Human hearing is not equally sensitive at all frequencies. We hear midrange frequencies (1–5 kHz) much more easily than very low or very high frequencies. Enter a frequency and SPL to see the perceived loudness in phons (equal-loudness level) and sones (a linear loudness scale where 2 sones = twice as loud as 1 sone).
Equal-loudness contour reference
Each row shows the SPL required at different frequencies to produce the same perceived loudness (in phons). At 1 kHz, phons always equal dB SPL by definition.
| Loudness | 63 Hz | 125 Hz | 250 Hz | 500 Hz | 1 kHz | 4 kHz |
|---|
What Are Equal-Loudness Contours?
Equal-loudness contours (historically called Fletcher-Munson curves, now standardised as ISO 226:2003) are lines on a frequency-vs-SPL graph connecting all frequency-SPL pairs that a typical listener perceives as equally loud. The reference frequency is 1 kHz, where the loudness level in phons is defined to equal the SPL in decibels: a 70 dB SPL tone at 1 kHz is 70 phons by definition.
At low frequencies, the contours rise steeply: to perceive a 50 Hz tone as equally loud as a 70 dB 1 kHz tone, the 50 Hz tone must be presented at roughly 85 dB SPL. Conversely, near 3–4 kHz—where the ear canal resonance amplifies incoming sound—lower SPL is needed to achieve the same perceived loudness.
Phons vs. Sones
The phon is an equal-loudness unit: it tells you the dB SPL at 1 kHz that would sound equally loud. Phons are useful for comparing loudness at different frequencies but are still logarithmic—40 phons is not “twice as loud” as 20 phons.
The sone is a linear loudness scale designed so that doubling the sone value represents a doubling of perceived loudness. By definition, 1 sone = 40 phons. Above 40 phons, each 10-phon increase doubles the sone value:
sone = 2(phon − 40) / 10 (for phon ≥ 40)
| Phons | Sones | Subjective loudness |
|---|---|---|
| 20 | 0.25 | Very quiet |
| 40 | 1 | Quiet (reference) |
| 50 | 2 | Twice as loud as 40 phon |
| 60 | 4 | Four times reference |
| 70 | 8 | Eight times reference |
| 80 | 16 | Sixteen times reference |
Why Equal Loudness Matters for Audio Engineering
The frequency-dependent sensitivity of human hearing has direct consequences for mixing, mastering, and monitoring:
- Monitor level affects perceived tonality. At low listening levels, bass and treble appear to recede (the contours are steeper at low SPL). A mix that sounds balanced at 85 dB SPL will sound thin and bass-light at 60 dB SPL. This is the phenomenon that “loudness” buttons on consumer electronics were designed to compensate for.
- Mix at a consistent level. To maintain a reliable frequency perception, mix at a consistent SPL—typically 79–85 dB SPL at the listening position (K-System calibration). Constantly adjusting the monitor volume changes your perception of the tonal balance.
- Low-end decisions are hardest. The steep rise of the equal-loudness contours below 100 Hz means small SPL changes in the sub-bass produce large changes in perceived loudness. This is why sub-bass balance is the most room-dependent and level-dependent aspect of any mix.
- A-weighting approximates hearing sensitivity. The dBA weighting curve is an inverted approximation of the 40-phon contour. It is used in noise measurements because it correlates better with perceived loudness than unweighted SPL.
The Ear Canal Resonance
The dip in the equal-loudness contours near 3–4 kHz corresponds to the quarter-wavelength resonance of the human ear canal (approximately 2.5 cm long, resonating near 3.4 kHz). This resonance amplifies incoming sound by 10–15 dB in that band, making us most sensitive to frequencies in the presence and sibilance region. This same resonance is why sibilance (2–8 kHz) in vocal recordings is so perceptually prominent and often requires de-essing.