We can measure everything we hear: Yes! Research at the Bio-Electronics-Lab of ETH Zurich

[fivepast8]

Hello Fellow ASR Members,

I am posting a video lecture by one of my friends (in memorandum of another friend at Georgia Tech who passed away this spring) that showcases the fantastic research happening today. Start at the 30min mark to see the circuitry deployed to measure neuro-stimulation at the neuron level, including following the firing along dendrites and axon towards neighboring cells. Understand SINAD required for proper measurement

Since hearing is a form of neuro-stimulation and if science and electronics can reliably measure single neuron firing than this should surely lay to rest the argument: ... but we cannot measure everything we hear! Yes we can, even down to the electro-potential along brain cells.

First Oliver Brand Memorial Lecture on Electronics and Nanotechnology, Sound starts after 2min

As most of us in any case believe in and rely on measurements, you will find this little lecture absolutely fascinating.
Enjoy!

Edit: Abstract below added:
"Microphysiological Systems and Highly Integrated Microelectrode Arrays"
Prof. Dr. Andreas Hierlemann Department of Biosystems Science and Engineering, ETH Zürich
Abstract: Recent technological advances in microfabrication techniques and the development of new biological model systems have enabled the realization of microphysiological systems capable of recapitulating aspects of human physiology in vitro with great fidelity. Using microfluidic, microtechnological and microsensor structures and representative in vitro models of human organs, robust microphysiological systems can be developed that accommodate high-resolution microscopy and integrated sensor readouts.
The functional characterization of electrogenic cell preparations can be performed by using high-density microelectrode arrays (HD-MEAs) featuring a very high spatial density (more than 5000 electrodes per mm2) of comparably small electrodes (diameters of 2-5 µm, center-to-center pitch of less than 15 µm). By using CMOS-based HD-MEAs it is possible to obtain comprehensive data sets across scales (subcellular resolution through single neurons to large networks) in various preparations, ranging from organotypic and acute slices to cultures of dissociated neurons and stem-cell-derived neurons. Applications include research in neural diseases and pharmacology.

Bio: Andreas Hierlemann got his college education in chemistry at the University of Tübingen, Germany and a Ph.D. degree in 1996. He held Postdoc positions in 1997 at Texas A&M University, College Station, TX and 1998 at Sandia National Laboratories, Albuquerque, NM, USA. He joined the Department of Physics of ETH Zurich in 1999, where he was appointed associate professor 2004. In 2008, he became full professor in the Department of Biosystems Science and Engineering of ETH Zurich in Basel. His research interests include the development and application of microsensor, microfluidic, and microelectronic technologies to address questions in biology and medicine with applications in the fields of systems biology, drug testing, personalized medicine, and neuroscience.

 

[Punter]

Wow! This tech can actually show the signal running through a neuron! Very exciting

 

[sergeauckland]

Very impressive technology indeed. However, neuron firing isn't hearing. What we hear is the result of the brain's processing, not just individual neurons firing. This is also a part of conciousness, and we're a long way from understanding how conciousness works.

I would love for there to be a clear understanding of how the brain works, down to the molecular level, and indeed, I expect eventually we will know, but it will be some time before I can know what you hear even with the same stimulus.

S.

 

[HarmonicTHD]

Hello Fellow ASR Members,

I am posting a video lecture by one of my friends (in memorandum of another friend at Georgia Tech who passed away this spring) that showcases the fantastic research happening today. Start at the 30min mark to see the circuitry deployed to measure neuro-stimulation at the neuron level, including following the firing along dendrites and axon towards neighboring cells. Understand SINAD required for proper measurement

Since hearing is a form of neuro-stimulation and if science and electronics can reliably measure single neuron firing than this should surely lay to rest the argument: ... but we cannot measure everything we hear! Yes we can, even down to the electro-potential along brain cells.

First Oliver Brand Memorial Lecture on Electronics and Nanotechnology, Sound starts after 2min

As most of us in any case believe in and rely on measurements, you will find this little lecture absolutely fascinating.
Enjoy!

PS: First part of the lecture is about growing cells for drug testing and looking at the blood brain barrier.

Great stuff

I admire your optimism however that this will put the arguments to rest. Always some science deniers left. But it would be nice though.

 

[Keith_W]

Yes, that is very cool ... but as the others have said, what comes out of the Organ of Corti is not necessarily what you perceive. For example, the brain does an excellent job at suppressing sounds that you don't want to hear. Even as you are reading this post, there are sounds around you that you are ignoring - e.g. the computer fan, running water, rustling leaves, and so on. You only hear them if you start actively listening to those sounds. Just like you can not perceive the sensation of your butt on your chair, your arms on your table, and so on until you actively start paying attention to them.

There are a multitude of other examples, such as the Haas effect (or the precedence effect) where your ears might report two sounds, but your brain perceives one. It may even be fooled into localizing the source of the sound incorrectly. Then there is the missing fundamental effect, where your brain reconstructs the fundamental note from the harmonics, even though your ears did not actually hear the fundamental. There are also binaural beats, where if you play 530Hz into the left ear, and 520Hz in the right, what you hear is a10Hz beat. As that Wikipedia article explains, this has to do with further processing in the auditory pathway that produces the illusion.

And then there are all the individual differences in hearing that we probably haven't studied yet. We know that HRTF varies between individuals. I am not aware of any studies that show if all of us localize sound the same way, with the same thresholds for volume and timing, and so on. So - even if we could come up with a complete model of hearing including the neuronal pathways of perception, this model may not apply to all individuals equally in the same way that everybody's HRTF is different.

 

[fivepast8]

Very impressive technology indeed. However, neuron firing isn't hearing. What we hear is the result of the brain's processing, not just individual neurons firing. This is also a part of conciousness, and we're a long way from understanding how conciousness works.

I would love for there to be a clear understanding of how the brain works, down to the molecular level, and indeed, I expect eventually we will know, but it will be some time before I can know what you hear even with the same stimulus.

S.

Yes, that is very cool ... but as the others have said, what comes out of the Organ of Corti is not necessarily what you perceive. For example, the brain does an excellent job at suppressing sounds that you don't want to hear. Even as you are reading this post, there are sounds around you that you are ignoring - e.g. the computer fan, running water, rustling leaves, and so on. You only hear them if you start actively listening to those sounds. Just like you can not perceive the sensation of your butt on your chair, your arms on your table, and so on until you actively start paying attention to them.

There are a multitude of other examples, such as the Haas effect (or the precedence effect) where your ears might report two sounds, but your brain perceives one. It may even be fooled into localizing the source of the sound incorrectly. Then there is the missing fundamental effect, where your brain reconstructs the fundamental note from the harmonics, even though your ears did not actually hear the fundamental. There are also binaural beats, where if you play 530Hz into the left ear, and 520Hz in the right, what you hear is a10Hz beat. As that Wikipedia article explains, this has to do with further processing in the auditory pathway that produces the illusion.

And then there are all the individual differences in hearing that we probably haven't studied yet. We know that HRTF varies between individuals. I am not aware of any studies that show if all of us localize sound the same way, with the same thresholds for volume and timing, and so on. So - even if we could come up with a complete model of hearing including the neuronal pathways of perception, this model may not apply to all individuals equally in the same way that everybody's HRTF is different.

Well said gents, of course I know that neurons firing is not hearing.... , I was making a tongue in cheek reference to the science deniers and pointing out that if we can measure down to such levels and details, surely we can master 20-20kHz signal processing. For the brain to process hearing, surely there are a few million neurons firing! I just love all the cool stuff these guys are doing

 

[Curvature]

And then there are all the individual differences in hearing that we probably haven't studied yet. We know that HRTF varies between individuals. I am not aware of any studies that show if all of us localize sound the same way, with the same thresholds for volume and timing, and so on. So - even if we could come up with a complete model of hearing including the neuronal pathways of perception, this model may not apply to all individuals equally in the same way that everybody's HRTF is different.

I think this raises a few central points. First is that all thresholds are averages and inherently assume variability. The papers investigating thresholds (for localization, see Blauert) experimentally show this in the data... it is very interesting to compare the various attempts to measure equal loudness contours, for example—they are all over the place. And those contours vary at all levels, not just for minimum audible pressure, and shift with age, hearing damage.

Second, a lot of the knowledge about the functional ear-brain physiology comes from animal testing and injuries and genetic disorders that people have. Direct human brain testing isn't performed because inserting probes, even single neuron probes, damages the brain. This has meant that psychoacoustic testing (experiments centered on subjective reports) have had a very important role in the development of auditory science. Though of course there are weaknesses to this approach because you are not testing the specific mechanisms and organs themselves. The major innovation discussed in the talk posted by @fivepast8 is the simulation of organ environments. Which is of course very interesting. I would equate it to the new discoveries about speaker drivers through simulation done by @René - Acculution.com.

Third, and this is very major, there are physiological/anatomical differences, and there are differences in experience, but despite differences, the cortical processing by the brain overcomes these to some extent. In other words, even though everyone has different HRTFs, which to be clear is about physical differences in the shape of your head, face, ears, and body, the functional goal is the same: to understand the aural environment. And so most people, despite their differences, roughly have the same abilities, which is reflected in the ranges found in experimental thresholds. However, what explains the variation in thresholds, whether gross or minor, is found in "aural diversity", referring to experience. The underlying fact is incontrovertible: everyone hears and experiences what they hear differently. We don't have a good sense of the range of that experience, or at least it's hard to map the differences meaningfully, but I would bet that, more likely than not, when we get to the point of more comprehensive measurement and individual aural simulation, whatever differences we uncover will be more or less intuitive and comprehensible.

I read a lot but I have no professional experience to lean on. Make of the above what you will.

 

[Anton D]

Wow! This tech can actually show the signal running through a neuron! Very exciting

This could go along was in support of subjectivism.

(Joke!)

 

[Keith_W]

Actually, your post reminds me of an experience I had many years ago when I was working in a remote community with Australian aborigines. I was taken on a hunting trip by one of them, and we were barrelling down an unsealed road in a Landcruiser. To me, the landscape was whooshing past but the driver was on the lookout for kangaroos to hunt. He suddenly braked and said "there is a goanna!" (a goanna is a type of lizard). Needless to say, I couldn't see anything. He pulled out his rifle and shot it. He said "Got him!" I still couldn't see anything. We walked up to it, and sure enough there was a dead goanna.

He told me later that city people can not see the way they can. Apparently they are accustomed to viewing the world as a horizontal plane, and they are very good at spotting vertical lines. He can also hear much better than me - later, when we were sitting around the fire, he could hear his friends coming. I heard nothing apart from the crackle of the fire and people talking. He pointed at the distance and sure enough, there was a tiny pair of headlights coming our way. I could only hear the 4WD when I saw the headlights. Needless to say, his survival skills were on another level and I would surely be dead if I tried to venture that far out without bringing my own supplies.