Limits...
A brain-computer-interface for the detection and modulation of gamma band activity.

Salari N, Rose M - Brain Sci (2013)

Bottom Line: The BCI incorporates modules for online detection of various artifacts (including microsaccades) and the artifacts were continuously fed back to the volunteer.The analyses revealed a high level of accuracy with respect to frequency and topography for the gamma-band modulations.Thus, the developed BCI can be used to manipulate the fast oscillatory activity with a high level of specificity.

View Article: PubMed Central - PubMed

Affiliation: NeuroImage Nord, Department of Systems Neuroscience, University Medical Center Hamburg Eppendorf, Martinistrasse 52, D-20246 Hamburg, Germany. n.salari@uke.de.

ABSTRACT
Gamma band oscillations in the human brain (around 40 Hz) play a functional role in information processing, and a real-time assessment of gamma band activity could be used to evaluate the functional relevance more directly. Therefore, we developed a source based Brain-Computer-Interface (BCI) with an online detection of gamma band activity in a selective brain region in the visual cortex. The BCI incorporates modules for online detection of various artifacts (including microsaccades) and the artifacts were continuously fed back to the volunteer. We examined the efficiency of the source-based BCI for Neurofeedback training of gamma- and alpha-band (8-12 Hz) oscillations and compared the specificity for the spatial and frequency domain. Our results demonstrated that volunteers learned to selectively switch between modulating alpha- or gamma-band oscillations and benefited from online artifact information. The analyses revealed a high level of accuracy with respect to frequency and topography for the gamma-band modulations. Thus, the developed BCI can be used to manipulate the fast oscillatory activity with a high level of specificity. These selective modulations can be used to assess the relevance of fast neural oscillations for information processing in a more direct way, i.e., by the adaptive presentation of stimuli within well-described brain states.

No MeSH data available.


Training success for each volunteer as percent change to day one for the power values separate for each frequency band. For the gamma band, four subjects showed a positive effect, two showed a slight negative effect and two volunteers showed no training effect. For the alpha band, two different volunteers were not successful in enhancing the power across days.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4061891&req=5

brainsci-03-01569-f004: Training success for each volunteer as percent change to day one for the power values separate for each frequency band. For the gamma band, four subjects showed a positive effect, two showed a slight negative effect and two volunteers showed no training effect. For the alpha band, two different volunteers were not successful in enhancing the power across days.

Mentions: Training success was analyzed by comparing the achieved gamma and alpha power values of the online estimation across training (ANOVA with repeated measures with factors frequency band and training). Results demonstrated that the volunteers learned to increase the power across days (training: F(1,7) = 6.5, p < 0.05), that the absolute power was larger in the alpha band (F(1,7) = 8.4, p < 0.05) and an identical learning across days (frequency band x training: F(2,14) = 0.5, n.s.). Individual data for each volunteer can be seen in Figure 4.


A brain-computer-interface for the detection and modulation of gamma band activity.

Salari N, Rose M - Brain Sci (2013)

Training success for each volunteer as percent change to day one for the power values separate for each frequency band. For the gamma band, four subjects showed a positive effect, two showed a slight negative effect and two volunteers showed no training effect. For the alpha band, two different volunteers were not successful in enhancing the power across days.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4061891&req=5

brainsci-03-01569-f004: Training success for each volunteer as percent change to day one for the power values separate for each frequency band. For the gamma band, four subjects showed a positive effect, two showed a slight negative effect and two volunteers showed no training effect. For the alpha band, two different volunteers were not successful in enhancing the power across days.
Mentions: Training success was analyzed by comparing the achieved gamma and alpha power values of the online estimation across training (ANOVA with repeated measures with factors frequency band and training). Results demonstrated that the volunteers learned to increase the power across days (training: F(1,7) = 6.5, p < 0.05), that the absolute power was larger in the alpha band (F(1,7) = 8.4, p < 0.05) and an identical learning across days (frequency band x training: F(2,14) = 0.5, n.s.). Individual data for each volunteer can be seen in Figure 4.

Bottom Line: The BCI incorporates modules for online detection of various artifacts (including microsaccades) and the artifacts were continuously fed back to the volunteer.The analyses revealed a high level of accuracy with respect to frequency and topography for the gamma-band modulations.Thus, the developed BCI can be used to manipulate the fast oscillatory activity with a high level of specificity.

View Article: PubMed Central - PubMed

Affiliation: NeuroImage Nord, Department of Systems Neuroscience, University Medical Center Hamburg Eppendorf, Martinistrasse 52, D-20246 Hamburg, Germany. n.salari@uke.de.

ABSTRACT
Gamma band oscillations in the human brain (around 40 Hz) play a functional role in information processing, and a real-time assessment of gamma band activity could be used to evaluate the functional relevance more directly. Therefore, we developed a source based Brain-Computer-Interface (BCI) with an online detection of gamma band activity in a selective brain region in the visual cortex. The BCI incorporates modules for online detection of various artifacts (including microsaccades) and the artifacts were continuously fed back to the volunteer. We examined the efficiency of the source-based BCI for Neurofeedback training of gamma- and alpha-band (8-12 Hz) oscillations and compared the specificity for the spatial and frequency domain. Our results demonstrated that volunteers learned to selectively switch between modulating alpha- or gamma-band oscillations and benefited from online artifact information. The analyses revealed a high level of accuracy with respect to frequency and topography for the gamma-band modulations. Thus, the developed BCI can be used to manipulate the fast oscillatory activity with a high level of specificity. These selective modulations can be used to assess the relevance of fast neural oscillations for information processing in a more direct way, i.e., by the adaptive presentation of stimuli within well-described brain states.

No MeSH data available.