Brain–Computer Interfaces - Index of

Dynamics of Sensorimotor Oscillations
in a Motor Task
Gert Pfurtscheller and Christa Neuper
1 Introduction
Many BCI systems rely on imagined movement. The brain activity associated with
real or imagined movement produces reliable changes in the EEG. Therefore, many
people can use BCI systems by imagining movements to convey information. The
EEG has many regular rhythms. The most famous are the occipital alpha rhythm
and the central mu and beta rhythms. People can desynchronize the alpha rhythm
(that is, produce weaker alpha activity) by being alert, and can increase alpha activity
by closing their eyes and relaxing. Sensory processing or motor behavior leads
to EEG desynchronization or blocking of central beta and mu rhythms, as originally
reported by Berger [1], Jasper and Andrew [2] and Jasper and Penfield [3]. This
desynchronization reflects a decrease of oscillatory activity related to an internally
or externally-paced event and is known as Event–Related Desynchronization (ERD,
[4]). The opposite, namely the increase of rhythmic activity, was termed Event-
Related Synchronization (ERS, [5]). ERD and ERS are characterized by fairly
localized topography and frequency specificity [6]. Both phenomena can be studied
through topographic maps, time courses, and time-frequency representations (ERD
maps, [7]).
Sensorimotor areas have their own intrinsic rhythms, such as central beta, mu,
and gamma oscillations. The dynamics of these rhythms depend on the activation
and deactivation of underlying cortical structures. The existence of at least three
different types of oscillations at the same electrode location over the sensorimotor
hand area during brisk finger lifting has been described by Pfurtscheller and Lopes
da Silva [8, 9–11]. In addition to mu ERD and post-movement beta ERS, induced
gamma oscillations (ERS) close to 40 Hz can also be recorded. These 40–Hz oscillations
are strongest shortly before a movement begins (called movement onset),
whereas the beta ERS is strongest after a movement ends (called movement offset).
G. Pfurtscheller (B)
Institute for Knowledge Discovery, Graz University of Technology, Graz, Austria
e-mail: pfurtscheller@tugraz.at
B. Graimann et al. (eds.), Brain–Computer Interfaces, The Frontiers Collection,
DOI 10.1007/978-3-642-02091-9_3, C○ Springer-Verlag Berlin Heidelberg 2010
47