In this talk, we explore the potential of sparse-electrode Brain–Computer Interfaces (BCIs) across a range of application domains, from entertainment and personalized information access to healthcare-related scenarios. Our research demonstrates how an AI-driven processing pipeline can extract meaningful signals from the apparent chaos of EEG data collected through commercial, low-density BCI devices.
Despite relying on only a handful of electrodes, these systems can support surprisingly rich interactions: depending on the task, as few as four electrodes can be sufficient to control a toy car on a track or infer a user’s emotional state. This challenges the common assumption that effective BCI applications necessarily require high-density EEG caps with 32 or even 64 electrodes.
By combining lightweight sensing technologies with advanced machine learning and signal processing techniques, our research demonstrates that sparse BCIs can provide reliable performance while significantly improving user comfort and accessibility. Compared to traditional full-cap systems, these devices are easier to wear and deploy, making them strong candidates for the next generation of everyday wearable neurotechnology. Through concrete examples from our recent projects, I discuss the opportunities and challenges of bringing practical, user-friendly BCI systems closer to real-world applications.

References : 

T. Colafiglio et al., "NeuroSense: A Novel EEG Dataset Utilizing Low-Cost, Sparse Electrode Devices for Emotion Exploration," in IEEE Access, vol. 12, pp. 159296-159315, 2024, https://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=10737340

Tommaso Colafiglio, Angela Lombardi, Tommaso Di Noia, Maria Luigia Natalia De Bonis, Fedelucio Narducci, Alice Mado Proverbio, Machine learning classification of motivational states: Insights from EEG analysis of perception and imagery, Expert Systems with Applications, Volume 275, 2025, 127076, https://doi.org/10.1016/j.eswa.2025.127076.