123129 research outputs found
Sort by
From Social Isolation to a New Normal: Examining Older Adults’ Post-Pandemic Social and Leisure Participation
Get PDFThe COVID-19 pandemic disrupted older adults’ participation in social and leisure activities that support healthy aging. This cross-sectional study examined reported post-pandemic changes in social and leisure participation and technology use among 82 community-dwelling older adults. The Activity Card Sort assessed reported changes in participation from pre-pandemic to present. Global cognition and self-reported technology use changes, functional disability, depressive symptoms, community integration, subjective cognition, and well-being were also assessed. Participants retained 82.5% of their pre-pandemic activities and used technology 17.3% more. Smaller increases in technology use, fewer functional limitations, and younger age significantly predicted higher activity retention, explaining 28.3% of the variance. A weak positive correlation was found between participation retention and well-being. The reported participation decline is similar to previous reports and may be explained by aging processes more than pandemic influences. The findings highlight the importance of offering a variety of accessible activities to support older adults’ participation.M.Sc
TEACHING WITH CONSCIOUSNESS: A NARRATIVE INQUIRY INTO SCHOLARS’ EXPERIENCES TEACHING ABOUT CONSCIOUSNESS BEYOND THE BRAIN
Get PDFHow might educators challenge the context of present-day mainstream science and philosophy in ways that help students open their minds to the possibility that consciousness may exist beyond the brain? To explore this question, I conducted a narrative inquiry with nine scholars from diverse disciplines, cultural, and philosophical orientations, who share an interest in this question in their personal, professional, and scholarly lives and have experience teaching about consciousness beyond the brain. The stories of their experiences provide insights into consciousness education, a field of inquiry defined as education about perspectives on the source and nature of consciousness and their implications for ways of being, knowing, teaching, and learning. Thematic analysis of the research findings led to six themes relevant to the theoretical development and practice of consciousness education: the three strands of consciousness education—culture, identity, and data; the role of experience; the challenge of integration; awareness of the ‘story we are in’; why consciousness education?—expanding notions of truth, self, and wellbeing; and teaching with consciousness.Ph.D
Late Complications in the Descending Aorta Following Valve-Sparing Root Replacement (VSRR) in Marfan Syndrome (MFS) Patients: A Computational Analysis
Get PDFPatients with Marfan Syndrome (MFS) experience an elevated risk of aortic dissection in the descending aorta (DA) following root surgery. Geometric factors related to either native anatomy or the root surgery may alter hemodynamic factors in the DA, potentially predisposing it to dissection. This study uses computational fluid dynamics (CFD) simulations alongside statistical shape modeling (SSM) to investigate the relationships between aortic geometry, hemodynamic indices, and post-surgical dissection risk. We retrospectively analyze CT imaging of a cohort of MFS patients who underwent root surgery, divided into a “dissection group” and a “non-dissection group.” Due to the absence of 4D MRI data, we conduct sensitivity analyses to evaluate the feasibility of using modified generic waveforms as inlet boundary conditions. After establishing a reliable CFD model, we assess near-wall hemodynamic parameters to explore potential links between flow characteristics and dissection likelihood. Our results show that the “dissection” group exhibits lower time-averaged wall shear stress (TAWSS) and higher oscillatory shear index (OSI) and relative residence time (RRT) in the DA. Additionally, we observe a localized increase in TAWSS near the subclavian artery post-surgery in the “dissection” group, which is not observed in the “no dissection” group. We then apply SSM, including principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA), to identify key geometric features differentiating dissection-prone patients. These analyses highlight morphological factors, including average aortic diameter, descending aorta diameter, and ascending aortic bending angle, as significant discriminators. Virtual modifications of these geometric parameters, when feasible within clinical constraints, may influence hemodynamic patterns in the DA, presenting opportunities to enhance surgical planning from a hemodynamic perspective. Together, these findings suggest that both anatomical and surgically induced geometric features drive hemodynamic changes that may increase the risk of aortic dissection in the DA following aortic root surgery for patients with MFS. Ultimately, integrating CFD-based hemodynamic assessment with SSM provides a promising framework for improving risk stratification in patients with MFS and refining surgical strategies.Ph.D
Cyanobacterial blooms in a warming climate: Paleolimnological assessments of three Boreal Shield lakes in central Ontario, Canada
Get PDFPaleolimnological techniques were used to assess long-term water quality changes in three cyanobacterial bloom-impacted lakes in Algoma, Ontario, Canada. Since the 2000s, frequent cyanobacterial blooms have been reported in these lakes despite stable nutrient levels, allowing investigation of recent climate warming as a possible environmental driver of the blooms. While diatom and chironomid community changes varied among the lakes, accelerated warming (~1990) led to near-synchronous shifts in paleolimnological indicators, including increased sedimentary chlorophyll a, planktonic diatoms (Discostella stelligera, elongate taxa), and chrysophyte scales. Diatom-inferred total phosphorus and chironomid-inferred hypolimnetic oxygen models indicate oligo-mesotrophic conditions prior to ~1950, with slightly higher late-summer oxygen concentrations in Desbarats and Bright lakes. Our results suggest a shift towards enhanced thermal stability due to regional warming, supported by instrumental records documenting rising air temperature, reduced wind speed, and a longer ice-free period observed in the last half century. Our paleolimnological inferences indicate that climate-driven changes in fundamental lake physicochemical properties and internal nutrient loading may be contributing to, or even triggering, recent cyanobacterial blooms in these lakes.The presentation of the authors' names and (or) special characters in the title of the pdf file of the accepted manuscript may differ slightly from what is displayed on the item page. The information in the pdf file of the accepted manuscript reflects the original submission by the author
Modelling resting-state neural oscillations to uncover circuit mechanisms of aging and rTMS response in late-life depression
No full textUnderstanding the brain's intrinsic activity is essential for advancing treatments for psychiatric and neurological disorders. Resting-state oscillations are rhythmic neural patterns observed when the brain is not engaged in a specific task, measurable via magnetoencephalography (MEG) or electroencephalography (EEG). Alpha rhythms, a prominent feature of these oscillations, are altered in several psychiatric and neurological disorders, namely depression. However, their underlying mechanisms remain poorly understood, especially how they vary across brain regions and respond to therapeutic interventions such as transcranial magnetic stimulation (TMS). Gaining mechanistic insight into these dynamics holds critical potential for optimizing neuromodulation strategies and personalizing clinical care. In the first study, we investigated four neural population models (Jansen-Rit (JR), Moran-David-Friston (MDF), Liley-Wright (LW), and Robinson-Rennie-Wright (RRW)) representing the alpha rhythm within cortical and corticothalamic circuits, evaluating their strengths, limitations, and key parameter effects. This provided a deeper understanding of the modelled mechanistic circuits of alpha generation. In the second study, we fitted the previously explored neurophysiological corticothalamic model to resting-state MEG power spectra, to define the changes in spatial and age-related neural mechanisms. We found that corticothalamic activity was most prominent in occipital regions, where aging was associated with increased corticothalamic delays and a slowing of alpha rhythms. In contrast, frontal regions exhibited greater age-related changes in intrathalamic inhibition, suggesting that the mechanisms underlying alpha generation vary across brain regions, and that TMS targeting strategies should account for this spatial heterogeneity. The final study analysed resting-state EEG data from individuals with late-life depression before and after accelerated bilateral rTMS treatment. Successful TMS response was strongly associated with model-derived increased intrathalamic inhibition in the right frontal hemisphere, highlighting the potential role of inhibitory regulation in therapeutic outcomes for late-life depression. This research provides new insights into resting-state oscillations, namely alpha rhythms, in the context of healthy aging and rTMS response in late-life depression. More broadly, it presents a framework for integrating computational modelling with resting-state oscillations to investigate underlying neural circuit mechanisms across space, age, and pathology, showing its application for understanding neuromodulation outcomes and its potential for application across diverse clinical conditions beyond depression.Ph.D
Using Protein Language Model Embeddings to Identify Intrinsically Disordered Regions Driving Protein Functional Diversification
Get PDFThe functional divergence of proteins drives evolutionary novelty and biological diversity, yet the role of intrinsically disordered regions (IDRs) in this process is poorly understood. IDRs are widespread in the human proteome and play key roles in protein function and disease. However, it is difficult to identify the functional divergence of IDRs using standard approaches based on sequence alignments due to their lack of sequence conservation. Protein language models (pLMs) are an emerging deep-learning approach to study proteins and do not rely on sequence alignments. pLMs generate embeddings, which are numerical representations of a protein sequence that capture features related to protein structure, function, and evolution. Here, we leverage pLM embeddings as a novel, alignment-free approach for investigating IDR functional evolution directly from sequences. Using the ancient Cyclin-Dependent Kinase (CDK) family as a case study, we provide evidence that CDK IDRs exhibit functional divergence and that some have converged towards the functions of IDRs in other ancient protein families. Beyond the CDKs, we find that IDR functional diversification appears to be widespread across human protein families. Our results suggest that IDRs are an underappreciated driver of protein functional divergence. pLM embeddings provide an alignment-free framework for quantifying divergence and generating hypotheses about IDR-driven diversification across protein families.M.Sc
“The Bones of Christ”: The Ecclesiological Aspect of the Doctrine of Deification According to Thomas Aquinas’ Postilla super Psalmos
Get PDFThe purpose of this thesis is to examine the relationship how Thomas Aquinas’ expression of the relationship between Christ and his mystical body in the Postilla super Psalmos enriches our understanding of his mature thought on the doctrine of deification. By examining Aquinas’ final biblical commentary within its literary and historical context, this project will show that a significant development in Aquinas’ understanding of the doctrine of deification occurred within his Neapolitan regency (1272-1274).
This project will begin on the macro level by asking how modern scholarship understands deification and if that understanding leads to a bias against Aquinas’ theology. This question will be answered in the affirmative through an analysis of the Majority Position regarding deification in scholarship, which holds that in order to have an authentic doctrine of deification, one must accept their understanding of deification as rooted in the theological developments of Gregory Palamas. By examining the patristic origins of deification, this project will show that not only is the Palamite criterion a thirteenth-century development, but that during the patristic era, there was a common understanding of deification in the Greek East and Latin West. Based on this, the project will move forward to show that just as the Palamites’ authentic development occurred in the Greek East, so too could ideas such as created grace, divine simplicity and the Lumen gloriae, that undergird Aquinas’ concept of deification, develop authentically and organically within Latin theology.
After tracing these developments, the project will shift to focus more specifically on the further developments to Aquinas’ understanding of deification that occur within his Super Psalmos. It will begin by providing an in-depth analysis of the text’s prologue that shows that Aquinas has firmly situated his text in Augustinian eschatology and the triple way of Pseudo-Dionysius. The project will conclude by noting that Aquinas’ understanding of deification is developed in the Super Psalmos to include an ecclesial aspect whereby the viator, imitating Christ the exemplum, serves as an instrumental cause of Christ’s graced exemplum to others in a way that deifies them and invites them to follow Christ towards their final union.Doctor of Philosophy (PhD
Novel Statistical Methods for Mitigating Inter-site Variability in Neuroimaging Studies
Get PDFIn neuroimaging studies, combining data collected from multiple study sites or scanners is becoming common to increase the statistical power and the reproducibility of scientific discoveries. However, unwanted inter-site variabilities are commonly observed across these multi-site datasets, making data integration challenging. From a statistical perspective, such variabilities introduce biases in mean, variance, and covariance structures. While several harmonization methods have been proposed to address these issues, most rely on univariate approaches that fail to account for covariance heterogeneity—an essential consideration in modern multivariate analyses. To overcome these limitations, this thesis develops three multivariate harmonization frameworks, each tailored to distinct neuroimaging modalities and scientificobjectives, and all aimed at advancing the mitigation of inter-site effects while preserving biological signals.
First, RELIEF (REmoval of Latent Inter-scanner Effects through Factorization) introduces a low-rank matrix decomposition to separate and remove both explicit and latent scanner variations from shared biological variations. Applied to diffusion tensor imaging (DTI) data from the Social Processes Initiative in Neurobiology of the Schizophrenia (SPINS) study, RELIEF outperformed existing methods in homogenizing covariances, reducing scanner prediction accuracy and enhancing statistical power in simulations.
Second, SAN (Spatial Autocorrelation Normalization) tackles site effects in the vertex-level cortical thickness data by modeling spatial covariance heterogeneity through Gaussian processes. SAN preserves spatial smoothness and corrects local covariance distortions, demonstrating its utility in the SPINS dataset.
Third, LASERC (Latent Adjusted Site Effect Removals for Connectome) addresses site-induced variability in the functional connectivity. By leveraging a low-rank latent factor decomposition, it separates site effects from functional connectivity patterns. Evaluated on the Autism Brain Imaging Data Exchange (ABIDE) dataset, LASERC mitigates site-driven variability in connectivity matrices while retaining biological associations.
Together, RELIEF, SAN, and LASERC collectively provide a modular toolkit for harmonizing diverse neuroimaging data types—vectors, spatial surfaces, and connectivity matrices—while addressing unique data structures and domain assumptions. This work advances the integration of multi-site studies, enhancing reproducibility and enabling more reliable discovery of neuroimaging research.Ph.D
Essays on Economics of Entrepreneurship
Get PDFThis dissertation examines the economics of entrepreneurship through three studies on how entrepreneurs develop innovations under uncertainty. I investigate: (1) how mentorship in startup accelerators creates value through learning and guidance, (2) how institutional designs shape outcomes when entrepreneurs and mentors hold different beliefs, and (3) how entrepreneurs use crowdfunding to learn about market demand. Using structural modeling, Bayesian frameworks, and empirical analysis, I study mechanisms that improve entrepreneurial decision-making and the welfare implications of institutional arrangements.
In the first chapter, “A Structural Model of Mentorship in Startup Accelerators,” I use data from the Creative Destruction Lab to estimate a dynamic structural model where mentors allocate mentorship over time and entrepreneurs decide whether to follow their advice. This model disentangles two key mentorship mechanisms: a direct effect improving startup quality, and a screening effect identifying high-potential ventures. I find that mentorship enhances outcomes through both channels and learning gains vary across sectors: traditional sectors show early learning, while emerging ones like quantum computing exhibit more gradual gains. Counterfactual analysis quantifies the value of mentors' strategic input and suggests sector-specific mentorship designs.
The second chapter, “Persuading Mentors: Heterogeneous Priors and Innovation Outcomes,” explores how combining advisory and funding roles affects innovation when mentors and entrepreneurs have divergent beliefs. I develop a Bayesian model where entrepreneurs may adopt mentors’ plans to enhance credibility, even if they believe their own plans would yield better outcomes. This dynamic can create inefficiencies and welfare loss, especially when belief divergence is large. I show that separating advisory and funding roles, as in some accelerator or VC models, can mitigate these tensions and improve outcomes in deep tech and other novel sectors.
In the third chapter, "Learning about Product Demand through Crowdfunding," I investigate how entrepreneurs strategically use crowdfunding platforms to learn about market demand while financing their projects. I develop a Bayesian learning model where entrepreneurs make pricing decisions that trade-off immediate profits against the informational value. Using data from Kickstarter's digital games category, I shows that entrepreneurs' pricing decisions are consistent with incentives for learning.Ph.D
Preparing for hibernation: above ground activity and body temperature of free-living golden mantled ground squirrels (Callospermophilus lateralis)
Get PDFThe annual cycle of obligate hibernators includes a prehibernation phase that remains poorly understood. To elucidate prehibernation behavior and physiology under natural conditions, we investigated patterns of above ground presence and body temperature (Tb) in free-living golden-mantled ground squirrels (Callospermophilus lateralis, GMGS, (Say, 1823)). We hypothesized that above ground presence and Tb, and transitions in daily patterns as animals approach hibernation, reflect energy conservation strategies that are important to survival. We found that daily above ground presence and Tb were predicted by day length, ambient temperature (Ta), and soil temperature (Tsoil), and that Ta and humidity explained variation in duration and timing of above ground presence. In the month prior to hibernation, animals reduced time spent above ground, reduced time spent at high Tb, and progressively decreased Tb. Daily and transitional patterns differed between animals that survived until hibernation and those that died prior. Compared to animals that died, survivors spent less time above ground per day and exhibited faster and more pronounced decreases in Tb. Our results do not provide evidence for causes of mortality for those that died, but an extended active season and above ground presence may expose animals to predation risks.The presentation of the authors' names and (or) special characters in the title of the pdf file of the accepted manuscript may differ slightly from what is displayed on the item page. The information in the pdf file of the accepted manuscript reflects the original submission by the author