Many approaches to solving tasks in the field of Natural Language Processing (NLP) use syntactic dependency trees (SDTs) as a feature to represent the latent nonlinear structure within sentences. Recently, work in parsing sentences to graph-based structures which encode semantic relationships between words—called semantic dependency graphs (SDGs)—has gained interest. This thesis seeks to explore the use of SDGs in place of and alongside SDTs within a relation classification system based on long short-term memory (LSTM) neural networks. Two methods for handling the information in these graphs are presented and compared between two SDG formalisms. Three new relation extraction system architectures have been created based on these methods and are compared to a recent state-of-the-art LSTM-based system, showing comparable results when semantic dependencies are used to enhance syntactic dependencies, but with significantly fewer training parameters.

Tout au long de cette thèse, l’évaluation musicale est traitée comme un objet d’étude latent visant à donner des outils pour l’apprentissage de la lecture musicale. Grâce à l’analyse des mouvements oculaires et les variables provenant de la performance, nous avons défini certaines variables qui rendent compte de l’expertise et des interactions entre différents groupes de niveaux d’expertise musicale lors de l’apprentissage d’un nouveau morceau de musique. De façon plus détaillée, nous avons observé la mise en œuvre de différentes stratégies de prise d’information, de traitement et de récupération de l’information musicale en fonction du niveau pianistique et souligné l’importance d’apprendre en interaction avec la classe et le professeur. Les stratégies sont en même temps ajustées par rapport à la confiance acquise au cours du processus d'acquisition de compétences (Bandura, 1997 ; McPherson et McCormick, 2006). En référence au débat actuel concernant la nature de la lecture de partitions, nous avons comparé les traitements musicaux et verbaux pendant une tâche de lecture « compréhensive » de textes et de partitions. Dans l’ensemble et au regard du modèle de Baddeley (1990), les ressources cognitives des musiciens pendant la lecture musicale seraient mobilisées en fonction de l’expertise et du type de style musical. / Throughout this thesis, evaluation of music performance is viewed as a latent object of study in order to provide tools for learning to read music. We have defined some variables from eye movements and music performance accounting for expert performance and interactions between skill groups when learning a new piece of music. In more details, we have observed the use of different strategies for music information intake, processes and information retrieval depending on musicians’ expertise and we have stressed the importance of learning through interaction. In the process of skill acquisition, when self-confidence is gained strategies are simultaneously adjusted (Bandura, 1997; McPherson and McCormick, 2006). In reference to the current debate about the nature of music reading, we have compared musical and verbal processing during comprehensive reading of texts and scores. On the whole, considering the model of Baddeley (1990), musicians’ cognitive resources during music reading would be mobilized depending on the expertise and the music style.

A presente pesquisa centra-se no estudo das relações entre memória de curto prazo, memória de trabalho, velocidade de processamento, processamento visuoespacial, funções executivas, atenção seletiva e alternada e desempenho matemático, uma vez que essas relações podem esclarecer as diferenças entre os alunos na aprendizagem da matemática. O trabalho teve como objeto de observação crianças da 2ª série do ensino fundamental. Consiste em um estudo com base correlacional e comparativa. A amostra desta pesquisa é composta por 40 alunos, com idade entre 8 e 9 anos. A investigação envolvendo o funcionamento neuropsicológico e o desempenho cognitivo na aprendizagem matemática é relevante, pois para desenvolver habilidades de cálculo matemático é fundamental que o aluno tenha construído o conceito de número e esta construção está vinculada ao desenvolvimento das funções neuropsicológicas. O desempenho matemático é avaliado através da Prova de Aritmética de Capovilla, Montiel e Capovilla (2007) e as funções neuropsicológicas, através de quatro subtestes da Escala Wechsler WISCIII, que são: Memória de Curto Prazo – teste de Dígitos de Ordem Direta; Memória de Trabalho – Dígitos de Ordem Inversa; Velocidade de Processamento - Procurar Símbolos e Códigos; Percepção Visuoespacial – teste de Cubos. A Função Executiva é avaliada através do Teste das Trilhas de Montiel e Capovilla (2007) e a Atenção, através do Teste de Atenção por Cancelamento de Montiel e Capovilla (2007). Observa-se uma correlação significativa entre Desempenho Matemático e Memória de Curto Prazo, Velocidade de Processamento, Função Executiva e Atenção. Neste estudo, não se encontrou uma correlação estatisticamente significativa entre as funções Memória de Trabalho e Processamento Visuoespacial. / This paper approaches the relationship amongst neuropsychological functions –Short Term Memory (STM), Working Memory (WM), Speed Processing (SP), Visual Spatial Perception (VSP), Executive Functions (EF), Attention (A) and Math Performance (MP) in 2nd grade children at a public Elementary School from Porto Alegre. It is a comparative and correlation study. The sample of this research is composed for 40 pupils with are between 8 and 9 years. Research involving neuropsychological functioning and cognitive performance in mathematics is pertinent, for it is essential that the student build the notion of number in order to develop mathematical skills and such construction relies on the expansion of neuropsychological functions. The mathematical performance is measured by the Arthmetic Test of Capovilla, Montiel and Capovilla (2007) and the neuropsychological functions are verified by subtests Wechsler WIS CIII; Short Term Memory is assessed by the Digits in Direct Order, while the Working Memory is established by the Digits in Reverse Order. The speed processing is assessed through the use of Find Symbols and Codes activities, and spatial perception is confirmed by the utilization of cubes. The Executive Function is assessed by the Trail Making of Montiel and Capovilla (2007) and attention is assessed by the Attention Test for Cancellation of Montiel and Capovilla (2007). There is a momentous correlation amid math performance and Short-Term Memory, speed processing, executive function and attention. This study one did not mett, however, the establishment of a statistically significant correlation between the Working Memory functions and the Visual Spatial Processing.

A presente pesquisa centra-se no estudo das relações entre memória de curto prazo, memória de trabalho, velocidade de processamento, processamento visuoespacial, funções executivas, atenção seletiva e alternada e desempenho matemático, uma vez que essas relações podem esclarecer as diferenças entre os alunos na aprendizagem da matemática. O trabalho teve como objeto de observação crianças da 2ª série do ensino fundamental. Consiste em um estudo com base correlacional e comparativa. A amostra desta pesquisa é composta por 40 alunos, com idade entre 8 e 9 anos. A investigação envolvendo o funcionamento neuropsicológico e o desempenho cognitivo na aprendizagem matemática é relevante, pois para desenvolver habilidades de cálculo matemático é fundamental que o aluno tenha construído o conceito de número e esta construção está vinculada ao desenvolvimento das funções neuropsicológicas. O desempenho matemático é avaliado através da Prova de Aritmética de Capovilla, Montiel e Capovilla (2007) e as funções neuropsicológicas, através de quatro subtestes da Escala Wechsler WISCIII, que são: Memória de Curto Prazo – teste de Dígitos de Ordem Direta; Memória de Trabalho – Dígitos de Ordem Inversa; Velocidade de Processamento - Procurar Símbolos e Códigos; Percepção Visuoespacial – teste de Cubos. A Função Executiva é avaliada através do Teste das Trilhas de Montiel e Capovilla (2007) e a Atenção, através do Teste de Atenção por Cancelamento de Montiel e Capovilla (2007). Observa-se uma correlação significativa entre Desempenho Matemático e Memória de Curto Prazo, Velocidade de Processamento, Função Executiva e Atenção. Neste estudo, não se encontrou uma correlação estatisticamente significativa entre as funções Memória de Trabalho e Processamento Visuoespacial. / This paper approaches the relationship amongst neuropsychological functions –Short Term Memory (STM), Working Memory (WM), Speed Processing (SP), Visual Spatial Perception (VSP), Executive Functions (EF), Attention (A) and Math Performance (MP) in 2nd grade children at a public Elementary School from Porto Alegre. It is a comparative and correlation study. The sample of this research is composed for 40 pupils with are between 8 and 9 years. Research involving neuropsychological functioning and cognitive performance in mathematics is pertinent, for it is essential that the student build the notion of number in order to develop mathematical skills and such construction relies on the expansion of neuropsychological functions. The mathematical performance is measured by the Arthmetic Test of Capovilla, Montiel and Capovilla (2007) and the neuropsychological functions are verified by subtests Wechsler WIS CIII; Short Term Memory is assessed by the Digits in Direct Order, while the Working Memory is established by the Digits in Reverse Order. The speed processing is assessed through the use of Find Symbols and Codes activities, and spatial perception is confirmed by the utilization of cubes. The Executive Function is assessed by the Trail Making of Montiel and Capovilla (2007) and attention is assessed by the Attention Test for Cancellation of Montiel and Capovilla (2007). There is a momentous correlation amid math performance and Short-Term Memory, speed processing, executive function and attention. This study one did not mett, however, the establishment of a statistically significant correlation between the Working Memory functions and the Visual Spatial Processing.

Muitas das interações que temos com o ambiente envolvem situações dinâmicas. Raramente estas situações são realizadas de maneira isolada. Em ações interceptativas, por exemplo, o sistema nervoso precisa estimar quando um alvo irá atingir uma determinada posição e constantemente ajustar os movimentos para diferentes contextos. Apesar de estudos psicofísicos terem investigado possíveis variáveis ópticas e modulações do contexto envolvidas na integração visuomotora, pouca atenção foi direcionada para compreender os mecanismos cerebrais da integração de informação visual de movimento e ações motoras em cenários dinâmicos. O objetivo geral desta tese foi investigar estes mecanismos e entender como o contexto modula a atividade cerebral em ações interceptativas em adultos jovens saudáveis. No primeiro experimento, foram investigadas quais áreas estão associadas com a integração entre informação visual de movimento e ação motora sincronizatória. Foi utilizado desenho relacionado à eventos em ressonância magnética funcional avaliando a atividade cerebral em tarefa de timing coincidente. Foi identificada uma rede fronto-parietal bilateral dorsal e ativação bilateral do braço ascendente do sulco temporal inferior, região funcionalmente definida como hV5+, e giro angular. No segundo experimento, foi testada a influência causal da área hV5+ e córtex pré-motor dorsal no viés comportamental que tentativa prévia exerce sobre a tentativa atual usando estimulação magnética transcraniana. Os resultados deste experimento sugerem que o efeito da tentativa prévia depende fortemente de um mecanismo de memória de curto prazo implícito no córtex pré-motor dorsal e moderadamente da área hV5+. No terceiro experimento, o objetivo foi verificar o efeito da repetição e expectativa de repetição da velocidade do estímulo nas áreas associadas com ações interceptativas. Para isso, os participantes interceptaram pares de alvos móveis com velocidade igual ou diferente em experimento relacionado à eventos em ressonância magnética funcional. A expectativa sobre a repetição da velocidade do estímulo foi manipulada pela probabilidade de repetição da velocidade do alvo em diferentes blocos. Os resultados comportamentais indicaram que a velocidade do primeiro estímulo enviesou o erro temporal dos participantes para responder ao segundo estímulo e que a manipulação da expectativa dos participantes não resultou em diferença no erro temporal. Não houve modulação da amplitude do sinal BOLD pela repetição do estímulo ou pela manipulação da expectativa sobre a repetição de velocidade. No quarto experimento, foram investigados os mecanismos cerebrais que permitem que a expectativa influencie o comportamento em uma tarefa de timing coincidente. Para isso, a expectativa sobre a velocidade do alvo foi manipulada a cada tentativa por meio de dicas em um experimento relacionado à eventos em ressonância magnética funcional. Os resultados deste experimento mostraram que a expectativa sobre a velocidade do alvo aumenta o sinal BOLD para expectativa válida em hV5+ e córtex pré-motor dorsal. Em geral, os resultados desta tese mostram que a integração visuomotora em ações interceptativas está associada com atividade em uma rede fronto-parietal dorsal e da área hV5+. Destas áreas, foi verificado que o córtex pré-motor dorsal e área hV5+ tem papel de armazenar informação da tentativa prévia que enviesa o comportamento na tentativa atual. Por fim, foi mostrado que a expectativa sobre a velocidade do alvo modula o sinal BOLD em áreas iniciais do processamento sensorial e de planejamento motor. Estes resultados avançam o conhecimento sobre os mecanismos neurais associados à integração visuomotora em ações interceptativas e mostram como o contexto em que a tarefa é realizada modula essa integração / Many of our interactions with the environment happen in dynamic situations. These situations are rarely experienced in isolation. In interceptive actions, for example, the nervous system needs to estimate when a target will arrive at a certain position and constantly adjust movements for different contexts. Although psychophysical studies have investigated the putative optical variables and contextual modulations in visuomotor integration, less attention has been devoted to understand brain mechanisms underlying the integration of visual motion information and motor actions in dynamic scenarios. The main goal of the present thesis was to investigate the brain mechanisms involved in visuomotor integration of interceptive actions in healthy young adults. In the first experiment, it was investigated which brain areas are associated with integration of visual motion information and timed motor action in an event-related functional magnetic resonance imaging experiment. Results showed greater BOLD signal in a bilateral dorsal fronto-parietal network, as well as hV5+ and angular gyrus. In the second experiment, it was tested the causal influence of area hV5+ and dorsal premotor cortex on the behavioral bias that previous trial exerts on the current trial using transcranial magnetic stimulation. The results of this experiment provide causal evidence that the previous trial effect is mediated to a large extent by an implicit shortterm memory mechanism in the dorsal premotor cortex, and to a lesser extent by hV5 +, in a visuomotor integration task with moving objects. In the third experiment, the goal was to verify the effect of repetition and expectation about repetition of speed in brain areas associated with visuomotor integration in interceptive actions. To that end, participants intercepted pairs of moving targets with either same or different speed in an event-related functional magnetic resonance experiment. Expectation about speed repetition of target speed was manipulated through the probability of speed repetition in different blocks. Behavioral results indicate that speed from the first stimulus biased participants\' temporal error for the second stimulus. However, manipulation of participants\' expectation did not affect temporal error. In addition, BOLD signal amplitude was modulated by neither stimulus speed repetition nor expectation. In the forth experiment, brain mechanisms that allow expectation to influence behavior in a coincident timing task were investigated. Participant\'s expectation about target speed was manipulated on a trial-by-trial basis by means of cues in an event-related functional magnetic resonance experiment. The results of this experiment showed that expectation about target speed increases BOLD signal for valid expectations in hV5+ and dorsal premotor cortex. Overall, the results in this thesis show that visuomotor integration in interceptive actions is associated with activity in a dorsal fronto-parietal network and hV5+. In addition, it was verified that dorsal premotor cortex and hV5+ have a role in storing information from previous trial that bias behavior on the current trial. Lastly, it was shown that expectation about upcoming target speed modulates BOLD signal in early stage visual motion processing as well as motor planning areas. These results advance knowledge about the brain mechanisms associated with visuomotor integration in interceptive actions and show how context modulates this integration process

La consommation de cannabis chez les adolescents et les jeunes adultes constitue un facteur de risque de développement de symptômes psychotiques. L’objectif de cette thèse était d’étudier les conséquences à l’âge adulte d’une exposition chronique aux cannabinoïdes pendant l’adolescence. Nous avons réalisé ces études chez le Rat et démontré que l’administration chronique d’un agoniste des récepteurs cannabinoïdes, le CP55, 940, durant l’adolescence (JPN 29 à 50) entraîne à l’âge adulte des déficits significatifs et durables de la mémoire à court terme et de la mémoire de travail spatiale. Le même traitement administré à l’âge adulte (JPN 70 à 91) n’entraîne pas de déficits cognitifs à long terme. L’apport de ces travaux et leur originalité a été de contribuer à la compréhension des mécanismes sous-jacents au développement de ces déficits cognitifs. Nous avons mis en évidence, chez ces rats à l’âge adulte, une diminution significative de l’amplitude de la PLT induite sur le circuit reliant l’hippocampe au cortex préfrontal, fortement impliqué dans les processus de mémoire et dans la physiopathologie de la schizophrénie. Ce déficit de PLT a été mis en relation avec des modifications de la morphologie des neurones pyramidaux de la couche II/III du cortex préfrontal et de certains marqueurs de la plasticité synaptique. Ces résultats confirment que l'adolescence est une période critique et vulnérable aux effets délétères des cannabinoïdes sur les processus cognitifs. L’étude des bases neurobiologiques de ces effets, en particulier en ce qui concerne les altérations des circuits préfrontaux, devrait améliorer notre compréhension des mécanismes de l'émergence de la psychose et conduire à de nouvelles stratégies thérapeutiques et de prévention. / Cannabis use among adolescents and young adults is a risk factor for developing psychotic symptoms. The aim of this thesis was to study the consequences at different levels (behavioral, functional and structural in adulthood) of chronic exposure to synthetic cannabinoids during adolescence in the rat. We showed that chronic administration of a cannabinoid receptor agonist, CP 55940, during adolescence (PND 29-50) in rats leads to significant long-term deleterious effects in short-term memory and in spatial working memory in adulthood. The same treatment administered in adulthood (PND 70-91) does not cause long-term cognitive deficits. We further demonstrated that cannabinoid during adolescence leads in adults to a significant decrease in the expression of LTP induced in hippocampal to prefrontal cortex synapses, a circuit directly involved in memory processes and in the pathophysiology of schizophrenia. This alteration could be associated with changes in the morphology of pyramidal neurons in layer II / III of PFC and/or in the density of synaptic markers that we also identified in these cannabinoid-treated rats. These findings demonstrate that adolescence is a critical and vulnerable period to the deleterious effects of cannabinoids on cognitive processes. The neurobiological basis for this effect, especially with regard to alterations to prefrontal circuitry, which we partially provide should improve our understanding of the emergence of psychosis and lead to new therapeutic strategies and prevention.

Aprender nuevas palabras en un idioma extranjero, es decir, el léxico necesario que fundamenta la posibilidad del desarrollo de las destrezas comunicativas, constituye uno de los problemas más complejos en el proceso tanto de enseñanza como de aprendizaje del español como lengua extranjera. En relación con el aprendizaje del vocabulario identificamos un posible problema; el riesgo de que el número de palabras aprendidas se olvide aumenta después de la prueba o los ejercicios. Si nuestros alumnos no pueden ampliar su vocabulario su competencia comunicativa tampoco va a desarrollar.Para poder entender por qué ocurre el problema y cómo se podría encontrar otros recursos didácticos que contribuyan a un cambio en el proceso, investigamos un fenómeno conocido por la psicología de la educación como el efecto de la memoria espaciada - un fenómeno cognitivo que se benéfica de las repeticiones, pero siempre distribuidas en el tiempo. Estrategias de enseñanza que utilizan dicho efecto se refiere como aprendizaje distribuido.Mediante un pequeño estudio analizamos el efecto de la memoria espaciada (ME) como método alternativa. De este estudio podemos inferir que existe un efecto de memoria espaciada tangible en el aprendizaje de los alumnos que estudiaron según un modelo distribuido, es decir con repeticiones.Pudimos constatar un resultado positivo en este pequeño estudio piloto. Los alumnos lograron recordar en la examinación el 85% de las palabras ejercitadas en la clase un mes después. Este resultado abre nuevas perspectivas de estudio e indica que puede haber alternativas didácticas en la enseñanza del vocabulario de ELE en el salón escolar sueco. / Studying and learning words in a foreign language in order to develop a vocabulary that promotes communicative competence, is a daunting task both for students and teachers of Spanish as a foreign language in Sweden and elsewhere. In this context we identify one problem; the possibility that words learned in class will be forgotten as soon as they have been tested on a quiz or exam. If our students cannot incorporate new words into their vocabulary it is quite possible that their communicative development will stop or slow down.       In order to understand this problem and find alternative ways to teach vocabulary we investigated a phenomena called ‘the spaced memory effect’. In the field of educational psychology this is when the learner study with repetitions distributed over time. Practices that build on the spaced memory effect are often called distributed learning.       In a small study we tested this effect as an alternative way of teaching and learning vocabulary. From this study we could conclude that the effect is possible to measure and that it is consistent in all the test subjects that followed a study model that was based on repetitions or a distributed learning model. It was shown that after one month the students were able to remember 85% of the words.       The results from this small study provides new perspectives for further investigation and suggests that there are alternative ways of teaching Spanish vocabulary in Swedish schools.

Doctorat en sciences psychologiques / info:eu-repo/semantics/nonPublished

Táto diplomová práca sa venuje automatickému rozpoznávaniu ludí v obrazových dátach s využitím konvolučných neurónových sieti na určenie polohy tváre a následnej analýze získaných dát. Výsledkom analýzy tváre je určenie pohlavia, emócie a veku osoby. Práca obsahuje popis použitých architektúr konvolučných sietí pre každú podúlohu. Sieť na odhad veku má natrénované nové váhy, ktoré sú vzápätí zmrazené a majú do svojej architektúry vložené LSTM vrstvy. Tieto vrstvy sú samostatne dotrénované a testované na novom datasete vytvorenom pre tento účel. Výsledky testov ukazujú zlepšenie predikcie veku. Riešenie pre rýchlu, robustnú a modulárnu detekciu tváre a ďalších ludských rysov z jedného obrazu alebo videa je prezentované ako kombinácia prepojených konvolučných sietí. Tieto sú implementované v podobe skriptu a následne vysvetlené. Ich rýchlosť je dostatočná pre ďalšie dodatočné analýzy tváre na živých obrazových dátach.

Dans les systèmes cognitifs, le rôle de la mémoire de travail est crucial pour le raisonnement visuel et la prise de décision. D’énormes progrès ont été réalisés dans la compréhension des mécanismes de la mémoire de travail humain/animal, ainsi que dans la formulation de différents cadres de réseaux de neurones artificiels à mémoire augmentée. L’objectif global de notre projet est de former des modèles de réseaux de neurones artificiels capables de consolider la mémoire sur une courte période de temps pour résoudre une tâche de mémoire et les relier à l’activité cérébrale des humains qui ont résolu la même tâche. Le projet est de nature interdisciplinaire en essayant de relier les aspects de l’intelligence artificielle (apprentissage profond) et des neurosciences. La tâche cognitive utilisée est la tâche N-back, très populaire en neurosciences cognitives dans laquelle les sujets sont présentés avec une séquence d’images, dont chacune doit être identifiée pour savoir si elle a déjà été vue ou non. L’ensemble de données d’imagerie fonctionnelle (IRMf) utilisé a été collecté dans le cadre du projet Courtois Neurmod. Nous étudions plusieurs variantes de modèles de réseaux neuronaux récurrents qui apprennent à résoudre la tâche de mémoire de travail N-back en les entraînant avec des séquences d’images. Ces réseaux de neurones entraînés optimisés pour la tâche de mémoire sont finalement utilisés pour générer des représentations de caractéristiques pour les images de stimuli vues par les sujets humains pendant leurs enregistrements tout en résolvant la tâche. Les représentations dérivées de ces réseaux de neurones servent ensuite à créer un modèle de codage pour prédire l’activité IRMf BOLD des sujets. On comprend alors la relation entre le modèle de réseau neuronal et l’activité cérébrale en analysant cette capacité prédictive du modèle dans différentes zones du cerveau impliquées dans la mémoire de travail. Ce travail présente une manière d’utiliser des réseaux de neurones artificiels pour modéliser le comportement et le traitement de l’information de la mémoire de travail du cerveau et d’utiliser les données d’imagerie cérébrale capturées sur des sujets humains lors de la tâche N-back pour potentiellement comprendre certains mécanismes de mémoire du cerveau en relation avec ces modèles de réseaux de neurones artificiels. / In cognitive systems, the role of working memory is crucial for visual reasoning and decision making. Tremendous progress has been made in understanding the mechanisms of the human/animal working memory, as well as in formulating different frameworks of memory augmented artificial neural networks. The overall objective of our project is to train artificial neural network models that are capable of consolidating memory over a short period of time to solve a memory task and relate them to the brain activity of humans who solved the same task. The project is of interdisciplinary nature in trying to bridge aspects of Artificial Intelligence (deep learning) and Neuroscience. The cognitive task used is the N-back task, a very popular one in Cognitive Neuroscience in which the subjects are presented with a sequence of images, each of which needs to be identified as to whether it was already seen or not. The functional imaging (fMRI) dataset used has been collected as a part of the Courtois Neurmod Project. We study multiple variants of recurrent neural network models that learn to remember input images across timesteps. These trained neural networks optimized for the memory task are ultimately used to generate feature representations for the stimuli images seen by the human subjects during their recordings while solving the task. The representations derived from these neural networks are then to create an encoding model to predict the fMRI BOLD activity of the subjects. We then understand the relationship between the neural network model and brain activity by analyzing this predictive ability of the model in different areas of the brain that are involved in working memory. This work presents a way of using artificial neural networks to model the behavior and information processing of the working memory of the brain and to use brain imaging data captured from human subjects during the N-back task to potentially understand some memory mechanisms of the brain in relation to these artificial neural network models.