NeuroCinema

NeuroCinema explores what happens when a viewer’s own brain activity becomes part of the editing process. A participant wears an EEG headset. Neural signals influence both the scene that appears and the pace of the edit, so each participant experiences a different version of the same visual material.

The goal was not to visualize brain data for its own sake, but to create an interaction audiences could perceive and understand.

My work began during an Independent Study with Dr. Griffin Milsap, a biomedical engineer at the Johns Hopkins Applied Physics Laboratory.

I assembled the G.tec Unicorn EEG hardware, connected processed neural data to TouchDesigner through OSC, and built a node-based prototype that translated physiological signals into video-editing decisions. The prototype established a foundational BCI-to-video workflow that informed later development.

The system explored two forms of agency. Participants selected among three projected scenes using SSVEP-based input. By focusing on a flickering visual target, the viewer could intentionally choose which scene appeared.

Early tests with subtle visual modulation proved unreliable once light from multiple projections mixed in the room. We switched to high-contrast checkerboard grids, which produced more dependable selection.

Once a scene was chosen, broader brainwave activity influenced editing speed. Higher alpha activity slowed the edit, while beta and gamma increased its pace. The result combined deliberate control with involuntary influence.

The participant neither fully controlled the film nor simply watched it. Scene selection was intentional. Edit pacing responded to broader changes in neural activity.

To make that relationship legible, I used thresholds and smoothing to translate noisy, continuous physiological signals into stable editing decisions. That design process reinforced an important principle: an interactive system is only meaningful if the audience can understand the relationship between their actions and what changes.