Open-source research and engineering at the intersection of computational neuroscience, adaptive intelligence, and neuroadaptive educational technology.

Noggin Labs develops computational systems, research frameworks, and accessible learning technologies informed by neuroscience, mathematics, and computer science.

Our work explores how intelligent systems can better understand learning, adapt to individual needs, and translate theoretical ideas into testable computational models.

We work across several interconnected areas:

• Computational Neuroscience — Neural dynamics, learning mechanisms, feedback timing, and computational models of cognition.
• Adaptive Intelligence — Systems that reason about learner responses, misconceptions, and instructional needs.
• Neuroadaptive Technology — Exploring how behavioral and physiological signals can inform adaptive learning systems.
• Scientific Computing — Numerical simulation, mathematical modeling, and reproducible computational research.
• Accessible Educational Technology — Building learning systems that accommodate different cognitive and learning needs.

Neural Feedback Optimization Theory (NFOT) investigates how feedback timing and its variability relate to performance in closed-loop learning systems.

Its central variable is the Write-Back Gap:

[ L = t_{\mathrm{fb}}-t_0 ]

Where:

• (t_0) represents a prespecified initiating biological or behavioral event.
• (t_{\mathrm{fb}}) represents the completion of feedback.

NFOT distinguishes the interval between an initiating event and completed feedback from ordinary system latency.

The current research version is v5.

The observational analysis uses the bigP3BCI dataset:

• 444 trials.
• 15 participants.
• Within-participant associations between Write-Back Gap and performance.
• Leave-one-subject-out analysis.
• Exploratory predictive comparisons.

The findings are observational. They do not establish causation, a universally optimal feedback delay, or a validated biological mechanism.

Prospective experimental validation, including the planned PTSP investigation of feedback-delay variability, remains necessary.

Read the NFOT research repository

Noggimigo is an open-source AI tutoring system designed around guided reasoning rather than simply supplying answers.

Its development brings together several complementary components:

• Socratic reasoning: Guiding learners through problems using structured instructional feedback.
• Misconception awareness: Identifying and responding to recognizable patterns of misunderstanding.
• Accessibility scaffolding: Supporting learners through adaptive instructional structures and smaller conceptual steps.
• Student simulation: Developing infrastructure for systematic evaluation against simulated learner behaviors.
• Hybrid AI direction: Exploring integration with pretrained models while maintaining distinct pedagogical and verification responsibilities.

Noggimigo's architecture separates pedagogical logic, reasoning and verification, model integration, and telemetry.

Educational interaction telemetry is distinct from NFOT's biological Write-Back Gap. They are not interchangeable measurements.

We believe scientific and engineering progress requires more than promising ideas.

Our development principles include:

1. Falsifiability — Research claims should be testable and open to disconfirmation.
2. Reproducibility — Experiments, computational methods, and results should be documented.
3. Scientific honesty — Observations, hypotheses, interpretations, and established findings must remain distinguishable.
4. Accessibility — Different learning needs should be considered in system design.
5. Iterative engineering — Implementations should be evaluated, improved, and validated rather than assumed to work.
6. Open research — We support accessible scientific tools and transparent development.

Our projects use technologies and methods from:

• Python
• JavaScript and TypeScript
• Computational modeling
• Machine learning and AI
• Numerical methods
• Scientific data analysis
• Open-source software development

Individual repositories document their own dependencies, installation procedures, and development requirements.

Noggin Labs aims to contribute to a future where computational intelligence, neuroscience, and accessible technology work together to improve how people learn and interact with intelligent systems.

We build, investigate, test, and refine.

Different ways of learning deserve serious scientific and engineering attention.

We welcome thoughtful contributions, scientific discussion, technical collaboration, and constructive criticism.

Explore our repositories:

Noggin Labs on GitHub

Research archive: Noggin Labs on Zenodo

_{© 2026 Noggin Labs and Folarera Kassim. Open-source licensing and contribution terms are specified in individual repositories.}