Imagination is a significant mental activity of human beings that can serve as a brain-computer interaction (BCI) paradigm, and visual imagery (VI) has become a relatively new BCI paradigm. VI is the ability to form visual representations in short-term memory without retinal input, offering users advantages like safety and rich content. It is anticipated to address the limitations of traditional BCI paradigms. However, further research is needed in the neural mechanisms, paradigm design, neural encoding, and decoding of VI-based BCI. Despite the existence of experimental studies on VI-BCI, there is a shortage of comprehensive reviews. This article provides a review of the neural mechanisms of VI, and explores paradigm designs for VI-BCI. Additionally, it covers neural encoding and decoding methods. Finally, the existing issues and suggested future research directions for VI-BCI are identified. The aim is to inspire further research in the field of VI-BCI.

With the advancement of non-invasive brain–computer interface (BCI) technologies, decoding high-level cognitive activity has become pivotal for expanding human–machine interaction. Visual imagery-based BCI (VI-BCI) enable voluntary activation of specific brain regions without external cue, offering novel pathways for immersive applications. However, research on the neural representation of such complex cognitive tasks is still limited, and most existing electroencephalogram (EEG) datasets primarily target motor imagery, hindering the development of robust VI decoding models. Here we present an EEG dataset recorded from 22 participants performing visual imagery tasks involving ten commonly recognized images across three categories: figures, animals, and objects. Each participant completed two sessions, with EEG recorded from 32-channels at 1000 Hz. This resource helps overcome data homogeneity issues in VI studies and provides a foundation for exploring neuroplasticity, adaptive decoding algorithms, and cross-subject generalization, facilitating the transition from controlled experiments to real-world applications.

Visual imagery is a cognitive process that generates internal visual representations in the absence of external visual stimuli and has become an important paradigm in brain-computer interface (BCI) research. In recent years, although publicly available simultaneous EEG-fNIRS datasets have been developed for tasks such as motor imagery and mental arithmetic, bimodal neural signal resources specifically designed for visual imagery tasks remain limited. Therefore, this study constructs and releases a simultaneous EEG-fNIRS dataset for three-class visual imagery tasks.The dataset initially collected neural signal data from 23 participants. After quality control, 22 valid participants were retained for subsequent analyses. During data acquisition, 32-channel EEG was used to record scalp electrophysiological signals, while 22-channel fNIRS was used to measure prefrontal hemodynamic responses. The dataset includes raw EEG and fNIRS signals, event markers, channel information, and example preprocessing codes.Furthermore, EEG-based, fNIRS-based, and EEG-fNIRS fusion baseline models were established to evaluate whether the dataset contains discriminative information for three-class visual imagery classification. The benchmark results demonstrate that the dataset contains neural information associated with visual imagery conditions and can support unimodal analysis, multimodal fusion, and BCI-related research.

Background: Visual imagery (VI) is a cognitive process that generates visual representations in the absence of external visual input; however, the spatiotemporal organization of its neural representations remains poorly understood. Methods: In this study, we combined electroencephalography (EEG) and functional magnetic resonance imaging (fMRI) with representational similarity analysis (RSA) and the Schaefer 2018 functional parcellation atlas to compare the correspondence between brain network representations and CORnet-S model representations during visual perception (VP) and VI. We further examined the cross-modal associations between EEG time–frequency representations and fMRI network representations during VI. Results: Significant model–brain representational correspondence was detected in the visual network during VP, and no significant correspondence was detected in any network during VI, yielding different descriptive significance profiles across the two tasks. Cross-modal EEG–fMRI RSA further revealed frequency- and time-dependent representational correspondence involving the frontoparietal, default mode, ventral attention, visual, and limbic networks. Significant cross-modal correspondence was observed across the alpha, beta, and gamma bands, with descriptive variation in the networks and temporal clusters involved in each frequency band. Conclusions: Collectively, these findings suggest that VI is not merely a reinstatement of VP but is associated with distributed representational patterns involving multiple functional networks. This study provides multimodal evidence for understanding the neural mechanisms underlying VI.

The large-scale, real-world deployment of brain–computer interface (BCI) is confronted with human factors bottlenecks, including reliability and usability. While modern human factors engineering has encompassed domains such as cognitive ergonomics, workload, adaptive automation, and human-in-the-loop systems, its theories and methodologies have historically placed their main emphasis on overt behavior and peripheral physiological measures, rather than on central neural activity as a primary control input. For the emerging interaction paradigm that takes central nervous activity as direct input and emphasizes long-term brain–computer co-adaptation, there remains a shortage of targeted conceptual and methodological frameworks. As a conceptual review, this paper discusses the necessity and disciplinary foundations for developing Human Factors Engineering for BCIs as an emerging interdisciplinary research direction. First, grounded in the defining features of BCI interaction—such as the implicit nature of input and brain–computer co-adaptation—it delineates the research boundaries and summarizes the core research objects as user neurocognitive characteristics, brain–computer collaborative adaptation mechanisms, multimodal feedback regulation, and contextual adaptation patterns. Second, it identifies three core scientific questions: individual differences and dynamic adaptation, human factors optimization and fault tolerance, and contextual generalizability principles, with ethics, privacy, and user rights and interests as an overarching dimension. Furthermore, it constructs a methodological system consisting of multimodal measurement and modeling, closed-loop validation, and multidimensional evaluation, and examines the supporting role of artificial intelligence approaches within this framework. Centered on the collaborative adaptation across neural, cognitive, and system levels, Human Factors Engineering for BCIs represents a deepening of the human-centered design philosophy. It provides theoretical reference for user-friendly BCI design and extends the research boundaries of human factors engineering.