Information Technology Reference
In-Depth Information
soni
cation should be straightforward in order to provide an intuitive correlation
between brain activity and sound. Control of EEG in soni
cation (and some
musi
cation) systems is largely passive, whereby the user has no direct control over
their EEG. EEG may be in
uenced external factors, such as tiredness or mood, but
in situations where brainwave control is not achieved by explicit choice.
In contrast to soni
cation, to
musify
data is to map the data into organised
musical form. This is rather different from soni
cation as one is not attempting to
understand the data through soni
cation per se, but rather attaching it to a musical
system. Therefore, musical structures are connected to the EEG information based
on the patterns or variables apparent within the data. For example, if the EEG
delivers
ve
parameters within a pre-designed musical piece. A common factor within EEG
musi
five distinguishable data,
then these can be directly mapped to
cation, BCMI
systems a passive approach to EEG control are generally used. EEG data are
generally limited in its meaning, and the shift in focus lies heavily on mappings
using advanced techniques of interpreting data in useful ways to grant musical
success. In summary, the difference between soni
cation is the use of generative musical approaches. In musi
cation and musi
cation are as
follows: (a) soni
cation produces sounds from EEG data, and the system would
normally control a sound synthesiser; (b) soni
cation is not, in principle, intended
for an artistic purpose, but rather as some sort of scienti
c auditory display of the
EEG behaviour.
Both soni
cation afford no explicit control of the sound of
music, and as such, strictly speaking, they could be regarded outside of the realms
of BCI research. This is because BCI research is based on the premise that a BCI
system allows for the
active
control of a device and/or software by the explicit
thought of the command, and the results of the mental activity are fed back to the
user in real time (Wolpaw and Birbaumer
2006
). This de
cation and musi
nition of BCI has been
harnessed within BCMI to the extent that subjective control over systems is now a
realisation. Here is where the challenge of being unable to translate musical thought
into direct action has been bypassed through embedding meaning into cognitive
processes. For example, where reading the explicit thought of
,is
not feasible, using learnt cognitive processes where a user understands the out-
comes may lead to a dedicated brainwave response that can be mapped to play the
note D#.
'
play the note D#
'
10.4 Observations on Musifying EEG
Musifying brainwave activity without a need for control can offer interesting
possibilities with regard to mapping data to music. Although musi
cation is not
really BCI, it is nevertheless a valid approach for BCMI for artistic purposes. For
instance, Miranda and Soucaret (
2008
) reported on a mapping method they
developed to produce melodies from the
'
topological
'
behaviour of the EEG across
a con
guration of electrodes on the scalp or
montage
. In this case, the EEG signal
