Music Psychology

Perceptual grouping

I love it when an author provides links to their papers. I think it is an excellent principle that allows quick and easy access to new works. I know it is not always possible, what with copyright restrictions, but luckily a number of the leading authors/labs in the music psychology world currently adopt this principle. Searches today of one of these sites, one by Dr Ani Patel, lead me to a new paper on auditory perceptual grouping.

 I got interested in auditory grouping during my PhD, when I was comparing the processing of language and music. The principles governing abstract perceptual grouping show common application in language and music, and these studies are hugely insightful when it comes to understanding how we come to interpret sounds.

Perceptual grouping describes how we psychologically segment a stream of sound into chunks.  So imagine a sequence of sounds where one beat is short and one is long (S-L-S-L-S-L). However the sequence starts, most English-speaking people segment this stream into an iambic pattern (S-L…S-L…). For a long time this tendency towards iambic grouping was thought to be innate and automatic, governed by laws known as the Iambic/Trochaic Laws whereby longer sounds mark group endings.

Another possibility however, is that English speaking people form this grouping preference as a result of exposure to the structure of the English language. In English phrases, the rhythm is typically iambic as a result of the fact that nouns follow pronouns (e.g. the CAT) and verbs precede objects.

One way to tease apart these two arguments as to how English speakers might pick up a preference for iambic grouping is to test individuals from a culture where the language phrase structure is more naturally trochaic (L-S). Japanese is such a language; in Japanese the word order is reversed (i.e. object before verb) and studies have shown that Japanese listeners do not group streams of SLSL beats into iambic patterns – their grouping is more heterogeneous with a tendency towards trochaic (Iverson et al., 2008).

In his recent paper (where he is third author) Patel’s group examined whether these differences in grouping between English and Japanese might be innate or learned, by testing grouping preferences in English-learning and Japanese-learning infants at 5-6 months of age and 7-8 months. They used a conditioned head turn procedure, where infants were first familiarised to a long SLSLS sound stream. They were then sat on their parent’s lap and their attention was drawn to a blinking light.  When they focused on the light either an iambic or a trochaic sequence was played and the authors timed how long they continued to look at the light. If the infants continued to look at the light (i.e. the direction of the sound stream) the theory goes that they perceived the stimulus to be novel and therefore different to the way they had naturally segmented the initial SLSLS pattern.   

As expected the 5-6 month olds had no looking time differences for either sound stream, suggesting they had no grouping preferences.  The 7-8 month old English-learning infants had a preference for the iambic stream, whereas the Japanese-leaning infants showed no preference.  The latter findings mirror those found in English speaking and Japanese speaking adults (Iverson et al., 2008).

These results provide evidence that non-linguistic grouping perception in sound develops by 7-8 months of age in tandem with native linguistic structure. The authors suggest that universal principles of grouping should not be totally disregarded however; rather that these principles are likely to be interpreted according to the language environment as opposed to guiding all development in an automatic way.

So there you go. And all this new knowledge I gained this lunchtime would not have been possible without free access to this article, since I am not at my University desk. Three cheers for free access!

Paper: Yoshida, K., Iversen, J.R., Patel, A.D., Mazuka, R., Nito, H., Gervain, J., & Werker, J. (2010). The development of perceptual grouping biases in infancy: A Japanese-English cross-linguistic study. Cognition,115:356-361. (http://vesicle.nsi.edu/users/patel/Yoshida_Iversen_Patel_et_al_2010_Cognition.pdf)

Ref: Iversen, J.R., Patel, A.D., & Ohgushi, K. (2008). Perception of rhythmic grouping depends on auditory experience. Journal of the Acoustical Society of America, 124: 2263-2271. (http://vesicle.nsi.edu/users/patel/Iversen_Patel_Ohgushi_JASA_2008.pdf )

Comments Off on Perceptual grouping