EEG biomarkers of cognitive load: Insights from incremental element encoding in short-term working memory

Cognitive load refers to the mental effort required to encode, maintain, and manipulate information. Although previous electroencephalography (EEG) research has examined spectral biomarkers of cognitive load, most studies employed static task paradigms that average neural activity across entire difficulty levels. Such an approach presupposes that cognitive load remains constant within each level, thereby neglecting transient fluctuations that may arise during information processing. To address this limitation, we implemented a novel EEG-based incremental encoding paradigm to track dynamic changes in cognitive load over time. EEG was recorded from 24 healthy young adults performing the Corsi Block-Tapping Test, a visuospatial short-term memory task with sequential stimulus presentation. Items were added one by one to working memory, simulating real-world cognitive demands. Spectral absolute power was estimated across theta (4–8 Hz), alpha (8–13 Hz), low beta (13–20 Hz), high beta (20–30 Hz), and gamma ( > 30 Hz) bands in frontal and parietal regions. Independently of the number of encoded elements, spectral power increased relative to rest: frontal theta by 80.52%, parietal theta by 139.66%, and frontal alpha by 17.72%, reflecting general attention engagement. In contrast, low beta power decreased consistently as more items were encoded ( p < 0 . 01 , r > 0 . 5 ), arising as the most reliable biomarker of incremental memory load. A spectral shift toward higher beta frequencies was also observed with increased load. These results challenge the conventional understanding of theta as a biomarker of working memory and highlight beta-band dynamics as key to real-time cognitive monitoring in adaptive systems. • Identified spectral EEG biomarkers for cognitive load during memory encoding. • Proposed a novel framework for analyzing EEG in dynamic cognitive load scenarios. • Reduced inter-subject variability using resting-state baseline EEG normalization. • Demonstrated frontal low-beta desynchronization as a marker of cognitive demands. • Established theta and alpha bands as indicators of general cognitive engagement.