Assessing the feasibility of online SSVEP decoding in human walking using a consumer EEG headset
Yuan-Pin Lin, Yijun Wang, Tzyy-Ping Jung1
1Swartz Center for Computational Neuroscience, Institute for Neural Computation, University of California, San Diego, La Jolla, CA, USA. jung@sccn.ucsd.edu.
Background:
Bridging the gap between laboratory brain-computer interface (BCI) demonstrations and real-life applications has gained increasing attention nowadays in translational neuroscience. An urgent need is to explore the feasibility of using a low-cost, ease-of-use electroencephalogram (EEG) headset for monitoring individuals' EEG signals in their natural head/body positions and movements. This study aimed to assess the feasibility of using a consumer-level EEG headset to realize an online steady-state visual-evoked potential (SSVEP)-based BCI during human walking.
Methods:
This study adopted a 14-channel Emotiv EEG headset to implement a four-target online SSVEP decoding system, and included treadmill walking at the speeds of 0.45, 0.89, and 1.34 meters per second (m/s) to initiate the walking locomotion. Seventeen participants were instructed to perform the online BCI tasks while standing or walking on the treadmill. To maintain a constant viewing distance to the visual targets, participants held the hand-grip of the treadmill during the experiment. Along with online BCI performance, the concurrent SSVEP signals were recorded for offline assessment.
Results:
Despite walking-related attenuation of SSVEPs, the online BCI obtained an information transfer rate (ITR) over 12 bits/min during slow walking (below 0.89 m/s).
Conclusions:
SSVEP-based BCI systems are deployable to users in treadmill walking that mimics natural walking rather than in highly-controlled laboratory settings. This study considerably promotes the use of a consumer-level EEG headset towards the real-life BCI applications.
