Research Update #2

Using this website from the University of New South Wales, I measured my audio response to various frequencies. This process is known as "audiometry", and the results, when compiled into a graph, are called a "Loudness Contour".



The point of such a task is to measure where a person's (in this case mine) audio sensitivity peaks. The loudness is measured in negative decibels (dB)... the closer to zero, the louder the sound. (-3 dB is MUCH louder than -90 dB).



The decibel scale is a logarithmic scale which is set up so that an increase of 3 dB sounds to the human ear as a doubling in loudness. Ex: -27 dB is twice as loud as -30 dB; -24 dB is twice as loud as -27 dB and thus four times as loud as -30 dB; -21 dB is twice as loud as -24 dB, four times as loud as -27 dB, and eight times as loud as -30 dB... etc. Remember, this is how it "should" sound to a human... to me, personally, I can't really tell when a sound is "twice as loud"... I just perceive louder and quieter relatively (i.e. much louder, a little louder, a little quieter, a lot quieter, etc).



At the bottom of the chart that's posted right under this paragraph, I have several notes related to some of the figures that I got. Make sure you read them. Also, remember that if you do this same test, at the same website, but using a different computer and different speakers or headphones, your results could be drastically different even if your ears are exactly as good (or bad) as mine. Things like speaker response, room acoustics (if using speakers rather than headphones or earphones), and your computer's soundcard all affect the results of this. The test is relative to yourself--you can see where your peaks and valleys are, but don't compare this test person-to-person unless they are using the same equipment in the same environment. Below this chart, which reports my results, is the chart of a friend's results; he took the test in a different environment, on a different computer, using different hardware and software.


Hz ; dB level when heard


30 ; -3*
45 ; -12*
60 ; -21*
90 ; -33
125 ; -45
187 ; -51
250 ; -57
375 ; -57
500 ; -60
750 ; -63
1k ; -72
1.5k ; -75
2k ; -81
3k ; -87
4k ; -90
6k ; -90
8k ; -78 (-90)**
12k ; -72 (-90)**
16k ; -52***


*These very low numbers, especially for 30 and 45 Hz, are probably due to the low-end limit of the earphones I was using. (iPod headphones)

**These two tones did something odd: at the listed numbers, -78 and -72 respectively, I could hear the listed tone. BUT, from -90 dB up to -78 or -72 dB, I could hear an octave undertone. I'm not sure of the reason for this other than something happening with the earphones as they maybe approached their upper limit, a problem with the computer's soundcard not being able to properly process the signal (I don't know much about computers, but I do know that soundcards have an effect on the performance on this test.), non-linear effects in my ear somehow producing a Tartini tone an octave lower, or a neuro-processing 'error'.


***This frequency is actually about the pitch that my ears were ringing at. So, until the dB level reached -52 dB, I was unable to tell if it was my ears ringing, or the speakers playing. I thought this was a funny phenomenon.


Here are my friend's results:

30 ; -30

45 ; -27

60 ; -30

90 ; -33

125 ; -37

187 ; -33

250 ; -30

375 ; -48

500 ; -54

750 ; -45

1k ; -54

1.5k ; -63

2k ; -51

3k ; -75

4k ; -69

6k ; -66

8k ; -78

12k ; -66

16k ; -39

This unit of research has raised a new question:

4) How could the human acuity to hearing certain frequencies alter the way humans hear and neurally process chord intervals and Tartini tones at different octaves?