Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
How the brain perceives rhythm (news.mit.edu)
72 points by benbreen on Jan 8, 2017 | hide | past | favorite | 33 comments


Without a more cross-cultural study this falls more into the domain of cultural bias (still interesting) than what the title suggests. I'd love to see what the results look like incorporating data from cultures whose musical traditions are rhythmically complex and better documented. Godfried Toussaint's work on Euclidean Rhythms (http://cgm.cs.mcgill.ca/~godfried/publications/banff.pdf) makes for a pretty convincing survey of "evenly spaced" beat distribution, albeit sometimes over uneven lengths. This doesn't even get into things like the polyrhythm and syncopation in much African music or the complex rhythmic cycles found in Indian percussion.


That's a good point, and we are on it: http://journal.frontiersin.org/article/10.3389/fnins.2016.00... We are working on a follow up in this direction with more cultures. (N.J)


Nice! Looking forward to seeing where this leads :)


The wording of the summary at the top is hilarious:

A team of neuroscientists has found that people are biased toward hearing and producing rhythms composed of simple integer ratios — for example, a series of four beats separated by equal time intervals.

Breaking news: the preponderance of rhythms in 4 and 3 beat patterns in virtually every music across the globe isn't a freak coincidence.


Yep. This article seems like yet another of MIT's attempts to aggrandize its "scientific reputation" in every field, even when the 'insights' gained are obvious or non-novel.


Can somebody explain what they mean by 1:1:1? Their description suggests a four-on-the-floor pattern, like in house music, but I'm not really sure.


They're describing the space in between the notes... so 4 equally spaced notes would have 3 gaps of 1:1:1 . It is indeed a bit confusing, as it says nothing about the duration of the last note, which to my ears would be an intrinsic part of the pattern.


1:1:1 is an isochronous (equal) rhythm. Most of the rhythms in the paper are played repeatedly, so the last note has the same duration.


Yikes that is a lousy notation. How would they designate the Bo Diddley beat: 0.75: 0.75: 0.75: 0.5: 0.5?


or 3:3:3:2:2?


Thank you, I thought I was going crazy.


I hoped the piece would comment on the role rhythm plays in visual designs. Is auditory rhythm more "complex"? Are other rhythms favored by the eye?

In auditory perception, rhythm seems very natural and is easily explained. We find the same term used in visual art, but it's meaning is less intuitive than "proportion".

I hoped for a moment a scholarly source would include both, although I'm not surprised. I imagine the complexity of drawing conclusions would cause any such reseach to be preceived as suspect. meh.


Have you heard of the architect Christopher Alexander? His life's work seems to be close to what you are describing. I'd highly recommend the 4 volume series 'The Nature of Order.'

I've always felt that there is a whole other math based explanation to art and music, that may even bridge the two, hiding in plain site. Alexander is the only person I've read who has come close to describing this math.

Any body else here know what I mean?


Not exactly what you are looking for but... visually speaking, Paul Klee's notebooks are incredible resources and online as PDFs.

https://monoskop.org/log/?p=10127


I'm reading "A Pattern Language" right now and it's pretty revelatory. It certainly suggests that structural properties of natural languages are prevalent throughout culturally-transmitted behaviors


I'm really looking forward to checking out these books, thanks for recommending!

My favorite Math/Music text is "The Harmonic Experience", by W.A. Mathieu.


Any body else here know what I mean?

Yes, I've read them all. Truly changed the way I look at and think about things in all sorts of contexts.


The brain seem to have a dozen mechanisms for perceiving various types of rhythm:

http://rstb.royalsocietypublishing.org/content/royptb/364/15...


In the end, the question in title was not answered. Also, it would be interesting to see and compare rhythms they came up with.


Nit: it's not a question. "How does the brain perceive rhythm?" would be a question. "How the brain perceives rhythm" is a promise to explain it to us.

However, you are correct: there is no satisfactory explanation here.


Yeah, the only conclusion we find here is that cultures with music that follow certain beat patterns are biased towards recognizing those beat patterns. Not exactly surprising.


Agreed, the article flirts with both biological and cultural reasons for rhythm perception. Puzzling.

400-500 years of musical history has engrained cultures in particular rhythms. Not much of a biological argument, IMO.


If you want a match in framing and claims, read the original article, not the linkbait university press release.


They don't talk about what kind of musicians. I quite distrust the notion of integer rhythms for them. Rubato is non integral, same goes for swing which constantly shifts the momentum of the beat. In my non researcher mind, momentum is the closest definition of what rhythm is (and much more telling than empty space between stimuli)


I agree with your momentum theory but furthermore it seems to also have a strong synesthetic element. As I mature as a musician these synesthetic experiences become more vivid while in the midst of making music.

The best unscientific metaphor I can think of to describe it would be like casting the Patronus Charm from Harry Potter.


I don't know HP enough to get your analogy. When I'm in flow, music is like surfing a multi dimensional curved space. If my body isn't tuned enough and I'm not relaxed enough I lose "sight" of the surface and energy loss occurs, otherwise you just follow the valleys or hills as you see fit.


I taught music for many years. When people can't accurately reproduce a rhythm it usually means that they simply arent yet capable of reproducing the pattern. This study suggests that they can reproduce it but are imposing a bias on it.


I mean, what would the priors look like for this dude?

https://www.youtube.com/watch?v=biZBmELy1Dc


It still seems like they're keeping to a steady "integer" rhythm, but being clever about what they fit inside that rhythm.


transfer of culture through beat rhythm priors distributions is a pretty neat idea. For cultural awareness, choose bayes theorem. Reminds me of Aumann's agreement theorem!


And I'm wondering whether this has something to do with the rhythm of language... The heart-beat would give the baseline but language would create different patterns...??


Last year the lead author was a post-doc in the lab I work in and we discussed it... Nori and colleagues are certainly interested in how prosodic differences between languages may affect the priors. But it's probably not going to be a very straightforward story... the rhythms in languages are considerably more complex than the ones they find people have a preference for.


Some of our preliminary research about speech rhythm did get to the Current Biology paper. Figure 6 in the paper uses the method to study prosody of 5 sentences in English. (N.J)




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: