Page 64 - Fall2020

MUSIC TRAINING CHANGES THE BRAIN
spine. More importantly, the more time the individual had spent in childhood practicing piano, the better the white matter axons align with one another, thereby pro- viding better infrastructure for faster communication across brain regions.
These results were replicated when researchers followed a group of children for up to 29 months and compared children who started music training between 5 and 7 years of age with controls who did not receive any music training during this period. Significantly, structural dif- ferences between the two groups only started to emerge after 15 months of training, speaking to the idea that structural changes in the brain require prolonged, inten- sive training (Hyde et al., 2009).
Functional Differences in Music Processing
The first questions researchers examined regarding functions involved how musicians’ brains responded to different elements of music in comparison to the brains of nonmusicians. In a series of studies, Pantev and colleagues (1998, 2001) demonstrated that musi- cians, compared with nonmusicians, exhibited stronger neural activity in the brain in response to musical sounds, especially if the spectral profile is similar to that of their primary instrument. For example, trumpeters responded to musical sounds made by trumpets more than to musical sounds made by violins, and vice versa for the violinists. In addition, researchers also observed that musicians detected pitch changes better than nonmusi- cians but only when the pitch changes were embedded in a context, such as intervals, chords, and melodies (Koelsch et al., 1999; Fujioka et al., 2004). Longitudinal studies following children starting to take violin lessons also showed similar effects in sound processing. For example, after a year of violin lesions, musically trained children demonstrated stronger neural activation for violin sounds than nontrained children but not for white noise (Fujioka et al., 2006).
Quite a few studies have also examined differences between musicians and nonmusicians in the neural processing of musical rhythm. Most studies focused on two hierarchical levels of musical rhythm: the beat-level rhythm that gives a regular pulse like a metronome and the meter-level rhythm that further groups the regular beats into patterns, such as strong-weak-strong-weak in a marching rhythm (Fitch, 2013). In earlier studies, rhythm
processing was largely examined by measuring the neural response to the disruptions of a regular rhythm, with a larger response indexing better detection of the disrup- tion. From better detection of disruption, we can infer better tracking of rhythm (Zhao et al., 2017).
Using the same approach as above, researchers manipu- lated music excerpts to contain occasional disruptions that either disrupt the beat-level rhythm or the meter- level rhythm. To illustrate, imagine a waltz rhythm with three beats equally spaced in a group, strong-weak-weak. A beat-level disruption would move the last weak beat closer to the middle weak beat and farther from the next stronger beat. A meter-level disruption would eliminate the last weak such that it becomes strong-weak-strong- weak-weak. When musicians’ and nonmusicians’ neural responses to these disruptions were measured, research- ers showed that the two groups can detect beat-level rhythm disruptions equally well, whereas musicians can detect meter-level disruptions much better (Vuust et al., 2005; Geiser et al., 2010).
In the last few years, a new approach has been devel- oped to allow direct measurements of neural tracking of beat- and meter-level rhythms without the occasional disruptions. Using this approach, a recent study showed that musicians’ neural tracking of beat- and meter-level rhythms were both enhanced. Furthermore, how well the neural signal tracks rhythm is correlated with years of training, suggesting a music training related effect (Doel- ling and Poeppel, 2015).
Functional Differences in Speech Processing: Evidence for Near-Transfer Effects
Speech processing is considered closely related to music processing because of the similar characteristics of the two acoustic signals and their common ubiquitous nature in the world’s cultures. Whether early music training can generalize its effect to speech processing, which is called the near-transfer effect, became a heavily studied question because speech processing is crucial to human communication. Over the years, converging evidence has shown that early music training indeed affects speech processing (Patel, 2014).
Researchers first approached the “near-transfer effect” question by examining how musicians and nonmusicians process pitch information in speech. Various studies
64 Acoustics Today • Fall 2020