Aptly named: consists of slow shifts in the EEG
(often 5-10 seconds) over cortex.
The SCP refers to a variety of EEG components:
CNV (contingent negative variation)
PINV (post imperative negative variation)
SPN (stimulus preceding negativity)
RP (readiness potential)
Patients can produce positive or negative shifts in
the EEG with extensive training (operant
The SCP can be modulated over different areas of
the brain. At first, users report that emotional or
mental imagery is important for changing the SCP.
It is often largest over central electrode sites.
Subjects generally have a 2 second “baseline” phase
followed by a 2 second “active” phase. The SCP
during the active phase may be more negative, more
positive, or the same as seen during the baseline
Before using the BCI to communicate, subjects
typically train until they can complete a three-event
(three cogneme) sequence with 70% accuracy.
Training often requires more than 100 training
sessions taking 3-5 months.
Some versions of the TTD sought to attain
multidimensional control (e. g., Kubler et al. 1999)
mapped positive or negative shifts over Cz onto
vertical cursor movement, while horizontal
movement was based on SCP asymmetry between
C3 and C4.
Like mu BCIs, multidimensional control was very
difficult, and later versions did not implement it.
The TTD has been extensively tested with disabled
users. It generally works - it was initially reported
that three patients who were locked-in at the onset
of training could not learn the TTD. Since then,
most locked in patients have learned SCP control.
Its authors claim that the SCP is more universal
than mu; it can be modulated by nearly everyone.
Further research is needed to determine which
individuals can use different systems.
A famous 1999 Nature paper from the Birbaumer
group showed a letter written by a locked in patient
using the TTD.
Accuracy for subject A Accuracy for subject B The letter written by subject A
The 4 direction ball
The binary language
support program (LSP)
The three-layered menu
for disabled users
(shown here, from
Kaiser et al. (2001)).
TMS and SCPs : Kubler et al (2002) showed that
TMS (transcranial magentic stimulation) can
influence SCP activity. Further research is needed to
explore how TMS could best affect training and/or
In a recent study, subjects trained in the TTD could
produce different activity for positive vs. negative
SCP shifts that were visible in an fMR.
While fMRs are not viable
imaging technologies for
BCIs, this work helps us
understand the neural
SCP activity. Perhaps this
will be useful someday
with improved imaging
The Birbaumer and Wolpaw labs are currently
collaborating on a “hybrid” BCI using both mu and
SCP. This is implemented in BCI2000, with
auditory, visual, and tactile feedback.
Birbaumer et al. are also exploring other
applications of the SCP, such as studying autism
and global aphasia.