Group-Based Computer-Assisted Training
Training without feedback is barren.
-----------------------------------------------------------------------------------------
----------------------------------------------
Carving figures out of marble can be painful and frustrating if you haven't learned the right skills the right way. Unfortunately, growth and learning often arrive the hard way, through experience, as we say, at times merely through mistakes. And from that you might conclude that it's important to make them. Maybe so. And maybe we do learn from them. But we don't make the mistakes deliberately in order to learn. On the contrary, it's the mistakes we make -- and we make plenty of them, which means we waste a lot of marble -- that motivates our studies. We don't like to blunder, and we can't afford it. Nor can we depend on serendipity to save us in our ever-changing worlds. Skills have to be learned properly, and this calls for careful study.
Athletes in particular, at every level of their sport, from novice to expert, have to be very cautious in how they go about learning their skills. For the novice a bad start can result in irreparable damage -- habits (established from early "learning" and re-enforcement) can become serious permanent faults. Even for the professional athlete, with well-developed and competent abilities, attempts to improve further can be very risky -- metamorphosing a working system could easily be chaotic and lead to serious loss of function. But they can't rest on their laurels, either. Action and risk are inevitable and uncompromising.
------------------------------------------------
The Group Approach to Learning
The saving grace, if there is one, is that over the thousands of years of human history, man has invented and devised countless assistive tools and techniques and developed many new skills in consequence of our various needs. For instance, evolution of the alphabet led to book reading skills and thereby improved communication. Similarly, invention of the needle and thread called for sewing skills to put together or repair clothing. And design and development of the automobile has required driving skills for better transportation. You can write your own long list of such items. While remarkable in themselves, these inventions and developments nevertheless draw attention to our underlying lack of readiness to cope with life's difficulties. We still need gadgets. We still have much to learn and we need new machinery to assist us.
It is here that a computer-assisted group perspective for study and training can be helpful. Using tennis as an example, this learning tool consists of two complementary and interactive ingredients:
Each component is meant to provide feedback to the other, in an interactive process. On one side, the court-based measurements are intended to fine-tune the theoretical model by improving theoretical expectations. And on the other side, theoretical studies are meant to guide the player's trial-based understanding of trajectories through heightened perception and sharpened production of the trajectories. Ultimately, this combination of theory and practice should yield better results than either of the two parts could yield separately.
----------------------------------------------
As one component of the group approach, computer-based simulation can be helpful, because it facilitates meaningful action and mitigates risk -- this aid gives you a good way to look at skills in order to improve them, through practice. In tennis, for instance, before even hitting the ball, you can work out problems in advance and find solutions for situations you might encounter at the court. Just as with testing a rocket launch in simulation before powering the thing aloft, so too you can test a skill before practicing it. Such a simulation makes it possible to run trials in the operational environment to try to learn exactly what is best (optimal) for you to do before taking action. You thus reduce the chances of making errors and developing poor habits, not to mention that you can minimize injuries. That's what simulation provides, and that's what I try to furnish in my own computer-based programs for tennis.
Many investigative techniques are available, some better than others. But conducting studies on a digital computer seems to have an edge when dealing with severely non-linear problems, such as skill development. This approach uses the computer to run science-grounded diagnostic tests of the skills, to help you understand and try to bring your actions closer to what is best for you, personally, without creating ineffective practices impossible to break. You need to assess your own capabilities and potential.
The method depends on simulation to conduct the studies and can be applied to all skills. It is this method that l use when looking specifically at the tracking, interception, and hitting skills of tennis (perception and action skills that are obviously fundamental to the game). The setting (turf) for the simulation is the tennis court itself, which is to be understood as the diagnostic study context. It is within that context that you apply your tennis techniques. It is this learning option that we explore.
The basic idea for the simulation is simplicity itself. The tennis activity occurs at the tennis court, so we first lay out the physical court quantitatively, in simulation. The next step is then to embed the tennis student/player in that context, again in simulation. And of course the individual naturally brings with him a perception and action potential, so the court is to be presented in perspective, to reflect the player's perception. (This would be analogous to the real person setting up at the actual court to begin hitting the ball in a practice session.) Finally, test trajectories in simulation (first having been formulated and made available in the design) can now be launched, one at a time, to the simulated hitter to begin the studies, just as a coach might hit them to a student.
The trajectories in the simulation are formed en mass using equations of motion that involve starting conditions and include a gravitational constant together with coefficients to deal with air resistance and the friction that occurs between the ball and the surface of the court. In programs that I've written I selected (shaped, designed) subsets of the trajectories to be available to the student as practice vehicles. One at a time, then, the student selects a trajectory and responds to it in some way, as one step in an inductive learning process.
Similar systems can be devised for many other operational settings, such as a busy street, a hospital ward, a manufacturing workstation, a concert hall or schoolroom, a soccer field, the interior of an airplane, boat, or train, a crime scene, or outer space,...,wherever. You can first put together the objective structure of the context. This becomes the ground in which you embed the appropriate skilled agent (nurse, teacher, director, athlete, driver, detective, worker, astronaut, and so on). Then you can proceed with your studies using relevant transaction exchange materials for the context.
Tennis simulation lets you run provisional trials in the skills context to discover ways to deal with particular problem situations that might arise during play. The procedure is meant to strengthen your skills by raising your level of perceptual competence. And this is meant to occur through heightened awareness and better understanding of trajectories and the mechanics of producing the trajectories. The more clearly you can see the true state of affairs, the more effectively you can act on it.
Naturally, in tennis, you want to improve the way you hit your shots; hitting, after all, is the crux of the game, the reason for being out on the court in the first place. To be competitive you need to understand how the ball is to be hit, for the striking action affects its path through the court. (Reading books like The Physics and Technology of Tennis can help you here.) But that's not enough. The need to grasp the underlying ideas applies not only to hitting but also to tracking and interception. You need to track the ball accurately and intercept it properly to make it possible to swing your racket freely and meet the ball correctly to make planned shots. Seeing, recognizing, and intercepting trajectories are a necessary part of the process -- it is the ongoing perception and action (read and react) procedure that predominates.
Even this isn't enough. You still have to know precisely how to meet the ball with your racket before you can correct your strokes. And there is much to understand if you wish to play the game well. For any shot to be good, of course, the ball has to stay within the court boundaries, and you have to learn how to keep it there. But what is more important, you need to aim accurately to reach very specific (or pinpoint) targets. The ultimate purpose of the simulation would therefore be to sharpen your shots and let you access target areas that presently, at your current skill level, might not be reachable.
This investigative, computer-assisted methodology is comparable to actual laboratory techniques, in the sense that each computer trial corresponds to a single laboratory trial (such as you might run in chemistry, say, or biology). It also corresponds to a trial you might execute at the tennis court itself. For each new test on the computer you would re-set inputs for the enquiry and run the simulation again, just as you'd set things up at the lab or at the court. But it's a lot easier to do on the computer, in a simulation.
-----------------------------------------------
Still with the tennis example, we can use computer simulation as an assistive tool at the real tennis court for studies. Now, though, adopting the group approach, the members of the group can operate special feedback equipment at different stations at the court to help conduct training. There would have to be a hitter, of course -- the trainee. And someone to deliver the ball to be hit, either by means of a ball machine or by means of a racket. And someone at the sideline, say, would conduct appropriate studies to support and guide the hitter on each shot to be taken. In other contexts, similar stations might be set up.
The same technique could be applied in other skills, particularly when at least some of the events occur in place. Golf is a very good example, because every shot is in fact hit from a standing, stationary position. In this application any of an almost infinite range of shot locations and a wide variety of local environments could be simulated for study. And feedback equipment could be made available to record real shots.. The golfer would be able to play many different regulation rounds of golf in simulation on real and imagined courses and could vary his or her clubs according to highly specific circumstances, just as might be the case at the live links. 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.
Feedback in the form of video clips and speed measurements would help make corrections in the way the ball is hit. For instance, speed-measuring devices could measure racket or club speed at the moment of impact with the ball and the ball speed just after being hit. This data could be fed back to the computer to make simulation adjustments and set conditions for more study.
A video with a known shutter rate of speed could be used to give you an estimate of the instantaneous ball or striking speed. But this would take a bit of calculating at the court. You know the number of frames running per second, so you would count the number of frames it took the ball to go a certain distance -- a fraction of a second. This would give the time it took to go that distance, so you would have the average speed for that interval. If the ball or racket is blurred, as would be the case for higher speeds, you would measure the length of the blur for a given interval.
To get the average speed of a shot over a longer distance at the court, say, one way would be to use the sound recorded by video. You could measure the interval between the moment the ball was hit by one racket to the time it bounced off the court or hit the other racket.
Radar is another possibility for measuring speed, as you might have learned from its use in automobile traffic regulation. The principle behind radar is the so-called doppler effect. Pulses of energy of known frequency are sent out by a radar gun, strike a target, and bounce back to the gun at a different frequency, depending on the speed and direction of the target. The change in frequency depends on how you line up with the target direction of motion., i.e., on how the speed appears to you. So, if you look exactly in the direction of motion of the target, you get its actual speed. Otherwise you apply the trigonometric cosine rule to get the true value. If you look precisely at right angles to the motion, there will be no frequency shift.
When measuring speed at the moment of ball/racket contact, though, problems arise because of the interactive effects of the ball and racket speeds on the readings. The two speeds would have to be isolated one from the other. For the ball, at least, this could be done by taking the speed measurement outside the reach of the racket. The same for the racket, which would be just before contact with the ball.
Still another way to measure object speed is to use a pair of laser beams. For example, the BatMaxx system, as it is called, measures "when something interrupts two parallel laser beams shooting across a zone. Anything can interrupt the two beams and BatMaxx will calculate the speed of the object. It can be a hand, or a baseball bat - wooden or metal: it doesn't matter." In our case it would be the tennis ball moving toward or away from the racket. It could also be the racket swinging through the ball.
Here it's useful to know that a shorter swing gives you a bit more control over hitting the ball. But you have to swing harder to acquire the same contact speed.
Spin, now, is kind of a nasty beast to measure. But even if you have a neat way to get a numerical value for the rate, it probably won't help the hitter much. The problem is that the unaided eye is a poor measuring device for spin, so there's no good way to correlate its reading with the numerical result; the rate won't be understood qualitatively. So there can't really be an accurate adjustment with the feedback. You have to depend almost entirely on how your opponent hits the ball.
---------------------------------------------------
Decisions as to how to proceed with the training would most likely be thought out in advance by the group. Possibly with the aid of a knowledgeable overseer/director, whose job it would be to coordinate the feedback with the selection of shots to be practiced and the studies of the shots to be run on the computer. It would be necessary to work out station assignments and rotation tactics and the length of time to be spent at each station. During a training session, all group members would best be linked together audibly by a portable conference phone hookup. Finding answers to the technical problems would almost always be the responsibility of the group. The main point to understand is that neither the studies alone nor the practice alone are adequate to achieve best results. But together they could be very effective.
----------------------------------------------
