Examining the effect of repeating stimuli on brain activity is important for theories of perception, learning and memory. Functional magnetic resonance imaging (fMRI) is a non-invasive way to examine repetition-related effects in the human brain. However the Blood-Oxygenation Level-Dependent (BOLD) signal measured by fMRI is far removed from the electrical activity recorded from single cells in animal studies of repetition effects. Despite that, there have been many claims about the neural mechanisms associated with fMRI repetition effects. However, none of these claims has adequately considered the temporal and spatial resolution limitations of fMRI. In this thesis, I tackle these limitations by combining simulations and modelling in order to infer repetition-related changes at the neural level. I start by considering temporal limitations in terms of the various types of general linear model (GLM) that have used to deconvolve single-trial BOLD estimates. Through simulations, I demonstrate that different GLMs are best depending on the relative size of trial-variance versus scan-variance, and the coherence of those variabilities across voxels. To address the spatial limitations, I identify six univariate and multivariate properties of repetition effects measured by event-related fMRI in regions of interest (ROI), including how repetition affects the ability to classify two classes of stimuli. To link these properties to underlying neural mechanisms, I create twelve models, inspired by single-cell studies. Using a grid search across model parameters, I find that only one model (“local scaling”) can account for all six fMRI properties simultaneously. I then validate this result on an independent dataset that involves a different stimulus set, protocol and ROI. Finally, I investigate classification of initial versus repeated presentations, regardless of the stimulus class. This work provides a better understanding of the neural correlates of stimulus repetition effects, as well as illustrating the importance of formal modelling.

Matrix factorisation treats observations as linear combinations of basis vectors together with, possibly, additive noise. Notable techniques in this family are Principal Components Analysis and Independent Components Analysis. Applied to brain images, matrix factorisation provides insight into the spatial and temporal structure of data. We improve on current practice with methods that unify different stages of analysis simultaneously for all subjects in a dataset, including dimension estimation and reduction. This results in uncertainty information being carried coherently through the analysis. A computationally efficient approach to correlated multivariate normal distributions is set out. This enables spatial smoothing during the inference of basis vectors, to a level determined by the data. Applied to neuroimaging, this reduces the need for blurring of the data during preprocessing. Orthogonality constraints on the basis are relaxed, allowing for overlapping ‘networks’ of activity. We consider a nonparametric matrix factorisation model inferred using Markov Chain Monte Carlo (MCMC). This approach incorporates dimensionality estimation into the infer- ence process. Novel parallelisation strategies for MCMC on repeated graphs are provided to expedite inference. In simulations, modelling correlation structure is seen to improve source separation where latent basis vectors are not orthogonal. The Cambridge Centre for Ageing and Neuroscience (Cam-CAN) project obtained fMRI data while subjects watched a short film, on 30 of whose recordings we demonstrate the approach. To conduct inference on larger datasets, we provide a fixed dimension Structured Matrix Factorisation (SMF) model, inferred through Variational Bayes (VB). By modelling the components as a mixture, more general distributions can be expressed. The VB approach scaled to 600 subjects from Cam-CAN, enabling a comparison to, and validation of, the main findings of an earlier analysis; notably that subjects’ responses to movie watching became less synchronised with age. We discuss differences in results obtained under the MCMC and VB inferred models.

RACIONAL: A obesidade caracteriza-se por um excesso de tecido adiposo branco que causa morbidades metabólicas, mecânicas e a morte de aproximadamente 2,8 milhões de pessoas anualmente. Sua fisiopatologia envolve alterações em mecanismos que interagem no cérebro, sejam humorais, neurais e comportamentais, os quais podem ser eliciados por alimentos. Por sua vez, técnicas de neuroimagem funcional, como o Single-photon Emission Tomography (SPECT), surgiram como ferramentas para avaliar alterações funcionais in vivo. Por isso, questionamos se imagens de SPECT cerebral, alteradas após a ingestão de glicose em comparação a um teste controle com água, explicariam, pelo menos em parte, o comportamento alimentar da obesidade. OBJETIVOS: Comparar a perfusão (ativação) cerebral de imagens de SPECT, após ingestão de água vs. glicose, em mulheres obesas e com peso normal. Correlacionar a intensidade da perfusão encontrada com níveis sanguíneos de leptina, insulina e glicemia, bem como com a quantidade de gordura corporal. MATERIAIS E MÉTODOS: 10 mulheres com e 10 sem obesidade (n = 20) foram submetidas à SPECT cerebral duas vezes, após marcação pelo [99mTc]-ECD 30 minutos após ingestão de 300 mL de água e de uma solução com 75 gramas de glicose, em dias separados (40 SPECTs), tendo sido cada sujeito controle de si mesmo. As imagens foram comparadas intragrupo e entre grupos por meio do software Statistical Parametric Mapping. Modelos de efeitos mistos foram usados para avaliar correlações entre as variáveis. RESULTADOS E DISCUSSÃO: Mulheres obesas apresentam maior ativação em regiões da Default Mode Network e da Salience Network após o teste com água. Enquanto as obesas apresentam-se mais engajadas na percepção dos processos fisiológicos (como fome e sede) na situação basal, com água, apenas aquelas com peso normal parecem responder às alterações desses processos eliciadas pelo alimento oferecido. Após estímulo com glicose, apenas o grupo sem obesidade aumentou a perfusão em regiões relacionadas à recompensa e ao controle do comportamento, como corpo estriado e córtices orbitofrontal e pré-frontal. A perfusão nestas áreas apresentou correlação negativa com a interação entre leptina e insulina (Coef. = - 0,001, p = 0,003). Além disso, tais regiões recebem aferências dopaminérgicas e, por isso, temos como hipótese que déficits na sinalização da dopamina explicariam os achados encontrados. CONCLUSÃO: A ingestão de glicose eliciou respostas relacionadas à recompensa alimentar normal e a um controle apropriado sobre o apetite nas mulheres sem obesidade, não sendo observado o mesmo nas obesas, processo do qual parece participar a interação entre insulina e leptina e déficits na sinalização dopaminérgica. / RATIONALE: Obesity is an excess of white adipose tissue that causes mechanical, metabolic injuries and mortality of approximately 2.8 million people annually. Its physiology involves alterations in humoral, neural and behavioral mechanisms that interact in brain, which can be elicited by nutrients. Functional neuroimaging techniques, such as Single-photon Emission Tomography (SPECT), arise as tools to evaluate abnormalities in vivo. Therefore, we argue if changes in brain images, after intake of glucose, compared to a test with water, would explain, at least in part, the altered feeding behavior of obesity. OBJECTIVES: Compare perfusion (activation) of brain SPECT images after water vs. glucose intake in obese and normal weight volunteers. Correlate the intensity of perfusion found with blood levels of leptin, insulin and glucose, as well as with the amount of body fat. MATERIALS AND METHODS: 10 women with and 10 women without obesity (n = 20) underwent SPECT twice, after labeling by the [99mTc]-ECD 30 minutes after ingesting 300 ml of water and a solution of 75 grams glucose, on separate days (40 SPECTs) being each subject control of yourself. The images were compared between groups and intragroup using Statistical Parametric Mapping. Mixed effects models were used to assess correlations between variables. RESULTS AND DISCUSSION: Obese women have higher activation in regions of the Default Mode Network and Salience Network after test with water. While obese have become more engaged in the perception of physiological processes (such as hunger and thirst) at baseline, with water, this group do not seem to respond to changes of these processes elicited by the food offered. After glucose intake, only those with normal weight increased perfusion in regions related to food reward and behavioral control, such as striatum and orbitofrontal and prefrontal cortices. The perfusion in these areas was negatively correlated with the interaction between leptin and insulin (Coef = -. 0.001, p = 0.003). In addition, these regions receive dopaminergic afferents and therefore we hypothesized that deficits in dopamine signaling could explain the results observed. CONCLUSION: Glucose intake elicited responses related to normal food reward and the appropriate control over appetite in women without obesity, not being observed the same in obese volunteers, a process in which the interaction between insulin and leptin, as well as dopamine signaling seems to participate.

As disfonias neurológicas são distúrbios vocais que acompanham lesões ou alterações no sistema nervoso. O Acidente Vascular Encefálico (AVE) é a segunda causa de morte no mundo e os danos cerebrais causados podem afetar a comunicação do indivíduo em diversos aspectos. As alterações de voz características dessas lesões são pouco descritas quanto à localização e extensão do acometimento cerebral. Desta forma, torna-se importante compreender a interferência das alterações no Sistema Nervoso Central (SNC) na produção da voz, visando maior substrato para reabilitação. Os objetivos do trabalho foram caracterizar a população de pacientes acometidos por AVE conforme a topografia da lesão observada ao exame de Tomografia Computadorizada (TC) e relacionar tais achados com as características fonatórias encontradas. Participaram do estudo 10 idosos acometidos por AVE. Foram realizadas avaliação perceptivo-auditiva da voz por meio d o protocolo CAPE-V, análise do Tempo Máximo de Fonação e avaliação da diadococinesia (DDC) laríngea, por meio do programa Motor Speech Profile Advanced, da KayPentax. Os exames de neuroimagem foram classificados quanto a localização, extensão, lateralidade e território de vascularização da lesão cerebral. Foi observado um grupo homogêneo de cinco sujeitos que apresentaram AVEs extensos de acometimento da artéria cerebral média e outros cinco sujeitos que apresentaram AVEs de menor extensão e com localização variada no cérebro. Os resultados da avaliação de voz foram relacionados com os achados dos exames de imagem e foi observado que não houve relação entre a localização e extensão da lesão cerebral com as alterações vocais dos indivíduos. As vozes dos sujeitos mostraram predominantemente presença de rugosidade, instabilidade e pastosidade, além de velocidade reduzida e instabilidade na repetição de vogais, indicativas de alteração no controle motor laríngeo, bem como redução dos tempos máximos de fonação, indicativos de alteração no controle do fluxo aéreo. / Neurological dysphonias are vocal disorders accompanying injuries or changes in the nervous system. The Cerebrovascular accident is the second leading cause of death worldwide and the brain damage caused by it can affect an individual\'s communication in several aspects. The voice changes characteristical of these lesions are poorly described as location and extent of cerebral involvement. Thus, it becomes important to understand the influence of changes in central nervous system on voice production, aiming to increase the substrate for rehabilitation. The objectives were to characterize the population of patients who had stroke according to the topography of the lesion to the cerebral computed tomography and correlate these findings with the phonatory characteristics found. The study included 10 elderly patients whit stroke. The perceptual voice analysis through the CAPE-V protocol, the Maximum Phonation Time and larynx diadochosinesis (DDK), through the program Advanced Motor Speech Profile of Kay Pentax, were evaluated. Neuroimaging studies were classified according to location, extent and laterality of the vascular territory of brain injury. Were observed imaging studies from a homogeneous group of five patients who had strokes extensive involvement of the middle cerebral artery and another five patients who had less extensive strokes and varied location in the brain. The results of the voice analysis were related to the fidings of imaging studies. It was found no relationship between the location and extent of brain injury with individuals vocal changes. The voices of the subjects showed presence of roughness, instability and pastiness, changes in larynx motor control and air flow control.

Emotional intelligence (EI) is defined as an individual's capacity to accurately perceive, understand, reason about, and regulate emotions, and to apply that information to facilitate thought and achieve goals. Although EI plays an important role in mental health and success in academic, professional and social realms, the neurocircuitry underlying this capacity remains poorly characterized, and no study to date has yet examined the relationship between EI and intrinsic neural network function. Here, in a sample of 54 healthy individuals (28 women, 26 men), we apply independent components analysis (ICA) with dual regression to functional magnetic resonance imaging (fMRI) data acquired while subjects were resting in the scanner to investigate brain circuits (intrinsic resting state networks) whose activity is associated with greater self-reported (i.e. Trait) and objectively measured (i.e. Ability) EI. We show that higher Ability EI, but not Trait EI, is associated with stronger negatively correlated spontaneous fMRI signals between the basal ganglia/limbic network (BGN) and posterior default mode network (DMN), and regions involved in emotional processing and regulation. Importantly, these findings suggest that the functional connectivity within and between intrinsic networks associated with mentation, affective regulation, emotion processing, and reward are strongly related to ability EI.

The primary tauopathies are a group of neurodegenerative diseases affecting movement and cognition. In this thesis I study Progressive Supranuclear Palsy (PSP) and the Corticobasal Syndrome (CBS), two parkinsonian disorders associated with accumulation of hyperphos- phorylated and abnormally folded tau protein. I contrast these two disorders with Parkinson’s disease (PD), which is associated with the accumulation of alpha-synuclein but has a genetic association with the MAPT gene encoding tau. Understanding the tauopathies to develop effective treatments will require a better grasp of the relationships between clinical syndromes and cognitive measures and how the anatomical and neurochemical networks that underlie clinical features might be altered by disease. I investigate simple clinical biomarkers, showing that a two-minute test of verbal fluency is a potential diagnostic biomarker to distinguish between PD and PSP and that the ACE-R and its subscores could play a role in monitoring cognition over time in PD, PSP and CBS. I assess the implementation of network analysis in Functional Mag- netic Resonance Imaging (fMRI) data, introduce Maybrain software for graphical network analysis and visualisation. I go on to show an overlap between graph theory network measures and I identify three main factors underlying graph network measures of: efficiency and distance, hub characteristics, network community measures. I apply these measures in PD, PSP and the CBS. All three diseases caused a loss of functional connectivity in com- parison to the control group that was concentrated in more highly connected brain regions and in longer distance connections. In ad- dition, widely localised cognitive function of verbal fluency co-varied with the connection strength in highly connected regions across PD, PSP and CBS. To take this further, I investigated specific functional covariance networks. All three disease groups showed reduced connectivity between the basal ganglia network and other networks, and between the anterior salience network and other networks. Localised areas of increased co- variance suggest a breakdown of network boundaries which correlated with motor severity in PSP and CBS, and duration of disease in CBS. I explore the link between gene expression of the tau gene MAPT and its effects on functional connectivity showing that the expression of MAPT correlated with connection strength in highly connected hub regions that were more susceptible to a loss of connection strength in PD and PSP. I conclude by discussing how tau protein aggregates and soluble tau oligomers may explain the changes in functional brain networks. The primary tauopathies are a group of neurodegenerative diseases affecting movement and cognition. In this thesis I study Progressive Supranuclear Palsy (PSP) and the Corticobasal Syndrome (CBS), two parkinsonian disorders associated with accumulation of hyperphos- phorylated and abnormally folded tau protein. I contrast these two disorders with Parkinson’s disease (PD), which is associated with the accumulation of alpha-synuclein but has a genetic association with the MAPT gene encoding tau. Understanding the tauopathies to develop effective treatments will require a better grasp of the relationships between clinical syndromes and cognitive measures and how the anatomical and neurochemical networks that underlie clinical features might be altered by disease. I investigate simple clinical biomarkers, showing that a two-minute test of verbal fluency is a potential diagnostic biomarker to distinguish between PD and PSP and that the ACE-R and its subscores could play a role in monitoring cognition over time in PD, PSP and CBS. I assess the implementation of network analysis in Functional Mag- netic Resonance Imaging (fMRI) data, introduce Maybrain software for graphical network analysis and visualisation. I go on to show an overlap between graph theory network measures and I identify three main factors underlying graph network measures of: efficiency and distance, hub characteristics, network community measures. I apply these measures in PD, PSP and the CBS. All three diseases caused a loss of functional connectivity in com- parison to the control group that was concentrated in more highly connected brain regions and in longer distance connections. In ad- dition, widely localised cognitive function of verbal fluency co-varied with the connection strength in highly connected regions across PD, PSP and CBS. To take this further, I investigated specific functional covariance networks. All three disease groups showed reduced connectivity between the basal ganglia network and other networks, and between the anterior salience network and other networks. Localised areas of increased co- variance suggest a breakdown of network boundaries which correlated with motor severity in PSP and CBS, and duration of disease in CBS. I explore the link between gene expression of the tau gene MAPT and its effects on functional connectivity showing that the expression of MAPT correlated with connection strength in highly connected hub regions that were more susceptible to a loss of connection strength in PD and PSP. I conclude by discussing how tau protein aggregates and soluble tau oligomers may explain the changes in functional brain networks.

BACKGROUND: Early initiation of alcohol and other substance use is considered one of the most important risk factors for the later development of an alcohol use disorder. However, it is not yet well understood to what extent this increased risk reflects neurobiological changes driven by the use itself, and to what extent it reflects pre-existing traits and patterns. This dissertation therefore aims to identify neurobiological and neuropsychological markers that exist prior to the initiation of substance use and may confer risk for earlier use onset. Specifically, the research places a focus on the domains of inhibitory control and learning and memory. METHODS: Adolescents (n=81) were enrolled into the study prior to the initiation of alcohol or other drug use at 13-14 years old to complete baseline brain imaging. Neuroimaging included acquisition of structural magnetic resonance imaging (MRI) and task functional magnetic resonance imaging (fMRI) during response inhibition (emotional Go-NoGo) and spatial memory (virtual Morris Water Task) tasks. Participants also completed a neuropsychological battery that included the California Verbal Learning Test Children’s Version (CVLT-C). Subjects were then followed for up to three years via quarterly online surveys in order to assess initiation of alcohol and substance use. Those who went on to endorse initiating substance use prior to reaching 16 years of age were included in an initiating group (IG, n=21) and those who turned 16 having continuously denied substance use were included in a non-initiating comparison group (CG, n=24). RESULTS: Performance measures on the emotional Go-NoGo (NoGo trial accuracy, Go trial accuracy, Go trial reaction time) showed no significant group differences between IG and CG. Functional brain activation differences, however, were observed in bilateral inferior frontal gyrus (IFG), with CG showing greater activation relative to IG on inhibitory (NoGo) trials with negative versus neutral emotional background images (Negative NoGo>Neutral NoGo). Performance on learning trials for the virtual Morris Water Task, completed offline prior to scanning, showed a subtle learning difference between groups, but no performance or functional brain activation differences were observed on Retrieval or Motor control trials completed during scanning. Performance on the neuropsychological test of verbal learning and memory (CVLT-C) indicated worse learning and memory in IG relative to CG (fewer correct responses on both the Long-Delay Free-Recall and the Recognition trials). CONCLUSIONS: These findings help characterize neurobiological and neuropsychological patterns that exist prior to exposure to substances, and thus may help differentiate adolescents who go on to initiate substance use earlier in adolescence from those who do not. Results suggest brain activation differences in frontal regions may predate use, while activation differences in hippocampal memory systems (observed in some cross-sectional studies of alcohol use) may not. These data help clarify questions of causality and provide a foundation for informing strategies for prevention and intervention efforts in maladaptive alcohol and substance use.

Thesis advisor: Elizabeth A. Kensinger / Emotion enhances memory longevity and vividness. Perceiving an experience as emotional, as well as the autonomic and functional brain responses involved in initially encoding an emotional experience, have been theorized to “tag” these memories. Tagged memories may then be prioritized for consolidation during sleep. However, direct evidence supporting this theory is sparse. The aim of the present study was to determine which encoding-related indicators of memory tagging interact with post-encoding sleep oscillations to promote emotional memory retention and vividness. To test this, participants incidentally encoded positive, neutral and negative multisensory stimuli during 3T fMRI scanning with concurrent heart rate monitoring. Participants provided emotional intensity ratings after each stimulus presentation. Following a 120-min post-encoding nap opportunity recorded with polysomnography, participants completed a surprise memory test. Memory for emotional and neutral stimuli was equivalent, though emotional stimuli tended to be remembered more vividly. Perceived emotional intensity, but not heart rate deceleration (HRD) magnitude or functional brain activity, was diagnostic of later successful retrieval of emotional, but not neutral stimuli. Higher REM sleep theta power during the nap was associated with a greater emotional intensity (EI) subsequent memory effect (i.e., higher EI for later remembered compared to forgotten stimuli) for positive stimuli, which were also remembered more vividly. Higher NREM spindle density was associated with a greater EI subsequent memory effect for neutral stimuli and lesser EI subsequent memory effect for negative stimuli. Lastly, higher numbers of NREM spindle-slow oscillation coupling events predicted a negative relationship between perceived emotional intensity at encoding and memory vividness for negative stimuli. Taken together, the present findings suggest that subjective, rather than objective, encoding-related arousal responses acted as emotion “tags”. How subjective arousal impacted later memory varied as a function of the memory’s emotion category and REM and NREM-specific oscillations. Future work is needed to clarify the underlying mechanisms for these observed effects. / Thesis (PhD) — Boston College, 2022. / Submitted to: Boston College. Graduate School of Arts and Sciences. / Discipline: Psychology.

In many regions worldwide, older adults now form the fastest growing portion of the population. As such, aging research has seen tremendous growth in recent years, with a focus on identifying early biomarkers of age-related disease. However, crucial to understanding age-related disease is to identify what constitutes normal brain aging, and the life-course factors associated with positive outcomes in later life. In support of this goal, the current dissertation is comprised of three manuscripts that aim to investigate the functional and structural correlates of normal aging in a sample of community-dwelling younger and older adults, from both a multimodal and multi-analysis perspective. Study 1: The first study examined how cumulative cardiovascular risk and self-reported levels of physical, social, and cognitive activity are associated with differences in hippocampal volumes in early and later adulthood. Results indicated that greater cumulative cardiovascular risk was associated with smaller hippocampal volumes across age cohorts. Moreover, a negative association found between frequency of social activities and bilateral hippocampal volumes in older adults, suggesting that social activities with a low cognitive load may not be beneficial to structural brain outcomes in older age. Study 2: This study employed novel advances in functional magnetic resonance imaging (fMRI) to study fluctuations in the blood-oxygen-level dependent (BOLD) signal in relation to age and markers of brain health. Specifically, the study examined the relationship between resting-state BOLD variability and markers of both vascular health and lifestyle activity levels. Results indicated that resting-state BOLD variability is increased in older relative to younger adults. The findings also suggest that the association between BOLD variability and lifestyle activity levels may differ depending on age. Study 3: The final study aimed to further investigate the origins of the BOLD variability signal by examining the feasibility of combining functional near infrared spectroscopy (fNIRS) with fMRI brain signal fluctuation data. In addition to providing proof of concept of combining fNIRS hemoglobin metrics with fMRI BOLD variability maps, the results of this study also indicate that the patterns of regional association between resting hemoglobin concentrations and BOLD fluctuations may vary according to age cohort. Together, the three studies comprising this dissertation illustrate the value of adopting a multimodal, life-course perspective in the study of normal aging. These findings also support increasing evidence of a relationship between the BOLD variability signal and age. Given the limitations of cross-sectional designs for demonstrating change over time, longitudinal investigations with larger sample sizes across multiple age groups are needed to further the development of public health measures aimed at promoting successful aging from early adulthood. / Graduate / 2022-12-08

Indiana University-Purdue University Indianapolis (IUPUI) / Prodromal Alzheimer’s disease (AD) has recently been identified as a disease state where pathophysiological changes may progress despite the absence of significant clinical symptoms. Yet, the specific processes of neural dysfunction occurring during this preclinical phase remain unclear. Resting state fMRI (RS-fMRI) in combination with brain connectomic measurements may be able to provide ways to measure subtle connectivity changes in different neurological disease states. For instance, RS-fMRI scans allow us to determine functionally connected yet spatially distinct brain regions that can then be separated into resting-state networks (RSNs). More recently, the exploration of RSNs in disease states have proved promising since they have been reliably altered when compared to a control population. By using brain connectomic approaches to assess functional connectivity we can evaluate the human connectome from a different and more global perspective to help us better understand and detect prodromal neurodegenerative disease states.