Information Technology Reference
In-Depth Information
cant work in using brainwaves for music has been developed with other
forms of measurement of brain activity. For instance, fMRI (functional Magnetic
Resonance Imaging) has been used to translate brain data as input to of
Signi
ine musical
compositions, one example of which is discussed in Chap.
12
in this volume.
However, fMRI is currently impractical for developing a BCMI: it is expensive, not
portable and has poorer time resolution than EEG, to cite but three encumbering
factors.
10.3.5 Methods of Music Generation with Brainwaves
When looking back on research into music and brainwaves, we can separate sys-
tems into three categories: ones for
EEG soni
cation
, ones for
EEG musi
cation
and ones for
BCI control
. EEG soni
cation is the translation of EEG information
into sound, for non-musical and predominantly medical purposes. EEG musi
ca-
tion is the mapping of EEG information to musical parameters; however, the EEG
data are arbitrary and when possible can offer only loose forms of control. BCI
control is inherent in systems where direct cognitive real-time control of music is
achievable. In some systems, more than one of these approaches can be found, and
in others where one approach has been adopted for investigation of the technique,
the application could well be applied to another approach as a result.
It should also be noted that the mapping approaches discussed in this chapter are
not wholly comparative, as it charts development in a relatively infantile
eld,
where, as previously mentioned, progress is heavily reliant on the advances within
neuroscience. Where considered useful, areas are touched upon that draw parallels
between systems as a way of directing the reader through the different approaches
and ideas.
Although this chapter does not attempt to explicitly categorise the accuracy of
each system, due to the wide range of disparaging technologies and individuals
incorporated, it should be carefully acknowledged that accuracy plays a very
important part in the derivation of
meaning
within EEG data, and this is considered
of high importance.
The soni
cation of data offers an interesting way of to listening to the sounds of
non-musical sources of information. Data harvesting allows us to sonify a world of
unlikely information, such as the stock market or even the weather. In soni
cation,
we are concerned with the
sound
of the information relative to itself, and it is a
passive process and a way of hearing numerical or graphical data.
Sound has long been used as a way of interpreting biological information, from
the use of the stethoscope to the steady beeping of the heart rate monitor. Both of
these are methods of
hearing
the body, which when used in real time to help affect
control over the signal is known as biofeedback. The visual complexities of EEG
have given reason to sonifying its information as a method for understanding
activity through the simpli
cation and the natural intuition of discernably listening
to multiple elements contained within sounds. As such, the mappings for direct data
