Draft — CIFAR Azrieli Global Scholars Program 2026

MOTIVATION LETTER

Addiction is a dynamical systems problem. The brain does not store reward memories as static engrams; it consolidates them through a coupled process of dopamine-mediated prediction errors, NMDA-receptor-dependent long-term potentiation, and affective contrast. My Conjunctive Consolidation Threshold (CCT) model formalizes this tripartite interaction as a system of ordinary differential equations, calibrated with Bayesian MCMC against 1,847 records from the literature. All five pre-registered hypotheses were confirmed, with posterior super-additivity of 13 to 22 percentage points across model versions. This work is currently under review at three peer-reviewed journals as sole-authored preprints.

I apply to the CIFAR Azrieli Global Scholars Program because CIFAR explicitly funds researchers who cross disciplinary boundaries. My work sits at the intersection of pharmacology, computational neuroscience, and dynamical systems theory. I hold a B.Pharm from the University of Ibadan, am enrolled in the M.Sc. Digital Health at the Hasso Plattner Institute in Potsdam, and have built an independent research track record without a PhD. My collaborators include Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. These relationships demonstrate that my research direction is recognized by leaders in the field.

The CCT model has direct global health relevance. Nigeria, where I am from, has one of the highest untreated addiction rates in West Africa, yet almost no computational neuroscience research infrastructure exists there. My work is designed to be deployable: the model runs on a laptop, uses only publicly available data for calibration, and produces testable predictions about which pharmacological interventions could prevent memory reconsolidation in addiction. CIFAR's global network would allow me to connect this model with experimental collaborators who can test its predictions in animal models and eventually in human trials.

I am an independent computational researcher who has already built, tested, and submitted for publication a complete theoretical framework. The CIFAR program's emphasis on early-career independence and interdisciplinary collaboration matches my trajectory exactly. I seek the 100,000 CAD to support two years of full-time research: extending the CCT model to incorporate individual-level variability, building a public simulation platform, and traveling to collaborate with CIFAR network members.

RESEARCH STATEMENT

My research program addresses a single question: can we predict and prevent pathological reward-memory consolidation using computational models that integrate pharmacology, circuit dynamics, and machine learning?

The CCT model is the core of this program. It couples three axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. The model is implemented as a system of ODEs solved with RK45 and calibrated with Bayesian MCMC using PyMC's DEMetropolisZ sampler, with 14 free parameters and literature-elicited priors. All five pre-registered hypotheses were confirmed. The model predicts that conjunctive activation of all three axes above a threshold is necessary for memory consolidation, and that disrupting any single axis is insufficient for prevention. This has direct implications for pharmacotherapy: combination treatments targeting dopamine, NMDA, and affective processing simultaneously may be required.

Beyond the CCT model, I have developed neurocascade, a receptor-to-behavior brain-circuit simulation engine that couples pharmacokinetics to receptor binding to Wilson-Cowan circuit dynamics to behavioral readout. Three literature-calibrated receptor systems (mu-opioid, D2 dopamine, GABA-A) are implemented, with 62 of 62 tests passing. This engine allows me to test CCT predictions at the circuit level.

My computational methods work includes topological data analysis applied to protein-ligand interfaces. A pre-registered replication study on hERG cardiotoxicity showed that bipartite persistent homology does not beat a plain descriptor baseline (AUROC 0.8426 vs 0.8782), settling a question the literature had never properly tested. A subsequent study on drug-resistance prediction found that interface topology carries almost no signal (AUROC 0.425 and 0.485), leading me to develop TOPOLOGIX, an ESM-2 protein-language-model approach that achieves AUROC 0.804 on the Platinum benchmark while covering 100% of mutations versus approximately 18% for structure-limited tools.

Within CIFAR's Brain and Mind program, I propose to extend the CCT model in three directions. First, incorporate individual-level variability using Bayesian hierarchical modeling, allowing the model to predict which patients are most vulnerable to addiction. Second, connect the CCT model to neurocascade to simulate circuit-level effects of proposed combination therapies. Third, build a public web-based simulation platform that allows other researchers to run CCT simulations and test their own hypotheses.

LEADERSHIP AND COLLABORATION STATEMENT

I have built my research program entirely outside a traditional PhD structure. This required developing independent project management, peer review navigation, and collaboration-building skills from scratch. I have sole-authored three preprints currently under review, each at a different journal, each with pre-registered hypotheses and complete code and data repositories. I have co-authored a paper currently under review at Alcohol (Elsevier). I maintain four independent DuckDB-based ingest-to-analyze pipelines across life sciences, technology, and social science domains, and I self-host my own LLM serving infrastructure for research use.

My collaboration strategy is deliberate and documented. I reached out to Kent Berridge at Michigan because his work on incentive salience is foundational to the affective contrast axis of the CCT model. I sought endorsement from Samuel Gershman at Harvard for arXiv submission, which he provided. I have corresponded with Nathaniel Daw at Princeton about reinforcement learning models of addiction and with Marcelo Mattar at NYU about memory consolidation dynamics. These are not casual contacts; each interaction involved sharing specific model equations, simulation results, or preprint drafts and receiving substantive feedback.

Within CIFAR, I would bring a perspective that is currently underrepresented: computational pharmacology from an LMIC context. Nigerian researchers working on addiction neuroscience are almost nonexistent, and Nigerian pharmacists working on computational models are rarer still. I can help CIFAR build connections to West African research institutions and public health agencies, and I can mentor the next generation of African computational neuroscientists through the network.

I also bring technical infrastructure skills that are directly useful to collaborative projects. I can deploy and maintain shared computing resources, build reproducible analysis pipelines, and manage version-controlled code repositories. These skills make interdisciplinary collaboration work.

GLOBAL PERSPECTIVE STATEMENT

Nigeria has approximately 14 million people living with substance use disorders, according to the United Nations Office on Drugs and Crime. The country has fewer than 50 psychiatrists and essentially no computational neuroscience researchers. The addiction treatment infrastructure relies almost entirely on abstinence-based models with no pharmacological support, because the research base for combination pharmacotherapy does not exist in the Nigerian context.

My CCT model was designed with this reality in mind. It runs on consumer hardware. It uses only publicly available data. It produces testable predictions that can be evaluated with basic laboratory equipment. If the model is correct, it would suggest that Nigerian patients could benefit from combination treatments that are already approved for other indications, rather than requiring new drug development.

CIFAR's global network is essential for this work because the experimental validation cannot happen in Nigeria alone. I need collaborators with access to rodent models, human neuroimaging, and clinical trial infrastructure. CIFAR connects me to those collaborators while keeping my research anchored in the Nigerian context. The program's emphasis on researchers from diverse geographic backgrounds is not a checkbox for me; it is the structural condition that makes my research program viable.

I am enrolled in the M.Sc. Digital Health at Hasso Plattner Institute in Germany, which gives me access to European computational infrastructure and collaborators. But my research questions come from Nigeria, and my long-term plan is to build a computational neuroscience research group at a Nigerian university. CIFAR's support at this early stage would accelerate that timeline by years.

CHECKLIST

- [ ] Complete online application form at CIFAR Azrieli Global Scholars Program portal
- [ ] Upload motivation letter (this document)
- [ ] Upload research statement (this document)
- [ ] Upload leadership and collaboration statement (this document)
- [ ] Upload global perspective statement (this document)
- [ ] Provide CV with ORCID 0009-0001-9272-6735 and GitHub github.com/AmunRaPtah
- [ ] Provide list of publications: three sole-authored preprints under review, one co-authored paper under review at Alcohol
- [ ] Provide names and contact information for three references: Kent Berridge, Samuel Gershman, Nathaniel Daw
- [ ] Confirm eligibility: within 5 years of starting independent research position (pre-PhD independent researcher status may require explanation)
- [ ] Verify deadline: 2026-05-01

EDITOR NOTES

- Eligibility risk: The program states "within 5 years of starting an independent research position (PhD or equivalent)." Eniola does not have a PhD. His independent research began in 2024. This may be interpreted as meeting the criterion, but the application should include a brief explanation of what constitutes "equivalent" in his case. Consider adding a sentence to the motivation letter explicitly addressing this.

- Reference confirmation: Samuel Gershman provided an arXiv endorsement, but it is unclear whether he has agreed to serve as a formal reference. Confirm with Gershman before listing him. Same for Daw and Mattar.

- Publication status: The three sole-authored preprints are "in review" but not yet accepted. The application should note that they are under review and provide preprint DOIs from OSF or Zenodo. If any are accepted before the deadline, update the application.

- M.Sc. enrollment: Eniola is enrolled starting Winter Semester 2026/27, which begins in October 2026. The CIFAR deadline is May 2026. He will not yet have started the program. This should be stated clearly, and the application should frame the M.Sc. as complementary to, not a substitute for, his independent research.

- Amount: 100,000 CAD is approximately 74,000 USD or 68,000 EUR. Eniola should verify that this amount is sufficient for his proposed two-year research plan, including travel to CIFAR network meetings and potential experimental collaborations.

MOTIVATION LETTER

Addiction is a dynamical systems problem. The brain does not store reward memories as static engrams; it consolidates them through a coupled process of dopamine-mediated prediction errors, NMDA-receptor-dependent long-term potentiation, and affective contrast. My Conjunctive Consolidation Threshold (CCT) model formalizes this tripartite interaction as a system of ordinary differential equations, calibrated with Bayesian MCMC against 1,847 records from the literature. All five pre-registered hypotheses were confirmed, with posterior super-additivity of 13 to 22 percentage points across model versions. This work is currently under review at three peer-reviewed journals as sole-authored preprints.

I apply to the CIFAR Azrieli Global Scholars Program because CIFAR explicitly funds researchers who cross disciplinary boundaries. My work sits at the intersection of pharmacology, computational neuroscience, and dynamical systems theory. I hold a B.Pharm from the University of Ibadan, am enrolled in the M.Sc. Digital Health at the Hasso Plattner Institute in Potsdam, and have built an independent research track record without a PhD. My collaborators include Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. These relationships demonstrate that my research direction is recognized by leaders in the field.

The CCT model has direct global health relevance. Nigeria, where I am from, has one of the highest untreated addiction rates in West Africa, yet almost no computational neuroscience research infrastructure exists there. My work is designed to be deployable: the model runs on a laptop, uses only publicly available data for calibration, and produces testable predictions about which pharmacological interventions could prevent memory reconsolidation in addiction. CIFAR's global network would allow me to connect this model with experimental collaborators who can test its predictions in animal models and eventually in human trials.

I am an independent computational researcher who has already built, tested, and submitted for publication a complete theoretical framework. The CIFAR program's emphasis on early-career independence and interdisciplinary collaboration matches my trajectory exactly. I seek the 100,000 CAD to support two years of full-time research: extending the CCT model to incorporate individual-level variability, building a public simulation platform, and traveling to collaborate with CIFAR network members.

RESEARCH STATEMENT

My research program addresses a single question: can we predict and prevent pathological reward-memory consolidation using computational models that integrate pharmacology, circuit dynamics, and machine learning?

The CCT model is the core of this program. It couples three axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. The model is implemented as a system of ODEs solved with RK45 and calibrated with Bayesian MCMC using PyMC's DEMetropolisZ sampler, with 14 free parameters and literature-elicited priors. All five pre-registered hypotheses were confirmed. The model predicts that conjunctive activation of all three axes above a threshold is necessary for memory consolidation, and that disrupting any single axis is insufficient for prevention. This has direct implications for pharmacotherapy: combination treatments targeting dopamine, NMDA, and affective processing simultaneously may be required.

Beyond the CCT model, I have developed neurocascade, a receptor-to-behavior brain-circuit simulation engine that couples pharmacokinetics to receptor binding to Wilson-Cowan circuit dynamics to behavioral readout. Three literature-calibrated receptor systems (mu-opioid, D2 dopamine, GABA-A) are implemented, with 62 of 62 tests passing. This engine allows me to test CCT predictions at the circuit level.

My computational methods work includes topological data analysis applied to protein-ligand interfaces. A pre-registered replication study on hERG cardiotoxicity showed that bipartite persistent homology does not beat a plain descriptor baseline (AUROC 0.8426 vs 0.8782), settling a question the literature had never properly tested. A subsequent study on drug-resistance prediction found that interface topology carries almost no signal (AUROC 0.425 and 0.485), leading me to develop TOPOLOGIX, an ESM-2 protein-language-model approach that achieves AUROC 0.804 on the Platinum benchmark while covering 100% of mutations versus approximately 18% for structure-limited tools.

Within CIFAR's Brain and Mind program, I propose to extend the CCT model in three directions. First, incorporate individual-level variability using Bayesian hierarchical modeling, allowing the model to predict which patients are most vulnerable to addiction. Second, connect the CCT model to neurocascade to simulate circuit-level effects of proposed combination therapies. Third, build a public web-based simulation platform that allows other researchers to run CCT simulations and test their own hypotheses.

LEADERSHIP AND COLLABORATION STATEMENT

I have built my research program entirely outside a traditional PhD structure. This required developing independent project management, peer review navigation, and collaboration-building skills from scratch. I have sole-authored three preprints currently under review, each at a different journal, each with pre-registered hypotheses and complete code and data repositories. I have co-authored a paper currently under review at Alcohol (Elsevier). I maintain four independent DuckDB-based ingest-to-analyze pipelines across life sciences, technology, and social science domains, and I self-host my own LLM serving infrastructure for research use.

My collaboration strategy is deliberate and documented. I reached out to Kent Berridge at Michigan because his work on incentive salience is foundational to the affective contrast axis of the CCT model. I sought endorsement from Samuel Gershman at Harvard for arXiv submission, which he provided. I have corresponded with Nathaniel Daw at Princeton about reinforcement learning models of addiction and with Marcelo Mattar at NYU about memory consolidation dynamics. These are not casual contacts; each interaction involved sharing specific model equations, simulation results, or preprint drafts and receiving substantive feedback.

Within CIFAR, I would bring a perspective that is currently underrepresented: computational pharmacology from an LMIC context. Nigerian researchers working on addiction neuroscience are almost nonexistent, and Nigerian pharmacists working on computational models are rarer still. I can help CIFAR build connections to West African research institutions and public health agencies, and I can mentor the next generation of African computational neuroscientists through the network.

I also bring technical infrastructure skills that are directly useful to collaborative projects. I can deploy and maintain shared computing resources, build reproducible analysis pipelines, and manage version-controlled code repositories. These skills make interdisciplinary collaboration work.

GLOBAL PERSPECTIVE STATEMENT

Nigeria has approximately 14 million people living with substance use disorders, according to the United Nations Office on Drugs and Crime. The country has fewer than 50 psychiatrists and essentially no computational neuroscience researchers. The addiction treatment infrastructure relies almost entirely on abstinence-based models with no pharmacological support, because the research base for combination pharmacotherapy does not exist in the Nigerian context.

My CCT model was designed with this reality in mind. It runs on consumer hardware. It uses only publicly available data. It produces testable predictions that can be evaluated with basic laboratory equipment. If the model is correct, it would suggest that Nigerian patients could benefit from combination treatments that are already approved for other indications, rather than requiring new drug development.

CIFAR's global network is essential for this work because the experimental validation cannot happen in Nigeria alone. I need collaborators with access to rodent models, human neuroimaging, and clinical trial infrastructure. CIFAR connects me to those collaborators while keeping my research anchored in the Nigerian context. The program's emphasis on researchers from diverse geographic backgrounds is not a checkbox for me; it is the structural condition that makes my research program viable.

I am enrolled in the M.Sc. Digital Health at Hasso Plattner Institute in Germany, which gives me access to European computational infrastructure and collaborators. But my research questions come from Nigeria, and my long-term plan is to build a computational neuroscience research group at a Nigerian university. CIFAR's support at this early stage would accelerate that timeline by years.

CHECKLIST

- [ ] Complete online application form at CIFAR Azrieli Global Scholars Program portal
- [ ] Upload motivation letter (this document)
- [ ] Upload research statement (this document)
- [ ] Upload leadership and collaboration statement (this document)
- [ ] Upload global perspective statement (this document)
- [ ] Provide CV with ORCID 0009-0001-9272-6735 and GitHub github.com/AmunRaPtah
- [ ] Provide list of publications: three sole-authored preprints under review, one co-authored paper under review at Alcohol
- [ ] Provide names and contact information for three references: Kent Berridge, Samuel Gershman, Nathaniel Daw
- [ ] Confirm eligibility: within 5 years of starting independent research position (pre-PhD independent researcher status may require explanation)
- [ ] Verify deadline: 2026-05-01

EDITOR NOTES

- Eligibility risk: The program states "within 5 years of starting an independent research position (PhD or equivalent)." Eniola does not have a PhD. His independent research began in 2024. This may be interpreted as meeting the criterion, but the application should include a brief explanation of what constitutes "equivalent" in his case. Consider adding a sentence to the motivation letter explicitly addressing this.

- Reference confirmation: Samuel Gershman provided an arXiv endorsement, but it is unclear whether he has agreed to serve as a formal reference. Confirm with Gershman before listing him. Same for Daw and Mattar.

- Publication status: The three sole-authored preprints are "in review" but not yet accepted. The application should note that they are under review and provide preprint DOIs from OSF or Zenodo. If any are accepted before the deadline, update the application.

- M.Sc. enrollment: Eniola is enrolled starting Winter Semester 2026/27, which begins in October 2026. The CIFAR deadline is May 2026. He will not yet have started the program. This should be stated clearly, and the application should frame the M.Sc. as complementary to, not a substitute for, his independent research.

- Amount: 100,000 CAD is approximately 74,000 USD or 68,000 EUR. Eniola should verify that this amount is sufficient for his proposed two-year research plan, including travel to CIFAR network meetings and potential experimental collaborations.

AI Draft — CIFAR Azrieli Global Scholars Program 2026