The use of virtual reality technologies is becoming common in training, particularly where visual (sensorimotor) skills are concerned. These techniques not only allow trainees to acquire skills and gain competence through repetitive practice, but, in many cases, also provide a better training environment than is available in a real scenario.

Discussed in this paper is a novel technique for learning to execute motions with increasing levels of difficulty, which are in turn represented by increasing the degrees of freedom. The just follow me method used in this paper presents a first-person viewpoint to the trainee, as opposed to the third person view provided by many other systems. By superimposing the ghost master on the trainees virtual existence, and adding in the use of a head mounted display, head tracking devices, and hand tracking devices, trainees are provided with a display comparing their own movements with those of the expert.

The authors undertook a comprehensive study to determine the effect of their virtual reality technique, with respect to traditional indirect methods. The study presented in this paper provides a convincing result: the first-person expert-superimposed view of actions provides as good, and, in some cases, better results, in terms of retained skills. The authors interpret some of the negative feedback they received as the fault of the hardware, and see the need for further experiments using lightweight and more robust hardware to verify their results.

This study represents a step forward in research on the effective use of virtual reality in practical training environments.