Empirical Inference Conference Paper 2011

Fronto-Parietal Gamma-Oscillations are a Cause of Performance Variation in Brain-Computer Interfacing

In recent work, we have provided evidence that fronto-parietal γ-oscillations of the electromagnetic field of the brain modulate the sensorimotor-rhythm. It is unclear, however, what impact this effect may have on explaining and addressing within-subject performance variations of brain-computer interfaces (BCIs). In this paper, we provide evidence that on a group-average classification accuracies in a two-class motor-imagery paradigm differ by up to 22.2% depending on the state of fronto-parietal γ-power. As such, this effect may have a large impact on the design of future BCI-systems. We further investigate whether adapting classification procedures to the current state of γ-power improves classification accuracy, and discuss other approaches to exploiting this effect.

Author(s): Grosse-Wentrup, M.
Pages: 384-387
Year: 2011
Month: May
Day: 0
Publisher: IEEE
Bibtex Type: Conference Paper (inproceedings)
Address: Piscataway, NJ, USA
DOI: 10.1109/NER.2011.5910567
Event Name: 5th International IEEE/EMBS Conference on Neural Engineering (NER 2011)
Event Place: Cancun, Mexico
Digital: 0
Electronic Archiving: grant_archive
ISBN: 978-1-4244-4140-2
Links:

BibTex

@inproceedings{GrosseWentrup2011,
  title = {Fronto-Parietal Gamma-Oscillations are a Cause of Performance Variation in Brain-Computer Interfacing},
  abstract = {In recent work, we have provided evidence that fronto-parietal γ-oscillations of the electromagnetic field of the brain modulate the sensorimotor-rhythm. It is unclear, however, what impact this effect may have on explaining and addressing within-subject performance variations of brain-computer interfaces (BCIs). In this paper, we provide evidence that on a group-average classification accuracies in a two-class motor-imagery paradigm differ by up to 22.2% depending on the state of fronto-parietal γ-power. As such, this effect may have a large impact on the design of future BCI-systems. We further investigate whether adapting classification procedures to the current state of γ-power improves classification accuracy, and discuss other approaches to exploiting this effect.},
  pages = {384-387},
  publisher = {IEEE},
  address = {Piscataway, NJ, USA},
  month = may,
  year = {2011},
  slug = {grossewentrup2011},
  author = {Grosse-Wentrup, M.},
  month_numeric = {5}
}