Thursday, July 26, 2007

'Why do people sing so shamelessly out of tune?'

This week a national newspaper called me with this peculiar question. It reminded me immediately of a lecture that Isabelle Peretz (University of Montreal) gave this spring in the UK on amusia or tone deafness. In that she showed recent video material of a lab member who sang very much out of tune, but who was not aware of it. Surprising, because he has a degree in music education.

The reason I mention the example is that we often equal a talent for music to performance, such as being able to sing or play an instrument, and not so much to perception, for instance, being sensitive to subtle differences in pitch and timing when listening to music. When somebody sings out of tune, we might infer that he or she has no talent for music.

That is of course a misunderstanding. We can not simply judge someone’s musicality through the acrobatics of performance (Besides it needs years of training; see an earlier posting). More and more research is showing that mere exposure —not musical expertise as a result of formal training— has an influence on making sophisticated musical judgments.

With regard to performance, an intriguing study was done by Simone Dalla Bella and colleagues (just published in JASA). They asked occasional singers, recruited in a public park, to sing a well-known Quebecan birthday song. It was no surprise to find the professional musicians to reproduce the song much more precise than the ‘non-musicians’. However, when the ‘non-musicians’ were invited in the lab, and were asked to sing it again at a slightly slower pace, most sang it just as accurately as the professional singers. Another example that shows that musical skills are more common than we might think.

ResearchBlogging.orgDalla Bella, S., Giguère, J., & Peretz, I. (2007). Singing proficiency in the general population The Journal of the Acoustical Society of America, 121 (2) DOI: 10.1121/1.2427111

Monday, July 23, 2007

What makes a theory compelling?

Karl Popper was a philosopher of science that was very much interested in this question. He tried to distinguish 'science' from 'pseudoscience', but got more and more dissatisfied with the idea that the empirical method (supporting a theory with observations and experiments) could effectively mark this distinction. He sometimes used the example of astrology “with its stupendous mass of empirical evidence based on observation”, but also nuanced it by stating that “science often errs, and that pseudoscience may happen to stumble on the truth.”

Next to his well-known work on falsification, Popper started to develop alternatives to determine the scientific status or quality of a theory. He wrote that “confirmations [of a theory] should count only if they are the result of risky predictions; that is to say, if, unenlightened by the theory in question, we should have expected an event which was incompatible with the theory — an event which would have refuted the theory.” Popper, 1963).

Popper was especially thrilled with the result of Eddington’s eclipse observations, which in 1919 brought the first important confirmation of Einstein's theory of gravitation. It was the surprising consequence of this theory that light should bend in the presence of large, heavy objects (Einstein was apparently willing to drop his theory if this would not be the case). Independent of whether such a prediction turns out to be true or not, Popper considered it an important quality of ‘real science’ to make such ‘risky predictions’.

I still find this an intriguing idea. The notion of ‘risky’ or ‘surprising predictions’ might actually be the beginning of a fruitful alternative to existing model selection techniques, such as goodness-of-fit (which theory predicts the data best) and simplicity (which theory gives the simplest explanation). Also in music cognition measures like goodness-of-fit (r-squared, percentage variance accounted for, and other measures from the experimental psychology toolkit) are often used to confirm a theory.* Nevertheless, it is non-trivial to think of (existing) theories in music cognition that make surprising predictions. That is, a theory that predicts a yet unknown phenomenon as a consequence of the intrinsic structure of the theory itself (If you know of any, let me know!)

Well, these are still relatively raw ideas. I hope to be able to present them in a more digested format next week at the music perception and cognition conference (SMPC) in Montreal. Looking forward to it!

* If you want to read more on this topic, see here.

Sunday, July 22, 2007

Can newborns make sense of rhythm?

Last month our research group organized the annual EmCAP workshop in Amsterdam: A consortium of four European universities that collaborate in trying to understand how cognition might emerge in active perception. Or, in other words, how we accumulate knowledge in the world by actively being engaged with it. Music was chosen as the ideal domain to figure that out.

One of the big challenges of this project is to see whether, and if so to what extent, newborns have musical capabilities, and how exposure to music allows cognitive constructs like harmony or meter to emerge. More and more studies show that even a few month old babies have all kinds of perceptual and musical skills that allow them, for instance, to note the difference between violations in complex Balkan rhythms and (for us) more straightforward western rhythms, a difference that adult listeners, in general, find difficult to notice.

In the EmCAP project, in collaboration with the Bulgarian baby-lab, we planned to start this spring to have newborns —like those in the picture above—listen to syncopated and non-syncopated rhythms as a way to find out whether they are sensitive to the concept of meter as an emergent property. Something that could, alternatively, well be simply a learned music theoretical and/or cultural concept. We hope to find out ...

Wednesday, July 18, 2007

Does music have an alphabet?

We all learn the alphabet at school and are quite used to the idea that just twenty-six letters and a few punctuation marks are enough to communicate the wildest stories and can easily evoke the most vivid imageries and delicate feelings. The question is: can music also be reduced to an alphabet as effectively and meaningful as language? Can it be reduced to a set of discrete symbols with which the essence or meaning of music can be captured?

In some sense it is a rhetorical question. The presence of music notation shows that it is at least partly possible. But how close to our experienced, mental representation of music actually is music notation?

And that’s the other rhetorical part of the question: I would argue that music notation (as we know it) has little to do with music listening. While very useful for some purposes (e.g., sight reading or as a set of instructions of how to perform the music), as a reflection of the listening experience it fails miserably. This is why researchers like Marie Louis Serafine and Jeanne Bamberger often stressed the fact that music notation is merely an ‘after-the-fact’ notion of music, not to be taken too seriously: notation is as informative to listening as a cooking recipe is to tasting.

[More on the same topic in Dutch]

Tuesday, July 17, 2007

Do hours of musical exercise help?

Last week I received an email from an enthusiastic amateur musician who was wondering whether indeed his teachers were right in stating that ‘to get better in music is mainly a matter of exercise’. Apparently he was doubting his talent for music: Would he ever come close to the quality of his beloved musicians?

John Sloboda of the University of Keele did an elaborated study in the nineties in which he proposed a number of challenges to what he called the ‘Myth’ of musical talent. Maybe four of them can provide some comfort to the hardworking amateur:

First, in several cultures a majority of the people arrive at a level of expertise that is far above the norm for our own society. This suggests that cultural, not biological, factors are limiting the spread of musical expertise in our own society.

Second, the majority of top-ranking professional musicians were not child prodigies. In fact, studies reveal that very few able musicians showed any signs of special musical promise in infancy.

Third, there are no clear examples of outstanding achievement in musical performance (or composition) that were not preceded by many years of intense preparation or practice (N.B. a twenty-one year old musician has generally accumulated more than ten thousand hours of formal practice)

Fourth, many perceptual skills, required to handle musical input, are very widespread, develop spontaneously though the first ten years of life, and do not require formal musical instruction (for the full list, see Sloboda, 1994).

So a talent for music appears not so much to be constraint by our biology as it is by our culture. We all seem to have a talent for music. Nevertheless, if you want to become good at it—like most musicians— one has to spend hours and hours doing it.

Why does it sound slow?

We know that it is not simply the number of notes (or event-rate) that defines a listeners impression of tempo. There are quite a few musical examples that have a lot of notes but that are generally judged to have a slow tempo (e.g, Javanese gamelan music). The inverse, an impression of a fast tempo caused by only a few notes, is more difficult to find (but I’m sure some of you know of an example).

One correlate of tempo is the ‘metricallity’ of the music, especially the tactus, the rate at which events pass by regularly at a moderate tempo (typically around half a second or 120 bpm). Models of beat induction try to explain this: how listeners arrive at perceiving a beat or pulse in the music. Interestingly, the most salient pulse might not be explicitly present in the musical material itself. It can be ‘induced’ by music while listening (hence the term ‘beat induction’). It’s one of those classical examples that shows that cognition influences our perception of music.

At the recent Rhythm Perception and Production Workshop (RPPW) in Dublin tempo perception was one of the topics. Subjective judgments of duration were shown, once more, to be influenced by event density. Listeners had to continue tapping after hearing a regular beat with the intervals filled with soft random clicks, so-called ‘raindrops’. Participants tapped slower when more ‘raindrops’ were inserted. Apparently they judged the regular beat to be at a slower tempo when more events occurred between the beats. This is of course a relatively artificial setup, but the effect of event or note density on tempo judgments was also shown in more musically realistic contexts. What we can conclude from this is that tempo —defined as the subjective judgment of speed— is at least a product of two things: a sense of pulse (or tactus) and event density. It still is quite a challenge for music cognition researchers to come up with a model that actually can predict and explain these tempo judgments in real music, to, for instance, be able to predict when listeners will perceive music as nice and slow.

Does rhythm make our bodies move?

Why do some people dance more rhythmically to music than others? Are these differences genetically or culturally determined? These are some typical questions journalists, interested in rhythm research, do not hesitate to ask.

The link between musical rhythm and movement has been a fascination for a small yet passionate group of researchers. Early examples, from the 1920s, are the works by Alexander Truslit and Gustav Becking. More recently researchers like Neil Todd (University of Manchester) defend a view that makes a direct link between musical rhythm and movement. Direct in the sense that it is argued that rhythm perception can be explained in terms of our physiology and body metrics (from the functioning of our vestibular system to leg length and body size).

While this might be a natural line of thought for most people, the consequences of such theories are peculiar. They predict, for instance, that body length will have an effect on our rhythm perception, longer people preferring slower musical tempi (or rates), shorter people preferring faster ones. Hence, females (since they are on average shorter than men) should have a preference for faster tempi as compared to males.

To me that is too direct and na├»ve a relation. There are quite a few studies that looked for these direct physiological relations (like heart rate, spontaneous tapping rate, walking speed, etc.) and how these might influence or even determine rhythm perception. However, none of these succeeded in finding a convincing correlation, let alone a causal relation. In addition, they ignore the influence that culture and cognition apparently have on rhythm perception. Nevertheless it should be added that embodied explanations do form a healthy alternative to the often too restricted ‘mentalist’ or cognitive approach.

An intriguing study in that respect was done by Jessica Phillips-Silver and Laurel Trainor (Canada) a few years ago. They did an inventive experiment with seven month old babies, and showed that body movement (i.e. not body size) can influence rhythm perception. Although one could be critical on some important details, it is a striking empirical finding, and a small step forward in trying to underpin the relation between rhythm cognition and human movement.