212
G. D. Johnson and D. J. Krusienski
prove distracting and detrimental to usability and performance. If many stimulus
frequencies are used, sufficient computational resources are needed to maximize the
correlation between the sinusoidal templates and the EEG to avoid an appreciable
lag in the feedback.
9.4.3 User Versus System Adaptation
For generally stable responses such as P300 and SSVEP, BCI designs with static
processing and classification are typically adequate. For some paradigms such as
motor imagery, the user can learn to better focus and modulate brain responses via
training [33]. Additionally, the brain state and background brain activity of the user
can change over time, even within a session. Thus, static designs where the user is
forced to adapt to the BCI feedback can be suboptimal. Alternately, the BCI can
be designed to adapt its processing and classification to the user’s changes in brain
activity and/or performance. The challenge is that this system adaptation must be done
using periodic calibration sessions, or without calibration sessions in an unsupervised
or semi-supervised manner. Furthermore, it is imperative to select an appropriate rate
of adaptation, which can be user-dependent [23]. Ultimately, the implementation of
an adaptive BCI results in a co-adaptive system since the user will inevitably adapt
to the provided feedback. Co-adaptive systems can be highly prone to instability and
it is vital to carefully design and select the adaptation parameters.
References
1. S. Ahn, K. Kim, S.C. Jun, Steady-state somatosensory evoked potential for brain-computer
interface-present and future. Front. Hum. Neurosci 16, 832 (2015)
2. F. Aloise, I. Lasorsa, F. Schettini, A. Brouwer, D. Mattila, F. Babiloni, F. Cincotti, Multimodal
stimulation for a P300-based BCI. Int. J. Bioelectromagn. 9, 128–130 (2007)
3. C.W. Anderson, E.A. Stolz, S. Shamsunder, Multivariate autoregressive models for classification of spontaneous electroencephalographic signals during mental tasks. IEEE Trans. Biomed.
Eng. 45, 277–286 (1998)
4. A. Bashashati, M. Fatourechi, R.K. Ward, G.E. Birch, A survey of signal processing algorithms
in brain–computer interfaces based on electrical brain signals. J. Neural Eng. 4, R32–R57 (2007)
5. G. Bin, X. Gao, Y. Wang, Y. Li, B. Hong, S. Gao, A high-speed BCI based on code modulation
VEP. J. Neural Eng. 8, 025015 (2011)
6. G. Bin, X. Gao, Z. Yan, B. Hong, S. Gao, An online multi-channel SSVEP-based brain—
computer interface using a canonical correlation analysis method. J. Neural Eng. 6, 046002
(2009)
7. X. Chen, Y. Wang, M. Nakanishi, X. Gao, T.P. Jung, S. Gao, High-speed spelling with a
noninvasive brain–computer interface. Proc. Natl. Acad. Sci. USA 112, E6058–E6067 (2015)
8. J. Dien, K.M. Spencer, E. Donchin, Localization of the event-related potential novelty response
as defined by principal components analysis. Cognitive Brain Res. 17, 637–650 (2003)
9. G. Dornhege, B. Blankertz, M. Krauledat, F. Losch, G. Curio, K.R. Muller, Combined optimization of spatial and temporal filters for improving brain-computer interfacing. IEEE Trans.
Biomed. Eng. 53, 2274–2281 (2006)
G. D. Johnson and D. J. Krusienski
prove distracting and detrimental to usability and performance. If many stimulus
frequencies are used, sufficient computational resources are needed to maximize the
correlation between the sinusoidal templates and the EEG to avoid an appreciable
lag in the feedback.
9.4.3 User Versus System Adaptation
For generally stable responses such as P300 and SSVEP, BCI designs with static
processing and classification are typically adequate. For some paradigms such as
motor imagery, the user can learn to better focus and modulate brain responses via
training [33]. Additionally, the brain state and background brain activity of the user
can change over time, even within a session. Thus, static designs where the user is
forced to adapt to the BCI feedback can be suboptimal. Alternately, the BCI can
be designed to adapt its processing and classification to the user’s changes in brain
activity and/or performance. The challenge is that this system adaptation must be done
using periodic calibration sessions, or without calibration sessions in an unsupervised
or semi-supervised manner. Furthermore, it is imperative to select an appropriate rate
of adaptation, which can be user-dependent [23]. Ultimately, the implementation of
an adaptive BCI results in a co-adaptive system since the user will inevitably adapt
to the provided feedback. Co-adaptive systems can be highly prone to instability and
it is vital to carefully design and select the adaptation parameters.
References
1. S. Ahn, K. Kim, S.C. Jun, Steady-state somatosensory evoked potential for brain-computer
interface-present and future. Front. Hum. Neurosci 16, 832 (2015)
2. F. Aloise, I. Lasorsa, F. Schettini, A. Brouwer, D. Mattila, F. Babiloni, F. Cincotti, Multimodal
stimulation for a P300-based BCI. Int. J. Bioelectromagn. 9, 128–130 (2007)
3. C.W. Anderson, E.A. Stolz, S. Shamsunder, Multivariate autoregressive models for classification of spontaneous electroencephalographic signals during mental tasks. IEEE Trans. Biomed.
Eng. 45, 277–286 (1998)
4. A. Bashashati, M. Fatourechi, R.K. Ward, G.E. Birch, A survey of signal processing algorithms
in brain–computer interfaces based on electrical brain signals. J. Neural Eng. 4, R32–R57 (2007)
5. G. Bin, X. Gao, Y. Wang, Y. Li, B. Hong, S. Gao, A high-speed BCI based on code modulation
VEP. J. Neural Eng. 8, 025015 (2011)
6. G. Bin, X. Gao, Z. Yan, B. Hong, S. Gao, An online multi-channel SSVEP-based brain—
computer interface using a canonical correlation analysis method. J. Neural Eng. 6, 046002
(2009)
7. X. Chen, Y. Wang, M. Nakanishi, X. Gao, T.P. Jung, S. Gao, High-speed spelling with a
noninvasive brain–computer interface. Proc. Natl. Acad. Sci. USA 112, E6058–E6067 (2015)
8. J. Dien, K.M. Spencer, E. Donchin, Localization of the event-related potential novelty response
as defined by principal components analysis. Cognitive Brain Res. 17, 637–650 (2003)
9. G. Dornhege, B. Blankertz, M. Krauledat, F. Losch, G. Curio, K.R. Muller, Combined optimization of spatial and temporal filters for improving brain-computer interfacing. IEEE Trans.
Biomed. Eng. 53, 2274–2281 (2006)