Yes but is it valid ?
You see the 10kHz dip is only there on certain industry standard fixtures and peoples hearing may well have that at different frequencies.
Looking at a dip there for this does not make sense. If it would simply be a dip at a certain frequency then one would simply notch any headphone at that frequency and magically a depth or width or image stability would occur. This does not happen (I tried) at all so it isn't 'the dip'.
The pinna activation is likely to have something to do with it. But it is also pinna shape, angle and driver size/distance dependent.
Just have a look at different ears and how they sit on ones head. That can differ a LOT from the ones on test fixtures.
So... it is highly unlikely one can 'measure' the effect on some fixture and make that the same for individuals.
Nah... I'd like to see Harman type research in this.
I saw some reviewers claim a 'dip' in the 1kHz-5kHz range can create the illusion of being 'further away' which might bring an illusion of 'depth' for some people.
Making that change in a HD650 (EQ it to HD800S), however, does not magically turn it into a 'wide soundstage headphone) so is unlikely to be a tell tale sign.
But as said, not everyone hears 'depth' in headphones, not even with some trickery. How do you measure that and quantify it ?
obviously not, but I also believe that reducing a test to distortion and frequency response is like underestimate a much more complex topic.
The dip frequency, like you said, may be different for everyone, each head is different, each ear is different, and each canal is different, but when you place audio source in front, or above, each person is perfectly able to detect the direction, even in anechoic environment where there are no reflections.
This is due on how the pinna works, and if what you say is true, each person may have different perception of sound coming from different angles, not only with phones, but also in real life sounds.
So, there should be a common denominator that makes people pinpoint sound even with different ears, maybe the reflections may vary in intensity, but frequency range for perception is mainly the same for everyone despite different anatomy.
I also would not take into account a poll on how people hear the headstage because they are not controlled, and we don't know how they wear the headphones, thing that is mainly important for something like soundstage (as you also stated).
I quote from that thread:
HD 800 (S): Slightly in the back of my head with ultra-wide panning
How do I know he wears the phones to much behind the ear?
That's why we need controlled tests for something like this.
Changing the frequency response of the headphone not working is a good point, but due to the nature of the headphones, I think that also reflections should be taken into account, and that's why probably PRTF alone may not be enough.
It would also not be enough because each person have different shape, but at the end ears works exactly the same for everyone and that curve is just a model to try to evaluate things, and not to say that curve is the same for each person.
It would be very interesting to test PRTF on totally different shapes (constructed on real people anatomy) and see the differences, then compare also with those real people perception of the sound.
Anyway, that said, I underline again that I don't belive that PRTF is bulletproof, but I also don't believe that frq resp and thd may be enough, moreover, considering that there is a lot of science behind headphone spatialization that really works creating realistic sound from behind, there is not a way to understand and measure this topic.
Anyway again, what I hear when swapping to balanced is nothing related to what has been said above, soundstage is not that different, and, above all, frequency response is not altered.
It's more about imaging and focus, better boundaries of the instruments and control of the frequencies (mainly bass), it's something more related to things like group delay or crosstalk
