Dry EEG in Sports Sciences: A Fast and Reliable Tool to Assess Individual Alpha Peak Frequency Changes Induced by Physical Effort

Novel state-of-the-art amplifier and cap systems enable Electroencephalography (EEG) recording outside of stationary lab systems during physical exercise and body motion. However, extensive preparation time, cleaning and limited long-term stability of conventional gel-based electrode systems pose significant limitations in out-of-the-lab conditions. Dry electrode systems may contribute to rapid and repetitive mobile EEG acquisition with significantly reduced preparation time, reduced cleaning requirements and possible self-application by the volunteer but are known for higher channel failure probability and increased sensitivity to movement artefacts. We performed a counterbalanced repeated measure endurance cycling study to objectively validate the performance and applicability of a novel commercially available 64-channel dry electrode cap for sport science. A total of 17 healthy volunteers participated in the study, performing an endurance cycling paradigm comprising 5 phases: (I) baseline EEG, (II) pre-cycling EEG, (III) endurance cycling, (IV) active recovery, and (V) passive recovery. We compared the performance of the 64-channel dry electrode cap with a commercial gel-based cap system in terms of usability metrics, reliability and signal characteristics. Furthermore, we validated the performance of the dry cap during a realistic sport science investigation, verifying the hypothesis of a systematic, reproducible shift of the individual alpha peak (iAPF) induced by physical effort. The average preparation time of the dry cap was one third of the gel-based electrode caps. The average channel reliability of the dry cap varied between 80±15% (Phase I), 66±19% (Phase III), and 91±10% (Phase V). In comparison, the channel reliability of the gel-based cap varied between 95±3%, 85±9%, and 82±9%, respectively. No considerable differences were evident for the comfort evaluations nor the signal characteristics of both caps. A within-volunteers RM-ANOVA did not show significant effects of the electrode type on the iAPF (F(1,12)=1.670, p=0.221, ηp²=0.122, Power=0.222). However, a significant increase of the iAPF exists from Phase II to Phase IV and V due to exhaustive physical task. In conclusion, we demonstrated that dry electrode cap is equivalent to the gel-based electrode cap based on signal characteristics, comfort and signal information content thereby confirming the usefulness of dry electrodes in sports science and other mobile applications involving ample movement.