Un des principaux résultats de l'étude scientifique de la conscience concerne l'existence de deux phases distinctes du traitement visuel. La première est caractérisée par une propagation antérograde de l'activité évoquée par le cortex visuel primaire et n'est pas typiquement associée à une perception consciente. La seconde, en revanche, est souvent citée comme un corrélat neuronal de la conscience, et implique une réactivation du cortex visuel précoce par le cortex parieto-frontal. Cette dichotomie soulève plusieurs questions : premièrement, quelle est l'origine de ce phénomène de feedback, et deuxièmement, qu'est-ce qui distingue un stimulus ayant subi ce traitement supplémentaire du stimulus n'ayant pas provoqué une telle réactivation ? Au jour d'aujourd'hui, deux grandes théories ont été proposées. La première, que nous appelons la théorie "précoce et locale" pose l'hypothèse que l'accès conscient émerge lorsque la boucle de réactivation sensorielle est établie. Ceci implique que seuls les stimuli présentant une qualité hautement saillante dès leur apparition peuvent accéder à un traitement conscient, et de ce fait, que le rapport subjectif d'un stimulus dépend uniquement de l'activité locale du cortex sensoriel. La théorie "tardive et globale", par contraste, propose que la perception consciente est le résultat d'un routage informationnel à travers un réseau cortico-cortical distribué appelé le Global Neuronal Workspace (GNW). Cette théorie suggère que l'information sensorielle provenant de diverses régions corticales accède à cette infrastructure de routage par le biais d'un processus sélectif : l'attention. En 2013, Sergent et collègues ont testé l'une des prédictions dérivées de cette seconde théorie : en principe, un stimulus dont l'observateur n'a pas pris conscience peut néanmoins accéder au GNW suite à un amorçage attentionnel. Afin de tester cette prédiction, des participants humains ont visionné un stimulus placé au seuil perceptif ainsi qu'une amorce attentionnelle qui pouvait soit attirer l'attention vers la position de la cible, soit du coté opposé. Ces participants discriminaient plus finement les caractéristiques de la cible au sein des essais pour lesquels l'amorce dirigeait l'attention vers la position préalable de la cible, ce qui suggère que l'intervention rétrospective de l'attention déclenche un accès conscient pour des faibles traces mnésiques qui ne seraient normalement pas traitées par le GNW. Nous présentons des données de modélisation psychophysique ainsi que des données d'imagerie fonctionnelle qui suggèrent que l'attention joue un rôle causal dans l'émergence d'un percept conscient, et qui offrent des indices quand à la structure des représentations perceptuelles au sein du cortex sensoriel primaire. / A major finding in the scientific study of conscious perception has been the existence of two temporally-distinct phases of visual processing. The first, characterized by the feed-forward propagation of evoked activity in early visual cortex, is not typically associated with conscious perception. The second phase involves a reactivation of early sensory cortex by downstream regions and is often cited as a correlate -- if not a proximal cause -- of consciousness. This raises a few crucial questions: firstly, what causes this feedback process to emerge, and secondly, what distinguishes a stimulus representation that has undergone such feedback processing from one that has not ? At the time of writing, two competing theories have been proposed. The first theory, hitherto referred to as "early-and-local", posits that conscious access emerges from the very emergence of a feedback loop between high-level sensory cortex and its primary counterpart, and that this cortical resonance is driven entirely by upstream activations along the feed-forward chain. This implies that only those stimuli that exhibit high salience from the onset can become conscious, and by extension, that the stimulus' reportability is governed entirely by early evoked activity in primary sensory cortex. "Late-and-global" theory, by contrast, posits that conscious perception is the direct result of routing of information through a distributed cortico-cortical network called the Global Neuronal Workspace (hereafter GNW). By this account, visual information in various local cortical regions is given access to routing infrastructure by some selective process, namely attention. In 2013, Sergent and colleagues tested a prediction derived from this second model: that an arbitrary sensory representation that has initially failed to become conscious can be hooked into the GNW by means of an attentional manipulation. To do this, a low threshold target Gabor patch was presented, followed by an extrinsic cue either at the location in which the Gabor had been presented, or on the opposite side of the screen. Subjects were better at discriminating the orientation of the Gabor in trials where the cue had been presented on the same side as the target, and also reported seeing the target more clearly, suggesting that the retrospective intervention of attention was enabling a weak signal to gain access to the global neuronal workspace. We present data from psychophysical modeling and functional magnetic resonance imaging that point to a causal role for attention in the emergence of a conscious percept, with implications for the structure of perceptual representations in early sensory cortex.

Visual working memory (WM) is the ability to hold information in mind for a short time before acting on it. The capacity of WM is strikingly limited. To make the most of this precious resource, humans exhibit a high degree of cognitive flexibility: We can prioritize information that is relevant to behavior, and inhibit unnecessary distractions. This thesis examines some behavioral and neural correlates of flexibility in WM. When information is of particular importance, anticipatory attention can be directed to where it will likely appear. Oscillations in visual cortex, in the 10-Hz range, play an important role in regulating excitability of such prioritized locations. Chapter 4 describes how even spontaneous fluctuations in 10-Hz synchronization (measured by electroencephalography, EEG) before encoding influence WM. Chapters 2 and 3 describe how attention can be directed retrospectively to items even if they are already stored in WM. Chapter 3 discusses how retrospective cues change neural synchronization similarly to anticipatory cues. Behavioral and neural measures additionally indicate that the boosting of an item through retrospective cues does not require prolonged deployment of attention: rather, it may be a transient process. The second half of this thesis additionally examines how items are represented in visual WM. Chapter 5 summarizes a study using pattern analysis of magnetoencephalographic (MEG) and EEG data to decode features of visual templates stored in WM. Decoding appears transiently around the time when potential target stimuli are expected, in line with a flexible reactivation mechanism. Chapter 6 further examines separate cortical networks involved in protecting vs. updating items in WM, and tests whether task relevance changes how well WM contents can be decoded. Finally, Chapter 7 summarizes the thesis and discusses how attentional flexibility can merge WM with a wider range of sources of behavioral control.

BACKGROUND: Schizophrenia is a diverse disorder comprising positive, negative, and cognitive symptoms. While several cognitive domains have been alleged to influence symptom severity and quality of life, impaired information processing speed (IPS) is the most profoundly affected and is present even before a diagnosis of a psychotic disorder. Although extensively studied, reliable and reproducible pathological neural circuitry leading to cognitive dysfunction has yet to be fully identified or understood. OBJECTIVE: Our study seeks to identify the neurobiological connections responsible for diminished processing speed in a group of patients with psychotic disorders and a group of individuals at risk to develop a psychotic disorder. METHODS: This data-driven evaluation of resting-state fMRI data and cognitive testing considers two populations: one cohort of adults diagnosed with psychotic disorders and another cohort of individuals identified as being at risk for developing psychotic disorders (ShangHai At-Risk for Psychosis). The images were processed and evaluated using multivariate pattern analysis (MVPA) to compare brain connectivity to information processing speed. The MVPA calculations generated a map to determine the regions where brain connectivity has the strongest correlation to information processing speed as measured by the Brief Assessment in Cognition for Schizophrenia (BACS). RESULTS: The results of the cognitive assessments demonstrate that information processing speed (IPS) is the most impaired domain of cognition in adults with psychotic disorders. Multivariate distance matrix regression (MDMR) was used to identify the cerebellum as the specific brain location where connectivity was most strongly correlated to information processing speed. Specifically, the cerebellar (CBM) connectivity to the dorsal medial prefrontal cortex (dmPFC) exhibited the strongest positive correlation with IPS. The cohort of individuals diagnosed with a psychotic disorder (n = 103) exhibited a moderately strong relationship (r = 0.396, p = 0.000036) between the CBM-dmPFC coactivation and IPS. This finding was then replicated in the SHARP dataset. The clinical high risk sample (n = 137) also identified the CBM-dmPFC circuit as the strongest correlation to information processing speed (r=0.39, p = 0.000002057). Furthermore, the CBM-dmPFC circuit relationship to IPS became stronger as we considered the most impaired subjects. Individuals who would go on to be diagnosed with a psychotic disorder (converters, n = 19) exhibited the strongest relationship between CBM-dmPFC and information processing speed (r = 0.56, p = 0.01). CONCLUSIONS: The reduction in functional connectivity between cerebellum and dmPFC appears to be critically related to the generalized decrease in information processing speed in patients with psychotic disorders and patients at risk of developing a psychotic disorder in the future. The finding is reproduced in two culturally, linguistically, and geographically disparate populations. Interestingly, the finding demonstrates a link between the functional dysconnectivity and cognitive dysfunction in the years before and during the onset of a diagnosis of a psychotic disorder. The research indicates a diagnostic role for assessing information processing speed to identify prodromal patients at risk of developing a psychotic disorder. Finally, the identification of a specific neurological circuit related to impaired cognition would provide a therapeutic target to alleviate cognitive deficits in patients with psychotic disorders in the future.

京都大学 / 新制・課程博士 / 博士(情報学) / 甲第25426号 / 情博第864号 / 新制||情||145(附属図書館) / 京都大学大学院情報学研究科知能情報学専攻 / (主査)准教授 水原 啓暁, 教授 西田 眞也, 教授 熊田 孝恒 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DFAM

Menschen teilen mit vielen Spezies die Fähigkeit, aus einer Menge von Objekten deren Numerosität zu extrahieren. Es wird angenommen, dass diese Fähigkeit die Grundlage für die Menschen eigene, symbolisch-mathematische Fertigkeiten bildet. Daher ist ein besseres Verständnis der neuralen Charakteristiken dieser Fähigkeit von großer Bedeutung. Eines der einflussreichsten Modelle (das Triple Code Modell-TCM) nimmt an, dass dieser evolutionär alte Mechanismus in horizontalen Anteil des intraparietalen Sulcus verortet werden kann, der die Bedeutung von Anzahl in einer format- und modalitätsunabhängigen Art und Weise repräsentiert (d.h., Größencode). Zusätzlich wird angenommen, dass Subtraktion auf eben dieser Fähigkeit aufbaut, wohingegen Multiplikation stärker auf phonologischen Verarbeitungsmechanismen beruht (d.h., verbaler Code). Elektrophysiologische Untersuchungen beim Affen deuten auf einen gewissen Grad an Abstraktion von semantischer Größeninformation im parietalen Kortex hin. Jedoch wurde dies bisher nur für kleine Numerositäten ( / Humans as well as other animals are endowed with the capacity to extract the numerosity (i.e., the number of items) of a given set of objects. This capacity is thought to form the basis of human specific symbolic mathematical abilities. Hence, understanding its nature is of importance. One of the most influential models (The Triple Code Model) suggests that this evolutionarily ancient mechanism resides on the horizontal aspect of the intraparietal sulcus and represents number semantics in a format and modality independent fashion (i.e., magnitude code). In addition, subtraction is thought to rely more on this mechanism whereas multiplication relies more on phonological circuits (i.e., verbal code). Although there is evidence from non-human primate electrophysiology suggesting a certain degree of abstraction for number semantics in the parietal cortex, this was only found for small numerosities (

L’expression faciale est un vecteur de communication important dans l’expérience de douleur. Cependant, les corrélats neuronaux associés à cette manifestation de la douleur demeurent peu investigués. Le but de ce mémoire était de développer un modèle neurobiologique permettant de prédire l’expression faciale évoquée par des stimuli douloureux afin d’approfondir nos connaissances sur les mécanismes cérébraux de la douleur et de la communication non verbale. La signature cérébrale de l’expression faciale de la douleur a été élaborée sur un jeu de données d’IRMf acquis chez des adultes en santé en utilisant des algorithmes d’apprentissage machine pour prédire des scores d’expression faciale évoquée par des stimulations douloureuses phasiques (c.-à-d. de courtes stimulations) à l’échelle de la population. Les résultats suggèrent qu’il est possible de prédire ces réponses faciales à partir d’un patron d’activation multivoxels. Cette signature cérébrale se distingue, du moins partiellement, de signatures cérébrales prédictives de l’intensité et du caractère déplaisant de la douleur rapportée et de la valeur future de la douleur. Bien que d’autres études soient nécessaires pour examiner la spécificité et la généralisabilité de la signature cérébrale de l’expression faciale de la douleur, ce mémoire souligne l’existence d’une représentation cérébrale spatialement distribuée prédictive des réponses faciales en lien avec la douleur, et suggère l’importance de cette mesure comportementale dans l’expérience de la douleur comme étant complémentaire aux mesures autorapportées de l’intensité perçue. / Facial expression is an important communication vector in the experience of pain. However, the neural correlates associated with this manifestation of pain remain relatively unexplored. The goal of this thesis was to develop a neurobiological model to predict facial expression evoked by painful stimuli in order to expand our knowledge of the brain mechanisms of pain and non-verbal communication. The brain signature of facial expressions of pain was developed on a dataset including healthy adults using machine learning algorithms to predict facial expression scores evoked by phasic painful stimuli (i.e., short stimulation) at the population level. The results suggest that it is possible to predict the facial expression of pain from a multivoxel activation pattern. This brain signature of facial pain expression is at least partially distinct from other brain signatures predictive of reported pain intensity and unpleasantness, and future pain value. Although further studies are needed to examine the specificity and generalizability of the brain signature of facial expression of pain, this master thesis highlights the existence of a spatially distributed brain representation predictive of pain-related facial responses, and suggests the importance of this behavioural measure in the experience of pain as complementary to self-reported measures of pain intensity.

Wo und wie werden komplexe Regelsätze im Gehirn repräsentiert? Drei empirische Studien dieser Doktorarbeit untersuchen dies experimentell. Eine weitere methodische Studie liefert Beiträge zur Weiterentwicklung der genutzten empirischen Methode. Die empirischen Studien nutzen multivariate Musteranalyse (MVPA) funktioneller Magnetresonanzdaten (fMRT) gesunder Probanden. Die Fragestellungen der methodischen Studie wurden durch die empirischen Arbeiten inspiriert. Wirkung und Anwendungsbreite der entwickelten Methode gehen jedoch über die Anwendung in den empirischen Studien dieser Arbeit hinaus. Die empirischen Studien bearbeiten Fragen wie: Wo werden Hinweisreize und Regeln repräsentiert, und sind deren Repräsentationen voneinander unabhängig? Wo werden Regeln repräsentiert, die aus mehreren Einzelregeln bestehen, und sind Repräsentationen der zusammengesetzten Regeln Kombinationen der Repräsentationen der Einzelregeln? Wo sind Regeln verschiedener Hierarchieebenen repräsentiert, und gibt es einen hierarchieabhängigen Gradienten im ventrolateralen präfrontalen Kortex (VLPFK)? Wo wird die Reihenfolge der Regelausführung repräsentiert? Alle empirischen Studien verwenden informationsbasiertes funktionales Mapping ("Searchlight"-Ansatz), zur hirnweiten und räumlich Lokalisierung von Repräsentationen verschiedener Elemente komplexer Regelsätze. Kernergebnisse der Arbeit beinhalten: Kompositionalität neuronaler Regelrepräsentationen im VLPFK; keine Evidenz für Regelreihenfolgenrepräsentation im VLPFK, welches gegen VLPFK als generelle Task-Set-Kontrollregion spricht; kein Hinweis auf einen hierarchieabhängigen Gradienten im VLPFK. Die komplementierende methodische Studie präsentiert "The Same Analysis Approach (SAA)", ein Ansatz zur Erkennung und Behebung experimentspezifischer Fehler, besonders solcher, die aus Design–Analyse–Interaktionen entstehen. SAA ist für relevant MVPA, aber auch für anderen Bereichen innerhalb und außerhalb der Neurowissenschaften. / Where and how does the brain represent complex rule sets? This thesis presents a series of three empirical studies that decompose representations of complex rule sets to directly address this question. An additional methodological study investigates the employed analysis method and the experimental design. The empirical studies employ multivariate pattern analysis (MVPA) of functional magnetic resonance imaging (fMRI) data from healthy human participants. The methodological study has been inspired by the empirical work. Its impact and application range, however, extend well beyond the empirical studies of this thesis. Questions of the empirical studies (Studies 1-3) include: Where are cues and rules represented, and are these represented independently? Where are compound rules (rules consisting of multiple rules) represented, and are these composed from their single rule representations? Where are rules from different hierarchical levels represented, and is there a hierarchy-dependent functional gradient along ventro-lateral prefrontal cortex (VLPFC)? Where is the order of rule-execution represented, and is it represented as a separate higher-level rule? All empirical studies employ information-based functional mapping ("searchlight" approach) to localise representations of rule set features brain-wide and spatially unbiased. Key findings include: compositional coding of compound rules in VLPFC; no order information in VLPFC, suggesting VLPFC is not a general controller for task set; evidence against the hypothesis of a hierarchy-dependent functional gradient along VLPFC. The methodological study (Study 4) introduces "The Same Analysis Approach (SAA)". SAA allows to detect, avoid, and eliminate confounds and other errors in experimental design and analysis, especially mistakes caused by malicious experiment-specific design-analysis interactions. SAA is relevant for MVPA, but can also be applied in other fields, both within and outside of neuroscience.