Playing sounds backwards

You can play a video backwards and still recognise the scenes quite well. Can you do that with sound?

I tried it on this Bryan Adams clip of Summer of ‘69 (mp3). When played backwards (mp3), it almost sounds like Arabic!

Instruments sound weird backwards too, like the guitar played backwards and drums played backwards.

It’s seems obvious once you see the wave file. The picture below shows the guitar. The sounds are clearly not symmetric left to right.

Whereas this guitar is a lot more symmetric, and doesn’t sound too different backwards.

So how come we can’t recognise sounds played backwards, but can recognise video played backwards? (Initially, I thought it was a trivial question. But I couldn’t find a trivial answer. The question may be subtler than it looks.)

Comments

B anand Prasad 25 Sep 2006 12:35 pm: its nothing to do with the technology Mr. Anand let me give an example dnana ….you just spell it… it sounds something like (d-na-na) now if we know that reversing it could(possibly) result in Anand — dnanA, then we could recognize all backward sounds its just that we r not acustomed to hear sounds backward. now practice daily some words which are often used like “good morning” “thanks” now say them to someone in a correct situation…90% of them will recognize it.. thats it my friend….sorry for reversing “Anand” name….its my name too

S Anand 25 Sep 2006 1:48 pm: Anand, if I’ve understood you right, we don’t recognise sounds backwards because we’re not used to hearing them. Agreed. My reasoning was: we’re not used to seeing things backwards either. How come we’re so much better at recognising those? A part of the answer is, we recognise sounds mainly by their movement in time, but images by their spread across space. But I wonder if there’s more to it…

ND 26 Sep 2006 3:07 am: I think this may have to do with the fact that seeing consists of detecting multiple facets like motion, pattern etc. The brain may have more data points decompose and “meaningfully” analyze backward. While hearing consists of analyzing the frequencies and amplitude implying lesser data points to “meaningfully” analyze backward with the same speed as visual stimuli. Caveat - Just my guess!

S Anand 26 Sep 2006 6:23 am: Hmm… ND, that’s a possibility! Certainly, in terms of information content, there’s more coming in visually than aurally.

ND 27 Sep 2006 12:42 pm: I asked three intelligent folks (an american, a german and an indian) to read “eye”, “racecar”, “malayalam” and “never odd or even” and waited for their reaction. Got blank stares! But it was sitter when explained. Guess some humans find it difficult to recognize a palindrome while reading and speaking. Taking a few million huge steps to infer humans normally normally don’t even try to process sound or reading alphabets backward. Is it easier to recognize a popular word like one’s name or a tune like Happy B’day when spoken or sung backward? A related front and back concept - I have always wondered why the bar showing s/w installing moves from left to right. And the same when uninstalling also. Is it convention? Why not show something more perceptive like a container getting filled or getting drained?

Sudeep Nair 20 Nov 2006 5:40 am: Anand, I feel that the fundamental difference between a reversed text and reversed sound is that the text is still completely visible to the eye at once, and so there is a subconscious process where the brain is able to decipher the text by trying out some permutations, whereas in case of sound, the individual notes reach the ears and disappear. They do not remain “static” or “stationary” for the brain to be able to try some reverse permuations of individual notes and identify the original sound. This reminds me of another observation that in case of English text, the reader is able to decipher the words if the first and last letter of every word is kept fixed and all other letters in between are jumbled. Msot raeedrs wlil be albe to udnreatsnd tihs snetncee atlohugh the lteerts are jbulmed.

Anonymous 20 Nov 2006 6:15 am: If we extend this logic from text to video, I think there is some degree of “persistence of vision” which allows the brain to “refer back”, unjumble and make sense of video images, but there is no persistence of sound. So in addition to anand prasad’s point about “familiarity of words”, and ND’s point about “more meaningful data points”, I would think that “persistence of images” is a major factor helping us understand visual images better than sound.