Simulation-based training is becoming widely established for the purpose
of acquiring and training procedural skill. However, in most cases these
simulators offer technological sophistication at the expense of theory-based
design and the question what features of the task are important for the
acquisition of procedural skill in psychomotor tasks is remained unanswered.
Motor memory and motor schema theories suggest that these features should
involve full representation of the motor parts of the task. Yet, based on
several theories we argue that focusing on the spatial characteristics or the abstract
representation of the necessary actions should contribute to procedural skill
acquisition. We addressed this issue by comparing two alternative training
methods for procedural skill acquisition in a Lego assembly task, one focusing
on the spatial features of the task (the allocation of the bricks) while the
other focused on the abstract features (the verbal explanation of what should
be done). These two training methods were compared to two benchmark groups, a
Motor training group (actual performance of the task) and a control group who
experienced no training at any kind. The results revealed that the Motor
training group was superior to both Spatial and Abstract groups, in most
aspects, supporting past research on the impotence of motor training and motor
memory to performance. Yet, applying any of the two training methods would be
better to no training at all. Additionally, a comparison of the Spatial and the
Abstract groups revealed that the Spatial group spent more time on errors, but
less time on each correct construction stage; suggesting that the two training
methods have complementary advantages. These results are consistent with the
item-order hypothesis.
