Abstract

In comparison to all types of injury, those to the brain are among the most likely to
result in death or permanent disability. Some of these brain-injured people cannot
communicate, recreate, or control their environment due to severe motor impairment.
This group of individuals with severe head injury have received limited help from
assistive technology. Brain-Computer Interfaces have opened up a spectrum of assistive
technologies, which are particularly appropriate for people with traumatic brain injury,
especially those who suffer from “locked-in” syndrome. The research challenge here is
to develop novel interaction paradigms that suit brain-injured individuals, who could
then use it for everyday communications. The developed interaction paradigms should
require minimum training, reconfigurable and minimum effort to use.
This thesis reports on the development of novel interaction paradigms for Brain-Body
Interfaces to help brain-injured people to communicate better, recreate and control their
environment using computers despite the severity of their brain injury. The
investigation was carried out in three phases. Phase one was an exploratory study where
a first novel interaction paradigm was developed and evaluated with able-bodied and
disabled participants. Results obtained were fed into the next phase of the investigation.
Phase two was carried out with able participants who acted as development group for
the second novel interaction paradigm. This second novel interaction paradigm was
evaluated with non-verbal participants with severe brain injury in phase three. An
iterative design research methodology was chosen to develop the interaction paradigms.
A non-invasive assistive technology device named Cyberlink™ was chosen as the
Brain-Body Interface. This research improved previous work in this area by developing
new interaction paradigms of personalised tiling and discrete acceleration in Brain-
Body Interfaces. The research hypothesis of this study ‘that the performance of the
Brain-Body Interface can be improved by the use of novel interaction paradigms’ was
successfully demonstrated.