AI for Neuroergonomics: Decoding the Brain and the Body in the Wild

Neuroergonomics lies at the intersection of neuroscience, neuroengineering, artificial intelligence, and human factors. It aims to understand and support brain and body function during real-world tasks, where noise, movement, context changes, and safety constraints can limit the direct translation of laboratory findings into operational environments. In parallel, AI is rapidly reshaping how we analyze neurophysiological and behavioral signals, design user-friendly neurotechnology, and enable new forms of human–machine symbiosis and interaction.A key driver of this evolution is the maturation of brain–computer interfaces (BCIs) from laboratory prototypes toward end-user-ready systems. This shift demands AI methods that deliver high accuracy with minimal calibration, remain robust to artifacts and context changes, and generalize across users and sessions to help overcome long-standing limitations such as “BCI illiteracy” and performance variability.This Research Topic will bring together advances across these communities to accelerate deployable and trustworthy AI for neuroergonomics. We welcome submissions that advance methods, validation, and applications across domains such as transportation, aviation, healthcare, industry, education, and immersive/extended reality environments.Best Paper Award (ANITI Prize)A Best Paper will be selected from the accepted contributions in this collection by the Topic Editors. Where applicable, the selected paper will be considered in connection with the ANITI Prize in AI for Neuroergonomics, in accordance with the prize’s official eligibility rules and timelines.This collection aims to highlight work that:· Enables robust decoding and modeling of brain and psychophysiological responses, in the lab and in real-world settings· Advances robust AI for neuroergonomic data under noise, artifacts, missing data, and distribution shift· Accelerates end-user BCI readiness, including minimal-calibration / calibration-free pipelines, cross-user generalization, and reliability to reduce performance variability (including challenges often grouped under “BCI illiteracy”)· Enables multimodal inference of cognitive, affective, and physiological states (individual and team levels)· Supports closed-loop neuroadaptive systems, including human–AI teaming and adaptive automation· Strengthens trustworthiness and responsibility (interpretability, uncertainty, fairness, privacy, safety, and governance)We invite Original Research, Reviews/Mini-Reviews, Methods, Perspectives, and Data Reports, including but not limited to:1) Methods for real-world neuroergonomics· Artifact-robust learning and signal processing for wearable neuroimaging (EEG, fNIRS, etc.)· Cross-subject and cross-session generalization; personalization; domain adaptation; continual/online learning· Uncertainty estimation, calibration, robustness, and reliability under distribution shift· Benchmarking, reproducible pipelines, reporting standards, open datasets, and shared tasks2) BCI methods focused on end-user performance and usability· Minimal-calibration and calibration-free BCIs (transfer learning, self-/semi-supervised learning, foundation models, few-shot adaptation)· Robust decoding under motion/EMG/EOG artifacts, non-stationarity, and real-world task variability· Strategies to improve accuracy, stability, and inclusivity across users (reducing variability often referred to as “BCI illiteracy”)· Online learning and adaptive classifiers; co-adaptation; rapid personalization without increasing user burden· Human-centered BCI evaluation: usability, workload, comfort, training burden, long-term use, and drop-out prevention3) Multimodal modeling of human state· Multimodal fusion: neurophysiology + peripheral physiology + behavior + context/environment· Workload, attention, fatigue, stress, affect, situation awareness, decision-making, and performance prediction· Team neurodynamics, collaboration, and group-level modeling in operational settings4) Neuroadaptive and interactive AI systems· Passive, active, hybrid, and multimodal BCIs for applied neuroergonomics and clinical applications· Neuroadaptive interfaces, adaptive automation, shared autonomy, and closed-loop human–machine interaction· Human–AI teaming: transparency, trust calibration, and explainable decision support· Neurofeedback, adaptive training, and augmentation to support performance, learning, and well-being5) Trustworthy, responsible, and privacy-preserving AI· Interpretable, causal, and mechanistic approaches for neuroergonomic inference· Fairness and bias assessment across individuals, contexts, and populations· Federated learning, on-device inference, privacy-preserving analytics, and secure pipelines· Safety validation, risk management, and governance for high-stakes deployments Topic editors: Frederic Dehais, Tiago Henrique Falk, Fabien Lotte, Waldemar Karwowski, Hasan Ayaz Source: https://www.frontiersin.org/research-topics/77504/ai-for-neuroergonomics-decoding-the-brain-and-the-body-in-the-wild