34296 research outputs found
Sort by
Refining Risk: Statistical Characterization of Wordlist Memory Task Performance for Preclinical Differentiation and Prognosis of Alzheimer’s Disease
No full textVerbal episodic memory impairment (i.e., linguistic-mnemonic deterioration) appears as one of the earliest and most sensitive cognitive markers of preclinical Alzheimer’s disease (AD). However, standard scoring methods for wordlist memory tasks may not optimally capture the subtle cognitive processes most predictive of clinical progression. This dissertation holistically assessed whether metrics derived from Hierarchical Bayesian Cognitive Process (HBCP) modeling—latent cognitive variables aligning with features of a wordlist memory test—offer superior prognostic value for identifying individuals at higher risk of clinical decline over 36 months compared to traditional recall scores and serial position effect process metrics. The study used statistical modeling techniques to analyze and compare verbal memory data from two wordlist memory tasks (RAVLT, ADAS-Cog) in Alzheimer’s Disease Neuroimaging Initiative (ADNI) participants spanning cognitively normal, subjective cognitive impairment, and mild cognitive impairment diagnostic categories at baseline. Covariate-adjusted logistic regression models included demographic factors (age, sex, education) and baseline clinical status to improve prediction accuracy, with model performance assessed using various statistical tests and interpreted through a clinical lens. Results showed partial, task-dependent support for HBCP parameters’ cross-task generalizability (Aim 1). In the RAVLT sample (N = 303), all three cognitive parameters showed robust prognostic associations with large effect sizes, while in the ADAS-Cog sample (N = 301), only the latent probability of recall of an episodic memory on the delayed recall task remained significantly prognostic. Pooled generalized estimating equation models found no statistically significant Task × M interactions, which supports measurement invariance after standardization, though mixed findings warranted a nuanced interpretation. In the RAVLT sample (Aim 2.1), hierarchical model comparisons indicated that the HBCP M model had discrimination statistically equivalent to Traditional scoring but was significantly better than serial position effect (SPE) Process scoring. Information criteria favored Traditional over HBCP M, highlighting Traditional’s parsimony advantage, although the Full model had the highest discrimination at the expense of increased complexity. These findings emphasize that the prognostic value of HBCP parameters largely depends on task structural features. Within-region comparison (Aim 2.2) showed that quantitative cognitive process composites offered only slight benefits over SPE regional ratios in the primacy domain, and no benefits in recency. The between-region comparison (Aim 2.3) strongly supported primacy’s superiority over recency, aligning with theories and prior literature indicating the sensitivities of early-list encoding processes to incipient AD. This dissertation merged cognitive neuropsychological testing with various statistical methods to help inform optimal approaches for identifying and characterizing risk in individuals at the preclinical stages of AD. Future research directions include seeking external validation in independent, more demographically diverse cohorts and biomarker correlation studies to establish neurobiological construct validity of HBCP parameters across various clinical contexts
Replanting Hawthorns: Towards a Diachronic Translation of \u3cem\u3eÀ la recherche du temps perdu\u3c/em\u3e
No full textA novel as long and complex as Marcel Proust\u27s À la recherche du temps perdu poses a challenge of methodology to its translators: how should one proceed with rendering its unique style, its bifurcating sentences, its aesthetic philosophy, and the interplay between content and form? What should one prioritize? This dissertation proposes a method of translation that gives priority not only to sentence structure, as most English translators of the novel have done, but also to the repetition and diachronic development of images, themes, and phrases in order to capture and recreate certain recursive effects of the novel’s form. It focuses especially on the passages relating to hawthorns in Du côté de chez Swann, examining their central importance to the architecture of La Recherche as a whole, analyzing how imagistic and sonic themes are developed in them, and assessing how previous English translations have rendered these passages. Guided by this analysis, it also attempts a translation of its own
Somatosensory Cortex Responses to Illusory Touch in a Visual-Tactile Integration Task
No full textPerceptual illusions arising from multisensory integration reveal how the brain constructs subjective experience from sensory input, and provide a powerful means of allowing the investigation of neural processes that track perceptual interpretation rather than physical stimulation alone. While visual illusions induced by auditory or tactile stimuli have been fairly well studied, the neural mechanisms underlying visually-induced illusory touch remain unclear. This study investigated whether illusory tactile perception engages the somatosensory cortex in a manner similar to veridical touch. Twenty healthy adults participated in an event-related functional magnetic resonance imaging (fMRI) study while performing a tactile numerosity judgment task. Each trial consisted of an initial visuotactile stimulus followed 100 ms later by one of four conditions: no stimulus, visual stimulus only, tactile stimulus only, or a combined visuotactile event. Participants reported the number of touches perceived, regardless of visual input. Behavioral results show that participants reported illusory touch on 54% of trials containing only a second visual stimulus, with substantial inter-individual variability. fMRI whole-brain analyses contrasted illusory touch versus veridical touch under matched physical stimulation, and this contrast revealed significant activation in the insular cortex and the adjacent rolandic operculum, which is part of the secondary somatosensory cortex. However, there was no detectable primary somatosensory cortex activation at the group level. In contrast, veridical perception across different physical conditions produced widespread activation across sensory and task-related networks. These findings suggest that visually-induced illusory touch arises from top-down integrative processes in higher-order multisensory regions rather than from the stimulus-driven activation of the primary somatosensory cortex. The results provide support for models of multisensory perception that emphasize probabilistic inference and indicate that illusory and veridical tactile percepts may engage distinct neural mechanisms
Investigating the Role of Phonology and Morphology in the Processing and Acceptability of Bilingual Code-Switching
No full textThe primary aim of this dissertation was to investigate the often-neglected linguistic domains in the study of bilingual code-switching, namely phonology and morphology. Motivated by empirical and theoretical observations, this dissertation narrowed the focus of study to auxiliary-participle compounds in Spanish-English code-switching. This structure has attracted attention because progressive code-switched phrases are favored over perfect ones in both distribution and comprehension, a pattern that has been interpreted through usage-based explanations. Here, I examined the potential influence that phonological and morphological variables can exert on the asymmetry. Two studies were designed, each isolating a different element from the aspectual compound. Study 1 controlled for and evaluated the role of auxiliary length in different aspect (progressive, estaban vs. perfect, habían - Experiment 1) and same aspect (present perfect, han vs. past perfect, habían - Experiment 2) structures. Study 2 concentrated on participle-level characteristics that are incongruous between English and Spanish past participles, specifically, verbal inflection (regular vs. irregular - Experiment 3) and syllable length (monosyllabic vs. multisyllabic - Experiment 4). Using a combination of online (eye-tracking while reading) and offline (acceptability judgments) tasks, this dissertation revealed that the well-documented aspectual processing asymmetry stems from a methodological confound. The eye-tracking results from Experiments 1 and 2 combined indicate that the lack of preference to switch in [han + English past participle] structures is triggered by the short length of the Spanish auxiliary rather than by aspect, a finding hard to accommodate by experience-based accounts. Acceptability ratings reflect these processing patterns and also suggest a metalinguistic bias to (left) phrase-boundary switching, i.e., at the auxiliary. Even though no significant results were observed for participle length and regularity, a trend emerged that merits further investigation of verbal inflection type. Overall, these findings underscore the role of non-syntactic variables in bilingual sentence processing, with implications for the design of experimental materials and the generalization of results
Reimagining Safety and Justice: Experiences of and Resistance to Police Sexual Violence in New York City
No full textIn the era following both Black Lives Matter and #MeToo, police sexual violence (PSV) currently constitutes the second most commonly reported police complaint after excessive force. Police across the U.S. are caught for some form of sexual misconduct once every five days, with on-duty officers found to commit sexual assaults at more than double the rate of the general population. The gendered and racialized perpetration of sexual violence extends far beyond policing; however, it has been intensified within the masculine, militarized carceral structures designed to legally control and punish.
Accordingly, this dissertation utilizes an interdisciplinary, multi-method approach to offer what is, to my knowledge, the first in-depth study of the lived experience of sexual violence enacted by officers outside of prisons and jails, in the city with the nation’s largest police force. The project draws from 37 interviews and a survey of over 3,700 New Yorkers to examine the ways in which survivors experience, resist, and heal from these incidents. Conducted in partnership with a citywide coalition of over 200 community organizations, this study captures: (i) the lived experiences, contextual complexities, and cumulative consequences of the wide range of acts that constitute PSV—including everyday forms of sexual harassment, legal forms of frisking and strip searches, and criminal forms of sexual coercion and assault—as well as their accumulation in the lives of those navigating heavily policed environments; (ii) pervasive psychosocial processes of dehumanization, misogyny, and hegemonic masculine bonding that not only underlie a diverse spectrum of sexual violence within an institutional culture and legal landscape of impunity, but also reflect gendered and racialized hierarchies of humanity that shape differential vulnerabilities to both interpersonal and state harm; (iii) participants’ modes of creating safety through relational embodiment to illuminate healing as a fundamentally collective decolonial project, requiring an interplay of individualized psychological and somatic growth, along with social and systemic transformation; and (iv) a collective reimagining of justice oriented toward immediate repair, harm reduction, long-term prevention, and the development of healing-centered communities.
Ultimately, this dissertation reframes PSV as an ongoing force of terror(ism) and form of state betrayal that strengthens the widespread internalization of dehumanizing carceral logics, eroding our collective safety. Participants’ experiential knowledge sheds light on structural, ideological, psychosocial, and embodied ways to address the root causes of various forms of racialized and gendered harms. In doing so, this work seeks to inform existing abolitionist and intersectional feminist research, as well as social movements against both sexual violence and police harm
Genocide, Memory & Prevention: Ethical Inquiry and Critical Thinking through Research & AI
No full textThis assignment engages students in ethical inquiry and critical analysis by examining lesser-known or silenced genocides within a global and historical context. Using academic sources alongside responsible use of generative AI, students compare dominant and marginalized narratives, analyze recurring patterns and causes of genocide, and reflect on prevention strategies. The assignment emphasizes ethical AI use, historical memory, and the development of civic awareness, empathy, and critical thinking in the study of race, power, and inequality
Analyzing Headers with ChatGPT
No full textThis assignment invites students to examine how section headers function as rhetorical and organizational tools in research writing by analyzing headers generated by ChatGPT. Using a shared sample text, students prompt generative AI to produce multiple sets of headers that foreground different analytical emphases (e.g., race or language). Students then compare these AI-generated headers to identify how shifts in emphasis reshape the implied structure, focus, and interpretation of a text. Through analysis and reflection, students use these insights to inform their own planning for inquiry-based research writing, while critically evaluating the affordances and limitations of generative AI at an early stage of the writing process
Protein Allostery Probed by Ligands Across Sites
No full textAllostery is a pervasive regulatory principle throughout biology, yet the structural pathways by which distal inputs alter active‑site chemistry within protein structures remain incompletely defined and exploited. This dissertation uses protein tyrosine phosphatase 1B (PTP1B) as a tractable model to map those pathways with complementary experimental and computational tools. I integrate high‑resolution hydrogen-deuterium exchange mass spectrometry (HDX-MS), room-temperature crystallography, NMR, steady-state kinetics, crystallographic pseudo-ensembles, and machine-learning-guided ligand discovery. The working premise is that regulation reflects redistribution within a conformational ensemble rather than a binary switch. That orthogonal perturbations by small molecules, mutations, and protein partners can be used to both delineate and control the underlying network. HDX‑MS establishes a solution‑state baseline for PTP1B dynamics and quantifies long‑range responses to ligands, while pseudo‑ensembles provide structural context for interpreting those dynamics. Together, these methods create a host of sensitive readouts for coupling among catalytic loops and remote surfaces.
Applying this framework, I first compare an active‑site inhibitor (TCS401) with an established allosteric ligand (BB3). Both depress exchange in the Q and WPD loops, with BB3 eliciting broader, distal changes consistent with entropic compensation, thereby extending the known allosteric network beyond prior assignments. I then use biophysical and structural techniques to probe four rare PTPN1 gene variants identified from an outlying lean cohort and the UK Biobank, altering PTP1B function in cells. These biophysical experiments trace the most potent effects to remote structural determinants that perturb catalytic-loop behavior. Several high‑impact positions cluster on surface pockets that prove ligandable, nominating previously untapped regulatory footholds. In a physiological counterpoint, the adaptor Grb2 binds the PTP1B proline‑rich region (PRR) and directly increases kcat across multiple phosphopeptide substrates without substrate colocalization. NMR and proteomics support specific PRR-dependent engagement and stabilization of catalytic-domain elements, demonstrating scaffold-mediated allostery in a phosphatase that lacks a dedicated regulatory domain. Finally, I evaluate a distal pocket centered on Loop‑16 (L16). The P241G mutation increases kcat by approximately 25% with no change in Km, consistent with V‑type modulation transmitted from L16 to the active site. A multiconformer AtomNet screen yields eleven reproducible binders, dominated by an imidazo‑pyridine series; most are functionally silent, but Atw103 inhibits noncompetitively with an HDX signature marked by E‑, pTyr‑recognition, and Q‑loop protection, consistent with distal transduction. A thermally derived species, Atw47B, shows an active‑site‑centered HDX pattern and lot dependence, cautioning against misassignment.
These results support a model of PTP1B as a distributed regulatory system with multiple ligandable surfaces that can be read and written through ensemble redistribution, rather than simpler models of conformational switching. Conceptually, it concludes with thematic motivations that align with my future research interests in multivalency, aiming to convert weak recognition of individual small molecule ligands into precise situational control of disease targets. I outline translational strategies by others that include bidentate ligands bridging adjacent pockets to encode cooperativity and selectivity, and event‑driven proximity modalities such as PROTACs and PhosTACs to extend control across proteins. I also propose explicit tests for combinatorial allostery at multiple PTP1B sites using full concentration-response surfaces fit to two-site models with cooperativity terms, enabling logic-like integration when pockets are co-occupied. These principles generalize to other dynamic enzymes and provide a framework for selective pharmacology, especially in phosphatases
Analyzing the Use of AI to Transcribe an Interview in Spanish
No full textThis assignment explores the differences between oral and written language and the challenges of transcribing speech. Students conduct an interview with a selected person, using ChatGPT to create and refine interview questions. They record the interview and use Otter.ai to generate a literal transcription that captures natural speech features such as repetitions and hesitations. After reviewing and correcting the transcript against the audio, students format it with participants’ initials to focus on content rather than timestamps. A simplified transcription guideline is applied to mark key elements. Finally, students reflect on the transcription process, addressing what was most difficult to represent in writing, whether it is necessary to transcribe everything as spoken, how the activity could be improved, and the usefulness of AI in the task
Bureau of Prisons Visitation Policy
No full textThis assignment in Law and Institutional Treatment requires students to design a modern, equitable, and legally sound visitation policy for individuals housed in the Federal Bureau of Prisons (BOP). Using Microsoft CoPilot and ChatGPT as drafting tools, students will generate, edit, and refine AI-produced policy language to ensure it aligns with principles of nondiscrimination, institutional safety, and constitutional standards. Students must critically evaluate AI outputs, revise them to reflect course concepts, and compare iterative drafts to assess which version demonstrates stronger policy structure and legal grounding. A bonus component invites students to generate AI-created images representing an equitable visitation space and analyze how design reflects policy values. The assignment emphasizes applied legal reasoning, policy development, institutional equity, and critical engagement with AI tools in professional practice