CNS*2020 Online has ended
Welcome to the Sched instance for CNS*2020 Online! Please read the instruction document on detailed information on CNS*2020.
Back To Schedule
Sunday, July 19 • 9:00pm - 10:00pm
P89: Quantification of changes in motor imagery skill during brain-computer interface use

Log in to save this to your schedule, view media, leave feedback and see who's attending!

Feedback form is now closed.
Google Meet Link: meet.google.com/mqh-qfnn-aqq 

James Bennett
, David Grayden, Anthony Burkitt, Sam John

Oscillatory activity over the sensorimotor cortex, known as sensorimotor rhythms, can be modulated by the kinaesthetic imagination of limb movement [1]. These event-related spectral perturbations can be observed in electroencephalography (EEG), offering a potential way to restore communication and control to people with severe neuromuscular conditions via a brain-computer interface (BCI). However, the ability of individuals to produce these modulations varies greatly across the population. Between 10-30% of people are unable to influence their SMRs sufficiently to be distinguishable by a BCI decoder [2]. Despite this, it has been shown that users can be trained to improve the extent of their SMR modulations. This research utilised a data-driven approach to characterise the skill development of participants undertaking a left- and right-hand motor imagery experiment.

Two publicly available motor imagery EEG datasets were analysed. Dataset 1 consisted of EEG data from 47 participants performing 200 trials of left- and right-hand motor imagery within a single session [3]. No real-time visual feedback was provided to the participants. Dataset 2 contained EEG from two sessions of 200 trials each from 54 participants [4]. Visual feedback was provided to users in the second session but not in the first. Various metrics characterising mental imagery skill were calculated across time for each participant.

The discriminability of EEG in the 8-30Hz range from left- and right-hand trials was found to increase across time for both datasets. Despite the overall improvement, there was great variability in the change of motor imagery skill across participants. For Dataset 1, the average change across time of the metric representing the discriminability of classes was 6.0±21.9%. For Sessions 1 and 2 of Dataset 2, the discriminability increased by 11.8±44.0% and 17.4±30.7%, respectively. Session 2 of Dataset 2 contained visual feedback and produced a larger overall improvement in motor imagery skill with a lower variability compared with Session 1.

In this work, we investigated the level of motor imagery skill acquisition during BCI use. The results indicate a baseline level of skill improvement that can be expected, and also emphasise the large variability across participants commonly seen in BCI studies. Overall, we provide a useful reference of BCI skill acquisition for future research that seeks to increase the rate of skill improvement and decrease the amount of variability.

1. Pfurtscheller, G., Da Silva, F.L., 1999. Event-related EEG/MEG synchronization and desynchronization: basic principles. Clinical Neurophysiology, 110(11), p. 1842-1857.
2. Allison, B.Z., Neuper, C., 2010. Could anyone use a BCI? Brain-Computer Interfaces (p. 35-54). Springer, London.
3. Cho, H., Ahn, M., Ahn, S., Kwon, M., Jun, S.C., 2017. EEG datasets for motor imagery brain–computer interface. GigaScience, 6(7), p. gix034.
4. Lee, M.H., Kwon, O.Y., Kim, Y.J., Kim, H.K., Lee, Y.E., Williamson, J., Fazli, S., Lee, S.W., 2019. EEG dataset and OpenBMI toolbox for three BCI paradigms: an investigation into BCI illiteracy. GigaScience, 8(5), p. giz002.

avatar for James Bennett

James Bennett

PhD Candidate, Biomedical Engineering, University of Melbourne

Sunday July 19, 2020 9:00pm - 10:00pm CEST
Slot 02