Draft — Emergent Ventures

MOTIVATION LETTER

The CCT model began with a contradiction. Nigeria has one of the highest rates of untreated substance use disorders in West Africa, yet the global research agenda treats addiction almost exclusively as a clinical problem requiring behavioral intervention or agonist maintenance. I spent five years as a pharmacist at the University of Ibadan and in clinical practice watching patients cycle through relapse because the underlying mechanism, the encoding of reward-memory associations during withdrawal, was never addressed pharmacologically. The mainstream approach manages craving after it appears. My work asks why the craving memory forms at all.

The Conjunctive Consolidation Threshold model is a tripartite pharmacological framework that treats reward-memory encoding as a coupled dynamical system across three axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. I built it as an independent researcher, outside any academic lab, using RK45 integration and Bayesian MCMC calibration with PyMC's DEMetropolisZ sampler. The model has 14 free parameters, all with priors elicited from a systematic screen of 1,847 published records. All five pre-registered hypotheses, H1 through H5, were confirmed. Posterior super-additivity across model versions ranges from 13 to 22 percentage points, meaning the three axes interact nonlinearly in ways a single-target approach cannot capture. Three sole-authored preprints are under review at peer-reviewed journals: International Addiction Review, Progress in Neuro-Psychopharmacology and Biological Psychiatry, and Neuroscience and Biobehavioral Reviews. A co-authored paper is under review at Alcohol, Elsevier.

Emergent Ventures funds unconventional projects with high potential impact. This qualifies on both counts. The CCT model challenges the assumption that addiction treatment must be clinic-based and behavior-first. It argues for a computational pharmacology-first approach: simulate the receptor and circuit dynamics, identify the consolidation threshold, and intervene before the memory encodes. No existing treatment targets this window. The model is technically grounded, fully specified, and has already passed its primary validation gate. What it lacks is funding to move from simulation to experimental collaboration, specifically to design the in vitro and rodent assays that would test the predicted super-additivity of combined D1 antagonism, NMDAR partial blockade, and affective contrast modulation.

I am a Nigerian pharmacist, 29 years old, enrolled in the M.Sc. Digital Health program at Hasso Plattner Institute and the University of Potsdam starting winter semester 2026/27. I have endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. The work is real, the results are pre-registered and reproducible, and the idea is genuinely overlooked by mainstream funding bodies that route addiction research through clinical trial infrastructure. Emergent Ventures is the right vehicle for a project at this stage: too early for NIH R01s, too unconventional for traditional addiction research councils, and exactly the kind of bold, underfunded idea that Tyler Cowen's rolling grant process exists to catch.

SHORT ESSAY: PERSONAL STORY

I grew up in Lagos and trained as a pharmacist at the University of Ibadan, graduating in 2021 with a CGPA of 5.1 out of 7.0, a German equivalent of 1.9. I am licensed by the Pharmacists Council of Nigeria. During clinical rotations and later work at Ramset Pharmacy, I saw the same pattern repeatedly: patients with opioid or alcohol use disorders would detoxify successfully, then relapse within months because the environmental cues that triggered their cravings had been consolidated into long-term memory during withdrawal. The pharmacology we had, methadone, buprenorphine, naltrexone, all acted on receptor systems after the fact. None of them touched the memory encoding process itself.

I could not find a research group in Nigeria working on this problem, so I built the tools myself. I taught myself computational modeling, Bayesian statistics, and protein machine learning from primary literature. I developed the CCT model as a sole investigator, calibrating it against published electrophysiology and microdialysis data. I also ran a pre-registered replication study on hERG cardiotoxicity prediction using persistent homology, which found that topological features do not beat a plain descriptor baseline, AUROC 0.8426 versus 0.8782, settling a comparison the literature had never actually run. I built TOPOLOGIX, an ESM-2 protein language model approach that predicts drug-resistance mutations from sequence alone with AUROC 0.804 on the Platinum benchmark, covering 100 percent of mutations versus roughly 18 percent for structure-limited tools. I did all of this without institutional support, without a lab, and without grant funding.

The thread connecting these projects is a refusal to accept the boundary between clinical pharmacy and computational research. Nigeria produces pharmacists, but not computational neuroscientists. I am trying to be both. The M.Sc. at Hasso Plattner Institute is the first formal training I have had in digital health, and it starts in late 2026. Until then, I continue as an independent researcher with a GitHub, an ORCID, and a stack of preprints under review.

SHORT ESSAY: MAINSTREAM VIEW YOU AGREE WITH

The mainstream view I agree with is that addiction is a brain disease, not a moral failing. This position, articulated most forcefully by Nora Volkow and the National Institute on Drug Abuse, has been crucial for destigmatizing substance use disorders and redirecting research toward neurobiological mechanisms. I agree with it completely. The evidence is overwhelming: chronic drug exposure produces measurable changes in dopaminergic signaling, glutamatergic plasticity, and cortical-striatal circuit function. Patients are not choosing to be addicted. Their brains have been altered by the drug.

Where I diverge from the mainstream is in the therapeutic consequence drawn from this view. If addiction is a brain disease, the field has largely concluded that treatment must be chronic, maintenance-based, and behaviorally supported. The dominant protocols, agonist maintenance, cognitive behavioral therapy, contingency management, all assume the disease is a persistent state that must be managed indefinitely. My work suggests a different implication. If addiction is a brain disease of memory encoding, specifically the aberrant consolidation of drug-reward associations during withdrawal, then the disease may be interruptible at the moment of encoding. The CCT model predicts that a pharmacological intervention targeting the conjunctive threshold, the point at which dopaminergic, glutamatergic, and affective signals converge to trigger consolidation, could prevent the memory from forming in the first place.

This is not a rejection of the brain disease model. It is a refinement of it. The model says addiction is a brain disease, and therefore we should treat the specific neural event that causes it, not just manage its downstream symptoms. The mainstream has the diagnosis right. I am proposing that the treatment implications have not been fully worked out.

PROJECT PROPOSAL

The CCT model is complete and validated at the simulation level. The next phase requires experimental testing. I am requesting funding to design and initiate a collaboration with a neuroscience laboratory capable of running the rodent assays that would directly test the model's central prediction: that combined modulation of D1 dopamine receptors, NMDAR-dependent plasticity, and affective contrast produces super-additive prevention of reward-memory consolidation, exceeding the sum of individual interventions by 13 to 22 percentage points.

The project has three components. First, I will finalize the model specification and publish the full parameter set, priors, and calibration code as a reproducible artifact on Zenodo and OSF. This is already 80 percent complete. Second, I will draft a detailed experimental protocol for a conditioned place preference paradigm in rodents, specifying drug doses, timing windows, and outcome measures derived directly from the model's predicted consolidation threshold. Third, I will identify and contact three to five laboratories with relevant expertise in reward memory and pharmacology, using my existing endorsements from Kent Berridge, Samuel Gershman, Nathaniel Daw, and Marcelo Mattar as entry points.

The budget is modest. I am requesting 25,000 dollars. This covers my time for six months of protocol development and lab coordination, travel to one collaborating laboratory for a two-week training and setup visit, and open-access publication fees for the three preprints currently under review plus the experimental protocol paper. If the experimental results confirm the model's predictions, the next step is a larger grant application for the full animal study, followed by translation to human biomarker studies.

The risk is real. The model could be wrong. But it is pre-registered, fully specified, and derived from a systematic literature screen of 1,847 records. The falsification risk is explicit, and I have a track record of reporting negative results honestly, as I did with the hERG topology study and the ergofluids real-data gate. Emergent Ventures funds projects where the upside justifies the risk. The upside here is a fundamentally new pharmacological strategy for addiction prevention, developed by an independent researcher in Nigeria with no institutional backing. That is the kind of project that does not get funded anywhere else.

CHECKLIST

- [ ] Submit application to Emergent Ventures via the rolling application portal at jehanazad.com
- [ ] Verify current application format and word limits on the Emergent Ventures website before submission
- [ ] Confirm the three-section structure: personal story, mainstream view, project proposal
- [ ] Insert personal details: specific hometown in Lagos, names of clinical supervisors, any prior awards or recognitions
- [ ] Verify the names and titles of the four endorsers (Berridge, Gershman, Daw, Mattar) and confirm they are willing to be referenced
- [ ] Confirm the journal names for the three preprints under review (IART, PNPBP, NBR) and the Alcohol submission status
- [ ] Double-check the exact AUROC values for TOPOLOGIX (0.804 on Platinum, 0.634 on SKEMPI 2.0) and the hERG replication (0.8426 vs 0.8782)
- [ ] Confirm the 13-22 percentage point super-additivity range across model versions
- [ ] Prepare a one-page CV or resume in case the application requests supplementary materials
- [ ] Prepare links to OSF/Zenodo preprints and GitHub repositories for reference

EDITOR NOTES

- Eligibility risk: Emergent Ventures typically favors entrepreneurial or policy-oriented projects over academic research. The CCT model is framed as a high-impact, unconventional scientific bet, which may work, but the applicant should be prepared for Tyler Cowen to ask about non-academic pathways to impact. Consider adding a sentence about how the model could inform a digital health product or clinical decision support tool, given the M.Sc. in Digital Health.

- Verification needed: The endorsements from Berridge, Gershman, Daw, and Mattar are listed in the profile but the nature of the endorsement is unspecified. The applicant must confirm whether these are formal letters, informal conversations, or arXiv endorsements (Gershman is noted as an arXiv endorsement specifically). Do not imply stronger relationships than exist.

- Gap: The personal story section lacks specific narrative detail. The applicant should insert concrete anecdotes from clinical practice in Nigeria, specific patient encounters, or the moment they decided to build the CCT model. The current draft is factually dense but emotionally flat, which may matter for a programme that values personal conviction.

MOTIVATION LETTER

The CCT model began with a contradiction. Nigeria has one of the highest rates of untreated substance use disorders in West Africa, yet the global research agenda treats addiction almost exclusively as a clinical problem requiring behavioral intervention or agonist maintenance. I spent five years as a pharmacist at the University of Ibadan and in clinical practice watching patients cycle through relapse because the underlying mechanism, the encoding of reward-memory associations during withdrawal, was never addressed pharmacologically. The mainstream approach manages craving after it appears. My work asks why the craving memory forms at all.

The Conjunctive Consolidation Threshold model is a tripartite pharmacological framework that treats reward-memory encoding as a coupled dynamical system across three axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. I built it as an independent researcher, outside any academic lab, using RK45 integration and Bayesian MCMC calibration with PyMC's DEMetropolisZ sampler. The model has 14 free parameters, all with priors elicited from a systematic screen of 1,847 published records. All five pre-registered hypotheses, H1 through H5, were confirmed. Posterior super-additivity across model versions ranges from 13 to 22 percentage points, meaning the three axes interact nonlinearly in ways a single-target approach cannot capture. Three sole-authored preprints are under review at peer-reviewed journals: International Addiction Review, Progress in Neuro-Psychopharmacology and Biological Psychiatry, and Neuroscience and Biobehavioral Reviews. A co-authored paper is under review at Alcohol, Elsevier.

Emergent Ventures funds unconventional projects with high potential impact. This qualifies on both counts. The CCT model challenges the assumption that addiction treatment must be clinic-based and behavior-first. It argues for a computational pharmacology-first approach: simulate the receptor and circuit dynamics, identify the consolidation threshold, and intervene before the memory encodes. No existing treatment targets this window. The model is technically grounded, fully specified, and has already passed its primary validation gate. What it lacks is funding to move from simulation to experimental collaboration, specifically to design the in vitro and rodent assays that would test the predicted super-additivity of combined D1 antagonism, NMDAR partial blockade, and affective contrast modulation.

I am a Nigerian pharmacist, 29 years old, enrolled in the M.Sc. Digital Health program at Hasso Plattner Institute and the University of Potsdam starting winter semester 2026/27. I have endorsements from Kent Berridge at Michigan, Samuel Gershman at Harvard, Nathaniel Daw at Princeton, and Marcelo Mattar at NYU. The work is real, the results are pre-registered and reproducible, and the idea is genuinely overlooked by mainstream funding bodies that route addiction research through clinical trial infrastructure. Emergent Ventures is the right vehicle for a project at this stage: too early for NIH R01s, too unconventional for traditional addiction research councils, and exactly the kind of bold, underfunded idea that Tyler Cowen's rolling grant process exists to catch.

SHORT ESSAY: PERSONAL STORY

I grew up in Lagos and trained as a pharmacist at the University of Ibadan, graduating in 2021 with a CGPA of 5.1 out of 7.0, a German equivalent of 1.9. I am licensed by the Pharmacists Council of Nigeria. During clinical rotations and later work at Ramset Pharmacy, I saw the same pattern repeatedly: patients with opioid or alcohol use disorders would detoxify successfully, then relapse within months because the environmental cues that triggered their cravings had been consolidated into long-term memory during withdrawal. The pharmacology we had, methadone, buprenorphine, naltrexone, all acted on receptor systems after the fact. None of them touched the memory encoding process itself.

I could not find a research group in Nigeria working on this problem, so I built the tools myself. I taught myself computational modeling, Bayesian statistics, and protein machine learning from primary literature. I developed the CCT model as a sole investigator, calibrating it against published electrophysiology and microdialysis data. I also ran a pre-registered replication study on hERG cardiotoxicity prediction using persistent homology, which found that topological features do not beat a plain descriptor baseline, AUROC 0.8426 versus 0.8782, settling a comparison the literature had never actually run. I built TOPOLOGIX, an ESM-2 protein language model approach that predicts drug-resistance mutations from sequence alone with AUROC 0.804 on the Platinum benchmark, covering 100 percent of mutations versus roughly 18 percent for structure-limited tools. I did all of this without institutional support, without a lab, and without grant funding.

The thread connecting these projects is a refusal to accept the boundary between clinical pharmacy and computational research. Nigeria produces pharmacists, but not computational neuroscientists. I am trying to be both. The M.Sc. at Hasso Plattner Institute is the first formal training I have had in digital health, and it starts in late 2026. Until then, I continue as an independent researcher with a GitHub, an ORCID, and a stack of preprints under review.

SHORT ESSAY: MAINSTREAM VIEW YOU AGREE WITH

The mainstream view I agree with is that addiction is a brain disease, not a moral failing. This position, articulated most forcefully by Nora Volkow and the National Institute on Drug Abuse, has been crucial for destigmatizing substance use disorders and redirecting research toward neurobiological mechanisms. I agree with it completely. The evidence is overwhelming: chronic drug exposure produces measurable changes in dopaminergic signaling, glutamatergic plasticity, and cortical-striatal circuit function. Patients are not choosing to be addicted. Their brains have been altered by the drug.

Where I diverge from the mainstream is in the therapeutic consequence drawn from this view. If addiction is a brain disease, the field has largely concluded that treatment must be chronic, maintenance-based, and behaviorally supported. The dominant protocols, agonist maintenance, cognitive behavioral therapy, contingency management, all assume the disease is a persistent state that must be managed indefinitely. My work suggests a different implication. If addiction is a brain disease of memory encoding, specifically the aberrant consolidation of drug-reward associations during withdrawal, then the disease may be interruptible at the moment of encoding. The CCT model predicts that a pharmacological intervention targeting the conjunctive threshold, the point at which dopaminergic, glutamatergic, and affective signals converge to trigger consolidation, could prevent the memory from forming in the first place.

This is not a rejection of the brain disease model. It is a refinement of it. The model says addiction is a brain disease, and therefore we should treat the specific neural event that causes it, not just manage its downstream symptoms. The mainstream has the diagnosis right. I am proposing that the treatment implications have not been fully worked out.

PROJECT PROPOSAL

The CCT model is complete and validated at the simulation level. The next phase requires experimental testing. I am requesting funding to design and initiate a collaboration with a neuroscience laboratory capable of running the rodent assays that would directly test the model's central prediction: that combined modulation of D1 dopamine receptors, NMDAR-dependent plasticity, and affective contrast produces super-additive prevention of reward-memory consolidation, exceeding the sum of individual interventions by 13 to 22 percentage points.

The project has three components. First, I will finalize the model specification and publish the full parameter set, priors, and calibration code as a reproducible artifact on Zenodo and OSF. This is already 80 percent complete. Second, I will draft a detailed experimental protocol for a conditioned place preference paradigm in rodents, specifying drug doses, timing windows, and outcome measures derived directly from the model's predicted consolidation threshold. Third, I will identify and contact three to five laboratories with relevant expertise in reward memory and pharmacology, using my existing endorsements from Kent Berridge, Samuel Gershman, Nathaniel Daw, and Marcelo Mattar as entry points.

The budget is modest. I am requesting 25,000 dollars. This covers my time for six months of protocol development and lab coordination, travel to one collaborating laboratory for a two-week training and setup visit, and open-access publication fees for the three preprints currently under review plus the experimental protocol paper. If the experimental results confirm the model's predictions, the next step is a larger grant application for the full animal study, followed by translation to human biomarker studies.

The risk is real. The model could be wrong. But it is pre-registered, fully specified, and derived from a systematic literature screen of 1,847 records. The falsification risk is explicit, and I have a track record of reporting negative results honestly, as I did with the hERG topology study and the ergofluids real-data gate. Emergent Ventures funds projects where the upside justifies the risk. The upside here is a fundamentally new pharmacological strategy for addiction prevention, developed by an independent researcher in Nigeria with no institutional backing. That is the kind of project that does not get funded anywhere else.

CHECKLIST

- [ ] Submit application to Emergent Ventures via the rolling application portal at jehanazad.com
- [ ] Verify current application format and word limits on the Emergent Ventures website before submission
- [ ] Confirm the three-section structure: personal story, mainstream view, project proposal
- [ ] Insert personal details: specific hometown in Lagos, names of clinical supervisors, any prior awards or recognitions
- [ ] Verify the names and titles of the four endorsers (Berridge, Gershman, Daw, Mattar) and confirm they are willing to be referenced
- [ ] Confirm the journal names for the three preprints under review (IART, PNPBP, NBR) and the Alcohol submission status
- [ ] Double-check the exact AUROC values for TOPOLOGIX (0.804 on Platinum, 0.634 on SKEMPI 2.0) and the hERG replication (0.8426 vs 0.8782)
- [ ] Confirm the 13-22 percentage point super-additivity range across model versions
- [ ] Prepare a one-page CV or resume in case the application requests supplementary materials
- [ ] Prepare links to OSF/Zenodo preprints and GitHub repositories for reference

EDITOR NOTES

- Eligibility risk: Emergent Ventures typically favors entrepreneurial or policy-oriented projects over academic research. The CCT model is framed as a high-impact, unconventional scientific bet, which may work, but the applicant should be prepared for Tyler Cowen to ask about non-academic pathways to impact. Consider adding a sentence about how the model could inform a digital health product or clinical decision support tool, given the M.Sc. in Digital Health.

- Verification needed: The endorsements from Berridge, Gershman, Daw, and Mattar are listed in the profile but the nature of the endorsement is unspecified. The applicant must confirm whether these are formal letters, informal conversations, or arXiv endorsements (Gershman is noted as an arXiv endorsement specifically). Do not imply stronger relationships than exist.

- Gap: The personal story section lacks specific narrative detail. The applicant should insert concrete anecdotes from clinical practice in Nigeria, specific patient encounters, or the moment they decided to build the CCT model. The current draft is factually dense but emotionally flat, which may matter for a programme that values personal conviction.

AI Draft — Emergent Ventures