Jump to content


Photo

Pattern recognition experiment


  • Please log in to reply
113 replies to this topic

#101 brycelarson

brycelarson

    Show me the Data!

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 8,588 posts
  • Gender:Male


Posted 13 December 2010 - 03:15 PM

Seeing something when it is difficult = success.
Hearing something that isn't there = failure.


Running away from a tiger that's about to eat you = not dead.
Running away from no tiger = not dead. :)

but yes, this test was designed to provide evidence about why we use math and statistics to analyze our results.

#102 TK-421

TK-421

    Me So Horny! :D

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 25,917 posts
  • Gender:Male
  • Location:Austin, Texas

Posted 13 December 2010 - 03:35 PM


First!

It's a sailboat
Posted Image


I think this is the best response so far.



My answer is the best answer


First!

It's a sailboat
Posted Image


Beat ya to it. :P

#103 The_Economist

The_Economist

    It's business time.

  • BST Moderator
  • PipPipPipPipPip
  • 2,363 posts
  • Gender:Male
  • Location:NW Indiana

Posted 13 December 2010 - 05:38 PM

This was a great experiment. I can't believe no one else heard Christmas music.


#104 MrEeske

MrEeske

    Member

  • Members
  • PipPip
  • 88 posts

Posted 13 December 2010 - 05:53 PM

I had a crazy idea that there was some ulterior motive to this experiment. ;)

I don't mean to sound like a sore loser, but the random noise generator you used isn't very random. Besides being able to see patterns, there really is a noise that can be heard, and I also see that noise's pattern on a melody spectrograph. Every time I hear the sound file, I hear the exact same sound repeated approx. 0.5 - 2 seconds apart (random intervals throughout the sound file) throughout it.

Attached is an image of the melody spectrograph that I was looking at. The noise I'm hearing is outlined in green. If you take a few seconds to look, you'll also see other patterns in the sound, but most of them are "dead spots" where there is no frequency being generated.

[Edit:] Oh, and to explain what you're looking at, the X axis is time, the Y axis is frequency, and the color is amplitude in decibels (black = no sound, blue = quiet sound, red/yellow = louder sound).

Irregardless of all of the above, I think this was a great little experiment, and gave us a cool little bit of experiential learning. :)

Attached Files


Edited by MrEeske, 13 December 2010 - 06:05 PM.


#105 brycelarson

brycelarson

    Show me the Data!

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 8,588 posts
  • Gender:Male


Posted 13 December 2010 - 06:15 PM

I'm sure there's some garbage in there - could be my computer, could be your computer, could be the white noise generator I used. However, I think we can all agree that there isn't:

"starts out in the beginning as waves, and then turns to rushing water"
"the turner almost finding a station"
"the Ukrainian Bell Carol"
"heart beat at around +/- 20 Hz"
"choral arrangement"
"a helicopter"
"a motorcycle reving"
"a train going by really fast on train tracks"
"a sonar beep"
"a dryer with shoes in it"
"a high pitch ring" (prob just tennitis btw :) )
"a blender pulsing, waves crashing, wind (tornado)"
"birds chirping or the stereotypical sound of a squeaky wheel rocking back and forth in a ghost town"
"Helicopter"
"Three quick clicks on right side only"
"thumping in left ear, repeating 5 beeps that sounded like piano in right ear"


there are more, but you get the point. the spectrum anaisys does show a flaw in the white noise - which is exactly why we use tools like that in our tests. The point isn't that there wasn't anything to hear - there was a ton of audio information - simply that the brain wants to find, catagorize and label things. In the cases above those answers may have been related to your aprox 800 hz 3db bumps - but they certainly aren't accurate labels.

That's all.

#106 MrEeske

MrEeske

    Member

  • Members
  • PipPip
  • 88 posts

Posted 13 December 2010 - 06:25 PM

No argument here.

This experiment reminds me of my recent communications class, where much of our education came from experiential learning exercises. In this class we learned how we, and others, subconsciously react to social situations. It was a very eye-opening class, because in these situations, almost everyone responded in extremely similar ways, most of the time without being consciously aware of it.

I think these experiential learning exercises are super cool. :D

[Edit:] Another thought: the source of patterns may be due to the MP3 compression algorithm. That's kind of what the MP3 algorithm does, finds patterns (or close-enough fits) and replaces parts of the sound file with those "close-enough" patterns. If we were give a large WAV file (which wouldn't have been practical), I'm think that (1) I wouldn't hear the noise I'm hearing, (2) the melody patterns visible on the spectrograph wouldn't be there.

Edited by MrEeske, 13 December 2010 - 06:31 PM.


#107 sticktodrum

sticktodrum

    Paintball's Lovable Prince Of Parody

  • Players Club
  • PipPipPipPipPipPipPipPip
  • 11,910 posts
  • Gender:Male
  • Location:Queens, NYC


Posted 13 December 2010 - 06:38 PM

If we were give a large WAV file (which wouldn't have been practical), I'm think that (1) I wouldn't hear the noise I'm hearing, (2) the melody patterns visible on the spectrograph wouldn't be there.

If you're saying that people wouldn't have heard anything, then I have to disagree. This experiment has been done before, with pure white noise. Given the suggestion that there is a hidden audio signal, the mind will create patterns on its own. Nothing in nature is truly random, but that's not to say that there are rules governing the patterns that we assume we're perceiving.

Pareidolia is a well-documented phenomenon among many animals, and it happens with all senses. Visual pareidolia is why we also see odd things in random places (clouds, buttered toast, etc). Dirty white noise or perfectly clean white noise, I can assure you that people would have heard plenty of things. When you're processing a lot of information, the white noise can and does get misinterpreted and rationalized.

Then again, you could be talking just about that noise that appears on the spectrograph, at which point ignore the above.
Gerber, Surefire, TOPS, and Cold Steel all suck the big one. Buy quality products, not junk. :-)

#108 UV Halo

UV Halo

    Bringing the Big Guns to LLVI

  • Members
  • PipPipPipPipPip
  • 1,631 posts
  • Gender:Male
  • Location:Fairfax, VA

Posted 13 December 2010 - 06:52 PM

I certainly agree with the fact of human conciousness creating something out of nothing. One prime example- Hallucinations experienced by individuals undergoing sensory deprivation.

Here's an easy read article on Wiredabout such a thing.

#109 MrEeske

MrEeske

    Member

  • Members
  • PipPip
  • 88 posts

Posted 13 December 2010 - 07:53 PM

Then again, you could be talking just about that noise that appears on the spectrograph, at which point ignore the above.


I was just talking about the spectrograph image and seeing the visual patterns, but I do agree with everything else you said above. The movie Pi, creepy but good, talks a lot about this; they have a conversation about numerologists versus mathematicians, and how numerologists just look for patterns that probably aren't there.

#110 cockerpunk

cockerpunk

    All the Dudes

  • Moderator
  • PipPipPipPipPipPipPipPip
  • 9,116 posts
  • Gender:Male


Posted 13 December 2010 - 08:04 PM

i think its interesting from the standpoint that the human mind is AMAZING at seeing patterns. so much so that it can find patterns where there are none, and make connections and conclusions that simply are not there. i know i constantly must remind myself of how little i really know when talking about anything technical. from the mathematical standpoint white noise is by definition EVERY signal. i have no doubt that many of the sounds you herd you DID hear, the issue is that your mind filtered the white noise and filtered out the conflicting signals with whatever you herd.

an interesting sociological experiment, and a great justification for exactly why we are such sticklers about what the math says vs what we see.

Edited by cockerpunk, 13 December 2010 - 08:07 PM.

The ultimate truth in paintball is that the interaction between the gun and the player is far and away the largest factor in accuracy, consistency, and reliability.

And yes, Gordon is the sexiest manifestation of "to the front."


#111 Lord Odin

Lord Odin

    3 may keep a secret if 2 of them are dead

  • Members
  • PipPipPipPipPip
  • 2,129 posts
  • Gender:Male
  • Location:Oak Lawn, IL


Posted 14 December 2010 - 08:43 AM

Then again, you could be talking just about that noise that appears on the spectrograph, at which point ignore the above.


I was just talking about the spectrograph image and seeing the visual patterns, but I do agree with everything else you said above. The movie Pi, creepy but good, talks a lot about this; they have a conversation about numerologists versus mathematicians, and how numerologists just look for patterns that probably aren't there.


Pi is an awesome movie. :tup:

#112 betasniper

betasniper

    Self Proclaimed Mechanical Engineer

  • Members
  • PipPipPip
  • 381 posts
  • Gender:Male
  • Location:where the sun always shines

Posted 17 December 2010 - 11:44 AM

Then again, you could be talking just about that noise that appears on the spectrograph, at which point ignore the above.


I was just talking about the spectrograph image and seeing the visual patterns, but I do agree with everything else you said above. The movie Pi, creepy but good, talks a lot about this; they have a conversation about numerologists versus mathematicians, and how numerologists just look for patterns that probably aren't there.


The end number of every square'd number makes a patern of 0,1,4,9,6,5,6,9,4,1,0 and repeats. (0,1,4,9,16,25,36,49,64,81,100,121,144,169,196,225,256,289,324,361,400)
╔══╗╔╗────────╔═╗───╔╦╗Zdspb-Marker Animations Long list of working markers, visit if confused
║╬╬╠╝╚╦═╦══╦╦═╣╚╬╦╦╦╣║║
║╔╗╠╗╔╣╬║║║║║╠╬╗║─╣║║║║
╚╝╚╝╚╝╚═╩╩╩╩╩═╩═╩╩╩═╩╩╝
Mechanical Engineer(working on a degree in college) and Machinist(Riverside Community College Norco)

#113 MrEeske

MrEeske

    Member

  • Members
  • PipPip
  • 88 posts

Posted 17 December 2010 - 02:28 PM

The end number of every square'd number makes a patern of 0,1,4,9,6,5,6,9,4,1,0 and repeats. (0,1,4,9,16,25,36,49,64,81,100,121,144,169,196,225,256,289,324,361,400)


And your number pattern is symmetrical, meaning that it is the same in both directions. I bet you could also formulate the pattern as a super position of a couple sine and/or cosine functions. More patterns!

#114 betasniper

betasniper

    Self Proclaimed Mechanical Engineer

  • Members
  • PipPipPip
  • 381 posts
  • Gender:Male
  • Location:where the sun always shines

Posted 18 December 2010 - 03:52 AM

Hmmm.... I wonder what is the true base number system. If we look at octal, those same squares would be:

0,1,4,11,20,31,44,61,100,121,144,171,220,251,304,341,400,441,504,551,620

0,1,4,1,0,1,4,1,0,1,4,1,0,1,4,1,0,1,4,1,0

Hmmm... Hexa Decimal:

0,1,4,9,10,19,24,31,40,51,64,79,90,A9,C4,E1,100,121,144,169,190

0,1,4,9,0,9,4,1,0,1,4,9,0,9,4,1,0,1,4,9,0

Interesting...
╔══╗╔╗────────╔═╗───╔╦╗Zdspb-Marker Animations Long list of working markers, visit if confused
║╬╬╠╝╚╦═╦══╦╦═╣╚╬╦╦╦╣║║
║╔╗╠╗╔╣╬║║║║║╠╬╗║─╣║║║║
╚╝╚╝╚╝╚═╩╩╩╩╩═╩═╩╩╩═╩╩╝
Mechanical Engineer(working on a degree in college) and Machinist(Riverside Community College Norco)




2 user(s) are reading this topic

0 members, 2 guests, 0 anonymous users