MOTIVATION LETTER The Electronic, Photonic, Magnetic, and Quantum Devices programme at the National Science Foundation funds foundational research into novel device architectures. My work on the Conjunctive Consolidation Threshold model proposes a computational device for predicting and preventing addiction liability, built on Bayesian population dynamics and validated against preclinical data. The CCT model achieved an 85.8 percent reduction in encoding probability, from 0.855 to 0.122, with super-additivity of 12.8 percentage points across five pre-registered hypotheses. This is a device concept in the pharmacological sense: a predictive algorithm that functions as a screening instrument for reward-memory consolidation risk. I hold a B.Pharm from the University of Ibadan with a CGPA of 5.1 out of 7.0, German equivalent 1.9, and am a PCN-licensed pharmacist. I have built three operational platforms: IMPRINT for addiction-liability screening, TOPOLOGIX for topological data analysis of drug-protein interactions using persistent homology and bipartite simplicial complexes, and GATE for BCI neural-stimulation safety evaluation, released under Apache 2.0. My computational toolkit includes Python with scipy, numpy, ODE/RK45, PyMC for MCMC, R, Ripser and Gudhi for TDA, NEURON and Brian2 for neural simulation, AlphaFold, RDKit, ADMET/QSAR, GROMACS, AutoDock, and Nextflow for HPC pipelines on SLURM. I have three sole-authored preprints on OSF and Zenodo, a review article under review at Neuroscience and Biobehavioral Reviews, and a co-authored paper in Alcohol under review at Elsevier. I hold endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. A provisional patent on the CCT core architecture is filed for Q3 2026. I am an independent researcher based in Lagos, Nigeria. I am not yet enrolled in an MSc programme; I am applying for October 2026 start at the Medical University of Graz in Austria. I am eligible for early-career, pre-PhD, LMIC-track, and independent researcher programmes. My research is conducted without institutional overhead, using personal HPC resources and open-source infrastructure. The EPMQD programme requires a U.S. institutional affiliation. I do not hold one. I am writing to inquire whether a collaboration with a U.S.-based principal investigator would satisfy the eligibility requirement, or whether an exception exists for independent researchers from low- and middle-income countries who can demonstrate equivalent institutional capacity through published preprints, provisional patents, and endorsed research outputs. If a U.S. PI partnership is required, I am prepared to identify a collaborator within the programme's scope and submit a joint proposal. My Bayesian MCMC validation pipeline, the IMPRINT screening platform, and the provisional patent on CCT architecture are concrete assets for any joint submission. RESEARCH STATEMENT The Conjunctive Consolidation Threshold model addresses a specific gap in addiction neuroscience: the absence of a formal, mathematically specified framework for predicting when reward-memory encoding transitions from reversible to consolidated. Current pharmacological interventions target dopamine receptors or NMDA channels without a quantitative threshold for intervention timing. The CCT model defines this threshold as a tripartite function of dopamine burst magnitude, glutamate co-activation duration, and calcium-dependent plasticity kinetics. The model is specified in three sole-authored preprints. The foundational paper, available at OSF 10.17605/OSF.IO/KG7B5, defines the tripartite framework. The formal mathematical specification, at OSF 10.17605/OSF.IO/EMY4U, provides the differential equations and boundary conditions. The Bayesian population dynamics and clinical trial architecture paper, at Zenodo 10.5281/zenodo.20492472, validates the model using ODE/RK45 integration and Bayesian MCMC against simulated population data. All five pre-registered hypotheses H1 through H5 were confirmed. Encoding probability dropped from 0.855 to 0.122, an 85.8 percent reduction. Super-additivity of 12.8 percentage points indicates that the three components interact nonlinearly, producing greater suppression than any pair alone. The model has direct device implications. IMPRINT, the addiction-liability screening platform I built, implements the CCT threshold as a scoring algorithm for candidate compounds. TOPOLOGIX uses persistent homology on bipartite simplicial complexes to map drug-protein interaction topologies, with a validated MVP for hERG cardiotoxicity screening. GATE evaluates neural-stimulation safety for BCI applications. These platforms are not theoretical; they run on my HPC pipeline using Nextflow and SLURM, with data stored in Supabase and Postgres. The provisional patent filed for Q3 2026 covers the core CCT architecture as a computational method for predicting consolidation risk. The patent is filed in Nigeria, with PCT extension planned. This positions the CCT model as a device in the regulatory sense: a predictive algorithm that could be deployed as a clinical decision support tool in addiction medicine. For the EPMQD programme, the relevant device concept is the CCT algorithm itself, implemented as a software-defined pharmacological device. The Bayesian MCMC validation pipeline provides the statistical rigor required for device performance characterization. The super-additivity result demonstrates that the device achieves more than the sum of its parts, a property directly relevant to device design optimization. I seek funding to extend the CCT model from simulated populations to real-world clinical data. The next phase requires access to human reward-learning datasets, fMRI-based dopamine burst estimates, and longitudinal addiction outcome data. A partnership with a U.S. PI at a institution with access to human subjects data would enable this extension. I bring the mathematical framework, the validated codebase, the provisional patent, and the computational infrastructure. The U.S. partner would provide the clinical data, institutional review board approval, and patient recruitment infrastructure. PROJECT NARRATIVE Objective: Validate the Conjunctive Consolidation Threshold model against human reward-learning data and develop a clinical decision support device for addiction liability screening. Background: Addiction is characterized by the consolidation of reward-memory associations that drive compulsive drug-seeking behavior. Current screening tools rely on self-report questionnaires with limited predictive validity. The CCT model provides a mathematically specified threshold for when a reward-memory encoding event becomes consolidated, enabling quantitative risk stratification. Methods: The CCT model is implemented as a system of ordinary differential equations solved via RK45 integration. Bayesian MCMC using PyMC estimates population-level parameters from simulated data. The model will be calibrated against human reward-learning datasets from published studies, including probabilistic reward tasks and Pavlovian conditioning paradigms. The IMPRINT platform will be extended to accept fMRI-derived dopamine burst estimates and behavioral measures as inputs, outputting a consolidation risk score. Timeline: Months 1-3: literature review and dataset identification. Months 4-6: model calibration against three independent human datasets. Months 7-9: sensitivity analysis and parameter optimization. Months 10-12: validation against held-out data and preparation of clinical trial protocol. Deliverables: A validated CCT-based risk score with known sensitivity and specificity. A software implementation of the IMPRINT platform suitable for deployment in clinical settings. A provisional clinical trial protocol for a prospective validation study. Three manuscripts for submission to peer-reviewed journals. Budget: Personnel support for the PI at 50 percent effort for 12 months: USD 30,000. HPC computing credits: USD 5,000. Open-access publication fees: USD 3,000. Travel for collaboration meetings: USD 2,000. Total: USD 40,000. CHECKLIST - [ ] Confirm U.S. institutional affiliation or identify U.S. PI collaborator before submission - [ ] Verify whether NSF allows international co-PIs or subcontracts to independent researchers - [ ] Obtain letter of endorsement from Kent Berridge, Samuel Gershman, Nathaniel Daw, or Marcelo Mattar - [ ] Prepare CV with ORCID 0009-0001-9272-6735 and GitHub github.com/AmunRaPtah - [ ] Include links to three preprints: OSF 10.17605/OSF.IO/KG7B5, OSF 10.17605/OSF.IO/EMY4U, Zenodo 10.5281/zenodo.20492472 - [ ] Include provisional patent filing number and filing date - [ ] Prepare budget justification for USD 40,000 total - [ ] Submit through NSF FastLane or Research.gov with U.S. PI as submitting organization - [ ] Confirm deadline on programme website: https://www.nsf.gov/funding/opportunities/epmqd-electronic-photonic-magnetic-quantum-devices/pd26-1517 EDITOR NOTES - Eligibility is the primary risk. The EPMQD programme explicitly requires U.S. institutional affiliation. Eniola must either find a U.S. PI collaborator or confirm that NSF allows independent researchers from LMICs to apply directly. The letter should be framed as an inquiry, not a full application, unless a collaborator is confirmed. - The programme focuses on electronic, photonic, magnetic, and quantum devices. The CCT model as a software-defined pharmacological device is a stretch. Eniola should verify whether the programme accepts computational models as devices or whether it requires physical hardware. If not, the application may be rejected on scope grounds. - The budget of USD 40,000 is small for NSF. Most NSF grants start at USD 100,000. Eniola should check whether the programme has a minimum budget threshold and whether a smaller budget is acceptable for early-career or international applicants. - The provisional patent is filed in Nigeria. NSF may require U.S. patent filing or PCT application. Eniola should confirm patent eligibility requirements for NSF-funded projects. - Eniola should verify his eligibility for NSF funding as a non-U.S. citizen. Some NSF programmes restrict funding to U.S. citizens or permanent residents. The ISE or Global Centers programmes may be more appropriate alternatives.