Why do we like some music and not others? Why does music feel right and why does it evoke certain moods? The brain's ability to segment the continual stream of sensory information into perceptual chunks and extract meaning, “event segmentation” functions, have long fascinated researchers.

In a series of experiments, a team led by Vinod Menon of Stanford University School of Medicine asked subjects to listen to symphonies of the English composer William Boyce. The symphonies were chosen because they are relatively short and comprise well-defined movements - changes in tempo, tonality, rhythm, and pitch, and brief silences. The movement transitions were especially important, because the researchers used them as perceptual events that would enable the analysis of the brain circuitry involved in event segmentation.

As the subjects listened to the symphonies, the researchers scanned their brains using functional magnetic resonance imaging. This widely used technique involves using harmless magnetic fields and radio waves to measure blood flow in brain regions, which reveals brain activity in those regions.

“Studying event segmentation in real-world or “ecologically valid” stimuli is of particular interest for two reasons: first, such an investigation can reveal perceptual grouping processes that occur under natural conditions; second, there is growing evidence suggesting that neuronal populations behave differently under natural conditions than they do under controlled experimental conditions,” they wrote. “For instance, responses of neurons to simple, controlled stimuli are often not predictive of how they respond to more complex, natural stimuli. Currently, the brain systems underlying the segmentation of ecologically valid stimuli, particularly in the auditory domain, are poorly understood.”

What’s more, wrote the researchers, “Studying such segmentation processes in music may be a useful window into similar processes in other domains, such as spoken and signed language, visual perception, and tactile perception.”

The researchers’ analysis of the ten seconds surrounding movement transitions revealed that two distinct networks of brain regions were involved in perceiving the transitions. One was in the ventral region of the prefrontal cortex and one in the dorsal region. The researchers’ analysis revealed that activity in the ventral region preceded and caused subsequent activity in the dorsal region. Significantly, these active regions were in the right hemisphere of the brain, which is believed to be involved in processing music.

“Our finding of distinct dorsal and ventral networks in event segmentation provides a novel perspective on auditory scene analysis and significantly expands on our current knowledge of event segmentation processes in the brain,” they wrote.

“Although the right hemisphere has been implicated in music processing, its precise role in music perception remains poorly understood, and lateralization of activation for music processing has never been adequately tested. Our findings suggest that the right hemisphere plays a dominant role in the perceptual segmentation of salient, coarse-grained event boundaries in music.”

The researchers said their findings suggest that the two regions active in detecting movement transitions “form a tightly coupled network that plays an important role in directing and maintaining attention during the movement transitions and in the perceptual updating that ensues.”

They said that the ventral network appears to be involved in the “detection of salient events based on the sensory features of the stimulus stream.” This network acts as a “circuit breaker” to activate the dorsal network, which “turns the spotlight of attention to the event boundary.”

The researchers include Devarajan Sridharan and Vinod Menon of Stanford University School of Medicine in Stanford, CA; Daniel J. Levitin of McGill University in Montreal, QC, Canada; Chris H. Chafe and Jonathan Berger of the Center for Computer Research in Music and Acoustics in Stanford, CA

This research was supported by a Stanford Graduate Fellowship to D.S., by grants from the Natural Sciences and Engineering Research Council of Canada to D.J.L., the National Science Foundation (BCS-0449927) to V.M. and D.J.L., the Ben and A. Jess Shenson Fund to J.B., and the National Institutes of Health (NIH-HD047520) to V.M.

Sridharan et al.: “Neural Dynamics of Event Segmentation in Music: Converging Evidence for Dissociable Ventral and Dorsal Networks.” Publishing in Neuron 55, 521–532, August 2, 2007. DOI 10.1016/j.neuron.2007.07.003.