Conclusion and Discussion: I'm All Ears
Conclusion: My hypothesis was that "It is possible for people to use vibration to determine the direction and magnitude of sound." After running experiments using the apparatus that I built, I conclude that people CAN use vibration to determine the direction of sound. 92% of my tests to detect and pinpoint sound were succesful. My tests were less conclusive about magnitude of sound. That's because my tests only determined that that people wearing the apparatus could detect the buzzer at different distances, and could not detect the buzzer when it was muffled. My buzzers produced only one level of sound. I did not run controlled experiments on other sounds in the room, even though I observed them while using the gloves.
Discussion: I'm quite excited about the results of my experiment, which were better than I expected. I tried the apparatus out myself, and found that it was not only possible to notice when the motor was vibrating, it was almost impossible to ignore. I think that if a smaller, more sensitive version of this apparatus was built, people with hearing problems could use it to help them be safer in traffic, and so on.
My tests don't show whether the brain can rewire itself to accept the sensations from the hand as part of its sense of hearing. To run tests like that, I'd need volunteers to wear the apparatus for a few weeks or a few months, and then I'd have to be able to scan their brains with an MRI to determine if any changes had happened. However, based upon the studies done by other doctors, which were reported in Dr. Doidge's book The Brain that Changes Itself, I believe that there's a good likelihood that the brain could adjust to these new inputs.
Sources of Error: During the testing of the apparatus, I had to consider several variables and problems. First, I had to make sure that my test subject (Duncan, for example) couldn't hear the buzzer. I made several adjustments to the headphone/ear plug setup to solve that problem. My control test for that was that I switched the inputs to the glove, which meant that the subject would think the sound was coming from the left if it was coming from the right. This test was successful, in the sense that the test subject did the expected result.
Another variable that I had to control was the possibility that even if my test subject couldn't hear me, he might be able to detect my movement as I carried the buzzers around the room. I solved that problem by creating a Remote Dual Buzzer System. That way, I could sit on the sofa and control the buzzers, and the subject couldn't detect my footsteps. (Our floor is creaky and bouncy.)
The third major problem I ran into was that when I calibrated the microphone to be more sensitive, or increased the volume of the music coming over the headphones, the gloves would detect that music. I solved that problem by changing the music to a recording of the buzzers themselves. They could mask the real buzzers at a lower volume. But I don't think that this would be a problem in real life, because a deaf person wouldn't have to listen to loud music in order to use the apparatus.
One problem that I had was that my house is not that big, so I could only test as far away as 25 feet. Therefore, I could only test for conditions indoors. Although I did not run a controlled experiment for how the gloves worked for household noise, I noticed that the gloves did pick up a louder noise, like someone calling me or talking very closely, or a doorbell, but did not pick up on music coming from the basement or kitchen. This made me think that the glove worked quite well indoors. It would not be useful if it picked up every single sound. This was probably a result of the way that my pulse-width modulation works. It randomly creates a value between zero and 100%, and checks to see if the sound is louder than that. If it is, then the motor vibrates. Otherwise it does not. In addition, the motor will only vibrate if the noise is louder than 15%. This was one of the first things that I did when writing my program.
I think that this apparatus would need a lot more work before it could be used outside. The microphones would have to be able to vibrate at different speeds for very different levels of noise. The wearer would have to be able to tell the difference between a horn honking, a child calling, and the sound of a far away lawnmower. Currently, the apparatus does vibrate at different speeds for different levels of sound (it will pulse for speech and shake wildly for the buzzer), but the calibration and (and rate of input) would have to be much better.
Another engineering problem comes up if people want to use this machine outside. Right now, my circuit is attached to a breadboard and a computer. It is not difficult to attach it to a laptop and put the laptop and breadboard in a backpack. But if people move around a lot, the wires come out of the breadboard and even the vibrating motor is a little delicate that way. For real use, the circuits could obviously not be held together by alligator clips. Everything would have to be designed to be very light and very unbreakable. A person who was relying on the gloves to keep them safe on the road could end up getting hit by a car if the circuit came loose.