May 2016

Sex & Audio
A clear case of discrimination!
Article By Peter Mapp, PhD, FASA, FAEs
Article reprinted from Sound & Communications, March 2016

  Now, before the lawyers among you start thinking of "kerching" and of potential added income, it's not that sort of discrimination that I am talking about (nor probably that sort of sex, either!). Instead, I am referring to the results of various studies that have been carried out over the years that show clear differences between how males and females listen and perceive sound, and also audio system performance. 

In general, it would seem that, for the female of the human species, less is more, in that women generally prefer lower listening volume levels and less seasoning, (i.e., bass and treble) than men. It is also interesting to note that women generally exhibit less hearing loss than men, although I would stop short at saying that they actually have better hearing, because this would require a complex definition of what "better" means in this context.

However, it is interesting to note that most studies of basic hearing perception or hearing loss indicate clear differences. For example, by the age of 35, most men will have a 10dB loss in their hearing at 8kHz, while most women do not exhibit such a loss for at least another 10 years. This differential carries on so that, by the age of 45, most men will have lost 20dB of high-frequency hearing acuity while, at a similar age, most women will have only lost 10dB to 12dB. (Interestingly, other studies have found that married men seem to have a selective hearing loss corresponding to the female speech range (although this seemed to vary with activity), but that's another story entirely, albeit with some physiological, let alone psychological, basis.)

Now, where is this sexual digression leading? For a start, it might help to explain why I have noticed on many occasions that I prefer a higher volume when listening to hi-fi or movie sound tracks than female co-listeners. This then begins to beg the question whether we need separate dBm and dBf sound level scales for male and female listeners, although perhaps the dBA (Average or Androgynous?) could work for both. (I will let you figure out for yourselves what dBH might cater to).

The point of interest to me is how large these sexual differences are, and should they be considered when designing or operating a sound system of pro audio and hi-fi equipment? Equally, it begs the question whether movie theaters and other places of audio-based entertainment should have segregated audiences of male and female listeners, although perhaps the segregation should be ageist not sexist!

The reason I have recently been considering the sexual connotations of audio relate to the fact that movie sound levels are a point of both academic and practical interest to me, and, also, I have been considering buying a new pair of headphones. The latter factor may seem to be a bit of a non-sequitur, but I will explain.

This question comes to mind: "What should the frequency response be for an ideal or good pair of headphones?" Whereas for loudspeakers, this has been pretty much known for many years (i.e., a flat anechoic axial response together with a smooth off-axis frequency and power responses). Furthermore, the final, optimal "in-room" response has been reasonably well established, at least for small and medium-sized listening rooms. To the surprise of many, this is not a flat response but has a sloping characteristic with gradual reduction in level from low to high frequencies, as shown in the chart below.

Figure 1

When it comes to headphones, the situation would seem a lot simpler but, in fact, it is actually rather more complex. On the one hand, compared to rooms, our ears are pretty small (acoustically) and physically well-defined, so there's really nowhere for the sound to go: All you get is the direct sound and no "room" reflections. However, the volume of the ear canal comes into play, and a strong resonance due to this enclosed tube occurs. The shape of the ear itself can affect response both for circumaural and supra-aural headphones.

Equally, the simple act of just putting on, taking off and then putting back on a set of headphones can result in quite marked differences in the measured and perceived frequency response. Closed- or open-backed formats also have a profound influence on the perceived sound and image, with open-backed types tending to be more spacious with less of an "in the center of your head" feeling, though these are more prone to the effects of the local ambient background noise.

But, getting back to sex and an ideal world, what should the optimal response for a pair of good quality headphones be?

For a start, the response must be measured and set at the eardrum, because this is the only fixed reference point. The effects of the ear canal resonance and diffuse field collection of sound by the pinna also have to be accounted for... but then what? Should the response be flat or should it have some specific characteristics, e.g., bass or treble emphasis (or de-emphasis)? Until comparatively recently, there has been little research and virtually no agreement on this.

A few years ago, however, Dr. Sean Olive at Harman turned his attention to the subject and made some fascinating discoveries. Essentially, it would seem that there is a general preference for a smooth response that notionally follows the high quality "loudspeaker in a room" listening response, as illustrated in Figure 1, though with a little less bass.

Equally interesting was that the current generation of MP3 and personal stereo listeners prefers a smooth response to one with significant bass or treble emphasis. Compared to in-room listening, the same audience, when listening to headphones, preferred around 2dB less bass and 3dB less treble. Interestingly, untrained listeners preferred 1dB to 2dB more bass and 2dB to 3dB more treble than experienced/trained listeners.

And now for the sex bit! Olive's latest findings showed that the female listeners prefer around 1dB less bass and 2dB less treble than their male counterparts. Listener age also had an effect, with younger listeners (15 to 25 years old) preferring around 2dB more bass and fractionally more treble than older listeners. Although these relatively small differences of 1dB to 2dB do not seem like much, anyone who has adjusted the EQ of a sound system by just 1dB will appreciate the significant effect that this can have.

The fact that, last year, about 300 million earphone/headphone units were sold worldwide shows the significance of this market and the need to get the sound right, although advertising and peer pressure often negate the science. However, as with many things in life, it is interesting to find that flat is not necessarily the best choice (beer being a particularly good example.) Now, alcohol, sex and audio, that could provide some interesting potential for research!

About The Author
Technical Council member Peter Mapp, PhD, FASA, FAES, Principal of UK-based Peter Mapp Associates, is a chartered engineer and physicist, and Fellow of the ASA and AES. Mapp was named the 2009 NSCA University Educator of the Year and received the IEC 1906 award in recognition for his work on intelligibility standards and research. He received the AES bronze medal in 2014 in recognition of his significant contributions. Contact him at pmapp@testa.com.

Enjoy the Music.com would like to thank Sound & Communications for allowing us to reprint this article. Please visit their website at www.SoundAndCommunications.com.


Copyright © 2016 Sound & Communications