Draft — Wissenschaftskolleg zu Berlin Early Career Fellowships

MOTIVATION LETTER

The Wissenschaftskolleg zu Berlin offers something rare: a space where a researcher can pursue a single line of thought for a full year, surrounded by scholars who ask questions from outside one's own discipline. My research sits at the intersection of pharmacology, computational neuroscience, and dynamical systems theory. The CCT model, my tripartite framework for reward-memory encoding prevention in addiction, is precisely the kind of project that benefits from sustained interdisciplinary exchange. It is a mathematical model built from three coupled axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. It is also a philosophical argument about how memories become compulsive. Those two dimensions need each other.

I am a Nigerian-trained pharmacist with a B.Pharm from the University of Ibadan, currently enrolled in the M.Sc. Digital Health program at the Hasso Plattner Institute in Potsdam. I have spent the past two years building the CCT model as an independent researcher. The model is a system of ordinary differential equations solved with RK45, calibrated through Bayesian MCMC using PyMC's DEMetropolisZ sampler with 14 free parameters. The priors were elicited from a systematic screen of 1,847 records in the addiction neuroscience literature. All five pre-registered hypotheses, H1 through H5, were confirmed. The posterior shows super-additivity of 13 to 22 percentage points across model versions, meaning the combined effect of the three axes exceeds the sum of their individual contributions. Three sole-authored preprints are currently under review at peer-reviewed journals: IART, PNPBP, and NBR. A co-authored paper is under review at Alcohol (Elsevier).

The Wissenschaftskolleg's selection criteria emphasize academic excellence, interdisciplinary openness, and the quality of previous work. I meet those criteria with a documented record of pre-registered, falsifiable research. But I also meet the deeper criterion: the willingness to have my assumptions challenged by people who do not share them. My model makes claims about how dopamine, glutamate, and affective state interact to consolidate reward memories. A philosopher of mind might ask whether my definition of "affective contrast" is coherent. An ethicist might ask what it means to prevent memory encoding in human subjects. A social scientist might ask how such a model could be deployed responsibly in addiction treatment across different cultural contexts, particularly in Africa where I am from and where addiction research infrastructure remains thin. These questions are central to whether the CCT model can move from a mathematical framework to a clinically meaningful one.

The fellowship period would allow me to deepen the theoretical foundations of the CCT model, expand its implications for addiction neuroscience, and engage with the collegial community at the Wissenschaftskolleg. I am asking for the conditions to do the hardest intellectual work of my career so far: making a computational model of addiction strong enough to survive contact with scholars who think differently than I do.

RESEARCH STATEMENT

The Conjunctive Consolidation Threshold (CCT) model addresses a specific gap in addiction neuroscience: the field has strong molecular evidence about dopamine, NMDA receptors, and affective state, but no unified quantitative framework that explains how these three systems interact to consolidate reward memories into compulsive patterns. My research builds that framework and tests it against pre-registered hypotheses.

The model is structured as three coupled axes. The first axis is dopaminergic reward prediction error, formalized using temporal difference learning. The second axis is NMDAR-dependent long-term potentiation, modeled as a synaptic weight update rule with a saturation threshold. The third axis is affective contrast, defined as the difference between the emotional valence of a drug experience and the valence of the surrounding baseline state. These three axes are coupled in a system of ordinary differential equations. The key theoretical claim is that memory consolidation in addiction requires a conjunctive threshold: no single axis is sufficient, but when all three exceed their individual thresholds simultaneously, the system crosses into a regime where reward memories become disproportionately strong and resistant to extinction.

The mathematical implementation uses RK45 integration for the ODE system. Calibration was performed with Bayesian MCMC using PyMC's DEMetropolisZ sampler. The model has 14 free parameters, all with priors elicited from a systematic literature screen of 1,847 records. I pre-registered five hypotheses before running the calibration. All five were confirmed. The posterior distribution shows super-additivity of 13 to 22 percentage points across model versions, meaning the conjunctive effect of the three axes is greater than the sum of their individual effects. This is the central quantitative result of the project.

Three sole-authored preprints document this work, each currently under review at a peer-reviewed journal: IART, PNPBP, and NBR. A co-authored paper is under review at Alcohol (Elsevier). The preprints are archived on OSF and Zenodo with full reproducibility materials, including code, data, and analysis scripts.

The CCT model has clear limitations. The model is calibrated to literature-derived parameters, not to individual patient data. The affective contrast axis is the least well-constrained empirically, and I have been transparent about this in the preprints. The model makes predictions that could be tested in animal models, but I am a computational researcher, not a wet-lab scientist. The next phase of the work requires collaboration with experimentalists who can test the model's predictions about threshold interactions.

The Wissenschaftskolleg is the right environment for this next phase. The fellowship's emphasis on intellectual freedom and interdisciplinary exchange aligns with the CCT model's need for input from philosophy of mind, ethics, and social science. A philosopher could help sharpen the definition of affective contrast. An ethicist could help think through the implications of memory-encoding prevention. A social scientist could help design a research agenda for how the model might apply in diverse cultural contexts, including African settings where addiction treatment resources are scarce.

During the fellowship year, I would focus on three deliverables. First, a theoretical paper that formalizes the conjunctive threshold concept and situates it within the broader literature on memory reconsolidation and addiction. Second, a computational extension of the model that incorporates individual-level variability in receptor densities and affective baselines, moving the model closer to clinical applicability. Third, a collaborative workshop series within the Wissenschaftskolleg community, bringing together scholars from neuroscience, philosophy, and ethics to interrogate the model's assumptions and implications.

The CCT model is my primary research line because it is the one that most directly benefits from the Wissenschaftskolleg's specific strengths: sustained attention, interdisciplinary challenge, and a community that values rigor over speed. I am asking the fellowship to fund the intellectual work of making a computational model of addiction conceptually strong enough to matter beyond its own mathematical boundaries.

SHORT ESSAY: INTERDISCIPLINARY EXCHANGE

The Wissenschaftskolleg's selection criteria explicitly value openness to exchange beyond one's field. My record demonstrates this in practice, not just in principle.

My research has already produced one notable negative result that required intellectual honesty. In a pre-registered, powered replication study on hERG cardiotoxicity, I tested whether bipartite persistent homology could predict cardiotoxicity from protein-ligand interface geometry. The result was negative: topological features did not beat a plain descriptor baseline, with an AUROC of 0.8426 versus 0.8782. This finding settled a comparison the published literature had never actually run. I reported it directly rather than reframing it. The same class of topological constructs applied to drug-resistance prediction also showed almost no signal, with AUROCs of 0.425 and 0.485 on the Platinum benchmark. These results led me to pivot to a sequence-representation approach, TOPOLOGIX, which now achieves an AUROC of 0.804 plus or minus 0.025 on the Platinum benchmark, beating structure-based baselines while covering 100 percent of mutations versus roughly 18 percent for structure-limited tools.

This pattern, pre-register, test, report honestly, pivot when the evidence demands it, is the kind of scientific practice that benefits from interdisciplinary scrutiny. A philosopher of science could ask whether my pre-registration practices are sufficiently rigorous. A statistician could ask whether my Bayesian calibration choices are appropriate. A domain expert in addiction psychiatry could ask whether my model's assumptions about affective contrast match clinical reality. I want those questions. They make the work better.

The CCT model itself is inherently interdisciplinary. It draws on computational neuroscience for the dopaminergic reward prediction error axis, molecular pharmacology for the NMDAR-dependent LTP axis, and affective neuroscience for the contrast axis. The model's implications extend to ethics, particularly the question of whether preventing memory encoding in addiction treatment is morally acceptable. That is a question I cannot answer alone. It requires input from philosophers, ethicists, and clinicians. The Wissenschaftskolleg's collegial environment is the right place to have that conversation.

SHORT ESSAY: FEASIBILITY AND PLANNED OUTPUTS

The CCT model is at a stage where a fellowship year can produce concrete, high-quality outputs. The computational infrastructure is complete. The model is calibrated, the hypotheses are confirmed, and three preprints are under review. What remains is theoretical deepening, computational extension, and community engagement.

The first planned output is a theoretical paper formalizing the conjunctive threshold concept. This paper would situate the CCT model within the broader literature on memory reconsolidation, addiction, and dynamical systems approaches to psychiatry. It would address the philosophical objections to the model's assumptions, particularly the definition of affective contrast and the ethical implications of memory-encoding prevention. Target venue: Behavioral and Brain Sciences or a comparable journal that welcomes interdisciplinary theoretical contributions.

The second planned output is a computational extension of the model incorporating individual-level variability. The current model uses literature-derived parameters. The extension would introduce distributions over receptor densities, baseline affective states, and learning rates, allowing the model to generate predictions about individual differences in addiction vulnerability. This would move the model closer to clinical applicability and provide testable predictions for experimental collaborators. Target venue: PLOS Computational Biology or a comparable journal.

The third planned output is a workshop series within the Wissenschaftskolleg community. The series would bring together scholars from neuroscience, philosophy, ethics, and social science to interrogate the CCT model's assumptions and implications. The goal is not to defend the model but to stress-test it. The workshop would produce a collaborative position paper on the ethics and feasibility of memory-encoding prevention in addiction treatment.

These three outputs are feasible within a fellowship year because the computational work is already done. The remaining work is intellectual, not technical. It requires time, attention, and the right interlocutors. The Wissenschaftskolleg provides exactly those conditions.

CHECKLIST

- [ ] Verify current deadline and application portal access at call.wiko-berlin.de
- [ ] Confirm eligibility for Early Career Fellowships as an independent researcher without a PhD
- [ ] Confirm whether enrollment in M.Sc. Digital Health at HPI/Potsdam affects eligibility status
- [ ] Prepare CV in the format required by the Wissenschaftskolleg application portal
- [ ] Prepare publication list including three preprints under review (IART, PNPBP, NBR) and co-authored Alcohol paper
- [ ] Prepare links to OSF and Zenodo repositories for CCT model code, data, and preprints
- [ ] Prepare ORCID (0009-0001-9272-6735) and GitHub (github.com/AmunRaPtah) identifiers for application
- [ ] Draft project description specific to the Wissenschaftskolleg's required format, not exceeding stated word limits
- [ ] Secure two letters of recommendation from named endorsers (Berridge, Gershman, Daw, or Mattar)
- [ ] Confirm whether recommendation letters must be submitted through the portal or sent directly
- [ ] Prepare a one-page summary of the CCT model for non-specialist reviewers
- [ ] Verify all pre-registration documents and hypothesis confirmations are publicly accessible
- [ ] Confirm the Wissenschaftskolleg's policy on researchers enrolled in other degree programs during the fellowship period
- [ ] Prepare a budget outline if the application requires one, noting the fellowship amount is not specified on the programme page

EDITOR NOTES

- Research line selected: CCT model. This is the correct choice because the Wissenschaftskolleg's stated focus on interdisciplinary exchange and its openness to natural scientists aligns with the CCT model's need for philosophical, ethical, and social-science input. The other research lines (TOPOLOGIX, neurocascade, ergofluids) are more technical and less suited to a humanities-leaning institution. The negative hERG result is mentioned in the interdisciplinary essay as evidence of scientific honesty, which matches the programme's emphasis on rigor.

- Eligibility risk: The Wissenschaftskolleg's Early Career Fellowships typically target postdoctoral researchers. Eniola does not hold a PhD and is enrolled in an M.Sc. program. This is a significant eligibility risk that must be verified before investing time in the full application. The strategy notes acknowledge this risk but do not resolve it.

- Verification needed: The three preprints under review (IART, PNPBP, NBR) and the co-authored Alcohol paper must be confirmed as currently under review. The application should not claim they are accepted or published. The OSF and Zenodo links must be live and contain the full reproducibility materials.

- Personal detail gap: The application does not specify why Eniola chose the Wissenschaftskolleg specifically beyond the strategic fit. The motivation letter should include a concrete reference to a specific Wissenschaftskolleg fellow, event, or publication that influenced this choice. This detail must come from Eniola, not from the profile.

- Timeline risk: The fellowship period would need to accommodate Eniola's M.Sc. enrollment at HPI/Potsdam. The application must clarify whether the fellowship would be taken concurrently with the degree program or after completion. This is a factual detail that must be resolved before submission.

MOTIVATION LETTER

The Wissenschaftskolleg zu Berlin offers something rare: a space where a researcher can pursue a single line of thought for a full year, surrounded by scholars who ask questions from outside one's own discipline. My research sits at the intersection of pharmacology, computational neuroscience, and dynamical systems theory. The CCT model, my tripartite framework for reward-memory encoding prevention in addiction, is precisely the kind of project that benefits from sustained interdisciplinary exchange. It is a mathematical model built from three coupled axes: dopaminergic reward prediction error, NMDAR-dependent long-term potentiation, and affective contrast. It is also a philosophical argument about how memories become compulsive. Those two dimensions need each other.

I am a Nigerian-trained pharmacist with a B.Pharm from the University of Ibadan, currently enrolled in the M.Sc. Digital Health program at the Hasso Plattner Institute in Potsdam. I have spent the past two years building the CCT model as an independent researcher. The model is a system of ordinary differential equations solved with RK45, calibrated through Bayesian MCMC using PyMC's DEMetropolisZ sampler with 14 free parameters. The priors were elicited from a systematic screen of 1,847 records in the addiction neuroscience literature. All five pre-registered hypotheses, H1 through H5, were confirmed. The posterior shows super-additivity of 13 to 22 percentage points across model versions, meaning the combined effect of the three axes exceeds the sum of their individual contributions. Three sole-authored preprints are currently under review at peer-reviewed journals: IART, PNPBP, and NBR. A co-authored paper is under review at Alcohol (Elsevier).

The Wissenschaftskolleg's selection criteria emphasize academic excellence, interdisciplinary openness, and the quality of previous work. I meet those criteria with a documented record of pre-registered, falsifiable research. But I also meet the deeper criterion: the willingness to have my assumptions challenged by people who do not share them. My model makes claims about how dopamine, glutamate, and affective state interact to consolidate reward memories. A philosopher of mind might ask whether my definition of "affective contrast" is coherent. An ethicist might ask what it means to prevent memory encoding in human subjects. A social scientist might ask how such a model could be deployed responsibly in addiction treatment across different cultural contexts, particularly in Africa where I am from and where addiction research infrastructure remains thin. These questions are central to whether the CCT model can move from a mathematical framework to a clinically meaningful one.

The fellowship period would allow me to deepen the theoretical foundations of the CCT model, expand its implications for addiction neuroscience, and engage with the collegial community at the Wissenschaftskolleg. I am asking for the conditions to do the hardest intellectual work of my career so far: making a computational model of addiction strong enough to survive contact with scholars who think differently than I do.

RESEARCH STATEMENT

The Conjunctive Consolidation Threshold (CCT) model addresses a specific gap in addiction neuroscience: the field has strong molecular evidence about dopamine, NMDA receptors, and affective state, but no unified quantitative framework that explains how these three systems interact to consolidate reward memories into compulsive patterns. My research builds that framework and tests it against pre-registered hypotheses.

The model is structured as three coupled axes. The first axis is dopaminergic reward prediction error, formalized using temporal difference learning. The second axis is NMDAR-dependent long-term potentiation, modeled as a synaptic weight update rule with a saturation threshold. The third axis is affective contrast, defined as the difference between the emotional valence of a drug experience and the valence of the surrounding baseline state. These three axes are coupled in a system of ordinary differential equations. The key theoretical claim is that memory consolidation in addiction requires a conjunctive threshold: no single axis is sufficient, but when all three exceed their individual thresholds simultaneously, the system crosses into a regime where reward memories become disproportionately strong and resistant to extinction.

The mathematical implementation uses RK45 integration for the ODE system. Calibration was performed with Bayesian MCMC using PyMC's DEMetropolisZ sampler. The model has 14 free parameters, all with priors elicited from a systematic literature screen of 1,847 records. I pre-registered five hypotheses before running the calibration. All five were confirmed. The posterior distribution shows super-additivity of 13 to 22 percentage points across model versions, meaning the conjunctive effect of the three axes is greater than the sum of their individual effects. This is the central quantitative result of the project.

Three sole-authored preprints document this work, each currently under review at a peer-reviewed journal: IART, PNPBP, and NBR. A co-authored paper is under review at Alcohol (Elsevier). The preprints are archived on OSF and Zenodo with full reproducibility materials, including code, data, and analysis scripts.

The CCT model has clear limitations. The model is calibrated to literature-derived parameters, not to individual patient data. The affective contrast axis is the least well-constrained empirically, and I have been transparent about this in the preprints. The model makes predictions that could be tested in animal models, but I am a computational researcher, not a wet-lab scientist. The next phase of the work requires collaboration with experimentalists who can test the model's predictions about threshold interactions.

The Wissenschaftskolleg is the right environment for this next phase. The fellowship's emphasis on intellectual freedom and interdisciplinary exchange aligns with the CCT model's need for input from philosophy of mind, ethics, and social science. A philosopher could help sharpen the definition of affective contrast. An ethicist could help think through the implications of memory-encoding prevention. A social scientist could help design a research agenda for how the model might apply in diverse cultural contexts, including African settings where addiction treatment resources are scarce.

During the fellowship year, I would focus on three deliverables. First, a theoretical paper that formalizes the conjunctive threshold concept and situates it within the broader literature on memory reconsolidation and addiction. Second, a computational extension of the model that incorporates individual-level variability in receptor densities and affective baselines, moving the model closer to clinical applicability. Third, a collaborative workshop series within the Wissenschaftskolleg community, bringing together scholars from neuroscience, philosophy, and ethics to interrogate the model's assumptions and implications.

The CCT model is my primary research line because it is the one that most directly benefits from the Wissenschaftskolleg's specific strengths: sustained attention, interdisciplinary challenge, and a community that values rigor over speed. I am asking the fellowship to fund the intellectual work of making a computational model of addiction conceptually strong enough to matter beyond its own mathematical boundaries.

SHORT ESSAY: INTERDISCIPLINARY EXCHANGE

The Wissenschaftskolleg's selection criteria explicitly value openness to exchange beyond one's field. My record demonstrates this in practice, not just in principle.

My research has already produced one notable negative result that required intellectual honesty. In a pre-registered, powered replication study on hERG cardiotoxicity, I tested whether bipartite persistent homology could predict cardiotoxicity from protein-ligand interface geometry. The result was negative: topological features did not beat a plain descriptor baseline, with an AUROC of 0.8426 versus 0.8782. This finding settled a comparison the published literature had never actually run. I reported it directly rather than reframing it. The same class of topological constructs applied to drug-resistance prediction also showed almost no signal, with AUROCs of 0.425 and 0.485 on the Platinum benchmark. These results led me to pivot to a sequence-representation approach, TOPOLOGIX, which now achieves an AUROC of 0.804 plus or minus 0.025 on the Platinum benchmark, beating structure-based baselines while covering 100 percent of mutations versus roughly 18 percent for structure-limited tools.

This pattern, pre-register, test, report honestly, pivot when the evidence demands it, is the kind of scientific practice that benefits from interdisciplinary scrutiny. A philosopher of science could ask whether my pre-registration practices are sufficiently rigorous. A statistician could ask whether my Bayesian calibration choices are appropriate. A domain expert in addiction psychiatry could ask whether my model's assumptions about affective contrast match clinical reality. I want those questions. They make the work better.

The CCT model itself is inherently interdisciplinary. It draws on computational neuroscience for the dopaminergic reward prediction error axis, molecular pharmacology for the NMDAR-dependent LTP axis, and affective neuroscience for the contrast axis. The model's implications extend to ethics, particularly the question of whether preventing memory encoding in addiction treatment is morally acceptable. That is a question I cannot answer alone. It requires input from philosophers, ethicists, and clinicians. The Wissenschaftskolleg's collegial environment is the right place to have that conversation.

SHORT ESSAY: FEASIBILITY AND PLANNED OUTPUTS

The CCT model is at a stage where a fellowship year can produce concrete, high-quality outputs. The computational infrastructure is complete. The model is calibrated, the hypotheses are confirmed, and three preprints are under review. What remains is theoretical deepening, computational extension, and community engagement.

The first planned output is a theoretical paper formalizing the conjunctive threshold concept. This paper would situate the CCT model within the broader literature on memory reconsolidation, addiction, and dynamical systems approaches to psychiatry. It would address the philosophical objections to the model's assumptions, particularly the definition of affective contrast and the ethical implications of memory-encoding prevention. Target venue: Behavioral and Brain Sciences or a comparable journal that welcomes interdisciplinary theoretical contributions.

The second planned output is a computational extension of the model incorporating individual-level variability. The current model uses literature-derived parameters. The extension would introduce distributions over receptor densities, baseline affective states, and learning rates, allowing the model to generate predictions about individual differences in addiction vulnerability. This would move the model closer to clinical applicability and provide testable predictions for experimental collaborators. Target venue: PLOS Computational Biology or a comparable journal.

The third planned output is a workshop series within the Wissenschaftskolleg community. The series would bring together scholars from neuroscience, philosophy, ethics, and social science to interrogate the CCT model's assumptions and implications. The goal is not to defend the model but to stress-test it. The workshop would produce a collaborative position paper on the ethics and feasibility of memory-encoding prevention in addiction treatment.

These three outputs are feasible within a fellowship year because the computational work is already done. The remaining work is intellectual, not technical. It requires time, attention, and the right interlocutors. The Wissenschaftskolleg provides exactly those conditions.

CHECKLIST

- [ ] Verify current deadline and application portal access at call.wiko-berlin.de
- [ ] Confirm eligibility for Early Career Fellowships as an independent researcher without a PhD
- [ ] Confirm whether enrollment in M.Sc. Digital Health at HPI/Potsdam affects eligibility status
- [ ] Prepare CV in the format required by the Wissenschaftskolleg application portal
- [ ] Prepare publication list including three preprints under review (IART, PNPBP, NBR) and co-authored Alcohol paper
- [ ] Prepare links to OSF and Zenodo repositories for CCT model code, data, and preprints
- [ ] Prepare ORCID (0009-0001-9272-6735) and GitHub (github.com/AmunRaPtah) identifiers for application
- [ ] Draft project description specific to the Wissenschaftskolleg's required format, not exceeding stated word limits
- [ ] Secure two letters of recommendation from named endorsers (Berridge, Gershman, Daw, or Mattar)
- [ ] Confirm whether recommendation letters must be submitted through the portal or sent directly
- [ ] Prepare a one-page summary of the CCT model for non-specialist reviewers
- [ ] Verify all pre-registration documents and hypothesis confirmations are publicly accessible
- [ ] Confirm the Wissenschaftskolleg's policy on researchers enrolled in other degree programs during the fellowship period
- [ ] Prepare a budget outline if the application requires one, noting the fellowship amount is not specified on the programme page

EDITOR NOTES

- Research line selected: CCT model. This is the correct choice because the Wissenschaftskolleg's stated focus on interdisciplinary exchange and its openness to natural scientists aligns with the CCT model's need for philosophical, ethical, and social-science input. The other research lines (TOPOLOGIX, neurocascade, ergofluids) are more technical and less suited to a humanities-leaning institution. The negative hERG result is mentioned in the interdisciplinary essay as evidence of scientific honesty, which matches the programme's emphasis on rigor.

- Eligibility risk: The Wissenschaftskolleg's Early Career Fellowships typically target postdoctoral researchers. Eniola does not hold a PhD and is enrolled in an M.Sc. program. This is a significant eligibility risk that must be verified before investing time in the full application. The strategy notes acknowledge this risk but do not resolve it.

- Verification needed: The three preprints under review (IART, PNPBP, NBR) and the co-authored Alcohol paper must be confirmed as currently under review. The application should not claim they are accepted or published. The OSF and Zenodo links must be live and contain the full reproducibility materials.

- Personal detail gap: The application does not specify why Eniola chose the Wissenschaftskolleg specifically beyond the strategic fit. The motivation letter should include a concrete reference to a specific Wissenschaftskolleg fellow, event, or publication that influenced this choice. This detail must come from Eniola, not from the profile.

- Timeline risk: The fellowship period would need to accommodate Eniola's M.Sc. enrollment at HPI/Potsdam. The application must clarify whether the fellowship would be taken concurrently with the degree program or after completion. This is a factual detail that must be resolved before submission.

AI Draft — Wissenschaftskolleg zu Berlin Early Career Fellowships