rhythmboy

The Ears, Hearing and Perception Thread

21 posts in this topic

Ear%20Anatomy.jpg  ear_anatomy.gif

CochleaDiagram.jpginner_ear.gif

Call me strange but I think the ear is a small miracle of bioengineering, and I've long been fascinated by the human ear and how it functions - not to mention the notion of how the brain processes and interprets sound and music. Hearing damage is also a burning issue for all of us who live, work and play in loud environments every day.

So I thought it would be cool to have a thread devoted to such a fundamental part of our lives and careers.

I'd like to gather in one place everything we post to do with:

- Ear anatomy and ear problems

- Hearing, hearing tests, tinnitus, deafness

- Perception, Cognition, Analysis

Similar to the Compressors thread, I'll paste links and summaries to all current and future threads about ears and hearing dotted around the place here in the first post. There is some real gold being buried away in old pages!

Hope to make it easy for people to find answers and advice  :D

________________________________________________________

Robbie's community announcement:

MUSICIANS AND PRODUCERS - YOUR EARS ARE THE MOST VALUABLE PIECE OF KIT YOU WILL EVER HAVE.

THEY ARE LITERALLY PRICELESS AND EFFECTIVELY IRREPLACEABLE.

IF YOU WANT - OR HAVE - A CAREER IN MUSIC, TREAT YOUR EARS WITH THE GREATEST RESPECT:

KEEP IT DOWN

WEAR PLUGS

REST IN SILENCE

Thank you!

________________________________________________________

INTERACTIVE LECTURE AND PRAC ON EARS AND HEARING (Rhythm Boy) (0 replies)

Uni of Sussex Lecture notes and Excel prac exercises by Chris Darwin

http://www.soundpunk.com/index.php?topic=993.0

________________________________________________________

HEADPHONES AND HEALTHY EARS (TankF) (33 replies)

Good advice on headphone monitoring levels, SPL meters and headphones, bass perception in headphones 

http://www.soundpunk.com/index.php?topic=915.0

________________________________________________________

HEARING TESTS (echosystem) (3 replies)

How to interpret the results of your hearing test 

http://www.soundpunk.com/index.php?topic=872.0

________________________________________________________

HEARING (echosystem) (4 replies)

Advice and experiences of frequency response in hearing

http://www.soundpunk.com/index.php?topic=769.0

________________________________________________________

EAR PROBLEMS (Amz-Star) (46 replies)

A progressive account of poor Amz's journey through sinus infection into an inverted eardrum

http://www.soundpunk.com/index.php?topic=785.0

________________________________________________________

EAR FATIGUE (Vagrant Producer) (9 replies)

Advice on what causes ear fatigue and how to reduce it

http://www.soundpunk.com/index.php?topic=596.0

________________________________________________________

AUDITORY EMAIL LIST (Rhythm Boy) (8 replies)

Link to an email list for news on hearing and perception (very academic/psychological)

http://www.soundpunk.com/index.php?topic=239.0

________________________________________________________

COMPOSITION AND COGNITION (Spazmatron) (0 replies)

Link to a great article/essay on "The Craft of Musical Communication"

http://www.soundpunk.com/index.php?topic=950.0

________________________________________________________

WHAT IS TIMBRE (Rhythm Boy) (5 replies)

Discussion on what we define the 'timbre' of a sound to actually mean, and how we use it

http://www.soundpunk.com/index.php?topic=228.0

________________________________________________________

AUDIO ANALYZERS (Rhythm Boy) (27 replies)

Reviews of people's favourite Audio Analysis programs

http://www.soundpunk.com/index.php?topic=264.0

________________________________________________________

PROFESSIONAL TRAINING IN ACOUSTICS (Captain Terrific) (6 replies)

Advice and links for training in acoustics

http://www.soundpunk.com/index.php?topic=642.0

________________________________________________________

Share this post


Link to post
Share on other sites

HOW TO READ THE 'FLETCHER-MUNSON' EQUAL LOUDNESS CONTOURS

thought this would be handy as a getting started guide for those who know this picture is important but don't quite know how to interpret it:

FletcherMunson_EqualLoudness_thumb.jpg

The Fletcher-Munson curves are recognised around the world as the best approximation of humans hear loudness we currently have. Originally devised by Harvey Fletcher and WA Munson in 1933, it has since been updated and standardised by the International Standards Organisation (ISO) and is defined as ISO:266. The preferred term today is Equal Loudness Contour, but 'Fletcher-Munson' will be understood by most audio and sound folk

http://en.wikipedia.org/wiki/Fletcher-Munson_curves

The ISO version is here in red, with original Fletcher-Munson curves in blue:

256px-Lindos4.svg.png

The curves basically draw a direct correlation between Sound Pressure Level (SPL) and Loudness. There is a subtle but significant difference between the two.

Sound Pressure Level measures actual changes to Air Pressure as result of vibrating sound waves passing through it over a given area. SPL is measured in decibels, with 0dB arbitrarily defined as air pressure at with a force of 2 x 0.00001 Newtons per square meter at sea level.

In simple terms: 0dB = still air = silence

At the other end, we have the pain threshold, which is around 120-130dB, depending on which textbook you read. Thing is, pain is a variable quantity in people so some leeway is excused. So called because at this level we literally experience pain in our hearing and logically is another good benchmark. We can certainly measure SPL higher than 130dB but in relation to hearing it serves little benefit.

Because air pressure itself is a measurable quantity, we can also accurately measure SPL, using the good old SPL meter.

Loudness is the term given to our perception of SPL - in other words, how the ear and brain actually hear the changes in air pressure in a sound wave. For sine waves, loudness is measured in Sones. For complex waves (far more common) we measure loudness in Phons. Neither are actually used very much. SPL in dB is far more common.

Problem is, measuring how someone perceives sound is much less accurate. In sound experiments especially, researchers must rely on verbal responses from test subjects simply telling them what they perceive. Believe it or not, the original Fletcher-Munson curves were derived from listening tests on about 75 people! From this was defined a 'world standard'!

The other problem with loudness is our ears do a thing called  Temporary Threshold Shift - where after exposure to loud noise over time, the eardrum tightens up and the ear naturally compresses the incoming sound to prevent damage (for a while and to a point - overdo it and damage will occur). So our own perceptions change constantly too.

However, the Equal loudness contours do a pretty good job of helping us understand how we hear differences in SPL. But more importantly they allow to understand how loudly we hear different frequencies. This is much more valuable to us than just measuring 'how loud'. Why? Because everyday sounds and music are made up of thousands of frequencies that constantly change. In the world of speech, our ability to detect certain frequencies helps us understand each other and identify people in a crowd. In the world of music and audio, knowing how we respond to frequencies affects everything from mixing and composing to speaker and instrument design.

How to read the graph

fletcher_curve.jpg

It's pretty easy once you know how to look. I find the easiest way is to choose a curved line and trace your way along it. Each curved line represents a given level in Phons, from 10 to 120. The dotted line shows the Threshold of Audibility, or the lowest loudness possible for a given SPL and frequency. Below the dotted line we can't hear anything.

The horizontal axis is frequency from 0Hz to 20kHz. I've highlighted a few points on here for reference.

The vertical axis is SPL from 0dB to 130dB. I've circled a few values here too for reference.

Now let's look at the blue line I scrawled across 80 Phons:

- Starting near the middle at 1000Hz, we can see that 80 Phons = 80dB SPL as we trace the red line across to the dB scale. Looking at all the other lines we see a similar pattern. So rule number 1: At 1kHz the loudness level in Phons = the equivalent SPL. The curve has been deliberately calibrated this way to set a standard.

- Tracing the blue line across to 4000Hz, we see now that 80 Phons = 70dB SPL. Again the pattern is the same for other Phons levels. This tells us something about rule number 2: at 3-4kHz our perception of loudness is most sensitive. We need 10dB less SPL to hear 4kHz at the same loudness as 1kHz. 70dB SPL is about 5-6 times lower than 80dB, so it's a lot!

Some of the sensitivity to 3-4kHz is thought to come from the length of the ear canal, which naturally resonates at those frequencies. The extra resonance creates and internal reinforcement in the ear canal.*

- Tracing the line across to 9-10kHz, we see that the curve has spiked up, so now 80 Phons = 87-88dB SPL. So rule number 3 is: at 9-10kHz our perception of loudness is far less sensitive. In fact it is nearly ten times less sensitive than at 4kHz.

- Tracing the line to the left, we see a huge curve up as the frequency goes down. At 20Hz now 80Phons = ~112dB. This is whopping 42dB more than at 4kHz. Rule number 4: at low frequencies our perception of loudness is least sensitive. The lower you go, the more SPL you need in order to hear it at the same loudness.

This why bass speakers often need so much more power to drive them than mid-range speakers. Power generates SPL, and the simple truth is you need more power for bass. It's an irreversable law based on the response of our ears to SPL.

A few real-life observations to finish with:

- *Speech is the easiest of all things to hear, being centred largely on the middle frequencies. Our ears are thought to have evolved their shape and size to optimise listening to speech.

- Putting hi-shelf EQ boosts on mixes is so tempting because of your relative insensitivity to 9-10kHz :D

- the 'Loudness' button on hi-fi systems intentionally boosts (typically) 100Hz and 10kHz so at low levels you can still hear some bass and top end. Note that at 40-50dB SPL (TV turned down lowish), 50Hz is below audibility.

Hope this helps and makes sense to you! If there are any inaccuracies please let me know and I'm happy to correct!

Share this post


Link to post
Share on other sites

Of course, absolutely tops, mate. I apologise for checking so much good stuff on here via my phone, and never get back to reply via a real keyboard.

Questions I have:

re: "0dB = still air = silence"

How many adults can hear a fraction above 0dB? I expect I'd be pushing it. I know I'm fatigued when I'm groaning about the 'hisss' noise through my headphones to discover I've yet to plug them in. ;)

re: "the brain interprets sinusoidal waves more pleasantly than square waves and noise"

But isn't every wave made up of a collection of sine waves? :P

And this thread reminds me I must get back to constructing my play-doh ears to measure headphone SPLs.

May each day this week be filled with more plur than then last.  :-*

Share this post


Link to post
Share on other sites

re: "0dB = still air = silence"

How many adults can hear a fraction above 0dB? I expect I'd be pushing it. I know I'm fatigued when I'm groaning about the 'hisss' noise through my headphones to discover I've yet to plug them in. :D

My understanding is that very few humans, even with perfect or "child grade" hearing, can hear variations < 3dB. It's more commonly 5-6dB variations that people pick up, no matter how good they claim their hearing is.

Share this post


Link to post
Share on other sites

re: "the brain interprets sinusoidal waves more pleasantly than square waves and noise"

But isn't every wave made up of a collection of sine waves? :D

Correct Spect, I must clarify. A more accurate statement would be:

The brain interprets relatively simple waves with strong harmonic partials more pleasantly than very complex waves with inharmonic overtones. I.e. musical sounds and consonant musical intervals are more pleasant to the auditory nerve and the brain than noisy distorted sounds.

Proper tute with pics and audio coming, but in essence:

- Sine waves generate a sequence of impulses from the auditory nerve to the brain that is so simple and predictable the brain finds it unexciting and annoying. The visual equivalent would be staring at a plain dot on a white wall.

- Musical tones and consonant intervals (octaves, fifths, 3rds, etc) create interesting but synchronous impulses in the auditory nerve. The impulse patterns are simple enough for the brain to 'understand' them easily, but interesting enough to excite the brain too. The visual equivalent would be looking at a neatly landscaped garden.

- Harsh musical tones and dissonant intervals (2nds, 7ths, etc) create more complex, asynchronous impulse patterns in the auditory nerve. The complexity makes the brain work harder to make sense of the sound and it can become tiring. The visual equivalent of looking at an Escher picture or an optical illusion. But like the illusions, one can train the brain to understand and appreciate the complexity and even learn to enjoy it. Atonal/experimental music  is often called 'head' or 'cerebral' music because it makes the brain think harder, and to the untrained this is tiresome. But to the learned it can be more enjoyable than simple tunes (which become as predictable as sine waves)*.

- Noise creates a barrage of complex, random impulses in the auditory nerve. This is so complex the brain cannot make sense of any one frequency and instead blurs it's comprehension across the whole sound and hears it as one. If the noise is constant the brain will in fact disconnect from it and stop hearing it after a while - until it changes and our perception is excited again. Noise that is strong in the middle frequencies can excite the eardrum and basilar membrane so much that it overloads and hurts, further distorting the impulses to the brain - this can confuse us and make us angry and annoyed. The visual equivalent would be staring at an endless sandy desert in midday glaring sun. Ouch.

* Reminds me of a time In 1993 when I was staying with a family in England who had a six year old daughter with autism - her 'specialty' was words - she couldn't speak, but she would speed-read books and could type with a vocabulary of a postgrad literature student. Very complex mind. Being a bit of a hack on guitar, each evening I would play songs to her which she really enjoyed. Interesting though - if I played children's songs and nursery rhymes she'd start crying and screaming unstoppably. If played fast arpeggios, adult songs and classical pieces she'd smile and want more. Her brain was naturally attuned to complex music, at an age when other kids are still singing nursery rhymes. In fact, simple music made her upset and angry. It was my experience with her that first got me interested in this area.

Share this post


Link to post
Share on other sites

THE AUDITORY NERVES – PART 1 – PITCH AND FREQUENCY PERCEPTION

Much of what is written below is referenced from the site below, and is a lay person's attempt to put into plain English. Any inaccuracies found please let me know!

http://www.lifesci.sussex.ac.uk/home/Chris_Darwin/Perception/Lecture_Notes/Hearing2/hearing2.html#RTFToC10

Cochlea-crosssection.png

THE ORGAN OF CORTI – THE EAR’S ANALOG TO DIGITAL CONVERTER?

I think this is pretty cool - the pink section in the middle of the pic called the Organ of Corti. For me this is where my previous discussion left off - here I think is the key to what I'm talking about. The Organ of Corti, from what I understand, is responsible for converting the wave motion of swaying hairs on the basilar membrane into neurotransmissions sent down the auditory nerves. No point talking about the nerves until we look at this 'A to D converter' of sorts.

It seems primarily the Inner Hair Cell Nerves do the conversion. They sway in sync with the basilar membrane and generate a tiny electrical voltage in doing so. This voltage fires neurotransmissions down the Cochlear Nerve Fibres. These fibres are bundled together as the Auditory Nerve, actually a spiralled 'rope' of nerve fibres connected to (via some other organs – I’ll bypass for now) the Auditory Cortexes in the brain.  The chemical Glucamate is used to fire the neurons to the auditory coretexes. There is an auditory cortex on both sides of the brain, and information from each ear is sent to both sides. This helps us determine which direction sound is coming from as the brain processes all sound in stereo. Cool!

CHARACTERISTIC AND CODING FREQUENCIES

Each cochlear nerve fibre is particularly sensitive to certain frequencies, called Characteristic Frequencies. Like pipes in a church organ, it's like they are tuned to fire at specific frequencies, and specific loudness thresholds. Multiple frequencies spaced widely apart will fire off nerves some distance from each other, and we perceive two tones easily. Multiple frequencies in a similar band will potentially fire off neurons in adjacent nerve fibres, and even suppress others, causing the brain some confusion. We hear beats, dissonance, vague pitch perception, etc. This is very closely aligned with the Critical Bands found on the Basilar membrane.

The brain determines any frequency being heard by encoding the frequency from two sets of information. The first is where nerve fibres are being excited – the stimulating tone matches the characteristic frequency of a particular nerve as outlined above. The second information is timing – when the neurons are being fired. This introduces the concept of Phase Lock.

PHASE LOCK AND PITCH PERCEPTION

What's interesting to me is the synchronicity of all this - the frequency of motion in the basilar membrane is basically the same as the frequency of neurons firing to the brain. Taking it even further back:

Sound waves (air pressure waves) -> Ear drum & middle ear motion (mechanical waves) -> cochlear & basilar membrane (fluid waves) -> Organ of Corti (electrical conversion) -> cochlear nerve fibre (neurotransmission) all operating at the same frequency and in perfect phase.

At frequencies below 3kHz, the firing of neurons is actually phase-locked to the basilar membrane - in other words, the swaying hairs of the basilar membrane move in perfect sync with the inner hair dells, and in turn the firing of neurons. This helps us determine pitch in low and mid range quite well.

At frequencies above 3kHz, the neurons are decreasingly phase locked, and we begin to approximate pitch with decreasing accuracy the higher up we go. The Inner Hair Cells cannot move fast enough to accurately reproduce the required voltage changes – the faster they move, the worse it gets.

In the following example I have a sequence of sine waves ascending in pitch at 500Hz, 1000Hz, 2000Hz, 4000Hz, 8000Hz and 16000Hz. This equals an octave pitch change with each frequency.

[mp3=200,20,0,center]http://www.soundpunk.com/downloads/home/RBstuff/sine500-16k.mp3[/mp3]

Note that for the three frequencies below 3000Hz your perception of the octaves is quite accurate. However, in the three frequencies above 3000Hz, the perception of octaves gets thrown out - especially at 8000Hz, which appears to bend up in pitch a little. And how many of you can hear 16kHz? ;)

What you are hearing is the phase-lock of your neuron pulses drifting out of sync with the basilar membrane.

PITCH PERCEPTION AND LOUDNESS

Finally, a quick experiment to do with pitch perception and loudness. Our perception of pitch alters as the loudness of a tone changes.

Play the following 20 seconds of sine wave at 500Hz, and fade it up from virtual silence to maximum. Sweep the level up and down smoothly. You’ll notice it lowers in pitch as it gets louder. It is argued that there is an optimum amplitude threshold for each nerve fibre. When the amplitude goes below or above a certain threshold, our frequency perception is either suppressed or saturated respectively.

[mp3=200,20,0,center]http://www.soundpunk.com/downloads/home/RBstuff/500hz20sec.mp3[/mp3]

Given this, imagine a guitarist tuning by ear at low levels, then tuning again at loud levels. It’s likely at loud levels to be tuned flat. Imagine a singer overdubbing with really loud headphones on, and the engineer monitoring at moderate levels. The singer may well sound out of tune to the engineer, but sound fine in their own headphones. Hmmm?

Also, mixing at loud levels alters your perception of the actual pitch of the music. Again, a loud mix sounds flatter in pitch than a quiet one. Think of that next time you are pitch correcting something :P

So this is all good for sine waves, but what about real sounds? In part 2 I’ll try to cover a couple of issues to do with processing complex tones. First, what happens to the nerves when more than one tone is heard, and then why some intervals sound nice and others sound harsh.

Share this post


Link to post
Share on other sites

Fascinating as always, RB!

At normal listening levels, right now I can't hear 16kHz ;)

That pitch perception vs loudness is an, er, ear-opener. Just like exact time, there's no such thing as exact pitch, as far as us mere humans are involved.

Share this post


Link to post
Share on other sites

Has anybody tried Moulton Golden Ears?

http://www.moultonlabs.com/page/cat/Product/

I couldn't hear 16KHz either, the 8Khz sounded sharp but then again are these mp3 or wav? I'm also listening through crappy earbuds.

An 8Khz sine wave @ 44.1Khz Fs is about 5.5 samples per cycle, coupled with a dodgy reconstruction filter in the onboard sound card may also have an effect. I should try and look at it with an analog CRO when I get home.

Cheers

Share this post


Link to post
Share on other sites

Has anybody tried Moulton Golden Ears?

http://www.moultonlabs.com/page/cat/Product/

I couldn't hear 16KHz either, the 8Khz sounded sharp but then again are these mp3 or wav? I'm also listening through crappy earbuds.

An 8Khz sine wave @ 44.1Khz Fs is about 5.5 samples per cycle, coupled with a dodgy reconstruction filter in the onboard sound card may also have an effect. I should try and look at it with an analog CRO when I get home.

Cheers

16kHz is there but very low in level. Yes I agree some degradation due to mp3 conversion and few samples per cycle at hi-freq could have an impact. I achieved the same results when playing the 24 bit wav's in real-time before bouncing down, so I don't think mp3 has had much effect. Can't speak for the digitisation - look forward to seeing what analog signal does, please report back if you run the tests!

In any case the audio I supplied merely verifies what the literature says, not the other way around ;)

One of my teachers at uni has done the golden ears system and is pushing for all staff and technology students in our dept to do it too. I really should, just lazy or busy!

Share this post


Link to post
Share on other sites

Perhaps a little esoteric for such a serious topic but anyway.

Ear candling, is it bullshit? Someone recommended it to me saying that the before and after difference is quite startling and that you hear things with much greater clarity afterwards. Anyone had it done? I think it is essentially deep cleaning of earwax, which I understand the ear produces in response to fatigue/damage as a sort of buffer or protection for the ear, is this true?

Share this post


Link to post
Share on other sites

Also, when I had my ears tested for work (call centre) I was sad to see a huge dip at 4khz. Glad to see it's normal!

Share this post


Link to post
Share on other sites

Ear candling, is it bullshit? Someone recommended it to me saying that the before and after difference is quite startling and that you hear things with much greater clarity afterwards. Anyone had it done? I think it is essentially deep cleaning of earwax, which I understand the ear produces in response to fatigue/damage as a sort of buffer or protection for the ear, is this true?

http://altmed.creighton.edu/ear/experiment.htm

BullShit.gif

Share this post


Link to post
Share on other sites

Perhaps a little esoteric for such a serious topic but anyway.

Ear candling, is it bullshit? Someone recommended it to me saying that the before and after difference is quite startling and that you hear things with much greater clarity afterwards. Anyone had it done? I think it is essentially deep cleaning of earwax, which I understand the ear produces in response to fatigue/damage as a sort of buffer or protection for the ear, is this true?

I'd imagine the benefits , if any, come from lying with your head sideways for several minutes, thus dislodging any excess wax build-up  ;)

Otherwise it looks downright dangerous in the hands of an amateur and ineffective otherwise. The site Spectrum quoted shows overwhelming evidence against the practice, and the evidence makes perfect sense.

The World Health Organisation has stipulated that the most effective way to keep ears clean of excess wax and free of bacteria is plain old salt water. A sterile saline like Audiclean is all we need  :)

Share this post


Link to post
Share on other sites

So, the other night I couldn't sleep because of noisy people outside so I put in the old faithful earplugs to get some z's. Next day, I'm deaf in my right ear and it feels kinda "full" like I went up a mountain & it never 'popped'. I'd felt kinda flu-ish, & I've had this feeling before so I thought it would go away.

It didn't, so my girlfriend dragged me to an ear doctor a couple days later. After inspecting my ear he preceeded to pull a bunch hairs from my head out of my ear canal, that were pushed up against my ear drum!

The whole experience kinda freaked me out, of course I value my hearing so much (like everyone else), its really something you don't wanna fuck wit!

Share this post


Link to post
Share on other sites

^ Wow, lucky it was something so straightforward. Also amazing how something so small can feel so big - like a sore tooth  :-

When I was six I had one of these stuck - pointy end in - against my eardrum:

1939808380_aed7161191.jpg?v=0

It was about to germinate in there. I had to go under general anesthetic to make sure I didn't move or in case they punctured the ear drum trying to get it out - it scared the bejjesus out of me  :wtf:

Considering the surgical technology of the early 70's, I could have been deaf in my left ear at six. I can only imagine what turn my career would have taken.

Share this post


Link to post
Share on other sites

That's really interesting rhythmboy, especially the information about pitch perception and loudness. I never knew that, but will keep note of it.

I was able to hear the 16 in the first clip, but I'm still in my teens (just) so I'm sure in a few years time I'll lose that.

Also in response to your experience with the seed nearly germinating; I heard a story about a child who hid in a barrel of seed and was complaining about an ear ache. Turns out there was wheat growing inside his hear. Crazy stuff

Share this post


Link to post
Share on other sites

I just discovered this thread and I have to say that the content is very interesting. Thanks for sharing!

And thank god that bug got removed, sounds incredibly nasty

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now