It has been proposed that mirror drawing is
diYcult because of the conXict between visual and proprioceptive
signals from the arm. However, even without
proprioception, there should be diYculties in
planning movements to visual targets observed in a
mirror, as the mirror-reversed spatial information must
be translated into appropriate hand actions. Mirror
drawing tasks suggest these planning conXicts are likely
to be most obvious at corners, when encountering
sharp changes in direction. We have therefore tested
the speed of mirror drawing in a chronically deaVerented
man and in a control group of normal subjects,
and hypothesized that increases in template complexity
(number of corners) would result in reduced drawing
speeds in all subjects. Indeed, all subjects, including the
deaVerented man, showed movement durations that
increased linearly as the complexity of the drawings
increased. However, the deaVerented man was signiWcantly
faster than the control subjects at tracing curved
templates. We suggest that the major diYculty in mirror
tracking is in the visuo-motor planning of actions
based on mirror-reversed visual information, and is not
a conXict between visual and proprioceptive signals
about arm motion.