OmniNeuro: A Multimodal HCI Framework for Explainable BCI Feedback via Generative AI and Sonification

Objective: While recent advancements in Deep Learning (DL) have significantly improved the decoding accuracy of Brain-Computer Interfaces (BCIs), clinical adoption remains stalled due to the “Black Box” nature of these algorithms. Patients and clinicians lack meaningful feedback on why a command failed, leading to frustration and poor neuroplasticity outcomes. We shift the focus from pure decoding maximization to Human-Computer Interaction (HCI), proposing OmniNeuro as a transparent feedback framework. Methods: OmniNeuro integrates three interpretability engines: (1) A Physics Engine (Energy Conservation), (2) A Chaos Engine (Fractal Complexity), and (3) A Quantum-Inspired Engine (utilizing quantum probability formalism for uncertainty modeling). These metrics drive a multimodal feedback system: real-time Neuro-Sonification and automated Generative AI Clinical Reports. We evaluated the framework on the PhysioNet dataset (N=109) and conducted a preliminary pilot qualitative study (N=3) to assess user experience. Results: The system achieved a mean accuracy of 58.52% across all subjects, with responsive subjects reaching 62.91%. Qualitative interviews revealed that users preferred the explanatory feedback over binary outputs, specifically noting that the “sonification” helped them regulate mental effort and reduce frustration during failure trials. Significance: By prioritizing explainability and multimodal feedback over raw accuracy, OmniNeuro establishes a new HCI paradigm for BCI. These findings provide preliminary evidence (not clinical validation) for the utility of explainable feedback in stabilizing user strategy. Furthermore, OmniNeuro is orthogonal to