Functional Dissociation of Ongoing Oscillatory Brain States Revealed by a Custom-Developed Brain Computer Interface

The state of a neural assembly in the human brain preceding an incoming stimulus is assumed to modulate the processing of subsequently presented stimuli. The nature of this state can differ with respect to the frequency of ongoing oscillatory activity. Oscillatory activity of specific frequency range such as alpha (8-12 Hz) and gamma (30-45 Hz) band oscillations is hypothesized to play a functional role in visual object processing. However, the precise role of prestimulus alpha or gamma band oscillations for visual object processing is not completely understood. Therefore, a selective modulation of this prestimulus activity could clarify the functional role of these oscillations. We hypothesized that an increase in gamma band activity as compared to an increase in alpha band activity over the visual cortex by BCI manipulation would enhance subsequent visual object processing. In contrast to previous studies in which oscillations of prestimulus activity were correlated with visual performance, we attempted to put the volunteers directly in control of the oscillatory brain activity. To this end, we designed and implemented a non-invasive brain computer interface (BCI) method to train volunteers to selectively increase their alpha or gamma band activity in the occipital cortex. During training, oscillatory brain activity was estimated online and fed back to the volunteers to enable a deliberate modulation of alpha or gamma band oscillations. The visual stimuli were presented during specific brain states in an individually adapted manner. During the testing phase which followed the training phase, alpha or gamma band activity was classified online and at predefined levels of activity, visual objects embedded in noise were presented in order to assess the influence of frequency modulation on subsequent visual object processing. In the process of developing a BCI method based on gamma band oscillations, several important aspects had to be considered, including presence of artifacts, experimental design and topographical precision of BCI training. We therefore perfected our BCI method with online artifact control for artifact suppression, a special visual display design to avoid distraction yet motivate volunteers, and a source-based BCI method to limit training to a distinct neural area in the visual cortex. In a series of experiments, we first evaluated the accuracy of the BCI method and then explored the specific effect of gamma band training on visual object perception. Finally, we compared the specific effect of gamma band training to the well defined alpha band. Our results demonstrated that volunteers learned to selectively modulate alpha or