Item

Abstract

In order to pick up an object, its visual location must be converted into the appropriate motor commands. Introducing a discrepancy between the seen and felt locations
of the object (e.g., via prism goggles) initially impairs the ability to touch it. The
sensory system rapidly adapts to the discrepancy, however, returning perception and
performance to near normal. Subsequent removal of the discrepancy leads to a renewed
performance decrement - a negative aftereect (NAE). It is generally believed that the
process of adaptation consists primarily of \recalibrating" the transformation between
the visual and proprioceptive perception of spatial location (Bedford, The psychology
of learning and motivation, 1999). According to such a purely perceptual account of
adaptation, the movement to reach the object is not important. If, however, the transformation
from perception to action is altered, then it will be dependent on motion -
i.e. changing motion parameters will reduce or eliminate the NAE (see also Martin et
al., Brain, 1996). According to our hypothesis spatial visuomotor information is distributively
stored and changed by prism adaptation and it is not based on a centrally
organized spatial information system. We conducted seven experiments consisting of
four blocks each, in which participants had to touch a cross presented at eye level on
a touch screen. In the rst block the participants were introduced and familiarized
with the experiment. Blocks two and four were pre and post tests to measure the
NAE produced during the dierent experimental conditions in block 3 in which the
participants were wearing prism goggles: we tested the eects of dierent trajectories,
dierent starting points, weight, vertical generalization and dierent types of feedback.
A total transfer from an adapted to a non-adapted condition didn't turn up in any of
our experiments, although the trajectories where highly identical in some of them. It
rather seems that newly learned spatial information in prism adaptation experiments
is stored and retrieved distributively for dierent extremities, for dierent trajectories
and for dierent stress/strain conditions (e.g. weight). Furthermore, transfer seems to
become weaker with bigger dierences in location. Therefore we conclude that no visual
\recalibration" is taking place but a relearning of distributetively organized parameters
of visuomotor coordination.