29 June 2010

Daily Digest, 2010.06.29

Today: Reception history of the "Mozart effect"; a review of electrical brain stimulation; the phylogeny of languages; the secrets of Çatalhöyük (stay tuned); math vs engineering; the "Two Children Problem" and its variants; mix tapes for dogs.


An excellent short piece by Alix Spiegel at the NPR website (the story was apparently on Morning Edition yesterday) on the reception history of the "Mozart Effect". The "Mozart Effect" mania had its origins in a 1993 paper in Nature by Frances Rauscher, which purported to show that (for the 36 college students in the study) listening to 10 minutes of a piano sonata by Mozart provided a 10-15 minute window in which performance on a test of spatial reasoning was superior to that of subjects who had 10 minutes of silence or 10 minutes of a relaxation tape. (Rauscher's original paper is available here.)

The NPR piece does not mention that this was, on the face of it, a poor experimental design.  In fact, one wonders how it got published in Nature in the first place.  For the author clearly simply assumed (rather than testing the hypothesis) that the identity of the music—that it was a sonata by Mozart—had something to do with the result.  But that, of course, would only have been established had one done a follow-up experiment using different composers and different genres of music.  What about sonatas by Beethoven?  Or symphonies by Mahler?  Or string quartets by Vanhal?  Or songs by the Beatles?  Or sitar music Ravi Shankar?  Or blues by Robert Johnson?  Or....

Rauscher now realizes this too, according to Spiegel's piece:
Rauscher still stands by her original finding, but says subsequent research has shown that it's not really about Mozart. Any music that you find engaging will do the same thing, because compared to something like sitting in silence, the brain finds it stimulating.
"The key to it is that you have to enjoy the music," Rauscher says. "If you hate Mozart you're not going to find a Mozart Effect. If you love Pearl Jam, you're going to find a Pearl Jam effect."
The piece used in the experiment was K. 448, the Sonata in D for Two Pianos. Rauscher's original paper apparently misidentified the piece (or got the K. number wrong); at any rate, the pdf available through the link above has a hand cancellation and the correction "K. 448" written in the margin. Oddly, the paper does not specify which movement of the sonata was used for the experiment, although I have read elsewhere that it was "the first 10 minutes" (one assumes of the first movement).

In any case, the NPR piece is a good cautionary tale about why one should, as a rule, be extremely skeptical of science reporting in the mass media, and also (perhaps) why one should be skeptical of quite a lot of research in psychology.

The "Mozart Effect" is a fascinating and bizarre episode in the history and sociology of science, and I may write more about it.



Vaughan at Mind Hacks points to a free review article that surveys all studies that have described the results of direct electrical stimulation of the brains of living patients:
Aslihan Selimbeyoglu and Josef Parvizi (2010), "Electrical stimulation of the human brain: perceptual and behavioral phenomena reported in the old and new literature," Frontiers in Human Neuroscience.
Vaughan's summary makes the article sound fascinating, and I'm looking forward to reading it.



Bayes at Gene Expression discusses a new article that investigates the use of phylogenetic techniques applied to the typological features of languages in order to elucidate the deep historical relationships of language groups. The article is:
S. J. Greenhill et al. (2010), "The shape and tempo of language evolution," Proceedings of the Royal Society B: Biological Sciences. The abstract is here. The article is freely available for download.
The typological data is drawn from The World Atlas of Language Structures Online (WALS), which is likewise freely available, and looks to be an extremely useful resource. The article looks quite interesting, even though (according to Bayes) the bottom line seems to be that the technique doesn't really give better results than using the same techniques on lexical data.



Mark Liberman at Language Log has a wide-ranging follow-up piece on the article published last week on spatial orientation in Nicaraguan Sign Language (see my digest for last Tuesday). Perhaps the most interesting part of Liberman's post is his investigation of the etymology of the expressions "hay foot" and "straw foot" (for left and right).



Via John Wilkins at Evolving Thoughts, this press release at Archaeo News from the Anatolia News Agency: "Secrets of Çatalhöyük to be revealed this summer."

Çatalhöyük (or Çatal Höyük) in southern Anatolia (Turkey) is one of the earliest known towns. It was occupied from around 7500 BC to 5700 BC (that is, from around 9500 years ago).



Via BoingBoing, this comic by David Malki at Wondermark on the difference between math and engineering. (I thought my Dad might like this.)




Julie Rehmeyer at ScienceNews writes on the counterintuitive probabilities of the "Two Children Problem" and an even more surprising variant.

The classic Two Children Problem, posed by Martin Gardner in an 1959 article in Scientific American, can be formulated as follows (this from Rehmeyer's article):
Suppose that Mr. Smith has two children, at least one of whom is a son. What is the probability both children are boys?
Intuition would suggest that the answer is 1/2, but careful counting (the key to understanding probabilities) reveals that the answer is 1/3.

Rehmeyer discusses an even more counterintuitive example recently presented at a conference in Gardner's honor:
I have two children, one of whom is a son born on a Tuesday. What is the probability that I have two boys?
Here the answer, even more counterintuitively, is 13/27.

Read Rehmeyer's article to find out why.



At io9, "What kind of music do dogs enjoy listening to? Science has the answer!"
Bookmark and Share

No comments:

Post a Comment