21090 research outputs found
Sort by
A New Cumulative Air Toxics Risk Assessment for Mobile Sources Introducing Stochastic Human Health and Deterministic Ecological Methods
No full textAir toxics emitted from mobile and stationary sources continue to pose substantial risks to both human and ecological health, particularly in urban areas and environmentally sensitive regions. However, many existing risk assessment frameworks rely on deterministic assumptions, narrowly focus on select exposure pathways, and often overlook ecological impacts. This dissertation addresses these limitations by developing and applying both deterministic and stochastic methodologies to assess air toxics risk across diverse sources, spatial scales, and exposure scenarios.
One of the key objectives of this research is to introduce a new stochastic risk assessment method for mobile source air toxics (MSATs) that provides a more detailed and comprehensive characterization of health risks. Structured as a manuscript-based thesis, it comprises four peer-reviewed journal articles, each presenting a novel methodology demonstrated through real-world case studies. This structure ensures a logical progression from foundational risk assessment methodologies to more comprehensive, multi-source, multi-pathway, and probabilistic approaches, providing a holistic and integrated evaluation of air toxics risk.
The first article presents a new deterministic human health risk assessment methodology for MSATs. It integrates emissions modeling using MOVES, receptor-specific air dispersion modeling with AERMOD, and multi-pathway exposure estimation across inhalation and ingestion routes. In the Saint Paul, Minnesota case study, benzo(a)pyrene cancer risks exceeded the target threshold at two urban receptors: 4.59 × 10⁻⁵ (45.9 in 1 million) at the Saint Paul–Ramsey Health Center and 2.02 × 10⁻⁵ (20.2 in 1 million) at the Anderson Office Building. Ingestion contributed significantly to total exposure. These findings highlight the importance of cumulative risk analysis and the need for probabilistic methods to capture inter-individual variability.
The second article extends the methodology to assess cumulative, multi-pathway human health risks from stationary sources across Kuwait. It evaluates emissions from glycol dehydration units, wastewater treatment plants, and oil and gas operations using a national emissions inventory and dispersion and deposition modeling across three air quality zones. While most risk levels were within regulatory thresholds, cancer risk at the Ahmadi Hospital site exceeded the 1 × 10⁻⁵ target threshold for adults, with benzo(a)pyrene—the cancer risk driver—contributing a risk of 1.09 × 10⁻⁵ (10.9 in 1 million). This case study demonstrates the value of geographically resolved modeling for informing national-scale mitigation strategies.
The third article introduces the Ecological Health Assessment Methodology (EHAM), addressing a previously underexplored dimension of air toxics risk. EHAM combines fate and transport modeling, habitat-specific food web development, and bioaccumulation analysis to characterize ecological risks from air toxics. Applied across Kuwait’s air quality zones, EHAM revealed elevated risks in the coastal region, particularly for carnivorous shorebirds, with an ecological screening quotient of 3.12 × 10³ driven by the biomagnification of benzo(a)pyrene. By contrast, risks were negligible in the inland and production zones. The EHAM supports both screening-level and comparative ecological risk assessments and enables ecological evaluation at the national scale, extending beyond localized industrial sources to encompass broader emission landscapes.
The fourth article builds on the earlier methodologies and develops the first stochastic human health risk assessment methodology specifically tailored to MSATs. Using Monte Carlo simulation, it quantifies variability in key exposure parameters such as body weight, inhalation rate, exposure duration, and dietary intake. In the Saint Paul case study, the model produced full cancer risk distributions for adults and children. Computed cancer risks ranged from 5.47 × 10⁻⁴ to 4.98 × 10⁻² (95th percentile) for adults, and from 2.95 × 10⁻⁴ to 2.28 × 10⁻² (95th percentile) for children. To contextualize, the upper bound of 4.98 × 10⁻² for adults is comparable to the lifetime odds of dying from all preventable causes of death (1 in 19). The model identified high-exposure scenarios and vulnerable subpopulations often obscured in deterministic assessments.
Together, these contributions form an integrated risk architecture that advances emissions estimation, dispersion modeling, exposure characterization, and probabilistic risk quantification for human and ecological receptors. The methodologies developed in this research provide a comprehensive foundation for air quality engineers, public health scientists, and environmental regulators.
By improving the representation of variability, cumulative exposure, and pathway integration, this work strengthens the credibility, transparency, and practical utility of air toxics risk assessment to support informed environmental decision-making
Investigating Multimodal Tasks in Asynchronous Online German Instruction
No full textIn this dissertation, I investigate task-based language learning (Ellis, 2003; Willis & Willis, 2007) in asynchronous online German instruction. Following a Design-based Research methodology, the study involves the development and implementation of a learning module into an online introductory German course. The module is built around a task which draws on principles of foreign language pedagogy (Biebighäuser, 2021; Funk, 2010) and multiliteracies (New London Group, 1996). A review of previous literature reveals that most studies which employ instructional interventions in language learning focus on synchronous online or in-person instruction with little focus on asynchronous learning. Previous approaches to online teaching stem from early distance education and focused on self-study (Andrade & Bunker, 2009) and automating learning (White, 2003), and are associated with behaviourist pedagogies (Bates, 2019; Swaffar, 2014). Taking a multiliteracies perspective, I argue that asynchronous online language learning requires tasks that are situated in real-world online communicative practices such as those that have become a normal part of daily life (Mills, 2015). Such tasks should be at home in an asynchronous online environment as it is insufficient to simply add technology to learning activities that are normally carried out in person (Bates, 2019) or which are rooted in singular understandings of literacy with a strong focus on reading, writing, and grammatical accuracy. I argue that the multimodality of digital literary practices (Kress, 2003) is beneficial for learners who can lean on the combination of meaning making modes when interpreting and creating multimodal texts. Data from the study include survey questions, semi-structured interviews, and analyses of participant submissions. The results of this study inform future research and teaching in foreign language learning as well as online learning more generally
Development of Functional Binders and Li2S@Carbon Nanocomposites for High-Performance Lithium Sulfide Batteries
No full textLithium sulfide (Li2S) is a promising cathode material for lithium-sulfur batteries (LSBs) owing to its high theoretical capacity (1166 mA h g-1) and potential for safer, scalable battery architectures. In contrast to sulfur cathode, Li2S enables direct pairing with commercial anode materials, avoiding the safety risks of lithium metal. Despite these merits, practical application of Li2S is challenged by its hygroscopic nature, which forms insulating LiOH/Li2O surface layers that cause a large first-charge overpotential; its high melting point (~938 °C), which prevents melt infiltration into carbon frameworks; sluggish redox kinetics; severe polysulfide dissolution; poor conductivity. Addressing these challenges requires integrated advances in binder design, electrode engineering, and cathode nanostructuring.
The large first-charge overpotential due to the insulating LiOH/Li2O surface layer in Li2S-LSBs hinders activation and induces irreversible side reactions. Chapter 3 proposes mitigating the activation barrier by exploiting the reaction between polyvinylidene fluoride (PVDF) binder and LiOH/Li2O through dehydrofluorination. The overpotential was successfully reduced from 3.74 V with 30 min slurry grinding to 2.75 V by extending slurry stirring to 48 h. However, PVDF was also found to react with Li2S itself, partially consuming active material and lowering discharge capacity. Overall, this study provides mechanistic insights into the origin of Li2S activation overpotential and demonstrates the dual role of conventional PVDF binders, where slurry processing with PVDF can effectively reduce the first-charge barrier, while also highlighting the limitations of PVDF as a binder for Li2S electrodes.
Since PVDF proved unsuitable for Li2S electrodes, Chapter 4 investigates alternative binders capable of enhancing the electrochemical performance of Li2S-LSBs. A binder based on a zinc acetate triethanolamine (Zn(OAc)2·TEA) complex was developed, which not only provides strong polysulfide-trapping ability but also exhibits redox catalytic activity, leading to markedly improved capacity, rate capability, and cycling stability compared with PVDF. To further reinforce electrode integrity and improve dispersion stability, polyethylenimine (PEI) was incorporated to form a Zn(OAc)2·TEA/PEI hybrid binder. Electrochemical testing showed that Li2S cathodes employing Zn(OAc)2·TEA/PEI with 10 wt.% PEI achieved superior rate performance, high discharge capacity, and excellent long-term cycling stability. An additional advantage of these binders is their fluorine-free composition, which aligns with sustainability goals and complying with emerging regulations, including EU restrictions on per- and polyfluoroalkyl substances (PFAS).
In Chapter 5, an efficient precursor solution infiltration-decomposition strategy was invented to synthesize Li2S@Carbon nanocomposites under mild conditions, overcoming the challenges of Li2S’s high melting point, poor solubility, and the large particle size of commercial Li2S. In this approach, Li2S was first reacted with carbon disulfide (CS2) in ethanol at ambient temperature to form a highly soluble lithium trithiocarbonate (Li2CS3) precursor, which was readily infiltrated into mesoporous Super P carbon (SP). Subsequent thermal decomposition of Li2CS3@SP at 400 °C produced Li2S@SP-400 nanocomposites with a Li2S:SP mass ratio of 60:40, containing finely dispersed Li2S particles (~11 nm) uniformly confined within the Super P matrix. Electrochemical testing demonstrated that these nanocomposites delivered a high discharge capacity of 821 mA h g-1 (Li2S) at 0.1 C, equivalent to 1190 mA h g-1 (S), and exhibited superior rate capability and cycling stability compared to commercial Li2S, non-infiltrated Li2S nanoparticles, and melt-infiltrated sulfur composites (S@SP).
The thermal decomposition of Li2CS3 precursor releases a large amount of CS2 gas (~62 wt.% of the precursor), which creates internal voids and limits the in-pore Li2S loading. To address this, Chapter 6 builds upon precursor infiltration-decomposition method with a multi-cycle strategy, enabling higher Li2S content and in-pore loading. Using mesoporous Super P as the conductive host and Li2CS3 as the precursor, repeated infiltration-decomposition cycles progressively increased the pore filling factor (FF) and in-pore Li2S loading (IPL), from FF = 38% and IPL = 30% for Li2S@SP-1 (one cycle) to FF = 91% and IPL = 73% for Li2S@SP-5 (five cycles), while also raising the overall Li2S content to 70 wt.%. Direct structural evidence from XRD and SEM confirmed reduced crystallite size, suppressed external deposition, and uniform Li2S distribution in the optimized Li2S@SP-5. Electrochemical tests demonstrated that Li2S@SP-5 delivered an initial discharge capacity of 807 mA h g-1 (Li2S) at 0.1 C, 598 mA h g-1 (Li2S) in the first cycle at 1.0 C, and retained 376 mA h g-1 (Li2S) after 500 cycles at 1.0 C.
To construct high-performance cathodes, the functional binder from Chapter 4 was combined with the high in-pore loading Li2S@SP from Chapter 6. This attempt failed because Zn(OAc)2·TEA/PEI-based binders exhibited limitations with highly reactive nanoscale Li2S, resulting in diminished binding effectiveness. Chapter 7 therefore introduces a series of polyethylenimine-epoxy resin (PEI-ER) binders, where high-molecular-weight PEI anchors and catalyzes polysulfides while epoxy crosslinking reinforces mechanical stability, making this strategy particularly effective for stabilizing nanoscale Li2S composites. The in-situ crosslinking method further improved processing by removing the short crosslinking time window and enabling uniform networks without altering Li2S@SP morphology. Electrochemical tests showed the optimized in-situ crosslinked PEI-ER1:1 binder achieved 928 mA h g-1 at 0.05 C, 688 mA h g-1 in the first cycle at 0.5 C and retained 325 mA h g-1 after 1000 cycles at 0.5 C with stable Coulombic efficiency. SEM confirmed its compact structure, establishing in-situ PEI-ER crosslinking as a robust binder strategy for nanoscale, high-loading Li2S cathodes.
Chapter 8 serves as the culmination of these research projects, combining the optimized Li2S@Carbon cathodes from Chapter 6 and functional binders developed from Chapter 7 with commercial Si/C anodes to successfully assemble and evaluate lithium-anode-free full cells, with PVP used as a baseline comparison, thereby demonstrating their practical feasibility. The in-situ crosslinked PEI-ER1:1-based full cell batteries delivered 670 mA h g-1 at 0.1 C and retained 304 mA h g-1 after 100 cycles (~45% retention), outperforming PVP-based full cell batteries (582 to 250 mA h g-1, ~43%). At 0.5 C, the in-situ crosslinked PEI-ER1:1-based full cell batteries achieved 564 mA h g-1 after activation and maintained 377 mA h g-1 after 500 cycles (66.8% retention), while the PVP counterparts fell from 573 to 176 mA h g-1 (30.7%). These results underscore the binder’s role in stabilizing cathodes and mark the successful assembly of lithium-free-anode Li2S full cells with commercial Si/C anodes.
In summary, this thesis addresses the critical challenges of Li2S cathodes, including the large first-charge overpotential, the drawback of PVDF consuming Li2S, the large particle size of commercial Li2S, the high melting point and poor solubility that hinder conventional Li2S@Carbon composite fabrication, and the limitations of binders when applied to nanoscale Li2S, each identified in the process of resolving the preceding issue. By systematically investigating these problems, this thesis advances functional binder design, exploits precursor chemistry, and engineers nanostructured composites, concluding with the successful demonstration of lithium-anode-free full cell batteries. Further improvements could be achieved by employing more efficient carbon hosts with tailored structures, developing high-loading electrodes, integrating solid-state electrolytes to mitigate polysulfide dissolution, and incorporating catalytic components to accelerate Li2S redox kinetics, thereby pushing Li2S-LSBs closer to practical, high-energy-density applications
Differential vulnerabilities: Extreme heat, health and well-being of older adults in Sunbelt cities in the United States
No full textHeat impacts vary disproportionately across geographic regions and social groups. Older adults in Sunbelt cities of the United States (US) are particularly vulnerable to localized urban warming, which threatens their health and well-being. Prevailing scientific evidence suggests that extreme heat events account for higher fatalities among older adults in the US compared to other extreme weather events. Despite existing studies on the impacts of extreme heat on older adults, limited research has addressed the specific socio-demographic, health-related conditions, and strategic coping factors influencing heat stress experiences among older urban populations. As part of a larger study examining heat and ozone-related risks in Sunbelt cities, this thesis adopts a retrospective cross-sectional study design, leveraging the strengths of mixed-methodology to examine predictors of heat stress and coping strategies among older adults in Los Angeles, Houston, and Phoenix. Utilizing a cross-sectional household survey data from 909 older adults in the three cities, the thesis sought to: (1) examine the factors affecting heat stress among older adults; and (2) explore coping strategies among older adults during extreme heat events. Qualitative data were generated from the open-ended survey questionnaire from 67 respondents across the three cities. Data was analyzed using thematic qualitative coding and multiple binary logistic regression analyses. The results reveal that marital status, income, respiratory diseases, physical activity, cool showers, and city of residence were significantly associated with a higher risk of experiencing heat stress. Also, older adults aged 80 and older were less likely to experience heat stress compared to their relatively younger counterparts. Further, the qualitative results uncovered a mix of heat-related coping behaviors adopted by older adults to protect themselves during extreme heat events. These results underscore the complex interplay of geographic, socio-demographic, behavioral and health-related factors affecting heat stress. The study makes important contributions to theory, policy and practice. First, the study contributes to the literature on the geographies of ageing and the relational processes that shape ageing across place, space and scale. Secondly, the study highlights the importance of integrating varied approaches within the geographies of health and healthcare, hence advancing the methodological diversity of the sub-discipline. Finally, the study underscores the need for policies aimed at strengthening social support for older adults, emergency heat risk warning awareness and public cooling systems in Sunbelt cities. The study recommends that urban health strategies should prioritize these demographics, fostering micro-level coping strategies to enhance resilience and reduce vulnerability to heat stress among older populations
Dynamic Treatment Regimes for Within- and Between-Group Interference in Clustered and Hierarchical Datasets
No full textPrecision medicine is an interdisciplinary field that aims to tailor treatments based on an individual’s unique characteristics. Dynamic treatment regimes (DTRs) formalize this process through step-by-step decision rules that utilize patient-specific information at each stage of analysis and subsequently output recommendations for the optimal course of action. To date, much of the biostatistical literature has analyzed DTR estimation under the assumption of no interference; that is, a given individual’s outcome is not affected by the treatments received by others. However, this assumption is often violated in a variety of social and spatial networks, such as households and communities, and particularly in the context of infectious diseases or resource allocation. Although some recent developments have been made for DTR estimation for couples in households and in networks where general forms of interference are taking place, it has yet to be shown how DTRs can be estimated for individuals in clustered networks where interference may occur within and between predefined groups. Specifically, our attention shifts toward hierarchical networks, which provide a framework where interference can occur both within and between groups in the same hierarchy but not across hierarchies.
This thesis contains three main projects that contribute to DTR estimation in the context of service utilization within the healthcare system and under interference networks. In Chapter 3, we show how the dynamic weighted ordinary least squares regression (dWOLS) DTR methodology can be applied to determine whether a patient would benefit from being discharged to home or admitted to a hospital using data simulated from a retrospective cohort study for patients who experienced an opioid-related overdose in British Columbia. This project was inspired from collaboration with a provincial health authority in Canada, which has resulted in the draft of a manuscript that can be submitted for publication following an ethics approval. In Chapter 4, we demonstrate an application of dWOLS to hierarchical networks where both within- and between-group interference takes place, and through a series of simulation studies, we show how incorrectly assuming that there only exists within-group interference can result in lower rates of optimal treatments assigned, particularly as the number of subgroups (and thus the potential for between-group interference) increases. Our models assess the performance of “standard” overlap weights constructed from logistic mixed models with nested random intercepts as well as network weights that are constructed from a joint propensity score. We apply our findings to the simulated opioid overdose dataset, where our goal is to illustrate how sequences of recommended dispositions for opioid overdose patients can be optimized if interference occurs within and between different facilities. Finally, in Chapter 5, we propose several sets of simulation studies to assess the extent to which interference may be taking place between individuals to warrant use of modified dWOLS methodology that accounts for interference
Black hole perturbations beyond the leading order
No full textThis thesis employs numerical methods to study black hole perturbations beyond the leading order. This includes vacuum gravitational perturbations as well as superradiant scalar, vector, and tensor boson perturbations. Understanding and quantifying beyond leading order effects enables more accurate tests of General Relativity and of physics beyond the Standard Model.
First, we explore the nonlinear behavior of gravitational perturbations on a Kerr black hole. The ringdown gravitational wave signal, during the final stage of binary black hole mergers, contains important information about the properties of the remnant black hole, and can be used to perform clean tests of general relativity. However, interpreting the loudest portion of the ringdown signal requires understanding the role of nonlinearities and their potential impact on modeling this phase using quasinormal modes. Here, we focus on a particular nonlinear effect arising from the change in the black hole's mass and spin due to the partial absorption of a quasinormal perturbation. We estimate the size and characteristics of this third-order effect using numerical techniques. Quantifying these effects, we find that they may be relevant in analyzing the ringdown in black hole mergers.
Next, we discuss self-gravity corrections to beyond the standard model particle dynamics around black holes. Specifically, for scalar and vector bosons forming superradiant clouds. Oscillating clouds of ultralight bosons can grow around spinning black holes through superradiance, extracting energy and angular momentum, and eventually dissipating through gravitational radiation. This makes gravitational wave detectors powerful probes of ultralight bosonic fields. Here, we use fully general-relativistic solutions of the black hole-boson cloud systems to study the self-gravity effects of scalar and vector boson clouds. We calculate the self-gravity shift in the cloud oscillation frequency, which determines the frequency evolution of the gravitational wave signal, improving the accuracy of gravitational wave searches for physics beyond the Standard Model. We also perform an analysis of the spacetime geometry of these systems, we compute how the presence of the cloud changes the innermost stable circular orbit and light ring.
Lastly, we investigate the nonlinear phenomenology of the spin-2 boson's superradiant instability. Models that result in a massive spin-2 boson at low energies have been proposed as solutions to the dark matter problem or as modifications to general relativity. The existence of ultralight scalar and vector bosons is constrained using measurements of black hole spins, due to the mechanism of black hole superradiance, and attempts have been made to place constraints on the existence of spin-2 bosons using the same approach. However, those constraints so far have relied on the assumption that the spin-2 superradiant behavior matches that of lower-spin fields. Here we consider a particular nonlinear theory, quadratic gravity, to study the behavior of spin-2 particles that undergo superradiance. We find that the phenomenology is different from that of spin zero or one bosons, increasing the spin of the central black hole and resulting in an extremal black hole horizon
Artificial Intelligence, the Environment and Resource Conflict: Emerging Challenges in Global Governance
No full textArtificial intelligence (AI) embodies a system's capacity to autonomously collect and interpret data from its environment, learn from that data, and apply these insights to inform decision making, problem solving and actions that traditionally require human intelligence. At the heart of this technological process are data centres - facilities where AI models are trained, deployed and maintained. As AI infrastructures expand rapidly across the globe, they bring into sharper focus the often-overlooked costs of digital innovation, particularly their entanglement with extractive economies and conflict dynamics
Modeling the Frequency Response of the Graphene/Electrolyte Interface in Electrochemical Systems
No full textThe graphene/electrolyte interface plays a central role in applications such as supercapacitors and biosensors. Traditionally modeled as two capacitors in series—the Debye capacitance of the electrolyte and the quantum capacitance of graphene—the interface is predominantly governed by the latter. While prior studies have focused on graphene’s voltage-dependent capacitance, its frequency response remains underexplored theoretically. This thesis develops a rigorous mathematical framework to model and analyze the frequency response of the graphene/electrolyte interface under diverse conditions, including finite-conductivity neutral graphene, charged graphene with infinite conductivity, and systems with room-temperature ionic liquids (RTILs).
We first examine a graphene electrode in a dilute electrolyte under small AC voltages. By linearizing and normalizing the Poisson–Nernst–Planck (PNP) equations, we derive analytical impedance expressions for graphene-metal and metal-metal systems, elucidating the role of quantum capacitance across electrolyte concentrations. For finite-sized graphene disk electrodes, we incorporate graphene’s intrinsic conductivity to obtain explicit quantum impedance expressions. The results reveal a transition from Warburg-type behavior at high frequencies to RC-circuit behavior at low frequencies, governed by quantum capacitance and conductivity.
The analysis extends to charged graphene electrodes under DC bias, using matched asymptotic expansions in the thin double-layer limit. We derive an analytical impedance expression that highlights the dependence of frequency response on ion concentration and bias voltage. Finally, we explore concentrated electrolytes with RTILs, introducing a new length scale to capture electrostatic correlations and characterizing their impact on low-frequency impedance
Towards satellite-assisted quantum communication with reconfigurable networks and frequency-bin qubits
No full textThe development of quantum networks will enable advanced applications in quantum cryptography, quantum-enhanced metrology, and quantum computing. Central to this vision is the ability to reliably distribute quantum entanglement between distant nodes. To date, quantum networks have been largely confined to metropolitan scales and a small number of nodes with predominantly static implementations, constraining their scalability and adaptability for broader deployment.
Satellites provide a promising platform for distributing entanglement over global distances. However, the integration of satellites into terrestrial quantum networks poses significant challenges, including intermittent connectivity and high link attenuation during orbital passes.
In this thesis, we first propose a reconfigurable quantum network architecture that dynamically adapts its topology according to satellite availability. This reconfiguration ensures efficient and continuous operation while accommodating satellite links within a metropolitan quantum network. We demonstrate this network architecture, using a correlated photon source specifically designed for Canada’s Quantum Encryption and Science Satellite (QEYSSat) mission. Using both frequency and time multiplexing, we demonstrate a linear performance improvement with minimal resource overhead. Moreover, we propose an integrated source design to facilitate deployment in realistic scenarios.
Second, we investigate frequency-encoded quantum communication over free-space channels. We propose a novel approach that leverages linear interferometry and time-resolved detectors to decode frequency-bins without any adaptive optics or modal filtering. Furthermore, we investigate the phase stability requirements so that frequency-bin encoding could be feasible for satellite to ground quantum links. A proof-of-concept experiment is conducted over a turbulent free-space channel.
Third, we distribute frequency-bin entangled photons over multi-mode channels and test their non-local correlations. We report, to the best of our knowledge, the first measurement of the joint temporal intensity between frequency-bin entangled photons revealing a rich temporal structure. By combining time-resolved detection with energy-correlation measurements, we perform full quantum state tomography and further certify our source's non-classicality via a violation of the time-energy entropic uncertainty relations. We extend this scheme to higher-dimensional frequency-bin states, opening new possibilities for high-capacity robust quantum communication and quantum information processing.
The concepts, protocols, and experimental demonstrations presented in this thesis establish new approaches to utilize frequency-bin entanglement and contribute to the development of satellite-assisted quantum networks thus paving the way towards the realization of a global quantum network
Electroretinograms following short-term chromatic light adaptation in high myopes and non-myopes
No full textPurpose: Worldwide, myopia cases are on the rise and the need for finding a definitive mechanism by which myopia develops has become more imperative than ever before. Axial elongation in myopic eyes is linked to short wavelength (λ) light-dependent increases in retinal dopamine (DA). DA is associated with enhanced electroretinogram (ERG) amplitudes. This thesis seeks to examine short-term adaptation to short and long λ light at moderate and strong levels. I hypothesize that short λ light exposure differentially affects full-field ERGs (ffERGs) and central ERGs in myopic eyes compared to controls.
Methods: In a multi-visit cross-over design, we compared ERGs before and after 20 minutes of full-field adaptation to long (red LED peak λ (λ 627 nm) or short (blue LED peak λ 448 nm) λ light (Espion ColorDome™, Diagnosys LLC). Human participants (ages 18-30) were healthy high myopes (≤-5 diopters (D)) and emmetropes (+1 D to - 0.25 D). Monocular, light-adapted (LA) ffERGs were recorded using skin electrodes and a handheld system that adjusted for pupil diameter in real time (RETeval™). ERG stimuli were LA standard white flash and flicker (85 Td.s), presented at 2 and 28.3 Hz, respectively. The a-wave, b-wave and flicker ERG amplitudes, and implicit times were compared as a function of pre/post adaption (time), adapting stimulus (λ), and refractive error (RE) group using a mixed model ANOVA. Monocular multi-focal ERGs (mfERGs; 61 hexagons) and pattern ERGs (PERGs; 15° field, 15’ reversing checks) were recorded with natural pupils in keeping with ISCEV standards using DTL electrodes. The Espion™ console and amplifier (Diagnosys LLC) was used for both ERG tests. The primary outcome measures analyzed were the amplitudes and implicit times of the mfERG central wavelet, mfERG
average of the surrounding rings and the PERG P50 peak and were compared as a function of pre/post adaption (time), adapting stimulus (λ), and RE group using a mixed model ANOVA.
Results: There were no significant differences between controls and myopes prior to adaptation (all p≥0.05) for ffERG tests. The luminance of the light had an effect such that changes in b-wave implicit time decreased with adaptation to 300 cd/m² light in controls (blue 30: -0.451 ± 1.28%; blue 300: -2.28 ± 2.56%; p≤0.001, η2G=0.14) but less so in the high myope group (blue 30: +0.05 ± 1.3%; blue 300: -1.26 ± 2.43%; p=0.01, η2G=0.05). Changes in flicker implicit time decreased with the stronger luminance level but were not different between refractive error groups (i.e. controls: blue 30: 0.15 ± 1.0%; blue 300: -1.05 ± 1.61%, myopes: blue 30: +0.44 ± 2.74%; blue 300: -1.1 ± 1.50%, p≤0.001, η2G=0.13). Stronger light conditions caused b-wave amplitudes to become less negative (smaller) (i.e. controls: blue 30: –5.38 ± 15.6%; blue 300: –3.50 ± 17.7%, myopes: blue 30: –10.9 ± 15.4%; blue 300: +0.85 ± 10.9%, p=0.02, η2G=0.05). Similarly, with v flicker amplitudes, they became less negative (smaller) with stronger luminance adaptation (i.e. controls: blue 30: –4.1 ± 13.4%; blue 300: –0.62 ± 17.1%, myopes: blue 30: –9.9 ± 18.3%; blue 300: +4.7 ± 12.0%, p=0.00, η2G=0.07). When comparing between controls and high myopes or between short and long wavelengths, changes in b-wave amplitudes did not differ. There was a significant difference between RE groups for the PERG P50 amplitude prior to adaptation (p=0.03; Cohen’s d=1.06) such that they were significantly smaller in myopes (2.63 ± 1.0 μV) than controls (3.95 ± 1.5 μV). The N95 amplitudes were also smaller (i.e. less negative) in myopes (-4.39 ± 1.82 μV) than controls (-6.58 ± 1.55 μV; p= 0.01; Cohen’s d=-1.3). The mfERGs showed no significant pre-adaptation RE differences (all p≥ 0.05; all η2G≤ 0.07). For the PERG, in both controls and myopes, the change in N95 amplitudes decreased more with long λ light (controls: -16.6 ± 23.4%; myopes: -24.7 ± 43.4%) than short λ light (controls: +10.4 ± 23.5%; myopes: +1.88 ± 26.7%; p≤ 0.001; η2G=0.541). There were no effects of adaptation on the P50 amplitude or implicit time. For mfERG, the change in N1P1 magnitude was not different between controls and myopes and was not different when comparing λs for any of the five rings (p≥ 0.05).
Conclusions: There is no evidence that chromatic adaptation has a differential effect on post-adaptation ffERGs in high myopes. Long λ adaptation, especially with stronger luminance, prolongs ERG implicit times, probably reflecting relatively reduced input from the faster long cone system. There is evidence to suggest smaller central retinal responses with the PERG P50 but not with the central wavelet of the mfERG, indicating altered retinal ganglion cell function but not altered central inner retinal function in young adults with myopia. Further studies should focus on confirming whether altered central retinal function persists in adult myopes and whether longer adaptation times would yield a greater RE difference