This platform is the backbone for building personalized neurofeedback applications. We separate core engineering from student-facing creativity. The result is a modular system: engineers build robust core libraries, while PhD/Masters students (or other developers) innovate apps on top.

1. Core platform for real-time signals, features, and states.

2. Construct axes as reusable dimensions.

3. SDK for student-led app development—defining new states, feedback, and interaction designs.

- Integrate EEG (MVP) first, with future support for fNIRS.

- Handle device streams, timestamps, channel metadata, and signal quality (impedance, artifacts).

- Preprocessing includes: filtering, rereferencing, artifact detection, and signal reliability scoring.

- Core library of evidence-backed markers (e.g. SMR, individualized alpha, theta/beta ratio, SCPs).

- Expandable registry of features (e.g. band power, coherence).

- Each marker is well-defined: modality, channels, latency, and evidence level.

- Axes are high-level dimensions derived from markers (e.g. Calm Focus, Task Engagement, Cognitive Control).

- Each axis fuses multiple markers.

- Axes are reusable across apps, acting as stable “control knobs.”

- Pre-defined states are combinations of axes (e.g. calm-focused, distracted, over-aroused).

- States are what apps respond to.

- Students combine existing axes to create new states relevant to their domain.

- Example: Define a “flow state” as a combination of high Task Engagement and Calm Focus.

- SDK exposes state listeners. Students define how feedback adapts when states change.

- Students can build interactive experiences—games, tasks, or interfaces that shift based on user brain state.

- Feedback can be visual, auditory, or task difficulty changes.

- Students create new neurofeedback apps by:

- Subscribing to states or axes.

- Designing novel feedback rules (e.g. when calm, show visual expansion).

- Creating interactive games or tasks that adapt to brain state shifts.