anyone familiar with audio cognition or ecological psychoaco

kensuguro · Post by **kensuguro** » Wed Mar 23, 2005 5:31 am

I'm working on a final paper and was having trouble finding information on audio cognition. I'll try to be more specific. This is for a class in cognitive science. Cognitive science has a model that says we think with visual images, and there's a lot of study of how we make mental representations (or internal representations) of images. But there's so little research on audio. This is probably because Gestalt research was that of vision.

The type of sound I want to deal with are non-speech, non-musical sounds, leaving us with generic audio, or every day sounds. Psychoacoustics deals with this, but the study is quite old. Much like other science that emerged during the early 20th century, psychoacoustics centers around trying to measure things. While measuring frquency, intensity of audio, and figuring out how these aspects got encoded into neural code, psychoacoustics has little to say about what happens to the audio once it is encoded, or inside the brain.

But instead of focusing on the chemicals and electric interaction between neurons, I wanted to deal with audio from a cognitive scientist's point of view. This means I would define how audio is represented in the mind, and what computational procedures we can perform on it. Basically, I'd work at a conceptual level, instaed of the highly physiological level of psychoacoustics.

I was wondering if anyone had books or papers that I could use, becuase there is not a lot of research in this direction. There's tons of pysiological info, but so little about what reall goes on in the mind. I think cognitive science provides the scientific and philosophical freedom that lets us deal with the mind, while psychoacoustics rules it out as an unwanted side effect. (becuase it's not physiological)

If this sounds interesting to you, do a search for Stephen McAdam's papers. You can read some of it online. It'll be interesting for us musicians to discuss this. Leaving scientists with lab coats to the research is somewhat scary.

I'm looking into the ways of:
Bregman, "Auditory Scene Analysis"
Neuhoff, "Ecological Psychoacoustics"
and various papers by Stephen McAdams

bosone · Post by **bosone** » Wed Mar 23, 2005 5:54 am

hi ken!
from the little test i'm currently doing (you can find the topic here in PlanetZ, some days ago), this field is extremely interesting.
i have no precise aim on my test, nevertheless it's emerging that, almost on some people, the brain "interprets" the audio signal that is coming and can even "create" things that are not even present in the recording.

(but mine is a comment from a point of view of one who doesnt have the appropriate backgound to judge)

"what" do our brain listens is actually a VERY complex subject...

kensuguro · Post by **kensuguro** » Sat Mar 26, 2005 11:22 pm

yeah, I'll make sure to check your experiment out. Sounds very interesting!

paulrmartin · Post by **paulrmartin** » Sun Mar 27, 2005 7:28 am

On 2005-03-23 05:54, bosone wrote:
...the brain "interprets" the audio signal that is coming and can even "create" things that are not even present in the recording.

The Mamas and the Papas were never happy with their vocal work until "Harvey" showed up. Harvey was an imaginary voice that they said they heard in their harmonies when they sang perfectly together.

I'll check out what McGill has, Ken.

paulrmartin · Post by **paulrmartin** » Sun Mar 27, 2005 7:53 am

Nice place to start here ->http://ccrma-www.stanford.edu/~malcolm/ ... minar.html

The contact list is quite nice if you read the messages

This is a nice link as well:
http://www3.sympatico.ca/philippe_auber ... stics.html

Check out this link for a book by a McGill professor:
http://www.amazon.ca/exec/obidos/ASIN/3 ... iacoust-20

<font size=-1>[ This Message was edited by: paulrmartin on 2005-03-27 08:01 ]</font>

braincell · Post by **braincell** » Sun Mar 27, 2005 11:16 pm

That could have been an LSD induced audio hallucination.

The Mamas and the Papas were never happy with their vocal work until "Harvey" showed up. Harvey was an imaginary voice that they said they heard in their harmonies when they sang perfectly together.

I'll check out what McGill has, Ken.

at0m · Post by **at0m** » Mon Mar 28, 2005 12:51 am

Ken wrote:
[...]measuring frequency, intensity of audio, and figuring out how these aspects got encoded into neural code[...]

K.U.Leuven works with this device that exists of a microphone with built-in chip, connected to the aural nerve(s). It mimics the function of the ear and translates it to the nerve system.

http://www.esat.kuleuven.ac.be/...

Cochlear implants are hearing aids or the profoundly deaf in which the auditory nerve is stimulated electrically. The electrical stimulation strategy tries to minimize the difference between the nerve response of an acoustically stimulated ear and an electrically stimulated ear. It is known that fractal effects are present in normal ear functioning. In our modelling study we show that ...

Not totally what you're after, but searching a bit more on this could lead to some physics information, instead of the psychological theories.

kensuguro · Post by **kensuguro** » Mon Mar 28, 2005 3:06 am

paul-> Yeah, my research did lead me to Stephen McAdams, who is at McGill. He's one of the most cutting edge researchers of this field, and is spot on with his approach. A bit too hard core for me to understand fully, but his approach is very well balanced between physiology and cognition, and is bound to make some much needed bridging between psychoacoustics and cognitive science.

Man, it's my first term paper and it seems like I've opened a can of worms!

emzee · Post by **emzee** » Mon Mar 28, 2005 3:45 am

Ken, you're really going out into new territory here. Are you sure you need to be so cutting edge.. the field is still quite small. Sounds like your paper could end up as primary research.........

kensuguro · Post by **kensuguro** » Mon Mar 28, 2005 11:12 am

I'm pretty sure it's going to be primary research.. or actually the paper will stop in pointing out that there needs to be more primary research in this field. I may continue on this topic even after the term paper though, because it calls for so much work to be done.

It's going to be in 2 parts, one part overviewing what we already know from psychoacoustics, and the later half presenting a hypothesis showing how what we know from psychoacoustics can translate into a cognitive scientific model of mental imagery. Mental imagery covers all senses, but most research is on vision so far. It's interesting because cognitive science has a model for how people "think" (problem solving, planning, learning, etc) in visual images, and I can draw some analogies to that in audio. Psychoacoustics rules out "thinking" in experiments because it's too complex to deal with. (and I always thought psychoacoustics covered all of this)

So it's either psychoacoustics starts looking into the mind, or cognitive science needs to get some researchers doing auditory imagery. Both of which is hard because we know so little about the physiological aspects of hearing. Much less than vision.

Oh yeah, did you all know that vision is so well studied because the series of studies by the gestalt scientists focused on vision?

Also, you'll have to excuse my thinking out loud. I think it's good for me because it really shapes up my thought. I mean, I am kind of just blurting things out.

paulrmartin · Post by **paulrmartin** » Mon Mar 28, 2005 12:00 pm

Ken, look into the newer models of hearing aids. Their construction(i.e. bandpass filters helping cognition) may have a hint as to what you're aiming at.

Just a shot...

emzee · Post by **emzee** » Mon Mar 28, 2005 6:38 pm

I was told by a photography student that people think visually...perhaps you could consider the different types of intelligence...

It's probable that people with high musical/aural intelligence would have to be treated differently within your paper. Just a thought.

mr swim · Post by **mr swim** » Mon Mar 28, 2005 9:24 pm

On 2005-03-28 11:12, kensuguro wrote:
cognitive science has a model for how people "think" (problem solving, planning, learning, etc) in visual images

bear in mind that this model itself is going to be quite metaphorical ... we don't have tv screens in our heads, nor do we have frame-by-frame cognition (if that makes any sense ... the idea is that we don't have linear processing on linear perceptions). Sounds trite I know but it is hard to get out of this way of thinking (from my own experience) - and treating it too literally can sometimes be detrimental further down the line.

I tend to think of it like this: perception comes in, then gets 'conceptualised' (parts 'interpreted' as instantiating properties etc.). Then it is the conceptualized information which can be structured and used in thought. I think this could have a straight analogy with aural processing (just think of linguistic processing, for instance). So it is dangerous to overstress the 'image' part precisely because this biases one to visual perception.

Interestingly, one of the reasons why I think models of cognition ARE biased to image-type models is simply because so much more research (as you point out) has been done into vision.

Which is also why your research is so exciting ! MUCH more needs to be done. Especially (for my interests) because language is so so important to us, and its main sense (in normal cases) is auditory.

Good luck with the paper,

Will.