---------------------------------------------

The advantage of simulation (or virtual reality) is that it can provide the proper setting (the real-world space) for performing the required skills, while reducing the risk of personal injury. As in traditional training methods, you can combine training with study (the former using the information provided by the latter). With simulation, practice sessions can occur in much the same context in which the skills being trained are normally performed. What's more, the training method can provide direct, detailed, and immediate feedback, something the real world can't offer -- it doesn’t stop to let you look at what just happened.) You can also repeat scenes to try to understand what occurred and make necessary adjustments.

The most obvious disadvantage of simulation is that it is time consuming and expensive to build, at least to the detail and accuracy needed to study and train advanced competencies. A simulation has to portray the environment of the skill being studied or trained. To be any good, it has to reflect the true state of affairs of the skills context, which is seldom well understood. (This is the problem I'm currently having with the development of my tennis programs: athletes manage to succeed in tracking and interception, but I don't know how they do it.) The simulation also has to provide all of the features of the real setting that influence the performance of the athlete’s skill, either positively or negatively. In other words it has to match the level of competence of the athlete and the degree of subtlety of the skill(s) in solving problems.

However, since the simulation only has to match the competence level of the skill being studied or trained, the simulation for a lower competence level skill needn't be as complex and as detailed as one for a higher competence level of the skill. Lower level simulations could then be first-stage vehicles and forerunners of the higher-level systems. This would make it possible to distribute costs and technical expertise more economically, over a range of skill levels. The argument is based on the fact that the degree of skill subtlety varies with the level of competence, so more subtle simulation detail is necessary to satisfy the training needs of better players.

Back to Index

------------------------------------------------

It's common to use training facilities in the form of a simulator that envelops the trainee in the appropriate environment for both the perceptual and the motor skills. Such a technique is employed to train airline or military pilots, for instance. Simulators are also used to train operators of military vehicles like tanks and trucks, and they can be used to train operating personnel of cars, trains, ocean liners, farm equipment, production machinery, and so on. These systems make good applications for simulation, because the trainee (usually seated) is stationary during real operations and the tools of his trade are at hand and fixed in place under normal conditions (like in a computer game), even though the system itself is moving. The trainee thus gets to practice real skills realistically. The usual procedure is to build a special simulator for each significantly different system, such as different types of aircraft, for instance.

So, if we furnish the trainee not only a simulation, but also the tools or instruments that are normally applied, and the trainee can use them realistically, we would have the training simulator. A flight trainer, for instance, houses a replica of a real airplane cabin, with appropriate displays, windows, and flight controls.

In the simulator, the pilot/trainee actually performs real-world perceptual-motor piloting skills, and you can measure the read/react responses directly. Typical training includes practice with operations like landing the aircraft, dealing with heavy turbulence, flying around or through a storm, or having to contend with a stall or with a bad engine. Visual displays at the windows depict the world "outside the moving aircraft" and real flight controls enable the trainee to "fly" the aircraft under a variety of flight conditions. The aircraft itself is "inside" the computer.

Unfortunately, though, we haven't reached that stage for sports, and certainly not for everyday skills; nor is it likely to happen any time soon. To provide an adequate format we would have to duplicate the arenas where games are played and the normal activities occur. For tennis training, specifically, we would need a chamber that was the same size as the court, or at least the trainee’s half of the court. (The player doesn’t perform his duties sitting on a chair, so the enclosure has to be large enough to accommodate the movement.) The simulator would then be equal in size to the real setting. But the playing arena (the court), like the aircraft in the flight simulator, would be in the computer.

Back to Index

-----------------------------------------------

Simulation (without the superstructures of simulators) is ideally suited to explore your skills space. The ideal is to exactly copy the relevant aspects of the objective setting in which the skills occur. But the selection of characteristics of the setting to be represented depends on the level of skill being learned, much as the content in traditional teaching varies with the grade level of the students. For example, you wouldn't give the intricacies of Einstein's relativity theory in a basic physics class. You would rather leave those details to a more advanced class. The same is true in simulation -- the more advanced the student, the subtler the skills and the more intricately the space has to be spelled out in order for the simulation to be useful. And that's where the real challenge is, or where the main problems lie: providing increasingly accurate or realistic simulation.

For the study of tennis skills, in particular, you need a simulation that accurately reflects what is occurring at the tennis court. This can be difficult enough, but at least you don’t need to get involved with the complexities of a training simulator for the game. You don’t need an enclosure for the training or objects like flight control sticks or headphones or gauges or the like. In order to conduct a study of tennis tactics, say, you only need an accurate representation of the game. This would also be suitable for optimization work

Back to Index

-----------------------------------------------

When a simulator costs too much to put together, the alternative is to use the real system itself as the training vehicle, like the SAGE system. In such a large-systems training approach, personnel would perform their normal duties in their natural operational setting. The training of a corporation, for instance, could be run in the natural setting of the corporate facilities, and employees would work system problems in the context of a virtual operating arena, using fabricated but realistic information.

Each situation would have to be judged on its own merits, but in cases where separate facilities aren't feasible, say where constraints are too severe, it may still be possible to use the normal operating arena itself as the training vehicle, using virtual reality.

In organizations like factories and power plants, personnel typically perform a variety of skills at any number of specific stations or offices designed to fulfill a group objective. While performing their duties, they interact with one another in any of a variety of ways, either face-to-face or via telephones, video, computers, fax machines, switches, gauges, or the like. The personnel could be trained even while the system is operational, and would be supplied with additional (simulation) information and material (appropriately marked as such). On-site training is particularly useful when the workstations of the system are equipped with computers, as in the SAGE system.

To gin up a virtual reality tennis trainer, I can imagine the following. First, the training "chamber" for the athlete would have to be large enough to allow movement as called for at the tennis court. So the court itself could be used. The athlete could be provided with a racket – instrumentation corresponding to a pilot’s flight control instruments. As part of the training, the player would have to be able to read a simulated situation as if it were actually happening – analogous to a pilot looking at his gauges and out the make-believe cabin window at a make-believe landing field. And he would have to be able to react to the situation in an appropriate and purposeful manner. Similar setups could be used for other active skills. The simulation itself would have to be presented by means of a device like a head set and goggles system to provide the visual images. Inertial body sensors (likely packaged in a suit with radio transmitters) would be needed to keep track of the player's movements and maintain the correct trainee perspective in the setting.

The same technique could be applied in other skills where at least some of the events occur in place. Golf is a good example, because every shot is hit from a standing position. In this application any of an almost infinite range of shot locations and a wide variety of local environments could be simulated and present a challenge to the player. The golfer would be able to play many different regulation rounds of golf in simulation on real and imagined courses and could vary his clubs according to highly specific circumstances. Every kind of shot would therefore be possible, from driving the ball off a tee to putting on the green and all the shots in between. A learning approach of this kind would certainly satisfy Bobby Jones’ training dictum that to become an expert in the game you should play on as many different courses as you can manage.

Back to Index

---------------------------------------------

Top of Page