MOTIVATION LETTER The intersection of addiction neurobiology and dementia risk in African populations remains almost entirely unmapped, yet it is where I have spent the last two years building quantitative tools. My Conjunctive Consolidation Threshold (CCT) model, a tripartite pharmacological framework for reward-memory encoding prevention, has confirmed all five pre-registered hypotheses (H1-H5) through Bayesian MCMC calibration (PyMC DEMetropolisZ, 14 free parameters, literature-elicited priors from a 1,847-record screen). The posterior super-additivity of 13-22 percentage points across model versions demonstrates that dopaminergic RPE signaling, NMDAR-dependent LTP, and affective contrast interact non-additively in reward memory formation. This matters for dementia research because the same glutamatergic and dopaminergic pathways implicated in addiction are central to neurodegeneration, and substance use is a modifiable risk factor that remains understudied in LMIC cohorts. My neurocascade engine extends this work from molecular pharmacology to circuit-level dynamics. It couples pharmacokinetics to receptor binding to Wilson-Cowan circuit dynamics to behavioral-readout ODE layers, with three literature-calibrated receptor/circuit systems (mu-opioid, D2 dopamine, GABA-A) and 62 of 62 tests passing. The Bayesian-calibrated parameters are explicitly labeled illustrative pending real behavioral-data fits, which is the honest status of the work. For a conference on brain ageing and dementia in LMICs, this engine offers a concrete method for simulating how chronic substance exposure alters the receptor-to-behavior pathways that overlap with dementia pathology. I am applying to the Brain Ageing and Dementia in LMICs 2026 conference because it explicitly prioritizes early-career researchers from LMICs and cross-cultural comparison. As a Nigerian pharmacist and independent computational researcher now enrolled in the M.Sc. Digital Health programme at Hasso Plattner Institute / University of Potsdam, I sit exactly at the intersection this conference targets. My prior work includes a pre-registered, powered replication on hERG cardiotoxicity topology that settled a comparison the literature had never actually run (topological features did not beat a plain descriptor baseline, AUROC 0.8426 vs 0.8782), and a drug-resistance study that ruled out interface geometry as the driver (AUROC 0.425 and 0.485 on the Platinum benchmark). I report negative results directly rather than reframing them, which I understand is the standard this conference expects. The proposed abstract, Computational modeling of addiction-related neurodegeneration risk in LMICs, directly addresses the conference theme of modifiable risk factors. It will present the CCT model and neurocascade engine as tools for simulating how substance use alters the dopaminergic and glutamatergic circuits that are also implicated in dementia, with specific attention to African populations where data are scarce and computational approaches can generate testable hypotheses before expensive cohort studies begin. I request consideration for an early-career researcher slot, and I am prepared to present either as a poster or in an oral session. RESEARCH STATEMENT The burden of dementia in LMICs is projected to rise faster than anywhere else in the world, yet the mechanistic models used to understand modifiable risk factors are built almost entirely on HIC data and HIC assumptions. My research program addresses this gap by building computational models of the receptor-to-behavior pathways that link substance use to neurodegeneration, calibrated to literature from both HIC and LMIC contexts, and designed to generate hypotheses that can be tested in African populations. The CCT model is the foundation. It is a tripartite pharmacological framework for reward-memory encoding prevention in addiction, implemented as a coupled three-axis ODE model (dopaminergic RPE, NMDAR-dependent LTP, affective contrast) solved with RK45. I calibrated it using Bayesian MCMC with PyMC DEMetropolisZ, 14 free parameters, and literature-elicited priors from a systematic screen of 1,847 records. All five pre-registered hypotheses were confirmed. The key finding is super-additivity: the three axes interact such that combined intervention produces 13-22 percentage points greater effect than any single axis alone, across model versions. This has direct relevance to dementia because the same NMDAR-dependent plasticity mechanisms are central to both addiction memory and neurodegenerative processes, and because chronic substance use is a known modifiable risk factor for cognitive decline. The neurocascade engine extends this to circuit and behavioral levels. It is a receptor-to-behavior brain-circuit simulation engine that couples four ODE layers: pharmacokinetics, receptor binding, Wilson-Cowan circuit dynamics, and behavioral readout. Three receptor/circuit systems are implemented and literature-calibrated: mu-opioid, D2 dopamine, and GABA-A. The full pipeline passes 62 of 62 tests. The circuit-layer parameters are explicitly labeled illustrative pending real behavioral-data fits, which I state plainly because the validation status is honest and incomplete. The value of neurocascade for this conference is that it provides a mechanistic bridge from molecular pharmacology to behavior, which is precisely the level of analysis needed to understand how substance use in midlife translates to dementia risk in later life. My methodological track record includes both positive and negative results, and I report both with equal rigor. The hERG cardiotoxicity topology study was a pre-registered, powered replication that tested whether bipartite persistent homology (opposition-distance metric, Ripser/GUDHI) predicts hERG cardiotoxicity from protein-ligand interface geometry. The result was negative: topological features did not beat a plain descriptor baseline (AUROC 0.8426 vs 0.8782). This settled a comparison the published literature had never actually run. The interface-topology-for-resistance study similarly ruled out interface geometry as the driver of drug resistance (AUROC 0.425 and 0.485 on the Platinum benchmark). My current TOPOLOGIX project, which uses ESM-2 protein-language-model delta-embeddings plus Morgan/ECFP fingerprints and a Random Forest classifier, achieves AUROC 0.804 +/- 0.025 on the Platinum benchmark (553 mutations) and 0.634 on SKEMPI 2.0, beating structure-based baselines (mCSM-lig ~0.70) while covering 100% of mutations versus ~18% for structure-limited tools. For this conference, I propose to present the CCT model and neurocascade engine as a unified computational framework for studying addiction-related neurodegeneration risk in LMICs. The abstract will describe how these models can simulate the effects of chronic substance exposure on the dopaminergic and glutamatergic circuits implicated in dementia, and how the resulting hypotheses can be tested in African populations where longitudinal cohort data are scarce. The work is at the hypothesis-generation stage, not the validation stage, and I will present it as such. ESSAY: RELEVANCE TO BRAIN AGEING AND DEMENTIA IN LMICS The conference's focus on modifiable risk factors and cross-cultural populations maps directly onto my CCT model and neurocascade engine. Substance use is a modifiable risk factor for dementia that is understudied in African populations, and my models provide a quantitative method for simulating how chronic exposure alters the specific receptor and circuit systems that overlap with dementia pathology. The dopaminergic RPE axis in the CCT model is the same mesolimbic pathway that degenerates in Parkinsonian syndromes and is implicated in vascular dementia. The NMDAR-dependent LTP axis is the same plasticity mechanism targeted by memantine in Alzheimer's disease. The affective contrast axis captures the emotional valence component that is increasingly recognized as relevant to both addiction and depression, which is itself a dementia risk factor. My geographic and professional position strengthens this fit. I am a Nigerian pharmacist with a B.Pharm from the University of Ibadan (CGPA 5.1/7.0, German equivalent 1.9) and a PCN license. I have worked as a clinical pharmacist at Ramset Pharmacy and as National Product Manager at Synthcare, giving me direct exposure to the clinical realities of substance use and cognitive decline in Nigerian populations. I am now enrolled in the M.Sc. Digital Health programme at Hasso Plattner Institute / University of Potsdam, which places me in a German academic context while maintaining my African research focus. This dual position allows me to serve as a bridge between HIC computational methods and LMIC research questions. The conference's explicit allocation of early-career researcher slots and prioritization of LMIC geographic representation aligns with my current career stage. I am 29, pre-PhD, and an independent researcher with a substantial publication record including three sole-authored preprints in review at peer-reviewed journals (IART, PNPBP, NBR) and a co-authored paper in Alcohol (Elsevier, under review). I have endorsements from Kent Berridge (Michigan), Samuel Gershman (Harvard), Nathaniel Daw (Princeton), and Marcelo Mattar (NYU). I am eligible for early-career, pre-PhD, LMIC-track, and independent researcher programmes, and I am seeking fellowships, research grants, training courses, and residencies that match this profile. CHECKLIST - [ ] Submit abstract (250-300 words) titled "Computational modeling of addiction-related neurodegeneration risk in LMICs" via the conference submission portal at https://conferences.ncl.ac.uk/advascular/ - [ ] Include title, author name (Eniola Ayodele Olutogun), affiliation (Independent researcher / Hasso Plattner Institute), and abstract text - [ ] Verify abstract deadline: 13 September 2026 (confirm on programme website) - [ ] Confirm whether ORCID (0009-0001-9272-6735) is required in the submission form - [ ] Prepare poster presentation materials in case abstract is accepted as poster rather than oral - [ ] Prepare 5-minute oral presentation summary in case selected for ECR oral slot - [ ] Confirm conference registration fee and whether LMIC/ECR fee waiver is available - [ ] Verify travel funding options for LMIC participants (conference website or organizer inquiry) - [ ] Confirm whether the conference requires a short bio or CV upload alongside the abstract - [ ] Check whether the conference offers mentorship or networking sessions for ECRs and register interest EDITOR NOTES - Eligibility risk: The conference is hosted by Newcastle University and the URL suggests a vascular dementia focus. The abstract framing emphasizes addiction-related neurodegeneration, which is adjacent but not identical to vascular dementia. The applicant should verify that the conference scope includes substance use as a risk factor, and if not, adjust the abstract to emphasize the vascular and cerebrovascular implications of chronic substance exposure (e.g., hypertension, stroke risk, endothelial dysfunction) which are direct vascular dementia risk factors. - Verification needed: The applicant's enrollment at HPI/Potsdam is listed as Winter Semester 2026/27, which may or may not have started by the conference date. The affiliation line on the abstract should reflect current status. If enrollment has not begun, use "Independent researcher" as the primary affiliation. - Gap to fill: The applicant profile does not specify any prior conference presentations or poster experience. If the applicant has presented at conferences before, that should be added to the bio or motivation letter. If not, the applicant should prepare a statement about being an independent researcher who has not yet had the opportunity to present at international conferences, which strengthens the case for ECR support. - Gap to fill: The applicant should insert a specific sentence about their connection to Newcastle or the UK if any exists (e.g., prior collaboration, UK-based co-authors, familiarity with UK research landscape). If none exists, the motivation letter should acknowledge this is a first UK conference application and express willingness to travel. - The abstract must be 250-300 words. The draft abstract in the research statement is currently longer than that and must be condensed. The applicant should draft the abstract separately and verify the word count before submission. - The conference selection criteria mention that most abstracts become posters. The applicant should prepare for a poster presentation as the most likely outcome and treat an oral slot as a bonus. The poster should include the CCT model diagram, the neurocascade architecture, and the key quantitative results (13-22pp super-additivity, 62/62 tests passing).