Assessing focus through ear-EEG: a comparative study between conventional cap EEG and mobile in- and around-the-ear EEG systems

Crétot-Richert, Gabrielle; De Vos, Maarten; Debener, Stefan; Bleichner, Martin G.; Voix, Jérémie · 2023 · Crossref

DOI: 10.3389/fnins.2023.895094

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the viability of mobile, wearable ear-EEG systems for monitoring mental focus and cognitive workload, addressing the need for inconspicuous neurotechnology that can operate in real-world settings. While traditional cap-EEG provides high-resolution data, its bulkiness limits practical application. The authors aim to determine if EEG signals recorded inside and around the ear can reliably differentiate between levels of attention, working memory, and cognitive load, thereby enabling future brain-computer interfaces (BCIs) that protect users’ focus by detecting mental fatigue or distraction. The researchers conducted a comparative experiment involving fifteen subjects who performed two tasks: an N-back task to assess working memory and a mental arithmetic task to evaluate cognitive workload. Data was recorded concurrently using a conventional 96-channel cap-EEG system and a mobile ear-EEG system comprising cEEGrids (around-the-ear electrodes) and TIPtrodes (in-ear electrodes). Signal processing included band-pass filtering, artifact removal via Independent Component Analysis, and Power Spectral Density (PSD) analysis across seven frequency bands. Classification models were trained using step-wise linear discriminant analysis (swLDA) on spectral features from twelve selected channels for both systems, as well as a two-channel model using only the in-ear electrodes. Results demonstrated that spectral features significantly differed between cognitive load conditions for both tasks. Single-trial classification accuracies were above chance for all subjects. For the N-back task, the twelve-channel models achieved mean accuracies of 96% for cap-EEG and 95% for ear-EEG. The two-channel in-ear model achieved 74% accuracy, comparable to the 76% achieved by the two-channel cap-EEG model. For the arithmetic task, the twelve-channel ear-EEG model slightly outperformed the cap-EEG model (85% vs. 82%), while the two-channel in-ear model achieved 69% accuracy versus 70% for the cap-EEG. These findings indicate that ear-EEG signals, particularly when utilizing multi-channel recordings, can reliably distinguish between varying levels of mental load. The study concludes that ear-EEG is a robust alternative to conventional cap-EEG for monitoring cognitive states such as working memory and workload. The comparable performance of ear-EEG, especially in multi-channel configurations, supports its integration into wearable devices for real-time mental state monitoring. This advancement facilitates the development of discreet BCIs capable of detecting focus levels in daily life, potentially enhancing workplace safety and productivity by mitigating interruptions during high-concentration tasks.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-nvidia on 2026-08-10; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• neuro workload indices

Information type

What kind of knowledge this paper contributes, grouped by family — independent of topic (what it is about) and method (how it was studied).

• Empirical Findings: physiological data