L’objectif de cette thèse est d’étudier l’impact de la fatigue du consommateur sur son traitement de l’information ublicitaire. Plus précisément, nous désirons comprendre comment la fatigue influence le processus de traitement visuel de la publicité, ainsi que les attitudes et la mémorisation. En nous appuyant sur la théorie d’adéquation des ressources, nous postulons que le traitement de l’information publicitaire, et donc l’efficacité de la publicité, est optimal quand le niveau de ressources cognitives disponible (niveau de fatigue du consommateur) correspond au niveau de ressources exigé (niveau de complexité de la publicité). Nos résultats mettent en lumière des effets de la fatigue différents selon les deux échantillons de notre recherche (étudiants et salariés). Les étudiants fatigués adoptent une stratégie visuelle de balayage pour compenser les effets de la fatigue et rendre le traitement de l’information plus facile. Les salariés fatigués adoptent une stratégie visuelle d’attention soutenue consistant à mobiliser le peu de ressources disponibles pour traiter l’information de façon efficace. Cette stratégie visuelle d’attention soutenue a un impact positif sur la mémorisation de la publicité. Nous montrons enfin que la théorie d’adéquation des ressources n’est pas le cadre théorique approprié pour étudier notre objet de recherche. En effet, les ressources cognitives s’avèrent être dynamiques : si elles y sont motivées, les personnes peuvent adapter leur niveaude ressources à la tache à accomplir. Les implications théoriques, méthodologiques et managériales de ces résultats sont discutées. / This research aims at studying the impact of consumers’ tiredness on the way they perceive and process advertising information. More specifically, we try to understand how tiredness influences advertisements’ visual processing as well as consumers’ attitudes and memorization. Relying on Resource Matching theory, we hypothesize that advertising informationprocessing, and therefore advertising efficiency, is optimal when the level of available resources (consumers’ tiredness) matches the level of required resources (advertisement complexity). Results show that the impact of tiredness differs according to sample under consideration (students or staff members). Tired students adopt a “screening” visual strategyin order to compensate for tiredness effects and ease their information processing. Tired staff members adopt a “sustained attention” visual strategy consisting in mobilizing their few available resources in order to efficiently process information. In this second case, sustained visual attention leads to better advertising memorization. We show that the Resource Matching theory is not the appropriate theoretical framework to study our research object. Indeed, we find that cognitive resources are dynamic:if motivated to do so, individuals are able to adapt their resource level according to the processing task. Theoretical, methodological and managerial implications are discussed.

A atenção visuoespacial pode ser considerada como a atividade neural que seleciona certas regiões do espaço em detrimento das demais. Existem dois tipos de orientação da atenção: atenção automática e atenção voluntária. Há evidências de que as áreas cerebrais associadas à estes dois tipos se sobrepõem, porém alguns autores observaram que há áreas distintas de ativação cerebral para atenção automática e voluntária. Estas diferenças resultaram em visões distintas a respeito de modelos cerebrais envolvendo a atenção visuoespacial. O objetivo de nosso trabalho foi identificar quais são as principais redes de ativação neuronal durante a mobilização da atenção automática e voluntária com Ressonância Magnética Funcional (RMf) relacionada à eventos, utilizando a tarefa de Posner modificada (1980). Participaram deste estudo vinte adultos jovens e saudáveis (17 mulheres). A tarefa foi programada no ambiente Eprime (Psylab, EUA) e os participantes treinaram antes de realizar o experimento de RMf. As condições foram pseudorandomizadas com otimização por algoritmos genéticos. Cada tentativa começava com um Ponto de Fixação na tela e dois anéis demarcados, um em cada lado. Após o aparecimento do Estímulo Precedente (E1), um Estímulo Alvo (E2) aparecia dentro de um dos anéis. Nestes experimentos foram consideradas as condições válida, neutra e inválida. Os dados foram adquiridos em 3 runs para cada Experimento. A aquisição das imagens foi realizada em um sistema de 3 tesla (Philips Achieva), acoplado à um sistema de detecção de movimentação ocular (eye tracker) compatível com ambiente de RM (Magconcept, USA). As imagens de RMf foram adquiridas com sequências BOLD, GRE-EPI com TR=2s, TE= 30s, voxels isotrópicos de 3mm. O tempo e a resposta ao estímulo foram registrados (Zurc&Zurc, São Paulo, Brasil). A análise das imagens de RMf foi realizada com o programa FSL 4.1 (Oxford University, Inglaterra) e incluiu a correção de movimento, filtro espacial, remoção de artefatos e transformação da imagem para o espaço comum. O mapas cerebrais representando o efeito de grupo (usando o modelo de regressão linear) foram produzidos com limiar de Z >2,3 e nível de significância de p<0,05 (corrigido para comparações múltiplas). O tempo de reação foi menor quando o E2 apareceu na condição válida em comparação com a condição inválida. Houve superposição dos mapas de ativação na condição de atenção automática e voluntária em extensa rede, envolvendo áreas: frontais (giros pré-centrais, frontais médios, frontais mediais e giro frontal inferior direito), parietais (bordas dos sulcos intraparietais, giros pós-centrais, angulares, supramarginais e pré-cuneo), temporais (giros temporais médios e no giro temporal superior esquerdo) e occipitais (giros fusiformes e no giro lingual direito. Observamos outras áreas: giros do cíngulo anteriores e posteriores, parahipocampais, ínsulas, colículos superiores e inferiores, e cerebelo). Concluímos que as redes frontoparietais foram detectadas tanto na mobilização da atenção visuoespacial automática quanto voluntária. No entanto, o giro frontal superior esteve presente apenas na mobilização da atenção voluntária. Estes resultados mostram que a distinção parece estar em áreas cerebrais envolvidas em subsistemas que incluem a orientação e reorientação da atenção automática e voluntária. / Visuospatial attention can be considered as the neural activity that selects certain regions of space at the expense of others. There are two types of orientation of attention: automatic attention and voluntary attention. There is evidence that the brain networks associated with these two types overlap, but some authors have observed that there are distinct areas of brain activation for automatic and voluntary attention. These differences resulted in different views about brains models involving visuospatial attention. The aim of our study was to identify brain networks associated with automatic and voluntary attention using Event-Related Functional Magnetic Resonance Imaging (fMRI) and a new modification of the Posner task (1980). The study included twenty young healthy adults (17 women and 3 men). The task was programmed in Eprime environment (Psylab, USA) and all participants trained before performing the experiment in the MRI machine. The conditions were pseudorandomized with the optimization method of genetic algorithms. Each trial began with a fixation point on the screen and two demarcated rings, one on each side. After the onset of the Prime Stimulus (S1), a Target Stimulus (S2) appeared inside one of the rings. In these experiments the valid, neutral and invalid conditions were considered. Data were acquired in three runs for each experiment. Image acquisition of this project was performed on a 3 Tesla system (Philips Achieva), coupled to an MR compatibel eye tracker system (Magconcept, USA). All fMRI images were acquired using BOLD sequences, GRE-EPI with TR= 2s, TE= 30s, with 3mm isotropic voxels. Response time and choice during fMRI experiment were recorded (Zurc&Zurc, Brazil). Image analysis of fMRI was performed using FSL 4.1 software package (Oxford University, England) including: motion correction, spatial filter, removal of artifacts and image transformation for the common space. Brain maps representing group effect (using the linear regression model) where threshold using a Z > 2.3 at cluster level and p < 0.05 (corrected for multiple comparisons). Reaction times were shorter when the S2 appeared in the valid condition compared with the invalid condition. Automatic and voluntary attention brain activation maps overlaped in frontal areas: precentral, middle frontal, medial frontal gyrus and right inferior frontal gyri; parietal areas: edges of intraparietal sulcus, postcentral, angular, supramarginal gyrus and pre-cuneus; temporal areas: middle temporal gyrus and left superior temporal gyrus; and occipital areas: fusiform gyrus and right lingual gyrus. Other areas was observed in anterior and posterior cingulate, parahippocampal gyrus, insulas, superior and inferior colliculus, and cerebellum. Superior frontal gyrus showed increased response in the voluntary attention when compared with automatic attention. We conclude that the frontoparietal networks were detected in both the mobilization of automatic and voluntary visuospatial attention. However the superior frontal gyrus showed response only in the voluntary attention. The distinction between these two subsystems involved in orientation and reorientation of automatic and voluntary attention.

La capacité de coordonner efficacement nos yeux avec nos mains est déterminante pour le succès de nos actions quotidiennes. En outre la capacité de prédire les conséquences sensorielles de nos propres actions est cruciale pour nos habilités motrices. Dans ce travail, à l’aide d’une tâche dans laquelle les participants doivent suivre avec leurs yeux une cible visuelle bougée par leur main, nous nous intéressons aux mécanismes prédictifs sous-tendant la coordination œil-main. Dans une première étude utilisant un protocole d’adaptation à une rotation visuomotrice, nous montrons que ces mécanismes prédictifs peuvent être mis à jour indépendamment de notre capacité à effectuer des mouvements précis de la main. Dans l’étude suivante nous cherchons à déterminer l’effet de la préférence manuelle, et montrons que malgré des différences évidentes en termes de précision concernant le contrôle manuel, la capacité d’anticiper les conséquences visuelles de nos actions reste identique que la cible soit bougée par la main droite ou gauche. Enfin, grâce à la stimulation magnétique transcranienne, nous testons l'hypothèse selon laquelle ces mécanismes prédictifs utilisent des signaux efférents de la main issus du cortex moteur primaire (M1). Nos résultats montrent que si cette contribution existe, elle doit se faire nécessairement en amont de M1. Au bout du compte nous proposons que la coordination œil-main soit sous-tendue par des mécanismes prédictifs similaires pour nos deux mains, situés vraisemblablement en amont de M1, et pouvant être mis à jour indépendamment du contrôle de la main. / The ability to coordinate efficiently eye and hand actions is central for humans in everyday activities. Furthermore it is argued that the ability to predict the sensory consequences of self-initiated movements is crucial for skilled motor behavior. Here by means of a task in which participants were asked to track with the eyes a visual target that was moved by their hand, we investigated the predictive mechanisms underlying eye-hand coordination. In a first study, using a protocol in which participants had to adapt to rotated hand visual feedback, we show that these predictive mechanisms can be updated independently of the ability to perform accurate hand movements. In a follow up study we tested the effect of hand dominance, and showed that, despite obvious differences in the accuracy of hand movement control, the ability to predict visual consequences of right and left hand actions was similar. Finally, by means of transcranial magnetic stimulation, we tested the hypothesis that those predictive mechanisms rely on hand efferent signals from the primary motor cortex (M1). However our results failed to support this view, and instead suggest that if such a contribution exists, it must be upstream of M1. Overall, we propose that eye-hand coordination relies on similar predictive mechanisms for both hands, possibly located upstream of M1, which can be updated independently of hand movement control.

Un défi majeur pour les organismes vivants est leur capacité d’adapter constamment leurs comportements moteurs. Dans la première étude de cette thèse, nous avons étudié le rôle des régularités statistiques et du conditionnement opérant sur la poursuite lisse d’anticipation (PLA). Nous avons démontré que la PLA est générée de manière cohérente avec la probabilité du mouvement attendu. De plus, nous avons montré pour la première fois que la PLA peut être modulée par les contingences de renforcement.Dans une seconde étude, nous avons créé un paradigme de poursuite, inspiré par l’Iowa Gambling Task (IGT), impliquant un choix entre deux cibles associées à différentes récompenses. Nous avons testé ce paradigme chez des patients Parkinson (PP), ainsi que des contrôles âgés ou jeunes. Seulement chez les participants jeunes, la latence du choix oculomoteur est fortement réduite quand celui-ci est associé à une règle de renforcement. Pour les PP le choix est fortement retardé dans toutes les conditions et cet effet n’est pas simplement attribuable à un déficit moteur. Autrement, la stratégie de choix s’avère de mauvaise qualité dans tous les groupes suggérant des différences avec les résultats de l’IGT classique.La dernière contribution de cette thèse fut de modéliser l’effet du biais directionnel sur la PLA que nous avons observé dans la première étude. Nous avons testé deux modèles incluant une mémoire de type intégrateur à fuite de la séquence d’essais, ou l’estimation Bayesienne adaptative de la taille optimale de mémoire. Nos résultats suggèrent que les modèles adaptatifs pourraient contribuer dans le futur à mieux comprendre l’apprentissage statistique et par renforcement. / One of the main challenges for living organisms is the ability to constantly adapt their motor behavior. In the first study of this thesis, we investigated the role of statistical regularities and operant conditioning on anticipatory smooth eye movements (aSPEM), in a large set of healthy participants. We provided evidence that aSPEM are generated coherently with the expected probability of motion direction. Furthermore, by manipulating reinforcement contingencies our findings demonstrated for the first time that aSPEM can be considered an operant behavior. In a second study, we designed a novel two-targets choice-tracking task, where a choice-contingent reward was returned, inspired by Iowa Gambling Task (IGT). We administered this new paradigm to Parkinson’s disease (PD) patients as well as age-matched control participants and young adult controls. For young participants, choice latency was clearly shortened in the IGT-pursuit task compared to the control-task. For PD patients choice latency was overall delayed and this difference could not be attributed to pure motor deficits. Overall the choice strategy performance was poor in all groups suggesting some possible differences between the standard IGT task and our IGT-pursuit task in probing decision-making. The last contribution of this thesis is an attempt to model the relation between aSPEM velocity and local direction-bias. Two models were tested to account for the trial-sequence effects, including either a decaying memory, or a Bayesian adaptive estimation of the efficient memory size. Our results suggest that adaptive models could be used in the future to better assess statistical and reinforcement learning.

In der vorliegenden Habilitationsschrift wurde eine Serie von Studien zusammengefasst, die menschliches räumliches Arbeitsgedächtnis, den "Visuospatialen Skizzenblock", untersucht haben. Dieses Kurzzeitgedächtnissystem ist häufig im Rahmen von Erkrankungen des frontalen Kortex und seiner mit ihm verbundenen Hirnareale, z.B. dem Morbus Parkinson, dem Morbus Alzheimer oder der Schizophrenie, beeinträchtigt und für einen relevanten Teil der kognitiven Defizite dieser Patienten verantwortlich. Wir untersuchten sowohl Gesunde als auch Patienten mit fokalen Läsionen des Gehirns mit Varianten des "Gedächtnissakkaden"-Paradigmas, einem etablierten okulomotorischen Verfahren zur Untersuchung von Raumgedächtnis. Es wurden sowohl behaviorale Aspekte von Arbeitsgedächtnis als auch mögliche anatomische Substrate dieses Gedächtnissystems sowie zeitstabilerer "Langzeit"-Gedächtnissysteme untersucht. Ziel war es, klarere Korrelationen zwischen messbarem Verhalten einerseits und Anatomie/Physiologie von Raumgedächtnis andererseits zu etablieren. Wir konnten erstmals zeigen, dass menschliches räumliches Arbeitsgedächtnis selektiv für aktuelles Verhalten relevante Wahrnehmungsinhalte repräsentiert. Der Zugang verhaltensirrelevanter Rauminformationen zu räumlichem Arbeitsgedächtnis wird offenbar durch effiziente (Aufmerksamkeits-) Filtermechanismen verhindert. Für die Existenz solcher Filtermechanismen gab es bislang nur elektrophysiologische Belege im Tiermodell. Da die Speicherkapazität von Arbeitsgedächtnis gering ist, erlauben diese Filtermechanismen möglicherweise einen effizienteren Umgang mit der Fülle und Komplexität unserer Umwelt. Umgekehrt lässt die in unserem Experiment sichtbar gewordene enge Verzahnung von Arbeitsgedächtnis und Aufmerksamkeit die Hypothese zu, dass eine gestörte Arbeitsgedächtniskapazität sowohl durch eine primäre Beeinträchtigung der Speichermechanismen selbst als auch durch Störungen der attentionalen Kontrolle derselben zustande kommen kann. Des weiteren konnten wir erstmals zeigen, dass menschliches räumliches Arbeitsgedächtnis eine klare Zeitgrenze hat, die für einzelne räumliche items bei ungefähr 20 Sekunden liegt. Jenseits dieser Zeitgrenze scheint eine vom Arbeitsgedächtnis unabhängige Raumrepräsentation für menschliches Verhalten bedeutsam zu werden. Der Begriff "Arbeitsgedächtnis" sollte also für Gedächtnisaufgaben reserviert bleiben, deren Gedächtnisphase 20 Sekunden nicht überschreitet. Unsere Befunde zeigen weiterhin, dass bei ansonsten konstantem Design einer Gedächtnisaufgabe, die Dauer der Gedächtnisphase bereits wesentlich darüber entscheidet, welches Gedächtnissystem untersucht wird. Die von uns durchgeführten Läsionsstudien an Patienten und neurophysiologischen Studien an Gesunden bestätigen, dass räumliches Arbeitsgedächtnis durch ein Netzwerk kortikaler Areale kontrolliert wird, das unter anderem den Dorsolateralen Präfrontalen Kortex, den Posterioren Parietalen Kortex und das Frontale Augenfeld umfasst. Innerhalb dieses Netzwerks nehmen diese Areale jedoch klar verschiedene kognitive Partialfunktionen wahr. Der Dorsolaterale Präfrontale Kortex und der Posteriore Parietale Kortex scheinen in erster Linie der Repräsentation von Raum in perzeptuellen Koordinaten, d.h. einem räumlichen "Wahrnehmungsbild" zu dienen, mit einer nur kurzfristigen Rolle des Posterioren Parietalen Kortex und einer dominierenden Rolle des Dorsolateralen Präfrontalen Kortex während der Gedächtnisphase einer Arbeitsgedächtnisaufgabe. Das Frontale Augenfeld scheint der Repräsentation von Raum in okulomotorischen Koordinaten zu dienen, d.h. der kurzzeitigen Speicherung einer geplanten okulomotorischen Antwort auf einen räumlichen Wahrnehmungsinhalt. Schließlich sprechen unsere Ergebnisse dafür, dass es mit dem Wechsel von Arbeitsgedächtnis zu einer zeitstabileren Raumrepräsentation bei Gedächtnisphasen von mehr als 20 Sekunden Länge auch zu einem Wechsel der anatomischen Substrate von Raumgedächtnis kommt. Die von uns durchgeführten Läsionsstudien zeigen, dass jenseits der Zeitgrenzen von räumlichem Arbeitsgedächtnis neokortikale Areale des Medialen Temporallappens eine aktive Rolle für Raumgedächtnis spielen. Hier konnten wir erstmals zeigen, dass der menschliche Parahippokampale Kortex eigenständige und vom Hippokampus unabhängige Raumgedächtnisfunktionen wahrnimmt. Möglicherweise ist diese Region das Substrat eines intermediären Gedächtnissystems zwischen räumlichem Arbeitsgedächtnis und Hippokampus-abhängigem Langzeitgedächtnis. Es wird ferner deutlich, dass in einer Gedächtnisaufgabe allein durch die Wahl verschiedener Dauern der Gedächtnisphase verschiedene anatomische Substrate von Gedächtnis untersucht werden können. Die in dieser Habilitationsschrift zusammengefassten Studien zeigen am Beispiel des räumlichen Arbeitsgedächtnisses, dass es möglich ist, mit einfachen physiologischen Paradigmen Gedächtnissysteme am Menschen zu untersuchen. Bestimmte mnestische Subfunktionen lassen sich mit den hier verwandten Paradigmen präzise quantifizieren und bestimmten Hirnregionen zuordnen. Wir glauben, dass dieser methodische Ansatz sowohl eine präzisere Diagnostik von kognitiven Defiziten bei Hirnerkrankungen erlaubt, als auch die Möglichkeit eröffnet, die Therapie von Gedächtnisstörungen effektiv zu kontrollieren. / This publication summarizes a series of experimental studies examining spatial working memory, the "visuospatial scratch pad", in humans. This short-term memory system is frequently affected in disorders involving the frontal cortex and connected subcortical structures, e.g. in Parkinson's disease, Alzheimer's disease or schizophrenia. Healthy human subjects and patients with focal cerebral lesions were tested with a series of "memory-guided saccade" paradigms, i.e. oculomotor spatial memory tasks. We examined both behavioural aspects and possible anatomical substrates of spatial working memory and more stable "long-term" memory systems. Our aim was to clarify the relationship between behavioural measures of spatial memory and its neuronal substrates. In a first experiment, we were able to show that visuospatial working memory selectively represents behaviourally relevant information. Access of irrelevant visuospatial information to working memory appears to be prevented by efficient attentional filters. Facing the limited storage capacity of spatial working memory, these filters may allow for successful behaviour in perceptually complex environments. Furthermore, the tight coupling of spatial attention and spatial working memory allows for the conclusion, that spatial working memory deficits in patients may likewise result from deficient storage systems and deficient attentional control. In a second experiment, we were able to demonstrate a clear temporal limit of about 20 seconds for spatial working memory. Beyond this temporal limit, an independent and more stable spatial memory system, less susceptible to the passage of time, becomes behaviourally relevant. Thus, the term "working memory" should be confined to spatial memory tasks where the memory delay does not exceed 20 seconds. In addition, these results show that selection of a certain memory delay in a given spatial memory task is a decisive factor when examining spatial memory systems. A third series of lesion studies in patients and neurophysiological experiments in healthy subjects confirmed that cortical control of spatial working memory involves dorsolateral prefrontal cortex, posterior parietal cortex and frontal eye field. Within this network, the dorsolateral prefrontal cortex and posterior parietal cortex appear to store spatial information in perceptual coordinates, with a transient role of the posterior parietal cortex at the very beginning of the memory delay and a dominating role of the dorsolateral prefrontal cortex for most of the delay. By contrast, the frontal eye field appears to store spatial information in oculomotor coordinates, i.e. to maintain a prepared eye movement to a remembered target location across a delay. A fourth series of lesion studies in patients showed that spatial memory for delays longer than 20 seconds is controlled by anatomical substrates distinct from those controlling spatial working memory. Beyond the temporal limits of spatial working memory, neocortical regions of the medial temporal lobe appear to contribute significantly to spatial memory. Within these neocortical regions, the parahippocampal cortex may carry spatial memory functions independent of the hippocampal formation and distinct from spatial working memory. We propose that this region is the neuronal substrate of an intermediate memory system, linking spatial working memory and spatial long-term memory both functionally and anatomically. Moreover, these results show that selection of a certain memory delay in a given spatial memory task is a decisive factor when examining neuronal substrates of spatial memory. Taken together, our experiments show that human memory can effectively be investigated with simple physiological paradigms. Spatial memory functions can precisely be quantified with oculomotor paradigms and related to defined anatomical substrates. This approach may allow for precise diagnosis of cognitive deficits and efficient monitoring of treatment of memory disorders.

OBJETIVO: Identificar áreas cerebrais ativadas durante tarefas de movimento ocular sacádico em crianças e adultos saudáveis. Por meio de registro de imagem com Ressonância Magnética funcional (RMf) e movimentos oculares com Eye tracking comparou-se a ativação neural em tarefas de sacadas guiadas externamente por um alvo e internamente pelo conhecimento prévio das características do movimento. MÉTODOS: Participaram do estudo 14 crianças saudáveis (7 meninos e meninas) com idade média de 11,1 (dp=1,2) e 14 adultos (9 homens e 5 mulheres) saudáveis com idade média de 23,9 (dp=3,3), todos destros. O paradigma de RMf incluiu tarefas de movimento sacádico em direção a um alvo. Foram apresentadas quatro condições: Previsível (P), Posição previsível (PP), Tempo previsível (TP) e Imprevisível (I). Durante a condição P, o alvo apareceu na tela em intervalos fixos de tempo (800 ms) e posição fixa, alternando entre dois marcos na linha horizontal. Na condição PP, a posição era fixa, o intervalo variado (400, 800 e 1200 ms). Na condição TP, o intervalo era fixo e o alvo aparecia em 5 posições no meridiano horizontal. Na condição I, posição e tempo eram variados. As aquisições BOLD foram realizadas em um sistema 3T (Philips Achieva) com uso integrado de Eye tracker (MagConcept), Tempo de repetição foi de 2 s, Tempo de eco 30 s, voxels isotrópicos de 3 mm. A análise de imagem incluiu correção de movimentos, filtro temporal, normalização para espaço comum, remoção de artefatos e inferência paramétrica (FSL versão 4.1 - FMRIB, Analysis Group, Oxford, UK). RESULTADOS: Análise dos dados de movimento ocular de crianças e adultos mostrou um padrão similar de latência sacádica com aumento da mesma da condição P para PP e TP. Apenas em adultos foi encontrada diferença nas condições TP e I. Adultos mostraram latência sacádica significativamente menor que crianças. Na condição P, em adultos, foi observada redução contínua da latência ao longo do bloco, já em crianças a latência se estabilizou. Em relação à ativação neural, em ambos grupos foi encontrado efeito BOLD nas áreas do campo visual frontal, campo visual suplementar, lobo parietal superior e inferior, córtex visual primário, V5, ínsula anterior, núcleos da base e cerebelo. Em crianças foi observada ativação mais difusa no córtex frontal. Na comparação direta dos grupos, crianças mostraram efeito BOLD no giro do cíngulo anterior e posterior. Correlação entre latência e ativação em crianças foi encontrada no giro do cíngulo anterior, giro frontal superior medial, giro orbitofrontal e giro temporal inferior. Em adultos, a correlação foi encontrada nas áreas comuns do circuito sacádico. CONCLUSÕES: crianças mostraram um padrão sacádico similar aos adultos, porém com menor aproveitamento da informação antecipada do alvo. O padrão de ativação em crianças diferiu daquele encontrado em adultos. Crianças ativaram mais na linha fronto-parietal medial e orbitofrontal, o que pode ser atribuído à diferente modulação dos mecanismos inibitórios e atencionais / OBJECTIVE: identify brain activation patterns in children and adults during predictive saccade task. We used functional Magnetic Resonance Imaging (fMRI) and Eye tracking to compare neural activation during saccades guided by external stimuli and internal monitoring of target movement properties. METHODS: the total of 11 children was assessed (9 boys and girls) with the mean age of 11.1 (SD = 1.2) and 14 adults (11 men and 5 women) with the mean age of 23.9 (SD=3.3). The fMRI paradigm was composed of different tasks for saccades generation. In total, four conditions were presented: Predictable (P), Position predictable (PP), Time predictable (TP) and Reflexive saccades (R). In P task, the target appeared in constants frequency oscillating between two positions. In PP, the target moved between constant positions with unpredictable timing (400, 800, 1200 ms). In TP, the frequency was constant but the target appeared in 5 different positions on horizontal meridian. In R, frequency and position were varied. The BOLD acquisitions were performed on 3T scanner (Philips Achieva) with integrated MRI compatible Eye tracker (Magconcept). Repetition time was 2 s, echo time 30 s and isotropic voxels 3 mm. Image analyses included a motion correction, time correction, artifacts removal, transformation of image into standard space and parametric analyses with FSL version 4.1 (FMRIB, Analysis Group, Oxford, UK). RESULTS: data analyses showed similar pattern of saccade latency in children and adults, with an increase in latency from P to PP and TP. The TP and R showed different latencies only in adults. Adults had significantly shorter latencies than children. In the P task, continuous reduction in latency was found along the block in adults only. The children showed latency stabilization. Regarding neural activation common to the both groups, the BOLD effect was found in the frontal eye field, supplementary eye field, superior and inferior parietal lobe, primary visual cortex, V5, anterior insula, basal ganglia and cerebellum. In children, diffuse activation was found in frontal cortex. In direct group comparison, the children showed BOLD effect in anterior and posterior cingulate gyrus. Correlations between saccade latency and activation were found in the anterior cingulate, frontal medial superior gyrus, orbitofrontal gyrus and inferior temporal gyrus. In adults, correlations were found in common saccadic circuitry. CONCLUSIONS: children showed a similar pattern of eye movements to adults, but with less efficiency in saccades guided by advanced knowledge of target movement. The activation patter differed between children and adults. Children activated more on medial line of frontoparietal cortex and orbitofrontal cortex. We attribute this pattern to the different processes of inhibitory and attention mechanisms in children

OBJETIVO: Identificar áreas cerebrais ativadas durante tarefas de movimento ocular sacádico em crianças e adultos saudáveis. Por meio de registro de imagem com Ressonância Magnética funcional (RMf) e movimentos oculares com Eye tracking comparou-se a ativação neural em tarefas de sacadas guiadas externamente por um alvo e internamente pelo conhecimento prévio das características do movimento. MÉTODOS: Participaram do estudo 14 crianças saudáveis (7 meninos e meninas) com idade média de 11,1 (dp=1,2) e 14 adultos (9 homens e 5 mulheres) saudáveis com idade média de 23,9 (dp=3,3), todos destros. O paradigma de RMf incluiu tarefas de movimento sacádico em direção a um alvo. Foram apresentadas quatro condições: Previsível (P), Posição previsível (PP), Tempo previsível (TP) e Imprevisível (I). Durante a condição P, o alvo apareceu na tela em intervalos fixos de tempo (800 ms) e posição fixa, alternando entre dois marcos na linha horizontal. Na condição PP, a posição era fixa, o intervalo variado (400, 800 e 1200 ms). Na condição TP, o intervalo era fixo e o alvo aparecia em 5 posições no meridiano horizontal. Na condição I, posição e tempo eram variados. As aquisições BOLD foram realizadas em um sistema 3T (Philips Achieva) com uso integrado de Eye tracker (MagConcept), Tempo de repetição foi de 2 s, Tempo de eco 30 s, voxels isotrópicos de 3 mm. A análise de imagem incluiu correção de movimentos, filtro temporal, normalização para espaço comum, remoção de artefatos e inferência paramétrica (FSL versão 4.1 - FMRIB, Analysis Group, Oxford, UK). RESULTADOS: Análise dos dados de movimento ocular de crianças e adultos mostrou um padrão similar de latência sacádica com aumento da mesma da condição P para PP e TP. Apenas em adultos foi encontrada diferença nas condições TP e I. Adultos mostraram latência sacádica significativamente menor que crianças. Na condição P, em adultos, foi observada redução contínua da latência ao longo do bloco, já em crianças a latência se estabilizou. Em relação à ativação neural, em ambos grupos foi encontrado efeito BOLD nas áreas do campo visual frontal, campo visual suplementar, lobo parietal superior e inferior, córtex visual primário, V5, ínsula anterior, núcleos da base e cerebelo. Em crianças foi observada ativação mais difusa no córtex frontal. Na comparação direta dos grupos, crianças mostraram efeito BOLD no giro do cíngulo anterior e posterior. Correlação entre latência e ativação em crianças foi encontrada no giro do cíngulo anterior, giro frontal superior medial, giro orbitofrontal e giro temporal inferior. Em adultos, a correlação foi encontrada nas áreas comuns do circuito sacádico. CONCLUSÕES: crianças mostraram um padrão sacádico similar aos adultos, porém com menor aproveitamento da informação antecipada do alvo. O padrão de ativação em crianças diferiu daquele encontrado em adultos. Crianças ativaram mais na linha fronto-parietal medial e orbitofrontal, o que pode ser atribuído à diferente modulação dos mecanismos inibitórios e atencionais / OBJECTIVE: identify brain activation patterns in children and adults during predictive saccade task. We used functional Magnetic Resonance Imaging (fMRI) and Eye tracking to compare neural activation during saccades guided by external stimuli and internal monitoring of target movement properties. METHODS: the total of 11 children was assessed (9 boys and girls) with the mean age of 11.1 (SD = 1.2) and 14 adults (11 men and 5 women) with the mean age of 23.9 (SD=3.3). The fMRI paradigm was composed of different tasks for saccades generation. In total, four conditions were presented: Predictable (P), Position predictable (PP), Time predictable (TP) and Reflexive saccades (R). In P task, the target appeared in constants frequency oscillating between two positions. In PP, the target moved between constant positions with unpredictable timing (400, 800, 1200 ms). In TP, the frequency was constant but the target appeared in 5 different positions on horizontal meridian. In R, frequency and position were varied. The BOLD acquisitions were performed on 3T scanner (Philips Achieva) with integrated MRI compatible Eye tracker (Magconcept). Repetition time was 2 s, echo time 30 s and isotropic voxels 3 mm. Image analyses included a motion correction, time correction, artifacts removal, transformation of image into standard space and parametric analyses with FSL version 4.1 (FMRIB, Analysis Group, Oxford, UK). RESULTS: data analyses showed similar pattern of saccade latency in children and adults, with an increase in latency from P to PP and TP. The TP and R showed different latencies only in adults. Adults had significantly shorter latencies than children. In the P task, continuous reduction in latency was found along the block in adults only. The children showed latency stabilization. Regarding neural activation common to the both groups, the BOLD effect was found in the frontal eye field, supplementary eye field, superior and inferior parietal lobe, primary visual cortex, V5, anterior insula, basal ganglia and cerebellum. In children, diffuse activation was found in frontal cortex. In direct group comparison, the children showed BOLD effect in anterior and posterior cingulate gyrus. Correlations between saccade latency and activation were found in the anterior cingulate, frontal medial superior gyrus, orbitofrontal gyrus and inferior temporal gyrus. In adults, correlations were found in common saccadic circuitry. CONCLUSIONS: children showed a similar pattern of eye movements to adults, but with less efficiency in saccades guided by advanced knowledge of target movement. The activation patter differed between children and adults. Children activated more on medial line of frontoparietal cortex and orbitofrontal cortex. We attribute this pattern to the different processes of inhibitory and attention mechanisms in children

O presente estudo teve por objetivo investigar a dinâmica temporal da percepção de tamanho de objetos inseridos em gradientes de textura, provenientes de desenho gerado computacionalmente (gradiente de perspectiva) e de fotografia com elementos naturais (gradiente fotografado). Além disso, verificou-se a ocorrência e padrão dos movimentos oculares nestas condições. Para isso, foram realizados 2 experimentos. No primeiro experimento, dois círculos negros eram apresentados no meridiano vertical e a tarefa dos 96 participantes foi indicar em qual parte do campo visual, superior ou inferior, foi apresentado o maior círculo. Usando o método das escadas duplas, estes círculos eram apresentados brevemente (50, 100, 150 ou 200 ms) em três condições de fundos de tela (sem textura, gradiente de perspectiva ou gradiente fotografado). Foram calculados a inclinação da função psicométrica e o ponto de igualdade subjetivo (PIS) como medidas da sensibilidade discriminativa e da distorção perceptiva, respectivamente. Os resultados mostraram que os participantes discriminaram melhor o tamanho dos estímulos à medida que o tempo de exposição aumentava, independentemente do gradiente de textura. A análise do PIS revelou que há uma forte distorção do tamanho aos 100 ms para o gradiente de perspectiva e aos 150 ms para o gradiente fotografado. No segundo experimento, 24 participantes realizaram uma tarefa semelhante ao do primeiro experimento e utilizou-se um eye tracker para registrar o movimento ocular. Os resultados mostraram que, em apresentações de 150 e 200 ms, os movimentos oculares ocorreram em menos de 10% das tentativas em todos os fundos de tela. Isto indica que a redução das distorções de tamanho após os 100 ms observada no primeiro experimento não é explicada completamente pela alocação da atenção. Tomados em conjunto, estes resultados sugerem que o processamento e integração das informações de tamanho e profundidade são mediados principalmente por mecanismos bottom-up. Ademais, em uma condição de menor restrição temporal (1500 ms), a presença de informações de profundidade afetou a dimensão espacial das fixações e sacadas, mas não a dimensão temporal. Os participantes olharam preferencialmente para o estímulo apresentado no campo visual superior, o que indica que a atenção foi alocada às posições de maior profundidade. Os resultados também indicaram que os estímulos maiores capturaram tanto a atenção transitória quanto a atenção sustentada. Assim, este estudo permitiu um maior conhecimento sobre os mecanismos perceptivos e atentivos envolvidos no processamento das informações de tamanho e de profundidade. / This study aimed to investigate the temporal dynamics of object size perception under texture gradients from line-drawing (perspective gradient) and photograph with natural elements (photographed gradient). Additionally, it was analyzed the occurrence and the pattern of eye movements in these conditions. For this, we carried out 2 experiments. In the first experiment, two black circles were displayed in the vertical meridian and 96 participants reported whether the bigger one was presented at the lower or upper visual field. By using double staircase psychophysics method, these circles were briefly presented (50, 100, 150 or 200 ms) in three background conditions (no texture, perspective gradient or photographed gradient). The slope of the psychometric function and the points of subjectivity equality (PSE) were calculated and used as discrimination sensitivity and size distortion measures, respectively. The results showed a greater sensitivity to size discrimination as exposure time increased, regardless the texture gradient. The analysis of PSE indicated greater size distortions in perspective gradient at 100 ms and in photographed gradient at 150 ms. In the second experiment, 24 participants performed a task similar to that in Experiment 1 while an eye tracker recorded their eye-movements. The results showed that, at 150 and 200 ms, eye movements occurred in less than 10% of trials in all backgrounds. It indicates that the size distortion reduction after 100 ms observed in the first experiment cannot be fully explained by attention allocation. Taken together, these results suggest that the process and integration of size and depth information are mainly mediated by bottom-up mechanism. Moreover, in a less time constrain condition (1500 ms), depth cues affected the spatial measures of fixations and saccades, but not the temporal ones. The participants looked more frequently at the stimulus on the upper visual field, indicating that attention has been allocated to positions of greater depth. The results also indicated that large stimulus capture both transient and sustained attention. Thus, this study provides a better understanding of perceptual and attentive mechanisms involved in size and depth process.

在閱讀文句的過程中，隨著文字內容的不同語言特性，眼睛凝視位置與時間也會有所不同；而使用眼動儀(eye tracker)可以精確記錄眼睛凝視的位置和停留時間，研究者可藉此瞭解閱讀歷程中不同的訊息處理機制。此外，眼動誘發呈現(eye movement contingent display)技術精確且快速的刺激呈現控制，能根據眼球凝視的位置，在特定時間與特定位置立即更換螢幕呈現的文字，如此對於閱讀歷程中訊息處理的動態運作，能夠提供更準確的掌握。實驗一以移動視窗的方式，測量中文閱讀的知覺廣度，發現眼睛每次凝視能知覺並處理的文字範圍，包括凝視點及右側四個中文字。實驗二亦採用移動視窗的方式，並操弄視窗週邊文字遮蔽的時間，發現週邊文字在凝視開始的90毫秒內即被提取；週邊文字遮蔽時間的長短不會影響凝視時間，而是影響在下個凝視點訊息整合之凝視時間。實驗三以邊界技術的方式，探討週邊文字被提取的特性，操弄預視字與目標字在字形與字音的相似或相異；結果顯示在一定的條件下，能獲得字音訊息在早期階段被提取之證據，而字形訊息在整合不同訊息的階段產生明顯的影響。實驗四則以快速促發作業的方式，探討凝視位置促發字的文字特性對目標字訊息整合的影響；結果顯示促發字呈現時間較短的情況下，字音效果只在形異的情況下產生，同樣獲得字音訊息在早期階段被提取之證據。本論文應用所發展的眼動誘發呈現技術進行一系列實驗，分別就中文閱讀的知覺廣度、凝視點對於週邊訊息之處理時序、以及週邊與凝視點位置對文字提取之特性和訊息整合機制加以探討，藉此推論中文閱讀時眼動所引發的動態文字處理歷程。 第一章　眼球運動與閱讀歷程的關係 I 第二章 閱讀的眼動控制之運作與影響因素 10 第三章 中文閱讀的眼動研究回顧 20 研究目的 30 第四章 眼動誘發呈現技術的發展 31 第五章 中文閱讀的知覺廣度與呈現改變的影響 34 實驗一 39 第六章 凝視點對週邊訊息的提取時間 45 實驗二 48 第七章 凝視點對週邊文字提取的特性 54 實驗三 55 第八章 凝視點上文字的語音提取 62 實驗四Ａ 64 實驗四Ｂ 65 綜合討論 68 參考文獻 76 附錄一：實驗一與實驗二使用的文章材料 109 附錄二：實驗三與實驗四的句子與操弄單字 113 / In the process of reading, the location and duration of eye fixations varied accroding the language properties of the text. Using the eye tracker nowdays, it precisely records the location and fixation duration of eye movements. That gives an opportunity to understand the cognitive operations of reading. Furthermore, the eye movement contingent display technique is capable to change the display with both the speed and precision. It changes the text on the computer screen at specific location and time, according the location where the eyes are fixating. This advanced technique provides the possibility to explore the nature of dynamic processing of reading. Experiment One uses the moving window paradigm to measure the perceptual span of Chinese reading. It showed that the range of perceived characters in reading includes four characters to the right of fixations. Experiment Two also uses the moving window paradigm but the time of showing "window" is varied. It showed that the characters in peripheral were perceived within 90 msec in the beginning of fixations. The time to mask the peripheral text did not affect the fixation duration of display change, but the time of next fixation. Experiment Three uses the boundary paradigm to show the type of character information can be retrieved in peripheral and integrated with the information on the next fixation. It showed the phonological code can be retrieved in peripheral, but the orthographic code is dominated when integrating information across fixations. Experiment Four uses the fast priming paradigm to show the type of information can be retrieved on fixations. It showed the same pattern for both the third and forth experiment. The phonological effect can be obtained under certain circumstances, when the availability or validity of orthographic information is reduced. The dynamic processing of eye movements in reading and a general model for cognitive operation are discussed.

本論文進行兩個眼動實驗，探討語境及多義特性在一詞多義處理歷程中扮演的角色。Frazier與Rayner(1990)認為一詞多義由於詞義間有重疊的語義屬性(sense overlap)，因此處理詞義之初，會先以部分語義(partial specification)來理解，不需立即進行完整詞義的解讀，支持部分語義解讀假設(Immediate Partial Interpretation Hypothesis)。在Frisson 與Pickering(2001)所提出詞義未定模型(Underspecification Model)中，進一步說明語境在一詞多義中扮演的角色，認為一詞多義的語義提取會先以未定意義(underspecified meaning)解釋，在語義提取完畢後，語境才會介入詞義的選擇。本論文操弄語境導引和詞義重疊性，來探討一詞多義的詞義提取與選擇歷程。實驗一操弄句子中目標詞的語境預測力(可預測語境vs.不可預測語境)與詞義數量(單詞義vs.多詞義)，記錄並分析目標詞與目標詞後區域的凝視時間。結果顯示：目標詞上的凝視時間多詞義顯著短於單詞義，支持部分語義解讀假設。然而，實驗一在目標詞後區域並未看到多詞義因延遲語義解讀(delayed semantic commitments)而造成抑制或競爭效果。實驗二就此現象與「一詞多義本身的詞義重疊程度」與「語境對一詞多義有無發生語義解讀的必要性」進行進一步探討。實驗二(同實驗一)操弄目標詞的語境預測力，與詞義的重疊程度(單義詞vs.中度詞義重疊vs.高度詞義重疊)，並控制語境偏向次要詞義的情形。結果顯示：目標詞上的晚期眼動指標在中度詞義重疊(中vs.單)與高度詞義重疊(高vs.單)皆看到與語境預測力發生交互作用，顯示語境預測力的影響在晚期階段發生；目標詞後區域觀察到中度與高度詞義重疊(中vs.單；高vs.單)皆在不可預測語境下發生詞義抑制效果，符合詞義未定模型預期之延遲語義解讀效果，並在目標詞晚期眼動指標看到中度詞義重疊效果與高度詞義重疊效果(單-中 vs. 單-高)受語境影響有不同的效果，顯示詞義重疊會影響詞義選擇的必要性。總結本論文的結果：首先，一詞多義的語義提取符合部分語義解讀的假設，而語境介入的影響支持詞義未定模型的看法，即語境在語義提取之初並未介入影響，在語義未定的情況下，若語境內容有進行語義解讀之必要時，則會發生延遲語義解讀。其次，詞義的促進效果與一詞多義的詞義數量多寡有關，詞義數量減少時，詞義促進效果也隨之消失。最後，當語境偏向一詞多義的次要詞義時，延遲語義效果才會發生。 / This thesis conducted two eye movement experiments with the aim to investigate the role of context and multi-sense feature in processing polysemy. Frazier and Rayner (1990) suggested that, at the beginning of semantic processing, polysemy is comprehended with partial specification. There is no immediate need to process the complete word sense due to the sense overlap between senses. This viewpoint supports the Immediate Partial Interpretation Hypothesis. In Frisson and Pickering’s (2001) Underspecification Model, further elaboration were made on the role of context in the process of retrieving and selecting one of the word senses of a polysemy. The sense that is first retrieved from a polysemy is considered to be an underspecified meaning. It is after the semantic retrieval is finished that context is involved in selecting a word sense. This thesis manipulated context guidance and sense overlap, to further research on processing polysemy in terms of word sense retrieval and selection. Experiment One manipulated contextual predictability (predictable context vs. unpredictable context) and number of senses (one-sense, monosemy vs. many-sense, polysemy) of the target words in sentences. Fixation times of the target words and post-target areas were recorded and analyzed. Results showed that the fixation times on target words were significantly shorter for polysemy than for monosemy, supporting the Immediate Partial Interpretation Hypothesis. However, in Experiment One, there is no inhibitory or competitive effect on the post-target area, indicating that there is no effect of delayed semantic commitments while comprehending polysemy. In Experiment Two, we further investigated how this phenomenon is connected with the degree of sense overlap and whether context is necessary to activate semantic commitment for polysemy. Contextual predictability of target words and the degree of sense overlap (monosemy vs. moderate-sense-overlap vs. high-sense-overlap) were manipulated, with the former designed as in Experiment One. Specifically, the context was controlled to bias toward the subordinate sense. The results showed that there were interactions of sense overlap degree (both moderate-sense-overlap vs. monosemy and high-sense-overlap vs. monosemy) and contextual predictability on target words for later-stage indices. This suggests that contextual predictability effects at later stages. On the post-target areas, there were inhibitory effects found for moderate-sense-overlap vs. monosemy and for high-sense-overlap when the context is unpredictable. This finding supports the delayed effect of semantic commitments in the Underspecification Model. Moreover, effects of sense overlap (polysemy-moderate vs. polysemy-high) were modulated by contextual predictability on target words for later-stage processing, showing that sense overlap affects the necessity of semantic commitments. In conclusion, the semantic retrieval of polysemy can be best explained by the Immediate Partial Interpretation Hypothesis and the involvement of contextual constraint supports the Underspecification Model. That is, context does not affect the beginning phase of semantic retrieval. Since the senses are underspecified, delayed semantic commitment occurs if it is necessary to make semantic commitment in the context. Furthermore, the facilitation of senses is related to the number of a polysemy’s senses. As number of senses decreases, facilitation of senses wanes and disappears. Finally, delayed semantic commitment occurs only when the context biases towards a subordinate sense of the polysemy.