Virtual training of the myosignal

Myocontrol is the control of an external device through EMG signals produced by
the muscles (the myosignals) and is used in myoelectric arm and hand
prostheses. Modern prostheses require a high level of myocontrol. Hence,
knowledge about characteristics of training that improve myocontrol might be
helpful for amputees to increase their control of these prostheses.
Conventional training involves actual prostheses, but virtual training methods
could provide the same results, as improvements in myocontrol have been shown
using virtual training methods (e.g. Radhakrishnan et al. 2008, Perez-Maldona
et al. 2010, Bouwsema et al. 2010). Furthermore, virtual training methods have
multiple benefits over training myocontrol with an actual prosthesis, as they
are easy to apply and use, low in cost and allow for training and testing of an
amputees* myocontrol ability before acquiring an actual prosthesis. Although
virtual training methods have been shown to produce positive effects on
myocontrol, it is currently unknown which type of virtual training method has
the best result. Therefore this study examines different virtual training
methods on their effects on myocontrol.

The main objective of the study is to investigate which type of virtual
training has the highest impact on learning myocontrol.
Secondary objectives are to investigate whether the progress in a participants*
myocontrol can be predicted by his or her scores on three manual dexterity
tasks, and to examine whether the dominant and the non-dominant hand differ in
the type of training that has the best result.

pre-test/post-test intervention

One group will train myocontrol with their produced myosignal as instantaneous
feedback. The second will train using a virtual myoelectric hand and the third
group will train using a videogame. All groups will be trained on three
consecutive days.

The myosignal will be measured with the same electrodes as are used in standard
myoelectric prostheses. Hence, essentially the tasks in the experiment are
comparable to handling a prosthesis. With regard to the load of the tasks,
participants train myocontrol for three consecutive days and tests will be
performed on day one and three. Moreover, at the start of day one and at the
end of day three, three non-invasive manual dexterity tests will be
administered. Each day the session will take less than one hour. There are no
risks associated with participation.