Biomedical Engineering Reference
In-Depth Information
Figure 12.2 Feed back-error learning model (Kawato and Wolpert, 1998).
process, which is supposed to be activated by the formation of some kind of
internal maps, cannot be observed directly, but can only be inferred from the
observation of subject's performance. Such internal models seem to be created
and reinforced through the repetitive execution of the task for understanding
the relationships between the action outcome and the motor control parameters
required ( Figure 12.2 ) . In fact, such relationships cannot be necessarily found in
the same way for all humans, which mean that each person learns in a different
manner and different learning experiences to draw from.
Even if the study of motor learning which analyzes behavioral, neurophysio-
logical, and imaging data has demonstrated interesting results, the unobservability
of the brain's internal processes makes it difficult to assess the validity of the
theories. Therefore, the existence of a widely accepted theory describing all the
learning processes is still unproven. Notwithstanding this, several approaches
can be formulated in order to strengthen theories related to specific aspects of
motor learning. Hence, it becomes necessary to find additional tools that may help
us to understand these specific aspects of the learning process under repetitive
controlled circumstances.
From this purpose, RT can be seen as powerful tools not only for under-
standing how the skill transfer occurs but also to enhance learners' performances.
Such systems offer the possibility of controlling accurately the conditions of the
learning process (Solis, 2004; Solis et al ., 2007) and moreover, meanwhile subjects
are practicing the skill to be learned, their performances can be measured and
analyzed by means of different performance indexes (Solis et al ., 2007; Solis et al .
2008).
It is worth to mention that in order to conduct effective learning experi-
ments; the robotic system must actively interact with humans at the same logi-
cal/perceptual level through all the stages of the learning process. This means that
robots should analyze the exchanged information between human and robot to
evaluate learner's performance and furthermore, to provide some kind of feedback
 
 
Search WWH ::




Custom Search