Tracking and Intercepting a Tennis Ball
To see or not to see -- that is the question.
------------------------------------------------------------------------
-----------------------------------------------
We are concerned here with a player's perception of, and motor response to an oncoming tennis ball in the course of a tennis game -- a read-and-react principle. When tracking a ball, the player recognizes a pattern and acquires information about the motion. The goal is to assess the trajectory in order to overtake the ball and return it. Developing a computer simulation to study the process is the subject of my doctoral thesis.
Unfortunately, a major difficulty in the modeling is that we just don't know how this remarkable feat is accomplished. What's more, it is difficult, if not insurmountable, to observe, because the very act of trying to see what's happening uses the very principles and mechanisms that we are trying to see and we can't get around them to learn directly what is occurring. We more or less have to guess and see if the model works. Click here for more on this.
The simulation presumes the ability to:
The simulation also presumes an interaction between the sensory and motor processes and between the player and the physical environment. It is the environment that's perceived, the player being an intimate part of it. The player draws on personal and environmental mechanisms (characteristics) to perform the skills. Without the mechanisms there can be no tennis.
We take it for granted that the player has to "keep his eyes on the ball." However, we don't know how tracking is done or what information is extracted. It is a pattern recognition task. The type of information extracted depends on the kind of model the player uses in the trajectory assessment. Different perceptual rules likely yield different results, some better than others. Different models generate different repositories of information.
The changing position of the ball alters the player's line of sight to the ball, and so too does the player's movement towards an intercept point. This alters the view of the moving ball -- of its trajectory. The new data, in turn, can affect the estimate of the trajectory, which could then alter the player's estimate of its intercept point and therefore the player's own speed and direction of motion. We know relatively little about this process.
----------------------------------------------
In simulating the player's track and intercept experience for my thesis, I represented both its subjective and objective aspects, which identify, respectively, what the player judges the situation to be and what the situation really is. He works in his perspective space and guesses about the non-perspective space. Expressed another way, he guesses what the trajectory should have looked like in his perspective space (the only way he can see the ball). Physical science is a formal extension of this process and presumes to deal with the objective (or non-perspective) space. Perception is a theory/facts process -- we apply the theory and generate the facts.
In my simulation I distinguished between the player's estimates of context properties and the objective (or "true") values of those properties, as determined by accepted scientific principles. That is, I distinguished between the physical and the perceptual aspects. The dual values (subjective/objective) describe such things as the position and speed of the ball and the player's own speed and distance from the ball, all working within the context mechanisms.
I presumed that a difference between what the player perceives to be the case and what actually is the case leads to actions short of best track and intercept performance, to a degree that depends on the player's abilities -- his track and intercept (read and react) competence.
To account for the differences, I presumed a certain level of player competence in both the track and intercept functions and developed logic to meet the skill levels. In other words, I applied the player's presumed competence level to establish the accuracy and precision of presumed track and intercept skills. The idea was to depict what the player thinks is the case about the trajectory properties, or what he sees them to be. And I used physical principles to define conditions as they really are in the objective world of the tennis court environment. So I had to set up standards of competence.
As you might expect, because of the nature of the subject matter, the ultimate product (a basic program and three extensions, as it turned out) had to be a greatly reduced version of the real sensory-motor process, which is severely non-linear and immensely complex. I am currently trying to extend this work and integrate it with my hitting games.
---------------------------------------------