Cognitive Daily

Learning to identify musical style

Listen to these two short music clips.

Music Clip 1
Music Clip 2

Now, can you identify the musical style of each clip?

If you said “Classical,” you’re technically only correct for the first clip. The second clip is actually in the Romantic style (bonus points for identifying the works and composers in the comments!). While both are examples of the classical genre, classical music is also divided into styles corresponding roughly to historical periods: Baroque, Classical, Romantic, and Post-Romantic. Traditionally, only trained musicians have been regarded as being able to easily distinguish between these styles.

But is that ability merely due to musicians’ familiarity with individual musical compositions, or is there something about the underlying structure of the music that enables musicians to tell the difference more readily than non-musicians? If the musical structure accounts for the difference, then can non-musicians easily be trained to recognize it, or is extensive musical training required?

Simone Dalla Bella and Isabelle Peretz found an innovative way to address those questions. Simply playing clips like the ones above gives trained musicians an unfair advantage, because they are more likely to be familiar with the specific musical composition. Instead, the researchers had a professional composer write four new compositions in each of the four major styles of classical music. They analyzed each piece for musical similarities and differences, and then played them for three different groups of volunteers: non-musicians familiar with Western music (Canadian college students); trained musicians familiar with Western music (music students at the University of Montreal); and non-musicians unfamiliar with Western music (exchange students from China who had spent less than two years in Canada).

They played the clips, about 30 seconds long each, in pairs. Participants were asked to rate each pair on a scale for similarity, with 1 being “very different” and 7 being “very similar.” If training offered a special advantage, then Western musicians should more be able to more readily observe the differences between different musical styles. Further, they should rate music that comes from more distant historical periods as more different than non-musicians. Here are the results:

i-2facead57c464dc4b8e8f38fac1b1145-musicclip.gif

As you can see, Western musicians did identify the most differences between styles, but even non-Western non-musicians were able to successfully see larger differences between styles that were more historically distant.

A deeper analysis of the data found that all participants were using the same musical criterion to distinguish between styles: the variation in duration of notes, which was measured in two ways. First, the researchers measured the length of each note in a composition and then calculated the standard deviation of this length (a range around the average note length in which most notes fell)—the larger this measure, the more variation in note length occurred. Next, they measured the variability of the difference in length between neighboring notes. Again, the larger this measure, the more variability between notes. The similarity ratings of experts and novices alike correlated strongly to these two measures.

Western musicians with extensive musical training did rely to a certain extent on tonal differences, but even without this training, non-musicians can easily identify different musical styles. So it appears that everyone can discern the differences between musical styles with a minimum of training.

Dalla Bella, S., & Peretz, I. (2005). Differentiation of classical music requires little learning but rhythm. Cognition, 96, B65-B78.

Comments

  1. #1 Scott Spiegelberg
    December 1, 2005

    First one is the Mozart Clarinet Concerto. The second one is the Brahms Violin Concerto (I think, its been a while since I heard it. Definitely Brahms). But to prove the point, I have an extensive Western Classical music education.

    I wonder about the ability of the composer to authentically compose in the four styles. By this I mean did he/she invest the same aesthetic involvement in each piece. I have a suggested variant at my blog.

  2. #2 Dave Munger
    December 1, 2005

    You’re right about the first one, but not the second. Keep those guesses coming!

    Also, to anyone who hasn’t yet checked out Scott’s blog, make sure you do—it’s fascinating!

  3. #3 Scott Reynen
    December 1, 2005

    “A deeper analysis of the data found that all participants were using the same musical criterion to distinguish between styles: the variation in duration of notes”

    Am I understanding this correctly, that the amount of variation in the duration of notes has a direct relationship with the age of the music? So does newer music have more variation or less?

  4. #4 Dave Munger
    December 1, 2005

    Scott Reynen:

    Newer music has more variation.

    Scott Spiegelberg:

    Nope. Not Bruch either.

  5. #5 Scott Spiegelberg
    December 1, 2005

    Ah, I think it is the Bruch violin concerto. I always get those two mixed up, as my first recording paired them . The Brahms is considerably longer, but they both do share than late Romantic angst.

  6. #6 John Dey
    December 2, 2005

    Mendelssohn, Violin Concerto

    Independent of musical structure (chord and scale use, rhythmic elements, thematic structure, instrumentation, etc.), the overall coloration and sentiment of music seems to vary from one period to another making it fairly easy to distinguish one from another to this listener’s ears.

  7. #7 Dan H
    December 2, 2005

    I was struggling with #2 as well. I have both Mendelssohn and Brahms on my iPod, and have this piece on my iPod, but left it at home today so I can’t check. My vote goes to Mendelssohn.

    There’s a gray area around some composers, however. Is Beethoven classical or romantic? Is a contemporary composer who writes a baroque (or neobaroque, some would say) peice easily identified as new? Paganini did some wild stuff akin to rock music, but because it’s a violin, we’d be more likely to say Romantic era. I’m also amazed with how we’re performing old music now: I’ve heard a couple violin students perform Bach unaccompanied works with a romantic twist. This sort of interpertation at recording session time would skew a non-music scholar’s classification of the work.

    Best trick in the book for repertoire tests is to know the period instrumentation well. Harpsichord, organ = Baroque, Lute, Polyphonic voices = Renaissance, etc. So there’s a short list of facts like this to get you mostly right, if you’re trained to hear what a Lute is.

    This article is eye-opening for me, that these kind of rules or feelings about how music is written (or how the music sounds now) can be a part of human nature. Thanks.

  8. #8 R J Keefe
    December 2, 2005

    The next step would be to produce a fifth snippet, this one in the neo-classical style that ranges from Dvorak’s and Brahms’s serenades through Stravinsky. In any case, I’d like to know if a return to less-note-variation is part of the neo-classical style. (As opposed to, say, instrumentation, an obvious factor.)

  9. #9 Dave Munger
    December 2, 2005

    Just to settle it once and for all, the second clip is indeed the Mendelssohn Violin Concerto.

    Dan H, you make some excellent points.

    Yes, the borderline composers definitely make these determinations different; for the sake of this study those problems were avoided by creating compositions that exemplified the style. But I don’t think it would be a problem to include borderline-style works—after all, shouldn’t we still rate Beethoven as closer to Mozart than Bach?

    I should point out that the instrumentation in each snippet in this study was the same: piano—which obviously isn’t an authentic baroque instrument, but certainly can be used to play baroque-style music.

  10. #10 John Richardson
    December 10, 2005

    Why was there no sample of Non-Western Musicians? That would have helped to isolate the effects of musical training from knowlege of the repertoire.

  11. #11 terry kelley
    December 10, 2005

    Greetings, as i listened to these clips i had NO idea who either of them were.i definitly enjoyed the first more then the second.you folks have a knowledge base i wish i had. i may not know what im hearing but i know what i like. the soft tones of the woodwinds can carry me away while the furious/pained sounds of the violin drive me away! i do like the deeper strings.now im going to buy a
    copy of the mozart piece.

  12. #12 Max Erick Busse-Grawitz
    December 11, 2005

    Hi,
    Mendelssohn is a particularly bad example for romantic music because in his time he was said to compose “old-fashioned”, that means, from our perspective, too classical. BTW the second part of your clip is actually a citation of Beethoven’s 5th. Ok, some might argue Beethoven was already early-romantic…

  13. #13 Todd Rockhold
    December 19, 2005

    While this thread is about perception of musical style, nevertheless I’ll mention here that I think Diana Deutsch has done some interesting work on perception of musical tones. See

    http://psy.ucsd.edu/~ddeutsch/

    Maybe her work on music is a little dated by now. I’m neither a cognitive psychologist nor a musician.

The site is currently under maintenance and will be back shortly. New comments have been disabled during this time, please check back soon.