• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

EQ Expectation Bias

Soandso

Senior Member
Joined
May 30, 2022
Messages
393
Likes
1,060
Many ASR posts have IEM and headphone frequency charts, which often include comparisons to others; as well as overlayed charts to a Harmon Curve. In quite a few instances commentators suggest EQ adjustments to those stock frequencies for improving some aspect or another. I propose this sets up conditions for expectation bias and then when commentators report on their "good" improvement results a degree of confirmation bias is involved.

To support my assertion I will use my own year old professionally administered hearing test results which demonstrate we can have different dB responses in frequency perception between our left and right ears. Meaning that when we EQ based on a frequency chart there is no subjective way to be sure our ears are both actually responding as we presume.

In both the context of dB SPL and SNR (signal to noise ratio) what our baselines in left and right ears were before EQ can diverge. Therefore we can't know whether what we think is an improvement is accurate since IEM and headphones activate both ears, not just the left or right, and what musical production we listen to doesn't always prioritize the same frequencies going to the same ear. Furthermore, one-size-fits-all EQ apparently can accentuate noise by worsening SNR of a particular frequency in one ear relative to the other ear.

I will not attempt to discuss any particular EQ suggestions that have seen on ASR. If anyone recalls a specific EQ proposed they might want to get an idea of what that frequency adjustment would actually cause in the left and right ears of someone with my hearing pattern. So, for any frequency EQ by frequency EQ study, here's my personal geriatric (2022) extrapolated data (source charts below) comparing different frequencies' dB SPL for the right and left ears (tested at 65/55dB SPL input); plus, in brackets, each ear's respective SNR at different frequencies. I wish to point out there does not seem to be a constant integer (mathematical multiple) of either hearing loss (dB SPL) or SNR decline (higher indicates lower noise) at specific frequencies in the right and left ears making it impossible to create a personal EQ ratio for the respective ears.

(At) 1kHz: right ear ("r") = +5 dB SPL [19.8 SNR] vs. left ear ("l" = +9 dB SPL [20.4 SNR]; 1.5 kHz: r = +11 [25.2] vs. l = +10 [23.9]; 2 kHz: r = +10 [27.8] vs. l = +9 [22.6]; 3 kHz: r = +2.5 [[22.2] vs. l = +1 [19.7]; 4 kHz: r= -5 [14.6 vs. l = -3 [14.3]; 5 kHz: r= -5 [20.11] vs. l = -10 [21.7]; 6 kHz: -11 [25] vs. l = -18 [15.5]; 7 kHz: r = -5 [20.9] vs. l = -12 [19.5]; 8 kHz: r = -9 [17] vs. l = -20 [11.4]; 9 kHz: r = -18 [15.6] vs. l = -26 [6.8]; and last data point 10 kHz: right = -20 dB SPL [8.5 SNR] vs. left = more than -35 db SPL [-8.3 SNR]
0248CAEE-4E0A-4396-A626-454BB5596953.jpeg
AE22F1C3-FC4F-4C95-B586-011FF3F30CC5.jpeg
 
Top Bottom