Attention-BrainVerse: A Self-Adaptive Hybrid BCI-VR Interaction Model Driven by Attention

Brain-computer interface (BCI) is an emerging interactive technology that, when integrated with virtual reality (VR), exhibits significant potential for diverse applications. However, widespread adoption of BCI-VR systems remain hindered by challenges such as poor usability, high hardware costs, and a lack of established design guidelines. To address these issues and enhance user experience, this study proposes an attention-driven self-adaptive hybrid BCI-VR interaction model, Attention-BrainVerse. This model innovatively redefines the three core components of BCI-VR interaction: attention-centered input, self-adaptive processing, and feedback-based output. It incorporates our proposed hybrid input modalities, self-adaptive processing with dynamic threshold and dual-factor continuous difficulty adjustment, and output mechanisms based on attention visualization techniques and visual perspectives. To evaluate the proposed interaction model, three rounds of user experiments were conducted. Results demonstrate that, Attention-BrainVerse achieves a user experience comparable to that of traditional VR control methods. The main contribution of this work is the establishment of Attention-BrainVerse as a novel design framework that bridges critical gaps in BCI-VR interaction through its innovative integration of adaptive processing and hybrid input-output mechanisms, ultimately providing both theoretical guidance and practical design pathways for future BCI-VR interaction systems to facilitate the broader adoption of this integrated technology.