NERV: A comprehensive framework for rapid, reproducible, and hardware-synchronized neuroscience experiment design and...

Behavioral neuroscience experiments require precise stimulus control, millisecond timing, hardware integration, and robust data provenance. The growing use of 3D environments and multimodal recordings increases challenges for development, accessibility, and reproducibility. Fragmented tools often separate presentation, synchronization, and logging, creating workflow inefficiencies.

The Neuroscience Experimental Runtime by Vanderbilt (NERV) is a Unity-based C# framework that unifies experiment design, execution, and data logging. Custom Unity Editor tools automate scene and script generation, state management, and hardware-synchronized event timing via TTL pulses. A modular ExtraFunctions system enables plug-and-play modules like photodiode markers, gaze tracking, and reward control, while an automated archival process captures all code, configurations, and logs for complete provenance. The open-source framework follows a "low floor, high ceiling" design that supports both no-code use and full extensibility.

Across 500 trials, Unity-to-TTL delay was 2.10 ± 1.21 ms, TTL-to-photodiode delay was 28.93 ± 0.76 ms, and Unity-to-screen delay was 31.04 ± 1.41 ms, confirming stable millisecond precision and frame-locked timing for reliable alignment of neural, behavioral, and visual events.

Existing frameworks involve trade-offs between timing precision, accessibility, hardware support, and 3D capability. NERV integrates millisecond precision, modular open-source design, and full provenance in a single platform, accelerating development, reducing workflow fragmentation, and enabling reproducible, scalable experiments.

NERV provides an accessible and extensible framework that unites rapid development, robust data provenance, and millisecond precision, establishing a scalable foundation for next-generation neuroscience research.