Taking a glimpse at this chart, if you called this the Fletcher Munson Curve, you would be right… back in 1937. What is the Robinson Dadson Curve? that answer would be correct if we were on Jeopardy, and if it existed back in 1956.
Today, the once experimental test with the results coined by the researchers Fletcher and Munson, is now known as The Equal-Loudness Contour; also known as ISO 226: 2003. What is this exactly? Why was this test important to audio engineers of the past and present? How is this information useful to us today? I’m here to cover all of this in laymans terms.
1933 : The Fletcher Munson Curve
In 1933 Harvey Fletcher and Wilden A. Munson wanted to define loudness by classifying a way to measure and calculate levels of relativeamplitude(loudness) among different frequencies. Harvey and Wilden would select subjects, give them a pair of headphones, and then play a pure steady tone at a selected frequency followed by a 1khz tone at specified volume. Subjects would then choose a sound level for the test frequency that matched the constant variable, the 1khz tone. After the selection was made by the subjects and collected for the current frequency, they would continue through the human hearing spectrum stopping at the next specified frequency and repeating the test. The chosen sound levels, given by the subjects, were pooled and averaged for each frequency eventually charting the curve. Fletcher and Munson displayed their consolidated findings in the first chart produced in 1937.
1956 : The Robinson Dadson Curve
From 1937 to 1956, different scientists tried to “recalibrate” the findings of Fletcher and Munson by conducting their own tests on the measurement of loudness. Most of their tests proved to be unsuccessful due to the inconsistencies within their findings. Yet, one group did prove to be somewhat successful and were considered to have superseded the then known Fletcher Munson Curve. D. W. Robinson and R. S. Dadson wanted to redetermine the measurement of loudness by isolating their subjects in the center of an anechoic chamber and having their subjects focus on a specific sound source from a distance by placing one loudspeaker in front of them. The specification for selection of tones and constant variable stayed the same from the Fletcher Munson tests.
The results yielded a major difference in the lower frequency spectrum when compared to the Fletcher Munson Curve and the then previous standard Harvey Fletcher and Wilden Munson compared their results to. Many came to the assumption that the Robinson Dadson test had more room for error and considered many different factors as to the difference in results from the comparison. Even with its speculation towards its accuracy, The Robinson Dadson Curve did supersed the Fletcher Munson Curve and was adopted by the International Organization for Standardization (ISO) as the standard chart for human equal loudness in 1956. It was renamed to ISO 226.
Present : The Equal-loudness Contour
ISO 226 was considered definitive until recently when a study group, using the current standard as a constant for their research, started to consolidate much different results. The vast difference in findings called for the ISO to revisit the current standard established by Robinson and Dadson.
Several studies from across the globe conducted the research to finalize a new standard, ISO 226: 2003. The Fletcher Munson Curve (Even though the curve was based off Robinson and Dadon’s research) was renamed to The Equal-Loudness Contour as a way to avoid confusion from previous research. The current contour is really the best of both previous contours. It pairs the information of the Fletcher Munson Curve’s accuracy of the low level frequencies with that of the Robinson Dadson Curve’s accuracy at high level frequencies.
The current tests showed that headphones are the best choice for testing low frequency equal loudness because it allows the subject to hear via “side presentation”. With a pair of headphones that are sealed to the ear, a subject could attest to higher accuracy of equal-loudness without resonance from the room or the masking affect caused by the face which tends to roll off low frequencies. At high frequencies, using headphones is a bit of an issue because the wave lengths are so short that they have the ability to resonate or reflect off of our pinna (outer ear) only to then resonate off the sealed earphone causing the amplitude to magnify twice over which tricks the subjects ears into thinking the frequency is twice as loud. For that portion of the test, a frontal presentation was used which was allowed to have both ears focus on a sound source from a distance minimizing resonance from the ear. High frequencies suffer very little from the masking caused by our face because of their short wave lengths so a frontal presentation of sound is better for a flat frequency response in the high end.
This explained why the Robinson Dadson Curve couldn’t account for low level frequencies and the Fletcher Munson curve seemed more accurate.
So How Is This Helpful To You?
When I looked back at the Equal-Loudness Contour curve, I initially took the straight forward idea, high level frequencies tend to be louder with less amplitude. So, I should keep my level low on my high frequency sounding instruments like my flute, piccolo, hi hats or cymbals. Yet, actually the bigger lesson here is that it’s scientifically proven that during your mix, you should use both near field monitors at a good distance along with a pair of headphones for the purpose of auditioning your mix from the low-end perspective and then from the high-end perspective. If you wear headphone with a good seal, you can gain a more flat response as to how truly loud the low-end is in your mix. Same for monitors, a good pair of flat response monitors can clearly tell you how your high-end measures with the rest of your mix. Even though you test from both options during your mixing stage, you should still take the time to listen to your final mix in different settings by different playback devices in order to familiarize what levels are ultimately right and wrong universally in your mix.