Pseudoscience in audio - Milind N. Kunchur

[botrytis]

Actually, hearing is the most analog device one could think of. All of our senses are analog, actually.

[fas42]

15 hours ago, mocenigo said:

 

There are no tiny receptors in the hearing apparatus that are impacted by air molecules. There is a membrane moved by air pressure which is sensed by some bones, and these move the cochlea which is filled by a liquid that hits about 25000 nerve endings. It is the latter that in a sense “digitize” the signal.

 

Fair enough. The process is indirect; but ultimately what the brain receives is a 'digitised' version of what the membrane responded to - it's not a lovely analogue, electrical, 'waveform', as those who are in love with such, would prefer .

[mocenigo]

5 minutes ago, botrytis said:

Actually, hearing is the most analog device one could think of. All of our senses are analog, actually.

I agree. My “digitizing” comment was tongue-in-cheek.

[fas42]

https://physoc.onlinelibrary.wiley.com/doi/full/10.1113/JP279189,

 

Quote

 

Abstract

Ribbon-class synapses in the ear achieve analog to digital transformation of a continuously graded membrane potential to all-or-none spikes.

 

 

But then again, what would someone who has an article published in The Journal of Physiology know ...

 

[fas42]

Think I'm on a roll here ...

 

Curious, I looked further, and found, https://crev.info/2012/12/eye-retina-is-analog-to-digital-converter/

 

Quote

 

A new study by opthalmologists at the University of Tübingen, Germany has confirmed that the eye goes digital.  Bipolar cells at the base of the retina, long thought to send continuous analog signals to the retina, have been shown to generate action potentials, or spikes, that represent on-or-off conditions (the basis of digital programming).  The paper in Current Biology is technical; the interesting part is in the press release from the university, entitled “The end of a dogma: Bipolar cells generate action potentials.”  The article explains the advantages of digital processing:

Action potentials allow for much faster and temporally more precise signal transmission than graded potentials, thus offering advantages in certain situations.

 

 

So, the new question is, are any of the human senses analogue ... ?

 

[pkane2001]

16 minutes ago, fas42 said:

Think I'm on a roll here ...

 

Curious, I looked further, and found, https://crev.info/2012/12/eye-retina-is-analog-to-digital-converter/

 

 

So, the new question is, are any of the human senses analogue ... ?

 

 

I was looking for an I2S connection to hook-up streaming services directly into my brain. Bit-perfect playback all the way through would be nice! 😎

 

 

[botrytis]

I have one in my Pioneer Elite Universal player. We could try it.

 

3 hours ago, pkane2001 said:

 

I was looking for an I2S connection to hook-up streaming services directly into my brain. Bit-perfect playback all the way through would be nice! 😎

 

 

 

[Kal Rubinson]

4 hours ago, fas42 said:

Think I'm on a roll here ...

 

Curious, I looked further, and found, https://crev.info/2012/12/eye-retina-is-analog-to-digital-converter/

 

 

So, the new question is, are any of the human senses analogue ... ?

 

This is a superficial "pop" article written for the general reader and not a scientific article.  It is chock full of plainly incorrect statements that would evoke gasps (or chuckles) from anyone who works in the field.  In particular, the use of the term "digital" is sloppy and, imho, panders to the non-scientist.  

[pkane2001]

41 minutes ago, Kal Rubinson said:

This is a superficial "pop" article written for the general reader and not a scientific article.  It is chock full of plainly incorrect statements that would evoke gasps (or chuckles) from anyone who works in the field.  In particular, the use of the term "digital" is sloppy and, imho, panders to the non-scientist.  

 

And I was so looking forward to feeding DSD1048576 directly into my brain!

 

[Kal Rubinson]

10 minutes ago, pkane2001 said:

 

And I was so looking forward to feeding DSD1048576 directly into my brain!

 

Mebbe but not via the retina (which is, technically, part of the CNS).

[pkane2001]

3 minutes ago, Kal Rubinson said:

Mebbe but not via the retina (which is, technically, part of the CNS).

 

I would be OK with an optical feed. It might offer a better noise rejection 😉

 

[botrytis]

 

[fas42]

6 hours ago, Kal Rubinson said:

This is a superficial "pop" article written for the general reader and not a scientific article.  It is chock full of plainly incorrect statements that would evoke gasps (or chuckles) from anyone who works in the field.  In particular, the use of the term "digital" is sloppy and, imho, panders to the non-scientist.  

 

Let's go to the, ahem, proper thing then ... https://www.cell.com/current-biology/fulltext/S0960-9822(12)01315-2#Summary

 

Hmmm,

 

Quote

Moreover, ≥three fastest groups generated clear all-or-nothing spikes.

 

Last time I looked, 'digital' circuits in electrical hardware all over the place do nothing more than generate "all-or-nothing spikes" of analogue voltages; which, luckily, most of the time are understood as meaning data ... maybe we should just give up on using the word "digital", and go back to considering all PCM, etc, as really being analogue, .

 

Edit: Noting that the article was cited 160 times ... which may not be enough for some in the science community, perhaps ...

[kumakuma]

8 minutes ago, fas42 said:

Last time I looked, 'digital' circuits in electrical hardware all over the place do nothing more than generate "all-or-nothing spikes" of analogue voltages; which, luckily, most of the time are understood as meaning data ... maybe we should just give up on using the word "digital", and go back to considering all PCM, etc, as really being analogue, .

 

Or you could consider electricity itself digital as electrons move from one atom to the next.

 

Perhaps there is nothing analog in the universe.

 

It's all digital with just the speed of the "all-or-nothing" states changing.

[Kal Rubinson]

Unfortunately, it is simplistic to equate the "all or nothing spikes" of action potentials to a digital signal.  Those in the field appreciate that there are phenomena, such as presynaptic inhibition, that can modulate the effect of individual APs.  The distinction between analog and digital in the CNS is not as clear cut as it is in electronics.

[Kal Rubinson]

8 minutes ago, kumakuma said:

Or you could consider electricity itself digital as electrons move from one atom to the next.

Interesting but is the movement of an electron significant.  If not, it is not a meaningful signal.

[fas42]

I note that the phrase "ON and OFF cells" is also used to describe this behaviour - as in this article, https://www.cell.com/neuron/pdf/S0896-6273(20)30399-8.pdf

 

Quote

Here, we perform large-scale multi-electrode array recordings
from human retinas to characterize luminance contrast re-
sponses of ON and OFF midget and ON and OFF parasol gan-
glion cells and use computational modeling to analyze informa-
tion encoding of these cells in our natural environment

 

Gosh, this is sounding mighty "digital", to me ...

[Kal Rubinson]

12 minutes ago, fas42 said:

Gosh, this is sounding mighty "digital", to me ...

Or just a matter of polarity.

[Speedskater]

Note:

 Kalman Rubinson
Professor Emeritus of Neuroscience and Physiology at New York University School of Medicine

 

This is his area of expertise.

[Jud]

9 minutes ago, Speedskater said:

Note:

 Kalman Rubinson
Professor Emeritus of Neuroscience and Physiology at New York University School of Medicine

 

This is his area of expertise.

 

@Kal Rubinson, there's a topic I've been interested in for a while on which I haven't been able to find good peer-reviewed journal articles, and wondered whether you might point me toward any. I recall reading without citation that there were auditory cortex neurons sensitive to timing (transients, differences in arrival time) distinct from those sensitive to frequency. Any thoughts on where I might find some good published papers on the subject?

[Kal Rubinson]

35 minutes ago, Jud said:

 

@Kal Rubinson, there's a topic I've been interested in for a while on which I haven't been able to find good peer-reviewed journal articles, and wondered whether you might point me toward any. I recall reading without citation that there were auditory cortex neurons sensitive to timing (transients, differences in arrival time) distinct from those sensitive to frequency. Any thoughts on where I might find some good published papers on the subject?

It is not really my field (or was) but the way to find the stuff is the way you find anything else, a search that drives from source to source. 

Just searching for "auditory cortex timing" should get you started.  It took me straight to https://www.pnas.org/doi/10.1073/pnas.1012656107 and the references in that paper might provide a useful path for investigation.

 

The problem for non-academics is getting access to the original papers instead of relying on watered-down reports in the popular press.

[Jud]

1 hour ago, Kal Rubinson said:

 

The problem for non-academics is getting access to the original papers instead of relying on watered-down reports in the popular press.

 

Yes, though there's a browser extension called "Unpaywall" that often finds free versions, for example in Bioarxiv.

 

Thanks for the link. 👍🏼

[fas42]

The current "debate" is because people are so hung up, on separating "digital" from "analogue" ... as I said in the post which triggered the latest round of posts, "Reality is always a bit more complicated ..." - the computer memory you're using right now to read this, which most would consider as being as digital as it gets, is so analogue in its true nature that it almost instantly forgets what is stored - the proverbial goldfish is way more advanced in its, er, capabilities. The only reason that it works at all, is that it is constantly 'reminded' of what was there a fraction of time ago, at a furious rate ... as mentioned in that post, . And we worry about humans remembering a snippet of sound over some parts of a minute ...

 

Things generally work because the best "value for money" mechanism, either through evolution or because human designers figure out a "better way" over time, ends up winning against alternatives - audio could be the same, but most outsiders wouldn't see it that way, going by the 'ridiculous' cost of most of the apparatus supposedly needed for optimum listening ...