Work

Evaluating Real-Time Adaptive Interfaces through EEG-derived and Self-reported Mental Workload Indicators

Year: 2026

Type: conference-paper

Abstract: Adaptive information visualization (InfoVis) systems aim to reduce cognitive burden by tailoring the interface to the user’s needs. Yet their evaluation typically relies on subjective, post-task measures that lack temporal resolution and may not fully capture nuanced cognitive processes during interaction. We investigate the relationship between subjective workload and electroencephalography (EEG) indicators in the context of real-time adaptive InfoVis dashboards. We conduct a controlled user study (n = 20) comparing a baseline (non-adaptive) interface with a real-time adaptive variant, collecting both NASA-TLX ratings and continuous EEG signals. Our results show that adaptivity significantly reduces perceived workload, while EEG measures reveal a consistent tendency toward increased activity in the EEG theta band during task execution. These findings suggest that subjective and physiological measures capture complementary aspects of user cognition, such as perceived effort and sustained attention. We discuss implications for evaluating real-time adaptive InfoVis interfaces and argue for multimodal workload assessment in the engineering of interactive computing systems.

Source: ACM SIGCHI Symposium on Engineering Interactive Computing Systems (EICS)

DOI: 10.1145/3807968.3810932

Authors: Saravanakumar Duraisamy, Nuwan T Attygalle, Parvin Emami, Jean Vanderdonckt, Luis A. Leiva

Institutions: University of Luxembourg, UCLouvain

Language: English

Topic: Human-Automation Interaction and Safety

Subfield: Human Factors and Ergonomics

Field: Social Sciences

Domain: Social Sciences

Open Access status: hybrid

Funder: HORIZON EUROPE European Innovation Council

Award: Context-aware adaptive visualizations for critical decision making