I'm not advocating for Griesinger's view myself, but FWIW it's largely to do with phase as opposed to amplitude.
The idea is laid out in this
powerpoint presentation, and some more detail can be found in the papers on his
website. I think his ideas need to at least be taken seriously until they are experimentally tested (it's been a few years since I looked into this though, so the memory is a little hazy now).
EDIT: just found another interesting link, this time a
video.
I'm glad you posted this. I drift back to Griesinger's ideas once in a while. He's basing his points on Blauert's
Spatial Hearing. Specifically, for low frequencies, spaciousness occurs when ITD fluctuates between 3Hz and 10Hz. What Griesinger calls spaciousness is also called localization lag, blur or strength elsewhere if you want to look for more references. This idea came out of testing the effects of lagging sounds like reflections and masking (inability to localize precisely can be interpreted as a kind of post-masking), and the attempt to establish a minimum audible angle (MAA) on the azimuth (in the lab, about 1° horizontally and roughly 4° vertically). At low frequencies, since we are relatively insensitive to level changes but sensitive to the timing information, the key variation will be interaural timing difference or ITD.
To be completely clear, he is not saying that you can hear 3Hz. He's saying that any musical signal will have fluctuating levels and phase: the effect of this fluctuation at low frequencies translates into a type of semi-localization during which we cannot precisely pinpoint the sound but are aware that it is external, rather than being centered on the listening position in a mono setup.
He has a few further points about why this fact is not more well-known:
- Recording engineers have generally mono'ed the bass for various reasons (limitations of vinyl and FM radio, effects of phase on definition and clarity when mixing basslines and drums).
- Studios are designed to have mono bass. I'm not completely clear on his reasoning, but it's to do with low reverberation times and mode excitation with subwoofers.
- Home listening systems are not set up to allow stereo bass (e.g., single sub, corner placement, instead of two subs directly at the sides of the listening position).
But, as you said, I don't know of studies which have quantified spaciousness/envelopment specifically. Surround sound literature will likely have good sources.
Griesinger also has this Dropbox link to his test signals for spaciousness, the minimum criteria for which is a wandering sound rather than a fixed one. I'd be curious about your impressions.
I made a dropbox folder which has the audio examples, and the .wav files for a test CD that has test tones in 1/3 octave frequency bands. These are very useful for testing a room for stereo bass.
https://www.dropbox.com/sh/0xxrfcseeexrizn/AAB9nwIFNJgtYIwQ490g5txta?dl=0
I'm personally interested in this since I'd like to understand the limits of stereo precision. I've read before that localization for broadband sounds is also a function of spectral differences (this is how monoaural cues work, for example, when the sound source is directly in front of the listener), and that high frequency cues in general may be a better indicator of location than low. Though since these comments were more to do with studies of free-field acoustical events than localization control or reproduction in stereo, they aren't strictly speaking relevant.
