Developing effective Brain-Computer Interfaces (BCIs) based on Imagined Speech (IS) is a significant challenge, largely due to high inter-subject variability in neural patterns. This study introduces a novel analytical framework to address this issue by integrating functional, effective, and complex network analyses with a more naturalistic sentence-level experimental protocol. Our findings confirm that while IS connectivity networks are characterized by considerable variability across individuals, our methodology successfully identifies a core set of stable pathways that persist across subjects. Specifically, we identified three principal pathways: a motor-language network in the left hemisphere driven by delta-band activity ( C L → F R , C R consistent in 60% of subjects), a right-hemisphere network relayed to motor planning areas via gamma-band activity ( T R → C L in 40% of subjects), and a top-down visual-spatial network involving parietal regions ( P O L → C R in 60% of subjects). In parallel, complex network analysis reveals the gamma frequency band to be the most functionally integrated and robust spectral signature, exhibiting significantly higher mean connectivity strength compared to all other bands (e.g., p = 0.0015 vs. beta) and appearing consistently in 6/10 subjects. By distinguishing these stable neural markers from subject-specific activity, this work provides more reliable EEG-based signatures for the future development of advanced speech BCIs. To identify stable neural markers of imagined speech, we first acquired EEG data using a novel sentence-level protocol. We then processed the signals using an innovative analytical framework that integrates multiple connectivity methods. An area selection criterion, based on consensus across undirected metrics, was used to isolate a core set of reliable connections. These selected connections were then used as targets for directed connectivity (EEC) analysis, while Complex Network Analysis (CNA) was separately applied to characterize global network properties. This combined functional, effective, and network analysis revealed the core EEG-based signatures of imagined speech, including three novel processing pathways. • Novel consensus-based framework disentangles stable signatures from high IS variability. • Gamma band proves significantly more functionally integrated than lower bands ( p < 0.002 ). • Identifies robust Delta motor-language ( C L → F R ) and Alpha visual-spatial ( P O L → C R ) loops. • Directed analysis reveals a specific Gamma-mediated auditory-to-motor drive ( T R → C L ). • Establishes stable connectivity features as neurophysiological priors for sentence-level IS decoding.
Iacomi et al. (Fri,) studied this question.