Mental Workload and Performance in Motor Imagery-Based Brain-Computer Interfaces: A Rapid Systematic Review
Year: 2026
Type: preprint
Abstract: Motor Imagery based Brain-Computer Interfaces have emerged as a promising non invasive tool for neuromotor rehabilitation, offering users the ability to control external devices through the modulation of sensorimotor rhythms without relying on peripheral neuromuscular pathways. However, the translational success of these systems into clinical practice is heavily dependent on the user's cognitive state. Mental workload, defined as the dynamic balance between the cognitive demands imposed by a task and the individual's available neurocognitive resources, is intrinsically modulated by task complexity, sustained attention, and mental fatigue. Recent evidence suggests that cognitive factors and age account for significant variability in BCI performance, and prolonged use often leads to attenuated ERD responses and reduced classification accuracy. Despite a growing body of primary research, the specific relationship between mental workload and MI BCI performance remains fragmented and insufficiently synthesized, hindering the development of more robust adaptive systems for clinical rehabilitation. This rapid systematic review is designed to answer the following question: in users of motor imagery based brain computer interfaces, does exposure to high mental workload, when compared to conditions of lower cognitive demand, compromise interface performance? The primary objective is to systematically identify, evaluate, and synthesize the available evidence on this association. Secondary objectives include mapping the methodological approaches used to assess mental workload and performance, identifying individual neurophysiological and operational factors that modulate this relationship, and evaluating the methodological quality and risk of bias of the included studies to outline current limitations and strengths. We will conduct systematic electronic searches across five major databases: IEEE Xplore Digital Library, PubMed/MEDLINE, Web of Science, Scopus, and Google Scholar. To capture the current state of the art, the search will be restricted to literature published between January 2016 and June 2026, with language limits applied for English and Portuguese. We will employ a PICOT framework to guide eligibility. The population consists of users of MI BCIs, including both healthy individuals and patients undergoing neuromotor rehabilitation. The intervention or exposure is defined as high mental workload, fatigue, or cognitive demand during system operation. The comparator comprises low workload, control, or rest conditions. The primary outcomes are BCI performance metrics, including classification accuracy, Information Transfer Rate, error rate, reaction time, and ERD/ERS patterns. Study selection and data extraction will be performed independently by two reviewers using Rayyan software to manage deduplication and screening, and disagreements will be resolved through consensus. We will conduct a narrative synthesis to summarize the findings. The results will be structured around thematic axes, including user proficiency, mental fatigue, interface and feedback design, and methodological variability in workload assessment. Methodological quality and risk of bias will be critically appraised using the Mixed Methods Appraisal Tool, version 2018. The entire selection and review process will be documented and reported in accordance with the PRISMA 2020 guidelines using an adapted flow diagram. This synthesis will provide a comprehensive mapping of the association between mental workload and MI BCI performance based on the available evidence, describing how mental workload has been assessed across studies and distinguishing between subjective scales and objective physiological markers. The review will also identify individual, neurophysiological, and operational factors that may modulate this relationship, while systematically evaluating methodological quality and risk of bias to outline current strengths and limitations in the literature. By consolidating the existing evidence, this review aims to offer a clearer perspective to inform the development of more robust and user adaptive BCI systems. The findings are expected to support the clinical translation of MI BCIs by providing evidence that can guide the design of rehabilitation protocols, potentially contributing to improved functional outcomes for patients with stroke and other disabling neurological conditions.
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W7204198674