The following graph, G-9 from the OSHA PEL Sound Standards, is interesting. It shows the equivalent sound levels in terms of exposure damage risk - basic a frequency “contour map” - kinda like a topo map. The contours dip around 4K, meaning frequencies around there are considered most damaging.
Perhaps two things are going on. High frequency sound is more energetic than low frequency sound because you are packing more amplitude peaks into the same time frame. But the human ear canal is most resonant aroud 3.5 KHz, according to ReSound literature on feedback. So perhaps the dip in the contours around 4 K is like feedback in reverse. It’s easiest for sound to efficiently get into the ear around 4 K and after that as the sound rises to higher frequencies, there is less efficient delivery of the sound to the ear (is my guess) so the contours rise again.
https://www.osha.gov/laws-regs/regulations/standardnumber/1910/1910.95
Equivalent sound level contours. Octave band sound pressure levels may be converted to the equivalent A-weighted sound level by plotting them on this graph and noting the A- weighted sound level corresponding to the point of highest penetration into the sound level contours. This equivalent A-weighted sound level, which may differ from the actual A-weighted sound level of the noise, is used to determine exposure limits from Table 1. G-16.