the content of multisensory information. The neurophysiological data suggest that deficits in ASD emerge in attentional selection and resource allocation processes related to occipito-parietal and middle frontal regions. Most importantly, conflict monitoring subprocesses during RI were specifically modulated by content of multisensory information in the superior frontal gyrus. Conclusions. RI processes are overstrained in adolescent ASD, especially when conflicting multisensory information has to be integrated to perform RI. It seems that the content of multisensory input is important to consider in ASD and its effects on cognitive control processes.

Presented master's thesis deals with detection of brain wakefulness from scalp EEG data with higher order statistics. Part of the thesis is a description of electroencephalography, from the method of signal generation, sensing, electroencephraphy, EEG signal artifacts, frequency bands of EEG signal to its possible processing. Furthermore, the concept of mental fatigue and the possibility of its detection in the EEG signal is described. Subsequently, the principles of higher statistical methods of PCA and ICA and the specific possibilities of decomposition of EEG signal are described using these methods, from which the method of group spatial-frequency ICA was chosen as a suitable method for selection of partial oscillatory sources in EEG signal. In the next part there is described a method of acquisition of data, a the suggestion of solution with selected method and a description of the implemented algorithm, that was applied to real 256-lead scalp EEG data captured during a block task focused on subject allertnes. The absolute and relative power of the EEG signal was decomposed. From the achieved results, we observe that the fluctuations of the spatial frequency patterns of relative power (especially for theta and alpha bands) significantly more closely correspond with the change of reaction time and the error of the subjects performing the task. These observations appear to be relatively consistent with previously published literature, and the current study shows that spatial frequency ICA is able to blindly isolate space-frequency patterns whose fluctuations are statistically significantly correlated with parameters (reaction time, error rate) directly flowing from the given task.

This thesis deals with questions of scanning electric activity of brain,¬ so-called electroencephalograph (undermentioned EEG), methods of audiovisual stimulation (undermentioned AVS) and a data-measurement processing. Theoretical part of the thesis is engaged in a theory of EEG signal creation, history and even in current methods of purchasing and processing of the EEG signal, theory of AVS and a theory of biofeedback. For measuring EEG signal with or without an application of AVS methods has been used EEG diagnostic device by Alien company. Its attributes are described in the thesis as well. For elaboration and analysis has been created a programme aplication EEG_xhroze00.fig which realizes frequency spectrum analysis using Fast Fourier Transform algorithm (FFT) and another programme aplication brain_mapping.fig for mapping activity of brain using designed algorithm.