Audio Control

What To Do

This is an auditory version of the compensatory tracking task in the Nature of Control demo. In this case you will be trying to keep a tone close to a target pitch rather than a cursor close to a target. When you press the "Run" button a tone will be played and a vertical line will be displayed on the screen. Quickly move the cursor arrow so that it is directly under that line. Then close your eyes and listen to the tone. After about 5 seconds the pitch of the tone will start to vary. Your job is to keep the tone as close as possible to the starting pitch. You do this by moving the mouse as necessary -- to the right if you need to increase the pitch and to the left to decrease it. This is a fairly difficult task so you might have to do it several times before you are able to keep the pitch under control. One a trial is completed and the results presented in a graph you can run another trial by pushing the "Run" button.

It should be noted that not everyone is apparently able to control pitch. In a post to internet discussion group dedicated to the application of his control theory model to understanding behavior, William T. Powers wrote the following:

"People controlling the pitch of a sound were investigated. Most people could control the pitch accurately (within 5%) for a period of one minute after the tone to be reproduced was sounded. About a quarter of the people (roughly) could not control the pitch at all. All of these people said they could not sing or play an instrument -- they considered themselves "tone-deaf" (and so did their teachers!). The distinction between pitch controllers and noncontrollers was on the order of five standard deviations by a measure of "stability factor". It is not known whether the noncontrollers could ever learn to control pitch. "

So don't feel bad if you find that you are never able to control the pitch of the tone in the demo.

What It's About

This task shows that the basic variables and relationships between those variables are the same regardless of the type of variable being controlled. In the Nature of Control demo the variable controlled was a visual variable -- the position of a cursor. In this demo the variable controlled is an auditory variable -- the pitch of a tone. In both cases you must move the mouse appropriately to compensate for the constantly changing disturbance to the controlled variable. But this auditory version of a control task emphasizes the fact that the target or reference value of the controlled variable is specified in your head, not in the environment. In the visual task it is easy to succumb to the idea that the target line specifies where the cursor should be kept. You can only tell that that's not the case when the person controlling the cursor decides to keep it some distance to one side or the other of the target. But in this auditory demo there is nothing outside the person doing the controlling (you) that provides a "target" for the pitch of the tone. When you can control the pitch of the tone you are clearly keeping your perception of pitch as close as you can to an image in your head of what the pitch of the tone should be. Your behavior in this task -- keeping the pitch of the tone close to a target pitch -- is clearly an example of controlling a perception (of pitch) relative to an internally specified reference for what that pitch should be.

What To Notice

As in the Nature of Control demo the relationships between variables in this tracking task can be seen in the graph of the results. The numbers immediately below the data plot are measures of how well you controlled the pitch of the tone. RMS Error measures the average deviation of the tone's frequency in Hz (pitch) from the target value of 440 Hz (A above middle C). So an RMS error of 12 indicates that, on average, the tone was kept within 8 Hz of 440 Hz. Stability measures the ratio of expected to observed variation of the tone's frequency (pitch) in Hz -- the controlled variable. Expected variation is the amount the controlled variable would have varied if you had done nothing to control it; observed variation is the actual amount of variation in the controlled variable. If expected and observed variation are the same, there is no control and Stability is 1.0. If expected variation is much larger than observed variation of the controlled variable then this variable is under control and Stability is >> 1.0. The greater the Stability measure, the better your control of the tone's frequency (pitch). Finally, the Average Frequency (Hz) is the average frequency at which you kept the tone during the test trial. The closer this average is to 440 the more precise your control of pitch. If, over several trials, the Average Frequency is always somewhat above or below 400 it would be evidence of a tendency for your reference for the desired pitch of the tone to drift up or down, respectively.

Above the graphic display and RMS Error and Stability measures are the correlations between the variables in this control task. The number on the left is the correlation between tone frequency (controlled variable) and mouse movements (C-M). The number in the center is the correlation between mouse movements and disturbance (M-D). And the number on the right is the correlation between controlled variable and disturbance (C-D). Finally the number on the right is the correlation between your mouse movements (M) and the mouse movements made by a control model that performed the same tracking task.

As in the Nature of Control demo, when you are able to control the pitch of the tone, keeping that controlled variable close to 440 Hz, you will see that the controlled variable-mouse (C-M) correlation is rather low (usually between -.2 and .2). This is surprising if you think of variations in the pitch of the tone as the stimulus for the mouse movements (the response). You will also see that the disturbance-mouse (M-D) correlation is very high (usually greater than .99). This is also very surprising since the disturbance in this task is invisible. All you can perceive in this task is variations in the tone's frequency, which is at all times a combined result of disturbance and mouse movements. Nevertheless, mouse movements are strongly (negatively) correlated with the imperceptible disturbance rather than with the perceptible tone frequency variations. Next, you will see that the controlled variable-disturbance (C-D) correlation is very low, indicating that the disturbance has little effect on tone's frequency. This result shows that tone frequency is under control -- protected from the effects of disturbance.

Once you have learned to control the pitch of the tone as best as possible you can use this demo to compare your ability to control pitch to your ability to control the position of a cursor. That is, you can compare your ability to control an auditory perception to your ability to control a visual perception. First, write down (or remember) the Stability measure of control that you got on your last trial controlling the pitch of the tone. Now run the demo again but this time with your eyes open. After the 5 second tone plays a cursor appears and the target line disappears. At this point try to keep the cursor in the same position. Because the target line disappears you are in the same situation with respect to cursor position as you were with respect to tone frequency inasmuch as there is no external "target" value to use as a reference for the controlled variable. Once you have mastered this visual version of the auditory control task note the Stability measure for you best trial. I think you will find that the Stability measure for control of the visual variable (cursor position) is much larger than the Stability measure for control of the auditory variable, indicating better control of the visual variable. In my case, control of the visual variable is nearly twice as good as control of the auditory variable. There are several possible explanations for this. One is that in this demo there is both visual and auditory information available when you control the visual variable and the added auditory information may make control of the visual variable seem better. Another is that for some reason it is easier to maintain a stable (non-variable) internal reference specification for a visual than for an auditory variable; this would increase the observed variability of the auditory variable, increasing the denominator and, thus, the size of the Stability measure of control of the auditory variable. And still another is simply that the sensory resolution is higher for visual than for auditory variables, making it possible to more precisely control visual than auditory variables. It might be fun to try to think of ways to test which of these explanations of the superiority of visual to auditory control is best.

What Works Best

It may take a while to develop the ability to control the pitch of the tone skillfully. But the basic relationships between variables should show up even if you cannot keep the tone exactly at the target value of 440 Hz. You should practice this task until you are able to obtain a Stability measure of at least 4.0.

Last Modified: March, 2016

Richard S. Marken