Harvard study of time keeping and groove

Caz

Senior Member
Hi all, I thought some of you might be interested in these studies of time keeping tendencies for musicians and non-musicians.

http://news.harvard.edu/gazette/story/2012/07/when-the-beat-goes-off/

After reading this one and finding it interesting, I contacted and met up with Holger to find out more since we were both in the same place. A while later I ended up as one of the test bunnies for his next set of experiments, which has now been published:

http://www.nld.ds.mpg.de/~holgerh/articles/Hennig_2014_PNAS.pdf

Some really interesting things came up.

Caroline
 
Despite the claim 'Every beat a single (noninteracting) layperson or musician
plays is accompanied by small temporal deviations from the exact
beat pattern, i.e., even a trained musician will hit a drum beat
slightly ahead or behind the metronome (with a SD of typically
5–15 ms) of musical interaction' and Taylor Beck says: 10 to 20 milliseconds, one would think a huge discrepancy in the millisecond world, each and every person executes musical time differently.

Ones fractal patterns are IMO what constitute ones 'feel', the 'something that can't be taught' aspect of music. Like snow flakes, no two people execute exactly alike.

Then there's that millisecond world again, trained/pro musicians are consistently in that 5-15 ms club.

Drummers by default are training that yet to be determined part of the brain that meters the millisecond world.
 
I'd love some researchers to have a closer look at certain people around me that are before or after the beat by a whopping 500 milliseconds or more... 😵
 
I'd love some researchers to have a closer look at certain people around me that are before or after the beat by a whopping 500 milliseconds or more... ��

I had a bass player that had terrible delay with her playing, she wasn't 500 ms off but she was probably around 125-250 ms delayed.
 
all I could think of while reading was....

... I cannot believe someone wasted this much time & effort and I hope no money was wasted doing these studies
 
Interesting studies. The PNAS study (hardly bed time reading) is not for the uninitiated to stats and bioscience !

Many have commented how in general pre-click track music of decades ago has more feel and sounds better (less sterile, or more pleasing) than post-click track material. Some of the science here could explain that.
 
Of COURSE people are turned off by the attempted humanization of computer generated music!! And I have a suggestion for Holger Henning, Eric Heller and anybody else researching the most effective way to go about it: If you want to 'humanize' the music, how 'bout we add an actual 'human'? Then we can all just break for lunch to discuss how so called 'computer generated music' is an oxymoron (at best), as we continue to take your research funding and grants; and we'll all laugh and laugh . . . .
 
All this computer wizardry to duplicate what's already here, human time.

I rebel against loops, clicks and backing tracks. The minute I have to play live with a headset on, I'm done. Not for me.

Last night I felt like a racehorse who needed to run, but all I could do was tappity tap out schmaltzy Christmas songs. I really needed to cut loose and actually play but I felt like I was chained down. The looper was in full force last night. I developed a F it attitude when I went off. TBH, I don't care if I go off anymore. F that thing. I resent anything that tries to subtract the human element from songs.

I have to say, Tony's advice really helped me last night. I did what he said and just shut up and played.

Shut up and play....I really needed to hear that because my mouth gets me in trouble. Unfortunately my face is a dead giveaway when I'm not happy.

Lying helps too. "Hey Larry you alright?"

Yea! everything's great!

Lie.
 
best line in the article:
"the drummer plays ahead of the beat for 30 consecutive beats, while half a minute earlier, he tended to play slightly behind the metronome clicks. These trends are pleasant to the ear.”

I didn't know they played around with the exact mathematical (machine) time of electronic music, in what they called 'humanizing' it. Yeah improving that aspect sounds ridiculous. But some of the point of the research is understanding patterns of rhythm in the mind/body.

People worried about money spent on this - its trivial relative to some politician's yearly bonus. This kind of study is super cheap - some computers and free samples ( students) at a Uni. Lots of applied research builds itself on pure research done many years prior....the cathode ray tube, oscilloscope, LED.....
 
all I could think of while reading was....

... I cannot believe someone wasted this much time & effort and I hope no money was wasted doing these studies

Many scientific studies look like a waste of effort.... until someone gets a brilliant idea and commercialises a bagless vacuum cleaner, a nuclear reactor or a drum computer that sounds like a real drummer because it's off by a couple milliseconds in a pattern that's agreeable to the ears.

Next thing you know they'll analyze the secret of love. It's pretty scary :)
 
Of COURSE people are turned off by the attempted humanization of computer generated music!! And I have a suggestion for Holger Henning, Eric Heller and anybody else researching the most effective way to go about it: If you want to 'humanize' the music, how 'bout we add an actual 'human'? Then we can all just break for lunch to discuss how so called 'computer generated music' is an oxymoron (at best), as we continue to take your research funding and grants; and we'll all laugh and laugh . . . .

Well it fits in the 21st century trend of inventing something that replaces a million people's jobs so that one person gets insanely rich and a million others scratch their heads and wonder what to do now. Happens all the time now.
 
I think you're all missing the point of the research, it's not humans trying to replicate human time in a computer but determine why human time is better. Why does a chaotic, barely noticable imperfection sound more pleasant than a perfectly timed track.
 
I think you're all missing the point of the research, it's not humans trying to replicate human time in a computer but determine why human time is better. Why does a chaotic, barely noticable imperfection sound more pleasant than a perfectly timed track.

I don't think determining why human time is better is an objective by itself. I lost my pink glasses long ago. The objective is to make a lot of money with that knowledge.
 
I think you're all missing the point of the research, it's not humans trying to replicate human time in a computer but determine why human time is better. Why does a chaotic, barely noticable imperfection sound more pleasant than a perfectly timed track.

I agree with PICODON on this.

Its not the point of the research, but its the results, should they at all prove anything workable that will be be exploited for profit. Who wouldn't want to buy the better sounding drum machine? But that can't be the basis for getting the grant money of course.

Its pretty simple why a human generating music is more pleasing compared to a computer, there's 1000's of years of conditioning. A computer on the other hand would probably prefer computer generated music.

Computers currently remain hard at work trying to condition younger humans to accept their generated music as more pleasing.

So, if computers get their way, their music will be accepted as 'better'.

Generating lightning, mother nature is still king. Crunching numbers, computers rule. Enslaving populations with fiat currency, humans.
 
I lost my pink glasses long ago. The objective is to make a lot of money with that knowledge.

A cynical if not paranoid view. This is just curiosity driven research, like a lot of other scientific research.....not Big Brother trying to make a buck.
 
A cynical if not paranoid view. This is just curiosity driven research, like a lot of other scientific research.....not Big Brother trying to make a buck.

Isaac Newton did curiosity research, unfortunately now days, all research takes $.

Not big brother, little brother in search of grant money- the welfare of academia.
 
Interesting Caz. Never thought of it but errors would be fractal. Within long phase tempo variance trends there will have to be smaller phases with their own trends, and so on down to the microtiming of each note.

I agree with others that research it sure beats some of the things we spend so much on, like tax breaks for corporations.

It will add much complexity but it looks to me that the next step in the study is to take into account the time relationship with bassists. Also, you'd see changes that don't relate to any personal cycle but that the song is doing.
 
Back
Top