Proactive intention decoding remains a critical yet underexplored challenge in brain–machine interfaces (BMIs), especially under naturalistic, self-initiated behavior. Existing systems rely on reactive decoding of motor cortex signals, resulting in substantial latency. To address this, we leverage the common marmoset’s spontaneous vocalizations and develop a high-resolution, dual-region ECoG recording paradigm targeting the prefrontal and auditory cortices and a neural decoding framework that integrates shapelet-based temporal encoding, position-aware attention, frequency-aware channel masking, contrastive clustering and a minimum error entropy-based robust loss. Our approach achieves 91.9% accuracy up to 200 ms before vocal onset—substantially outperforming 13 competitive baselines. Our model also uncovers a functional decoupling between auditory and prefrontal regions. Furthermore, joint modeling in time and frequency domains reveals novel preparatory neural signatures preceding volitional vocal output. Together, our findings bridge the gap between foundational neuroscience and applied BMI engineering, and establish a generalizable framework for intention decoding from ecologically valid, asynchronous behaviors.

Brain disorders have been consistently associated with abnormalities in specific brain regions or neural circuits. Identifying key brain regional activities and functional connectivity patterns is essential for discovering more precise neurobiological biomarkers. However, previous studies have primarily emphasized alterations in functional connectivity while overlooking abnormal neuronal population activity within brain regions. To bridge this gap, we propose a novel Local-Global Coupling Spiking Graph Transformer (LGC-SGT) that jointly models both inter-regional connectivity differences and deviations in neuronal population firing rates within brain regions, enabling a dual-perspective neuropathological analysis. The global pathway leverages spike-based computation in LGC-SGT to model biologically plausible aberrant neural firing dynamics, while the local pathway adaptively captures abnormal graph-based representations of brain connectivity learned by local plasticity in the liquid state machine module. Furthermore, we design a shortcut-enhanced output strategy in LGC-SGT with the hybrid loss function to suppress outlier interference caused by inter-individual and inter-center variability, enabling a more robust decision boundary. Extensive experiments on three brain disorder datasets demonstrate that our model consistently outperforms state-of-the-art graph methods in brain disorder diagnosis. Moreover, it facilitates the extraction of interpretable neurobiological biomarkers by jointly analyzing regional neural activity and functional connectivity, offering a more comprehensive framework for brain disorder understanding and diagnosis.

Developing new diagnostic models based on the underlying biological mechanisms rather than subjective symptoms for psychiatric disorders is an emerging consensus. Recently, machine learning (ML)-based classifiers using functional connectivity (FC) for psychiatric disorders and healthy controls (HCs) are developed to identify brain markers. However, existing ML-based diagnostic models are prone to overfitting (due to insufficient training samples) and perform poorly in new test environments. Furthermore, it is difficult to obtain explainable and reliable brain biomarkers elucidating the underlying diagnostic decisions. These issues hinder their possible clinical applications. In this work, we propose BrainIB, a new graph neural network (GNN) framework to analyze functional magnetic resonance images (fMRI), by leveraging the famed information bottleneck (IB) principle. BrainIB is able to identify the most informative edges in the brain (i.e., subgraph) and generalizes well to unseen data. We evaluate the performance of BrainIB against three baselines and seven state-of-the-art (SOTA) brain network classification methods on three psychiatric datasets and observe that our BrainIB always achieves the highest diagnosis accuracy. It also discovers the subgraph biomarkers that are consistent with clinical and neuroimaging findings. The source code and implementation details of BrainIB are freely available at the GitHub repository (https://github.com/SJYuCNEL/brain-and-Information-Bottleneck).

Repetitive transcranial magnetic stimulation (rTMS) is a common non-invasive treatment for medication-resistant major depressive disorder (MDD). It utilizes continuous and adjustable magnetic stimulation to modulate neural circuits implicated in the pathogenesis of depression. Nevertheless, constructing a universal and effective predictive factor for forecasting treatment outcomes remains challenging. To address this, we first collect neuroimaging data and five depression scales from 26 medication-resistant MDD patients before and after rTMS treatment. Then we propose a novel framework for predicting treatment effects precisely, which combines open-loop control and neural manifold estimation. This framework utilizes the geometric information of the manifold as a biomarker to predict the therapeutic efficacy of rTMS. Experiments based on the clinical dataset demonstrate the effectiveness and robustness of our framework.

Graph Neural Networks (GNNs) have been widely applied to graph-structured data. However, the lack of interpretability impedes its practical deployment especially in high-risk areas such as medical diagnosis. Recently, the Information Bottleneck (IB) principle has been extended to GNNs to identify a compact subgraph that is most informative to class labels, which significantly improves the interpretability on decision. However, existing Graph Information Bottleneck (GIB) models are either unstable during the training (due to the difficulty of mutual information estimation) or only focus on a special kind of graph (e.g., brain networks) that suffer from poor generalization to general graph datasets with varying graph sizes. In this work, we extend the recently developed Brain Information Bottleneck (BrainIB) to general graphs by introducing matrix-based Rényi’s α-order mutual information to stablize the training; and by designing a novel mask strategy to deal with varying graph sizes such that the new method can also be used for social networks, molecules, etc. Extensive experiments on different types of graph datasets demonstrate the superior stability and generality of our model.