Equal-Loudness Contours and Why “the Same dB” Feels Louder at Different Colors
If white noise sounds louder than brown noise at the same volume setting, or a deep bass rumble at 80 dB SPL feels less intrusive than a tinny hiss at 60, the reason is a property of human hearing mapped in detail since the 1930s: the equal-loudness contour. The ear is not a flat-response microphone. Its sensitivity varies enormously with frequency, so perceived loudness is fundamentally different from acoustic energy.
The Original Fletcher-Munson Curves
In 1933, Harvey Fletcher and Wilden Munson at Bell Labs established the foundation for equal-loudness research. Listeners adjusted tones at different frequencies until they sounded as loud as a 1 kHz reference tone at a known sound pressure level (SPL). Repeating that comparison across levels and listeners produced a family of curves showing the SPL required at each frequency to match a reference loudness.
The unit on these curves is the phon: a sound is N phons if it sounds as loud as a 1 kHz tone at N dB SPL. At 1 kHz, dB SPL and phons therefore have the same number; at other frequencies, they diverge.
- The ear is most sensitive around 2–5 kHz. This region includes the ear canal’s natural resonance and much of the energy in speech consonants. A tone near 3 kHz needs less SPL than a 1 kHz tone to seem equally loud.
- The ear is less sensitive at low frequencies. At moderate levels, a 100 Hz tone needs substantially more SPL than a 1 kHz tone to match its loudness. Below 50 Hz, the difference becomes more extreme.
- The curves flatten at higher levels. Frequency matters somewhat less to loudness at high levels than at low ones. That is one reason a quiet sound system can seem thin while a louder one seems fuller.
ISO 226 and the Modern Standard
The ISO 226 standard formalizes normal equal-loudness contours for practical use. Modern contours refine the original measurements using a broader body of experimental data, but their qualitative shape is unchanged: human hearing is strongly frequency-dependent.
These contours also inform A-weighting, the filter behind dB(A) measurements used in many environmental-noise standards. A-weighting approximates the ear’s frequency sensitivity at a moderate level, producing a more useful single-number summary than unweighted dB SPL for many everyday sounds. It is still an approximation: perceived loudness depends on level, spectrum, duration, and the listener.
What This Means for Noise Color
The spectra of noise colors interact with the ear’s unequal sensitivity, making their perceived loudness differ even at the same SPL.
White noise has equal energy per hertz across the audible band. The ear gives its 1–6 kHz content substantial perceptual weight, which is why white noise is commonly heard as bright, hissy, or tinny. Many listeners prefer it at a lower level than other colors.
Pink noise has equal energy per octave, rolling off at roughly 3 dB per octave. It reduces high-frequency energy relative to white noise and often sounds fuller and less hissy.
Brown noise rolls off at roughly 6 dB per octave. Its reduced high-frequency content can make it sound warmer and quieter than white noise at the same SPL. Raising its level to match white or pink noise can still feel comfortable because more of the added energy is in a range where the ear is less sensitive.
The practical implication is simple: one SPL number does not characterize loudness across noise colors. Two tracks at the same dB SPL can sound very different, while two tracks matched by ear can have different SPLs. dB(A) is often more informative than unweighted SPL, though it too is only an approximation.
The Contour and Listening Fatigue
Spectrum also affects long-session comfort. Noise with substantial energy in the sensitive 2–5 kHz region can feel more demanding over time than a lower-tilted spectrum at a comparable perceived loudness. This helps explain why listeners who use noise for hours often gravitate toward pink, brown, or a custom blend rather than bright white noise.
A generative system such as the dpli noise generator lets the listener adjust the spectrum continuously. The comfortable point is personal and depends on hearing, the room, the task, and intended listening duration.
Age and the Drifting Contour
Equal-loudness contours measured on young adults do not apply identically to every listener. Age-related hearing loss commonly affects high frequencies first, reducing the perceptual weight of the upper band. A spectrum that sounds balanced to one listener can sound bass-heavy to another.
For noise-tool design, this is an argument for control rather than a one-size-fits-all preset. A slightly brighter spectrum may feel balanced to someone with reduced high-frequency hearing, while another listener may prefer a warmer spectrum for long sessions.
dB(A) in Sleep and Office Guidance
Environmental guidance for sleep, office acoustics, and hearing protection often uses dB(A). The weighting is intended to better reflect audible impact than raw SPL. For broadband generators, brown and white noise with the same dB(A) reading can be closer in perceived loudness than two sounds with the same unweighted SPL, even though their spectra remain different.
When tuning a noise generator, use dB(A) measurements when available and then make the final adjustment by ear at the actual listening position. A smartphone reading or a number on a volume slider cannot fully capture the room, playback device, spectrum, and listener together.
The Bottom Line
The equal-loudness contour is one of the central facts of psychoacoustics. The ear is most sensitive in the 2–5 kHz region, less sensitive at low frequencies, and changes its frequency response with overall level. Fletcher-Munson curves and ISO 226 make that behavior measurable; A-weighting turns part of it into a practical, if imperfect, measurement.
For noise generation, the spectrum and the listener’s ear matter as much as the dB number. White, pink, and brown noise can differ in both perceived loudness and long-session comfort at the same SPL. Tuning a noise environment by ear is not imprecise—it is accounting for the part of the system that a single volume number leaves out.
References
- Fletcher, H., & Munson, W. A. (1933). Loudness, its definition, measurement and calculation. Journal of the Acoustical Society of America, 5(2), 82–108.
- ISO 226:2003. Acoustics — Normal equal-loudness-level contours.
- Suzuki, Y., & Takeshima, H. (2004). Equal-loudness-level contours for pure tones. Journal of the Acoustical Society of America, 116(2), 918–933.
- Robinson, D. W., & Dadson, R. S. (1956). A re-determination of the equal-loudness relations for pure tones. British Journal of Applied Physics, 7(5), 166–181.
- Moore, B. C. J. (2012). An Introduction to the Psychology of Hearing (6th ed.). Brill.
- IEC 61672-1:2013. Electroacoustics — Sound level meters — Part 1: Specifications.
- World Health Organization. (2009). Night Noise Guidelines for Europe.