The Puzzle of Perception Page 2

Without sufficient information at your brain's fingertips for the map to be formed, it's hard to make sense of what you hear. I had firsthand experience of this when I was 14. I went to live in Paris for a few weeks to improve my command of French. Upon arriving, however, I was horrified to learn that despite my theoretical knowledge and the fact that I could read written French, I couldn't understand a word anyone was saying. It was impossible to detect the starts or ends of words; all I heard was a continuous string of mellifluous sound that had about as much meaning for me as birdsong. Only after a week or so did I start to comprehend what I heard, meaning gradually imposing itself on disordered noise. Similarly, the first time I heard the Vaughan Williams "London" Symphony, it was a confusing experience---the work sounded like an apparently arbitrary collection of meaningless musical noises (footnote 5). The second time I listened to the work, it began to come into focus, an evocative fragment of a folk tune here, a previously used phrase there. Eventually the work, with all its implications and richness, became familiar and well-loved.

When CD was introduced nearly 10 years ago, many audiophiles declared it to sound "perfect." And perfect it indeed sounded to them: When you are presented with something completely outside of your previous experience, you simply do not hear anything wrong. The wrongness doesn't fit into your map, so is ignored. Only after familiarity breeds awareness do you both hear what is wrong and develop a vocabulary to describe it.

To judge the quality of a hi-fi component, therefore, involves assessing the quality of the acoustic models it enables the listener to create. Too often, however, by focusing on the minutiae of those models---where and how well defined they are in space, how accurate is the midrange tonality, how much bass they have, how high the highs---audiophiles forget to examine the bigger picture: does the sound convey the musical meaning? It is an essential rule of thumb when judging hi-fi equipment that the better component is one that lets you get into the music more easily, that opens your ears to new music, and not the one that meets arbitrary standards of audiophile excellence (footnote 6).

This essay is veering dangerously near the semantic swamp of musicality vs euphony vs accuracy. I'll raise your blood pressures, therefore, by disinterring Ivor Tiefenbrun's decade-old statement that bad hi-fi components make it harder for listeners to follow the tune, and good components make it easier for them to hum along with the tune (footnote 7). Surely a "tune" is an abstract concept, you ask, its very lack of physical reality implying an immunity to the slings and arrows of outrageous hi-fi components. However, rather than the subject of the debate being something as well-defined as a "tune," it concerns the perception of "pitch," something very different.

Many people confuse pitch with frequency, regarding the two terms as interchangeable. Frequency is an objective property: it is both measurable and one-dimensional, in that it represents the number of soundwave crests passing a certain point every second. Pitch is wholly subjective, however. Not only does its perception differ from individual to individual---how many people do you know who are either "tone deaf" or who have "perfect pitch"?---it is multidimensional, frequency being only one of the physical parameters affecting its perceived value (footnote 8).

It is related non-linearly to loudness, for example. Play a 440Hz sinewave through your hi-fi system at a moderate level; you will hear a note with the pitch accepted as that of the note A above middle C. Increase the volume, however, and above a certain threshold (you may need headphones to get sufficient level) the pitch will droop. The frequency of the note won't have changed, but its pitch will have gone appreciably flat! (Before the advent of visual tuning machines, many rock guitarists would play woefully out of tune: standing in the strong soundfield given out by their stack, they would hear themselves as flat. They would therefore tune sharp of concert pitch, something noticeable to everyone else in a less intense soundfield.)

Listen to someone talking (footnote 9). You will notice that the pitch of their voice changes constantly, there being small up-and-down inflections that underline the meanings of the words. Now listen to a speech recording on poor loudspeakers. You will find that those natural pitch inflections are diminished, perhaps even absent. The voice acquires a more monotonous quality, its naturally varied formant structure being corrupted by drive-unit and cabinet resonances that pull the pitches of the recorded voice toward their own preferred frequencies.

When a Linn devotee asks you to "hum along with the tune," therefore, he is neither suggesting that a specific tune can be turned into another by a less-than-ideal piece of hi-fi equipment nor that hi-fi components are capable of changing the frequencies of the signals they process. Instead he is asking you to project your map of perceived pitches back on to reality, so that you can compare the fit. The places where it doesn't correlate with the absolute sound will be thrown into sharp relief. As Tony Gregory put it in The Absolute Sound, "Humming along with the tune is a straightforward attempt to get people to relate to reproduced music as music, in terms of one of its most basic characteristics, and to evaluate it that way."

Sounds natural. Doesn't it?

Footnote 5: I mention this work in particular because, to my surprise, I found it much less accessible than Mahler's almost contemporaneous Das Lied von der Erde.

Footnote 6: "[The] problem with focusing on...peripheral qualities of reproduced music [is that] one easily forgets to keep track of whether the sound remains faithful to the most fundamental qualities of music, like the tune, or even the beat."---Anthony M. Gregory of Audiophile Systems, The Absolute Sound, Issue 32, December 1983, pp.9-11.

Footnote 7: See "Views," Hi-Fi News & Record Review, February 1983, pp.21-22, and "How To Judge a Hi-Fi System," available from Audiophile Systems, 8709 Castle Park Drive, Indianapolis, IN 46256. Tel: (317) 849-7103.

Footnote 8: For a full discussion of the subjective nature of pitch perception, see Diana Deutsch's article (which includes a disc of examples) in the September 1983 issue of the Journal of the AES.

Footnote 9: I refer to a speaker of an Indo-European language such as English. In theory, speakers of Japanese emphasize their meaning by changing the word endings or by adding special particles, not by pitch inflection.


dalethorn's picture

We see upside down, so the brain has to turn that around, and then everything including motion that goes with it. We have 2 ears that each hear what they hear, then the brain has to interpret that as spatial information. Military exercises teach "off center" seeing at night, since when staring at something in near darkness it usually disappears. I've always assumed there's an audio "blind spot" that defeats a lot of blind testing of very subtle differences.