Navigation through the environment requires the ability to select relevant information from a multitude of irrelevant stimuli. Under conditions of processing conflict, attention and cognitive control processes bias sensory input based on internal goals. These processes are supported by the interplay of a fronto-parietal attention network that exerts a top-down influence on information processing and a superior temporal network that operates in parallel and that responds in a stimulus-driven manner to behaviorally salient stimuli. It is often reported that nicotine can enhance top-down attention control and reduce distraction. In experiments 1 and 2, the effects of increasing control demands on behavior were assessed using electrophysiological (EEG) and behavioral measures in an auditory number parity decision task with different levels of distraction. Participants made forced choice ‘odd’ or ‘even’ number decisions, while ignoring preceding or simultaneous novel distractors. A group of non-smokers was compared to overnight abstinent smokers (9 hours) and after nicotine intake via 2 mg nicotine tablet or via smoke-inhaled nicotine. The results revealed that preceding distractors impaired task performance due to orienting to and reorienting from the distractor. Simultaneous distractors did not cause orientation of attention (indicated by absence of a P3a Event-Related Potential) and produced smaller increments in response latencies. However, this type of complex novel stimulus initiated processes of memory updating that significantly impaired response sensitivity and accuracy. Nicotine withdrawal enhanced these distraction effects, whereas nicotine intake, particularly via smoking, normalized performance. In experiment 3, dichotic listening performance in a group of non-smokers was compared to abstinent smokers (12 hours) using behavioral, EEG and functional Magnetic Resonance Imaging (fMRI) measures. The perceptual salience of the stimuli was manipulated by systematically varying the Inter-aural Intensity Difference (IID) between them. The analysis pointed to distinct brain networks that differentially activate depending on the level of competition between sensory inputs and these effects were additionally modulated by nicotine withdrawal. Nicotine withdrawal impaired behavioral performance supported by evidence of enhanced use of memory and attention resources, and some evidence of task-independent default mode network activation. Overall, the findings suggest that withdrawal from nicotine, particularly in heavy smokers, is associated with impairments in cognitive control and that subsequent intake of nicotine serves mainly to normalize performance.

The rapid advancements in the field of machine learning and artificial intelligence has led to the emergence of technologies like the Brain Computer Interface (BCI), which has revolutionized rehabilitation protocols. However, given the neural basis of BCIs and the dependence of its performance on cognitive factors, BCIs may be used to characterize the functional capacity of the user. A resting state segment can also be considered for characterization of the functional network integrity, creating a two part framework that probes the functional networks and their cognitive manifestations. This thesis explores such a two part framework using a simultaneous EEG-fMRI setup on a healthy population. The BCI accuracies for all subjects increased over the course of the scan and is thought to be due to learning processes on the subject's part. Since such learning processes require cognitive faculties such as attention and working memory, these factors might modulate the BCI performance profile, making it a potential metric for the integrity of such cognitive factors. The resting state analysis identified four EEG Microstates that have been previously found to be associated with verbal, visual, saliency and attention reorientation tasks. The proportion of each microstate that composed the corresponding fMRI resting state networks (RSN) were identified, opening up the potential for predicting fMRI-based RSN information, from EEG microstates alone. The developed protocol can be used to diagnose potential conditions that negatively affect the functional capacity of the user by using the results from this study as healthy control data. This is the first known BCI based system for characterization of the user's functional integrity, opening up the possibility of using BCIs as a metric for diagnosing a neuropathology. / Thesis / Master of Applied Science (MASc)

The main objective of this diploma thesis is to describe simultaneous electroencephalography-correlated functional magnetic resonance imaging analysis using EEG data processing. It includes basic characteristics of EEG and fMRI recording and analysis and their combination as simultaneous EEG/fMRI analysis and deals with obstructions during its processing. The thesis includes a design of an experiment used for recording and analysis of simultaneous EEG/fMRI data using EEG source reconstruction for regressor construction. Thesis incorporates a software solution used for extraction of signal describing a source activity interpolated by EEG source reconstruction. The signal is then processed and used to construct a basic regressor. The thesis also deals with the software solution being used for a study focused on intracranial epileptic discharges localization using a simultaneous EEG/fMRI analysis in which it reveals source activity during ongoing epileptic spike and summarizes the results.

The simultaneous acquisitions of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) have been applied to improve the surgery planning of patients with drug resistant epilepsy. However, the classical approach of analyzing the EEG-fMRI data is inefficient in patients whom only few or non interictal epileptiforms discharges (IEDs) are detected during the simultaneous acquisition. Another issue of EEG-fMRI acquisition is related to its high sensitivity to motion, which decreases the quality of both data, even worse in non-cooperative patients. In this work we propose and discuss the application of two methods of analyzing fMRI data of patients with focal epilepsy: Independent component analysis (ICA) and two-dimensional temporal clustering (2dTCA). Each method was applied in a distinct group of patients and the results were compared to those obtained by the classic EEG-fMRI analysis. We have also proposed a method to improve the quality of EEG data using the head position measurements obtained, by a prospective motion correction (PMC) system, during the EEG-fMRI acquisitions. In the ICA study, we have used the electrical source images for selecting independent components (ICs) in EEG data of 13 patients with different spiking frequency. The method detected epilepsy-related BOLD activity in all the patients. Comparatively, the classic EEG-fMRI could be applied in 11 patients and epilepsy-related BOLD activities were found in seven of them. In the 2dTCA study, we have evaluated 20 patients and found epilepsy-related maps in 14 of them. Thirteen of the twenty patients have IEDs detected during the simultaneous acquisition; the classic EEG-fMRI provided maps related to the epileptogenic region in six of them. Finally we have verified in three health subjects that the proposed method for correcting motion-induced artefacts in the EEG data is effective for high amplitude and velocities (~1cm and 55mm/s). We concluded that the ICA and 2dTCA methods increase the sensitivity of using fMRI for mapping the epileptogenic region, mainly in patients presenting few or no IEDs in the EEG data simultaneously acquired to the fMRI. The PMC use during the fMRI acquisition does not degrade the quality of the EEG data acquired simultaneously. In fact, the motion information can be used for improving its quality by correcting motion-induced artefacts. / As aquisições simultâneas de dados de eletroencefalografia (EEG) e imagens funcionais por ressonância magnética (fMRI) vêm sendo utilizadas com intuito de melhorar o planejamento cirúrgico de pacientes com epilepsia refratária. Entretanto, o processamento classicamente usado nestes dados combinados não é possível em pacientes sem descargas epileptiformes interictais (IEDs) e possui baixa sensibilidade para aqueles em que poucas IEDs são detectadas durante a aquisição simultânea. Além disto, a técnica é sensível ao movimento dos pacientes durante as aquisições, o que reduz a qualidade dos dados, principalmente em pacientes não cooperantes. Neste trabalho é proposto e discutido o uso de dois métodos de processamento, baseados nas técnicas de análise de componentes independentes (ICA) e análise temporal de clusters em duas dimensões (2dtca), para se mapear regiões epileptogênicas. Cada método foi analisado em um conjunto diferente de pacientes e os resultados foram comparados com os obtidos pelo EEG-fMRI clássico. Finalmente, propomos um método que utiliza às medidas de posicionamento da cabeça, obtidas durante a aquisição das fMRI, para aumentar a qualidade dos dados de EEG adquiridos simultaneamente. No estudo usando ICA combinado com imagens de fontes elétricas analisamos os dados de 13 pacientes com diferentes frequências de descargas e observamos que este método encontrou ao menos uma componente independente relacionada à epilepsia em cada paciente. Comparativamente usando o processamento convencional foi possível avaliar 11 dos 13 pacientes, e em apenas sete deles os mapas resultantes foram considerados concordantes com a região epileptogênica (RE). No estudo utilizando 2dTCA avaliamos 20 pacientes e encontramos mapas relacionados com a RE em 14 deles. Neste conjunto de pacientes, 13 apresentaram IEDs durante as aquisições; neles o método clássico de processamento teve resultados concordantes com a RE em seis deles. Finalmente verificamos em três sujeitos saudáveis que o método aqui proposto para corrigir os artefatos induzidos no EEG devido ao movimento é efetivo para altas amplitudes e velocidades (~1cm e 55mm/s). Concluímos que os métodos ICA e 2dTCA aumentam a sensibilidade do uso de fMRI para mapear RE, principalmente em pacientes com baixa ou nenhuma detecção de IEDs durante às aquisições. Também concluímos que o uso da correção prospectiva de movimento em aquisições de fMRI não reduz a qualidade do dado de EEG adquirido simultaneamente e que às informações de movimento mensuradas podem melhorar a qualidade deste dado em situações de repouso e movimento do sujeito durante o experimento.

A epilepsia do lobo temporal (ELT) é a forma mais comum de epilepsia e a mais resistente ao tratamento medicamentoso. Existem diversos tipos de drogas anti-epilépticas usadas no controle das crises. Entretanto, em alguns casos, esse tipo de tratamento não é eficaz e a cirurgia para remoção da zona epileptogênica (ZE) pode ser uma alternativa recomendada. A ZE é definida como aquela onde as crises são originadas. Trata-se de um conceito teórico e, atualmente, não existem técnicas capazes de delimitá-la precisamente. Na prática, exames de EEG, vídeo-EEG, MEG, SPECT, PET e diversas técnicas de MRI, em especial as funcionais, têm sido usados para mapear zonas relacionadas à ZE. Contudo, em alguns casos, os resultados permanecem não convergentes e a determinação da ZE inconclusiva. Desse modo, é evidente a importância do surgimento de novas metodologias para auxiliar a localização da ZE. Assim, pois, o objetivo deste trabalho foi desenvolver dois métodos para a avaliação da ZE, ambos baseados na imagem funcional por ressonância magnética. No primeiro, investigamos possíveis alterações da resposta hemodinâmica (HRF) quando da modulação da pressão parcial de CO2. Para tanto, fizemos um estudo sobre 22 pacientes com ELT e 10 voluntários assintomáticos modulando a pressão parcial de CO2 sanguíneo cerebral por um protocolo de manobra de pausa respiratória e outro de inalação passiva de CO2/ar. Os resultados mostram que o tempo de onset da HRF tende a ser maior e a amplitude da HRF tende a ser menor em áreas do lobo temporal de pacientes com ELT quando comparados com os dados de voluntários assintomáticos. Além disso, os resultados mostram mapas de onset individuais coincidentes com exames de SPECT ictal. O segundo estudo foi baseado em medidas de EEG-fMRI simultâneo. Neste, avaliamos a relação entres as potências dos ritmos cerebrais alfa e teta (EEG) e o contraste BOLD (fMRI) de 41 pacientes com ELT e 7 voluntários assintomáticos em estado de repouso. A análise da banda alfa mostrou correlações negativas nos lobos occipital, parietal e frontal tanto nos voluntários quanto nos pacientes com ELT. As correlações positivas nos voluntários foram dispersas e variáveis em ambos hemisférios cerebrais. Por outro lado, encontramos forte correlação positiva no tálamo e ínsula dos pacientes com ELT. Na análise da banda teta observamos correlações positivas bilaterais nos giros pré e pós central de voluntários. Ainda, foram observados clusters no cíngulo anterior, tálamo, ínsula, putamen, em regiões parietais superior, frontais e giros temporais. Também, utilizamos um cálculo de índice de lateralização (IL) no lobo temporal em confrontos entre pacientes com ELT à direita, pacientes com ELT à esquerda e voluntários assintomáticos. Verificamos que os ILs, utilizando os clusters obtidos nas análises em teta, foram coincidentes com o diagnóstico clínico prévio da localização da ZE em todas as análises dos grupos de pacientes com ELT à direita, e na maioria do grupo de pacientes com ELT à esquerda. De forma geral, verificamos que o método de hipercapnia se mostrou ferramenta interessante na localização da ZE comprovada pelos coincidentes achados pela avaliação de SPECT. Inferimos que o maior tempo de onset e menor amplitude da HRF observadas nos pacientes em relação a voluntários possam estar relacionados a um stress vascular devido à recorrência de crises. Já o método de ritmicidade alfa e teta proposto parece promissor para ser usado na determinação da lateralização da ZE em pacientes com ELT. / Temporal lobe epilepsy (TLE) is the most common and resistant form of epilepsy to anti-epileptic drug. There are several types of anti-epileptic drugs used in seizure control. However, in some cases drug treatment is not effective and surgery to remove the epileptogenic zone (EZ) is a recommended alternative. EZ is a theoretical concept and there are many techniques that have been applied to enclose it precisely. In practice, EEG, video-EEG, MEG, SPECT, PET and various MRI techniques, especially functional MRI (fMRI), have been used to map areas related to EZ. However, in some cases, the results remain non-convergent and the EZ, undefined. Therefore, the use of new methodologies to assist the location of EZ have been proposed. Herein, our goal was to develop two methods for assessing the EZ. The first one was designed to access changes in the hemodynamic response (HRF) of the EZ in response to hypercapnia. 22 patients with TLE and 10 normal volunteers were evaluated by modulating the partial pressure of CO2 during the acquisition of fMRI in a breathing holding and a passive inhalation CO2/air protocols. The results show increased onset times and decreased amplitude of the HRF in the temporal lobe of TLE patients compared with asymptomatic volunteers. Moreover, most patients had onset maps coincident with ictal SPECT localizations. The second proposed study was based on simultaneous EEG-fMRI acquisitions. The relationship between powers of alpha and theta bands (EEG) and BOLD contrast (fMRI) was investigated in 41 TLE patients and 7 healthy controls. Alpha band results show a consistent negative correlation in the occipital, parietal and frontal lobes both in controls and TLE patients. In addition, controls show disperse positive correlations in both hemispheres. On the other hand, TLE patients presented strong positive correlations in the thalamus and insula. Theta band analysis, in controls, primarily show positive correlations in bilateral pre-and post-central gyri. In patients, robust positive correlations were observed in the anterior cingulate gyrus, thalamus, insula, putamen, superior parietal, frontal and temporal gyri. Moreover, the lateralization index (LI) indicates a coincidence between the side of the EZ evaluated by clinical diagnosis and clusters detected in the theta band. In conclusion, the hipercapnia study showed to be an interesting tool in locating EZ and the results are similar to SPECT findings. The longer onset and lower amplitude of the HRF observed in patients may be related to a vascular stress due to the recurrence of seizures. Furthermore, alpha and theta rhythms may be a promising tool to be used in determining the lateralization of EZ in patients with TLE.

O mapeamento das respostas BOLD (Blood Oxygenation Level Dependent) constitui etapa importante nos experimentos de imagem funcional por ressonância magnética (Functional Magnetic Resonance Imaging fMRI) e de EEG-fMRI simultâneo. Em sua grande maioria, a análise de dados de fMRI e de EEG-fMRI está baseada no modelo linear geral (General Linear Model GLM), que procura localizar as respostas BOLD por meio de modelos definidos a priori. Porém, em muitos casos, como em pacientes, variações na forma e/ou atraso podem reduzir a confiabilidade dos resultados. Desse modo, o primeiro objetivo deste trabalho foi explorar métodos clássicos e propor novos métodos para análise de dados de fMRI e de EEG-fMRI simultâneo. Neste trabalho, um método modificado baseado na distância de Kullback-Leibler generalizada (dKLg) foi desenvolvido. Diferentemente do GLM, essa abordagem não requer um modelo para a resposta. Dados simulados foram utilizados para otimizá-lo e compará-lo ao GLM sob diferentes condições de resposta como a relação sinal ruído e a latência. Em seguida, o dKLg foi testado em dados reais, adquiridos em 14 voluntários assintomáticos, submetidos a tarefas motoras e auditivas padrões. Os resultados mostram a equivalência entre o dKLg e o GLM. Em seguida, essa estratégia foi testada em 02 pacientes com com estenose carotídea unilateral. Neste caso, o dKLg foi capaz de detectar regiões significativas ipsilaterais à estenose, não detectadas pelo GLM, em virtude do atraso do sinal BOLD. Em seguida, esses métodos foram aplicados sobre exames de EEG-fMRI realizados em 45 pacientes com epilepsia. Para esse conjunto de dados, mais uma abordagem foi elaborada, que utiliza a Análise de Componentes Independentes (Independent Component Analysis ICA). Denominado ICA-GLM, ele permite extrair de modo semi-automático a amplitude, duração e topografia das descargas epileptiformes interictais (Interictal Epileptiform Discharges IED), favorecendo a inclusão de sinais do EEG de menor destaque. Além dessa vantagem, ele ainda permite incluir modelos do sinal BOLD com diferentes latências, aumentando a abrangência da variabilidade das respostas encontradas em pacientes com epilepsia. A eficiência do ICA-GLM também foi comparado à do GLM e dKLg nos exames de EEG-fMRI. Embora os resultados tenham demonstrado a robustez do GLM, em alguns pacientes o dKLg foi mais eficiente para localizar regiões concordantes que não foram detectadas pelo GLM. Ainda, em boa parte dos casos o ICA-GLM detectou regiões mais extensas e com maior valor estatístico, quando comparado ao GLM. De forma geral, nota-se que o dKLg e ICA-GLM podem ser ferramentas complementares importantes ao GLM, aumentando a sensibilidade dos exames de EEG-fMRI como um todo. Outra etapa importante nas avaliações de EEG-fMRI em pacientes com epilepsia tem sido a utilização de imagens de fontes elétricas (Electrical Source Imaging ESI). Neste trabalho, os mapas de ESI foram obtidos por dois métodos de solução inversa distribuída nunca usados no cenários da EEG-fMRI: Bayesian Model Averaging (BMA) e constrained Low Resolution Electromagnetic Tomography (cLORETA). Além da construção dos mapas de ESI, avaliamos a utilidade de combinar as técnicas de ESI e de EEG-fMRI para promover a diferenciação entre fontes primárias e de propagação temporal. Essa análise permitiu avaliar a concordância entre as regiões detectadas pelo ESI e EEG-fMRI e diferenciar as respostas BOLD relacionadas aos componentes iniciais e posteriores da IED. Embora os resultados ainda sejam preliminares para eleger qual método seria mais eficiente (cLORETA ou BMA), a distância encontrada entre o máximo ESI e o cluster de EEG-fMRI mais próximo foi consistentemente similar, em ambos, com os dados recentes da literatura. / Functional magnetic resonance imaging (fMRI) and combined EEG-fMRI usually rely on the successful detection of Blood Oxygenation Level Dependent (BOLD) signal. Typically, the analysis of both fMRI and EEG-fMRI are based on the General Linear Model (GLM) that aims at localizing the BOLD responses associated to an a priori model. However, the responses are not always canonical, as is the case of those from patients, which may reduce the reliability of the results. Therefore, the first objective of the present study was to explore the usage of classical methods, such as the GLM, and to propose alternative strategies to the analysis of fMRI and combined EEG-fMRI. A first method developed was based on the computation of the generalized Kullback-Leibler distance (gKLd), which does not require the use of an a priori model. Simulated data was used to allow quantitative comparison between the gKLd and GLM under different response conditions such as the signal to noise ratio and delay. The gKLd was then tested on real data, first from 14 asymptomatic subjects, submitted to classical motor and auditory fMRI protocols. The results demonstrate that under these conditions the GLM and gKLd are equivalent. The same strategy was applied to 02 patients with unilateral carotid stenosis. Now the dKLg was capable of detecting the expected bilateral BOLD responses that were not detected by the GLM, as a consequence of the response delay imposed by the stenosis. Those comparisons were now extended to the evaluation of EEG-fMRI exams from 45 patients with epilepsy. For this data set, an additional method was used, based on the use of Independent Component Analysis (ICA), which was called ICA-GLM. It allows extracting semi-automatically the amplitude, duration and topography of EEG interictal Epileptiform Discharges (IED), favoring the use of less prominent signals. Moreover, it also allows the use of BOLD response models with different delays, expanding the variability of the responses to be detected in patients with epilepsy. ICA-GLM was also compared to GLM and dKLg in these EEG-fMRI evaluations. Although in general the results have demonstrated the robustness of the GLM, dKLg was more efficient in detecting the responses from some pacients, while the ICA-GLM mostly detected broader regions with more significant results when compared to GLM. In general, dKLg and ICA-GLM seem to offer an important complementary aspect to the GLM, increasing its sensibility in EEG-fMRI as a whole. Another important aspect of EEG-fMRI applied to patients with epilepsy has been the inspection of Electrical Source Imaging (ESI) to evaluate some dynamical aspects of the IED. Herein, ESI maps were obtained from two inverse distributed solutions that were not applied so far to EEG-fMRI: Bayesian Model Averaging (BMA) and constrained Low Resolution Electromagnetic Tomography (cLORETA). Besides, we also evaluated the combined information from ESI and EEG-fMRI in order to differentiate from primary sources to temporal propagation of the signal. Such analysis allowed us to inspect for the correspondence between regions detected by ESI e EEG-fMRI and to separate BOLD signals whose sources are related to the initial and later components of the IED. Although the results are preliminary to determine which ESI method (cLORETA or BMA) is more efficient, the distance between the maximum ESI and the closest EEG-fMRI cluster was consistently similar with those reported in the literature.

O mapeamento das respostas BOLD (Blood Oxygenation Level Dependent) constitui etapa importante nos experimentos de imagem funcional por ressonância magnética (Functional Magnetic Resonance Imaging fMRI) e de EEG-fMRI simultâneo. Em sua grande maioria, a análise de dados de fMRI e de EEG-fMRI está baseada no modelo linear geral (General Linear Model GLM), que procura localizar as respostas BOLD por meio de modelos definidos a priori. Porém, em muitos casos, como em pacientes, variações na forma e/ou atraso podem reduzir a confiabilidade dos resultados. Desse modo, o primeiro objetivo deste trabalho foi explorar métodos clássicos e propor novos métodos para análise de dados de fMRI e de EEG-fMRI simultâneo. Neste trabalho, um método modificado baseado na distância de Kullback-Leibler generalizada (dKLg) foi desenvolvido. Diferentemente do GLM, essa abordagem não requer um modelo para a resposta. Dados simulados foram utilizados para otimizá-lo e compará-lo ao GLM sob diferentes condições de resposta como a relação sinal ruído e a latência. Em seguida, o dKLg foi testado em dados reais, adquiridos em 14 voluntários assintomáticos, submetidos a tarefas motoras e auditivas padrões. Os resultados mostram a equivalência entre o dKLg e o GLM. Em seguida, essa estratégia foi testada em 02 pacientes com com estenose carotídea unilateral. Neste caso, o dKLg foi capaz de detectar regiões significativas ipsilaterais à estenose, não detectadas pelo GLM, em virtude do atraso do sinal BOLD. Em seguida, esses métodos foram aplicados sobre exames de EEG-fMRI realizados em 45 pacientes com epilepsia. Para esse conjunto de dados, mais uma abordagem foi elaborada, que utiliza a Análise de Componentes Independentes (Independent Component Analysis ICA). Denominado ICA-GLM, ele permite extrair de modo semi-automático a amplitude, duração e topografia das descargas epileptiformes interictais (Interictal Epileptiform Discharges IED), favorecendo a inclusão de sinais do EEG de menor destaque. Além dessa vantagem, ele ainda permite incluir modelos do sinal BOLD com diferentes latências, aumentando a abrangência da variabilidade das respostas encontradas em pacientes com epilepsia. A eficiência do ICA-GLM também foi comparado à do GLM e dKLg nos exames de EEG-fMRI. Embora os resultados tenham demonstrado a robustez do GLM, em alguns pacientes o dKLg foi mais eficiente para localizar regiões concordantes que não foram detectadas pelo GLM. Ainda, em boa parte dos casos o ICA-GLM detectou regiões mais extensas e com maior valor estatístico, quando comparado ao GLM. De forma geral, nota-se que o dKLg e ICA-GLM podem ser ferramentas complementares importantes ao GLM, aumentando a sensibilidade dos exames de EEG-fMRI como um todo. Outra etapa importante nas avaliações de EEG-fMRI em pacientes com epilepsia tem sido a utilização de imagens de fontes elétricas (Electrical Source Imaging ESI). Neste trabalho, os mapas de ESI foram obtidos por dois métodos de solução inversa distribuída nunca usados no cenários da EEG-fMRI: Bayesian Model Averaging (BMA) e constrained Low Resolution Electromagnetic Tomography (cLORETA). Além da construção dos mapas de ESI, avaliamos a utilidade de combinar as técnicas de ESI e de EEG-fMRI para promover a diferenciação entre fontes primárias e de propagação temporal. Essa análise permitiu avaliar a concordância entre as regiões detectadas pelo ESI e EEG-fMRI e diferenciar as respostas BOLD relacionadas aos componentes iniciais e posteriores da IED. Embora os resultados ainda sejam preliminares para eleger qual método seria mais eficiente (cLORETA ou BMA), a distância encontrada entre o máximo ESI e o cluster de EEG-fMRI mais próximo foi consistentemente similar, em ambos, com os dados recentes da literatura. / Functional magnetic resonance imaging (fMRI) and combined EEG-fMRI usually rely on the successful detection of Blood Oxygenation Level Dependent (BOLD) signal. Typically, the analysis of both fMRI and EEG-fMRI are based on the General Linear Model (GLM) that aims at localizing the BOLD responses associated to an a priori model. However, the responses are not always canonical, as is the case of those from patients, which may reduce the reliability of the results. Therefore, the first objective of the present study was to explore the usage of classical methods, such as the GLM, and to propose alternative strategies to the analysis of fMRI and combined EEG-fMRI. A first method developed was based on the computation of the generalized Kullback-Leibler distance (gKLd), which does not require the use of an a priori model. Simulated data was used to allow quantitative comparison between the gKLd and GLM under different response conditions such as the signal to noise ratio and delay. The gKLd was then tested on real data, first from 14 asymptomatic subjects, submitted to classical motor and auditory fMRI protocols. The results demonstrate that under these conditions the GLM and gKLd are equivalent. The same strategy was applied to 02 patients with unilateral carotid stenosis. Now the dKLg was capable of detecting the expected bilateral BOLD responses that were not detected by the GLM, as a consequence of the response delay imposed by the stenosis. Those comparisons were now extended to the evaluation of EEG-fMRI exams from 45 patients with epilepsy. For this data set, an additional method was used, based on the use of Independent Component Analysis (ICA), which was called ICA-GLM. It allows extracting semi-automatically the amplitude, duration and topography of EEG interictal Epileptiform Discharges (IED), favoring the use of less prominent signals. Moreover, it also allows the use of BOLD response models with different delays, expanding the variability of the responses to be detected in patients with epilepsy. ICA-GLM was also compared to GLM and dKLg in these EEG-fMRI evaluations. Although in general the results have demonstrated the robustness of the GLM, dKLg was more efficient in detecting the responses from some pacients, while the ICA-GLM mostly detected broader regions with more significant results when compared to GLM. In general, dKLg and ICA-GLM seem to offer an important complementary aspect to the GLM, increasing its sensibility in EEG-fMRI as a whole. Another important aspect of EEG-fMRI applied to patients with epilepsy has been the inspection of Electrical Source Imaging (ESI) to evaluate some dynamical aspects of the IED. Herein, ESI maps were obtained from two inverse distributed solutions that were not applied so far to EEG-fMRI: Bayesian Model Averaging (BMA) and constrained Low Resolution Electromagnetic Tomography (cLORETA). Besides, we also evaluated the combined information from ESI and EEG-fMRI in order to differentiate from primary sources to temporal propagation of the signal. Such analysis allowed us to inspect for the correspondence between regions detected by ESI e EEG-fMRI and to separate BOLD signals whose sources are related to the initial and later components of the IED. Although the results are preliminary to determine which ESI method (cLORETA or BMA) is more efficient, the distance between the maximum ESI and the closest EEG-fMRI cluster was consistently similar with those reported in the literature.

The simultaneous acquisitions of electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) have been applied to improve the surgery planning of patients with drug resistant epilepsy. However, the classical approach of analyzing the EEG-fMRI data is inefficient in patients whom only few or non interictal epileptiforms discharges (IEDs) are detected during the simultaneous acquisition. Another issue of EEG-fMRI acquisition is related to its high sensitivity to motion, which decreases the quality of both data, even worse in non-cooperative patients. In this work we propose and discuss the application of two methods of analyzing fMRI data of patients with focal epilepsy: Independent component analysis (ICA) and two-dimensional temporal clustering (2dTCA). Each method was applied in a distinct group of patients and the results were compared to those obtained by the classic EEG-fMRI analysis. We have also proposed a method to improve the quality of EEG data using the head position measurements obtained, by a prospective motion correction (PMC) system, during the EEG-fMRI acquisitions. In the ICA study, we have used the electrical source images for selecting independent components (ICs) in EEG data of 13 patients with different spiking frequency. The method detected epilepsy-related BOLD activity in all the patients. Comparatively, the classic EEG-fMRI could be applied in 11 patients and epilepsy-related BOLD activities were found in seven of them. In the 2dTCA study, we have evaluated 20 patients and found epilepsy-related maps in 14 of them. Thirteen of the twenty patients have IEDs detected during the simultaneous acquisition; the classic EEG-fMRI provided maps related to the epileptogenic region in six of them. Finally we have verified in three health subjects that the proposed method for correcting motion-induced artefacts in the EEG data is effective for high amplitude and velocities (~1cm and 55mm/s). We concluded that the ICA and 2dTCA methods increase the sensitivity of using fMRI for mapping the epileptogenic region, mainly in patients presenting few or no IEDs in the EEG data simultaneously acquired to the fMRI. The PMC use during the fMRI acquisition does not degrade the quality of the EEG data acquired simultaneously. In fact, the motion information can be used for improving its quality by correcting motion-induced artefacts. / As aquisições simultâneas de dados de eletroencefalografia (EEG) e imagens funcionais por ressonância magnética (fMRI) vêm sendo utilizadas com intuito de melhorar o planejamento cirúrgico de pacientes com epilepsia refratária. Entretanto, o processamento classicamente usado nestes dados combinados não é possível em pacientes sem descargas epileptiformes interictais (IEDs) e possui baixa sensibilidade para aqueles em que poucas IEDs são detectadas durante a aquisição simultânea. Além disto, a técnica é sensível ao movimento dos pacientes durante as aquisições, o que reduz a qualidade dos dados, principalmente em pacientes não cooperantes. Neste trabalho é proposto e discutido o uso de dois métodos de processamento, baseados nas técnicas de análise de componentes independentes (ICA) e análise temporal de clusters em duas dimensões (2dtca), para se mapear regiões epileptogênicas. Cada método foi analisado em um conjunto diferente de pacientes e os resultados foram comparados com os obtidos pelo EEG-fMRI clássico. Finalmente, propomos um método que utiliza às medidas de posicionamento da cabeça, obtidas durante a aquisição das fMRI, para aumentar a qualidade dos dados de EEG adquiridos simultaneamente. No estudo usando ICA combinado com imagens de fontes elétricas analisamos os dados de 13 pacientes com diferentes frequências de descargas e observamos que este método encontrou ao menos uma componente independente relacionada à epilepsia em cada paciente. Comparativamente usando o processamento convencional foi possível avaliar 11 dos 13 pacientes, e em apenas sete deles os mapas resultantes foram considerados concordantes com a região epileptogênica (RE). No estudo utilizando 2dTCA avaliamos 20 pacientes e encontramos mapas relacionados com a RE em 14 deles. Neste conjunto de pacientes, 13 apresentaram IEDs durante as aquisições; neles o método clássico de processamento teve resultados concordantes com a RE em seis deles. Finalmente verificamos em três sujeitos saudáveis que o método aqui proposto para corrigir os artefatos induzidos no EEG devido ao movimento é efetivo para altas amplitudes e velocidades (~1cm e 55mm/s). Concluímos que os métodos ICA e 2dTCA aumentam a sensibilidade do uso de fMRI para mapear RE, principalmente em pacientes com baixa ou nenhuma detecção de IEDs durante às aquisições. Também concluímos que o uso da correção prospectiva de movimento em aquisições de fMRI não reduz a qualidade do dado de EEG adquirido simultaneamente e que às informações de movimento mensuradas podem melhorar a qualidade deste dado em situações de repouso e movimento do sujeito durante o experimento.

A epilepsia do lobo temporal (ELT) é a forma mais comum de epilepsia e a mais resistente ao tratamento medicamentoso. Existem diversos tipos de drogas anti-epilépticas usadas no controle das crises. Entretanto, em alguns casos, esse tipo de tratamento não é eficaz e a cirurgia para remoção da zona epileptogênica (ZE) pode ser uma alternativa recomendada. A ZE é definida como aquela onde as crises são originadas. Trata-se de um conceito teórico e, atualmente, não existem técnicas capazes de delimitá-la precisamente. Na prática, exames de EEG, vídeo-EEG, MEG, SPECT, PET e diversas técnicas de MRI, em especial as funcionais, têm sido usados para mapear zonas relacionadas à ZE. Contudo, em alguns casos, os resultados permanecem não convergentes e a determinação da ZE inconclusiva. Desse modo, é evidente a importância do surgimento de novas metodologias para auxiliar a localização da ZE. Assim, pois, o objetivo deste trabalho foi desenvolver dois métodos para a avaliação da ZE, ambos baseados na imagem funcional por ressonância magnética. No primeiro, investigamos possíveis alterações da resposta hemodinâmica (HRF) quando da modulação da pressão parcial de CO2. Para tanto, fizemos um estudo sobre 22 pacientes com ELT e 10 voluntários assintomáticos modulando a pressão parcial de CO2 sanguíneo cerebral por um protocolo de manobra de pausa respiratória e outro de inalação passiva de CO2/ar. Os resultados mostram que o tempo de onset da HRF tende a ser maior e a amplitude da HRF tende a ser menor em áreas do lobo temporal de pacientes com ELT quando comparados com os dados de voluntários assintomáticos. Além disso, os resultados mostram mapas de onset individuais coincidentes com exames de SPECT ictal. O segundo estudo foi baseado em medidas de EEG-fMRI simultâneo. Neste, avaliamos a relação entres as potências dos ritmos cerebrais alfa e teta (EEG) e o contraste BOLD (fMRI) de 41 pacientes com ELT e 7 voluntários assintomáticos em estado de repouso. A análise da banda alfa mostrou correlações negativas nos lobos occipital, parietal e frontal tanto nos voluntários quanto nos pacientes com ELT. As correlações positivas nos voluntários foram dispersas e variáveis em ambos hemisférios cerebrais. Por outro lado, encontramos forte correlação positiva no tálamo e ínsula dos pacientes com ELT. Na análise da banda teta observamos correlações positivas bilaterais nos giros pré e pós central de voluntários. Ainda, foram observados clusters no cíngulo anterior, tálamo, ínsula, putamen, em regiões parietais superior, frontais e giros temporais. Também, utilizamos um cálculo de índice de lateralização (IL) no lobo temporal em confrontos entre pacientes com ELT à direita, pacientes com ELT à esquerda e voluntários assintomáticos. Verificamos que os ILs, utilizando os clusters obtidos nas análises em teta, foram coincidentes com o diagnóstico clínico prévio da localização da ZE em todas as análises dos grupos de pacientes com ELT à direita, e na maioria do grupo de pacientes com ELT à esquerda. De forma geral, verificamos que o método de hipercapnia se mostrou ferramenta interessante na localização da ZE comprovada pelos coincidentes achados pela avaliação de SPECT. Inferimos que o maior tempo de onset e menor amplitude da HRF observadas nos pacientes em relação a voluntários possam estar relacionados a um stress vascular devido à recorrência de crises. Já o método de ritmicidade alfa e teta proposto parece promissor para ser usado na determinação da lateralização da ZE em pacientes com ELT. / Temporal lobe epilepsy (TLE) is the most common and resistant form of epilepsy to anti-epileptic drug. There are several types of anti-epileptic drugs used in seizure control. However, in some cases drug treatment is not effective and surgery to remove the epileptogenic zone (EZ) is a recommended alternative. EZ is a theoretical concept and there are many techniques that have been applied to enclose it precisely. In practice, EEG, video-EEG, MEG, SPECT, PET and various MRI techniques, especially functional MRI (fMRI), have been used to map areas related to EZ. However, in some cases, the results remain non-convergent and the EZ, undefined. Therefore, the use of new methodologies to assist the location of EZ have been proposed. Herein, our goal was to develop two methods for assessing the EZ. The first one was designed to access changes in the hemodynamic response (HRF) of the EZ in response to hypercapnia. 22 patients with TLE and 10 normal volunteers were evaluated by modulating the partial pressure of CO2 during the acquisition of fMRI in a breathing holding and a passive inhalation CO2/air protocols. The results show increased onset times and decreased amplitude of the HRF in the temporal lobe of TLE patients compared with asymptomatic volunteers. Moreover, most patients had onset maps coincident with ictal SPECT localizations. The second proposed study was based on simultaneous EEG-fMRI acquisitions. The relationship between powers of alpha and theta bands (EEG) and BOLD contrast (fMRI) was investigated in 41 TLE patients and 7 healthy controls. Alpha band results show a consistent negative correlation in the occipital, parietal and frontal lobes both in controls and TLE patients. In addition, controls show disperse positive correlations in both hemispheres. On the other hand, TLE patients presented strong positive correlations in the thalamus and insula. Theta band analysis, in controls, primarily show positive correlations in bilateral pre-and post-central gyri. In patients, robust positive correlations were observed in the anterior cingulate gyrus, thalamus, insula, putamen, superior parietal, frontal and temporal gyri. Moreover, the lateralization index (LI) indicates a coincidence between the side of the EZ evaluated by clinical diagnosis and clusters detected in the theta band. In conclusion, the hipercapnia study showed to be an interesting tool in locating EZ and the results are similar to SPECT findings. The longer onset and lower amplitude of the HRF observed in patients may be related to a vascular stress due to the recurrence of seizures. Furthermore, alpha and theta rhythms may be a promising tool to be used in determining the lateralization of EZ in patients with TLE.

Behavior is considered goal-directed when the actor integrates information about the subsequent outcome of an action (Balleine & O'Doherty, 2010; Dickinson & Balleine, 1994; Kiesel & Koch, 2012), potentially enabling the anticipation of consequences of an action. Thus, it requires prior acquisition of knowledge about the current contingencies between behavioral responses and their outcomes under certain stimulus conditions (J. Hoffmann & Engelkamp, 2013). This association chain enables events lying in the future to be mentally represented and assessed in terms of value and achievability. However, while neural correlates of instructed goal-directed action integration processes have already been examined in a functional magnetic resonance imaging (fMRI) study using this paradigm (Ruge & Wolfensteller, 2015), there has been no information if those processes are also reflected in Electroencephalography (EEG) and if so which specific EEG parameters are modulated by them. This dissertation set out to investigate neurocognitive mechanisms of instructed outcome response learning utilizing two different imaging methods, namely EEG and fMRI. Study 1 was an exploratory study to answer the question what kinds of learning-related EEG correlates were to expect. The O-R outcome integration specific EEG correlates identified in Study 1 served as regressors in a unified general linear model (EEG-informed fMRI analysis) in the co-registered EEG-fMRI study (Study 2). One of the key questions in this study was if the EEG signal could help to differentiate between BOLD pre-response activation associated with processes related to response preparation or initiation and activation associated with post-response outcome integration processes. The foundation to both studies of this work was an experimental paradigm of instructed S-R-O learning, which included a learning and a test phase. Stimuli were four abstract visual patterns that differed in each block. Each visual stimulus required a distinct manual response and was predictably followed by a distinct auditory outcome. Instructions were delivered via a “guided implementation” procedure in which the instruction was embedded within the first three successful behavioral implementation trials. In these first three trials, the visual stimulus was followed by an imperative stimulus highlighting the correct response. The guided implementation phase was followed by an unguided implementation phase where the correct response now had to be retrieved from memory. Behaviorally, the strength of acquired O-R associations can be analyzed via O-R compatibility effects measured in a subsequent outcome-priming test phase (Greenwald, 1970). In this test phase a previously learned outcome becomes an imperative stimulus that requires either the response, which produced that outcome in the preceding learning phase (O-R compatible), or a response, which produced a different outcome (O-R incompatible). The experimental design was embedded into an EEG recording setup in study 1 while study 2 comprised a simultaneous EEG-fMRI recording setup in which EEG scalp potentials were continuously recorded during the experimental session inside the MR scanner bore. Study 1 revealed various ERP markers correlated with outcome response learning. An ERP post-response anterior negativity following auditory outcomes was increasingly attenuated as a function of the acquired association strength. This suggests that previously reported action-induced sensory attenuation effects under extensively trained free choice conditions can be established within few repetitions of specific R-O pairings under forced choice conditions. Furthermore, an even more rapid development of a post-response but pre-outcome fronto-central positivity, which was reduced for high R-O learners, might indicate the rapid deployment of preparatory attention towards predictable outcomes. Finally, the study identified a learning-related stimulus-locked activity modulation within the visual P1-N1 latency range, which was thought to reflect the multi-sensory integration of the perceived antecedent visual stimulus with the anticipated auditory outcome. In general, study 2 was only partially able to replicate the EEG activity dynamics related to the formation of bidirectional R-O associations that were observed in study 1. Primarily, it was able to confirm the modulation in EEG negativity in the visual P1-N1 latency range over the learning course. The EEG-informed analysis revealed that learning-related modulations of the P1-N1 complex are functionally coupled to activation in the orbitofrontal cortex (OFC). More specifically, growing attenuation of the EEG negativity increase from early to late SRO repetition levels in high R-O learners was associated with an increase in activation in the OFC. An additional exploratory EEG analysis identified a recurring post outcome effect at central electrode sites expressed in a stronger negativity in late compared to early learning stages. This effect was present in both studies and showed no correlation with any of the behavioral markers of learning. The EEG-informed fMRI analysis resulted in a pattern of distinct functional couplings of this parameter with different brain regions, each correlated with different behavioral markers of S-R-O learning. First of all, increased coupling between the late EEG negativity and activation in the supplementary motor area (SMA) was positively correlated with the O-R compatibility effect. Thus, high R-O learners exhibited a stronger coupling than low R-O learners. Secondly, increased couplings between the late EEG negativity and activation in the somatosensory cortex as well as the dorsal caudate, on the other hand, were positively correlated with individual reaction time differences between early and late stages of learning. Regarding activation patterns prior to the behavioral response the results indicate that the OFC could serve as a (multimodal) hub for integrating stimulus information and information about its associated outcome in an early pre-stage of action selection and initiation. Learnt S-O contingencies would facilitate initiating the motor program of the action of choice. Hence, the earlier an outcome is anticipated (based on stimulus outcome associations), the better it will be associated with its response, eventually leading to stronger O-R compatibility effects later on. Thus, one could speculate that increased activation in response to S-R-O mappings possibly embodies a marker for the ongoing transition from mere stimulus-based behavior to a goal-directed behavior throughout the learning course. Post-response brain activation revealed a seemingly two-fold feedback integration stream of O-R contingencies. On one hand the SMA seems to be engaged in bidirectional encoding processes of O-R associations. The results promote the general idea that the SMA is involved in the acquisition of goal-directed behavior (Elsner et al., 2002; Melcher, Weidema, Eenshuistra, Hommel, & Gruber, 2008; Melcher et al., 2013). Together with prior research (Frimmel, Wolfensteller, Mohr, & Ruge, 2016) this notion can be generalized not only to extensive learning phases but also to learning tasks in which goal-directed behavior is acquired in only few practice trials. However, there is an ongoing debate on whether SMA activation can be clearly linked to sub-processes prior or subsequent to an agent’s action (Nachev, Kennard, & Husain, 2008). The results of this work provide additional evidence favoring an involvement of the SMA only following a performed action in response to an imperative stimulus and even more, subsequent to the perception of its ensuing effect. This may give rise to the interpretation that the SMA is associated with linking the motor program of the performed action to the sensory program of the perceived effect, hence establishing and strengthening O-R contingencies. Furthermore, the analysis identified an increased coupling of a late negativity in the EEG signal and activation in the dorsal parts of the caudate as well as the somatosensory cortex. The dorsal caudate has not particularly been brought into connection with O-R learning so far. I speculate that the coupling effect in this part of the caudate reflects an ongoing process of an early automatization of the acquired behavior. It has already be shown in a similar paradigm that behavior can be automatized within only few repetitions of novel instructed S-R mappings (Mohr et al., 2016).:Table of contents Table of contents II List of Figures IV List of Tables VI List of Abbreviations VII 1 Summary 1 1.1 Introduction 1 1.2 Study Objectives 2 1.3 Methods 3 1.4 Results 4 1.5 Discussion 4 2 Theoretical Background 7 2.1 Introduction 7 2.2 Theories of acquiring goal-directed behavior 9 2.2.1 Instrumental learning 9 2.2.1.1 Behavioral aspects 9 2.2.1.2 Neurophysiological correlates 14 2.2.2 Acquisition of goal-directed behavior according to ideomotor theory 16 2.2.2.1 Behavioral aspects 16 2.2.2.2 Neurophysiological correlates 22 2.3 Summary 25 2.4 Methodological background 26 2.4.1 Electroencephalography (EEG) 26 2.4.2 Functional magnetic resonance imaging (fMRI) 28 2.4.3 Co-registered EEG-fMRI 29 3 General objectives and research questions 34 4 Study 1 – Learning-related brain-electrical activity dynamics associated with the subsequent impact of learnt action-outcome associations 36 4.1 Introduction 36 4.2 Methods 39 4.3 Results 47 4.4 Discussion 60 5 Study 2 - Within trial distinction of O-R learning-related BOLD activity with the means of co-registered EEG information 64 5.1 Introduction 64 5.2 Methods 66 5.3 Results 86 5.4 Discussion 101 6 Concluding general discussion 109 6.1 Brief assessment of study objectives 109 6.2 Novel insights into rapid instruction based S-R-O learning? 109 6.2.1 Early stimulus outcome information retrieval indicates the transition from stimulus based behavior to goal-directed action 110 6.2.2 Post-response encoding and consolidation of O-R contingencies enables goal-directedness of behavior 112 6.3 Critical reflection of the methodology and outlook 116 6.3.1 Strengths and limitations of this work 116 6.3.2 Data quality assessment 117 6.3.3 A common neural foundation for EEG and fMRI? 119 6.3.4 How can co-registered EEG-fMRI contribute to a better understanding of the human brain? 121 6.4 General Conclusion 123 7 References 124 Danksagung Erklärung