Page 65 - Fall2020

quickly demonstrated converging results showing more advanced pitch processing for native speech in musically trained individuals (adults and children), such as detect- ing pitch changes in syllables as well as at the sentence level (Schon et al., 2004; Chobert et al., 2011). More strik- ingly, musicians were also able to process pitch variations in foreign speech sounds much better than nonmusicians. Tonal languages, such as Mandarin, provided an ideal window to examine this question as the pitch variation patterns (e.g., lexical tones) at the syllable level indicate word meaning. For example, in Mandarin, there are four different types of lexical tones: ma in Tone 1 (level pitch) means “mom,” whereas ma in Tone 4 (dropping pitch) means “scold.” To date, nontonal language-speaking musicians have exhibited an enhanced ability in pro- cessing and discriminating lexical tones behaviorally and neurally at the cortex and even at the subcortical brainstem (Wong et al., 2007; Marie et al., 2011a; Zhao and Kuhl, 2015).
Very few studies have examined differences in speech rhythm processing between musicians and nonmusicians and the results are mixed. On the one hand, it was found that musicians were better able to detect occasional shortened vowels when the vowels were played repeat- edly, both behaviorally and in the cortex (Chobert et al., 2011). In addition, when the relationships between syl- lable durations in words were manipulated in a sentence, such as “mama” with equal duration for both syllables changed to longer duration for the first syllable, musi- cians exhibited higher sensitivity to such manipulations (Marie et al., 2011b). On the other hand, when examining how the neural signal follows the amplitude modulation patterns in music and speech signals, Harding and col- leagues (2019) found a musician advantage in tracking the music rhythm but not in the speech rhythm. Given the mixed results, more studies are warranted for us to elucidate the effects of early music learning on speech rhythm processing.
Functional Differences in Other Domains: Evidence for Far-Transfer Effects
Limited research efforts have also been devoted to exam- ining whether the “transfer effect” from early music training generalizes beyond speech processing to more general domains that are not as closely related to music processing, such as memory and attention skills (i.e., far transfer). The results from these studies are mixed, and
the question regarding “far transfer” remains highly debated. For example, Chan and colleagues (1998) first observed that musically trained adults and children recalled, both immediately and after some delay, a signifi- cantly higher number of words from a list read to them compared with nontrained individuals. This original study was done with a Cantonese-speaking population using the Hong Kong Verbal Learning Test and the effects were later replicated using the California Verbal Learning Test-II with English-speaking musicians and nonmusi- cians (Jakobson et al., 2008). Strait and colleagues (2010) tested a group of musicians and nonmusicians on a series of different tasks that targeted both cognitive abilities such as auditory working memory (reverse digit recall) and auditory attention (detection of a “beep” while watching a video). The results demonstrated better per- formance by musicians on auditory attention but not on auditory working memory.
By contrast, George and Coch (2011) observed enhanced working memory in musicians by conducting a more comprehensive set of standardized tests. Indeed, much more work is needed to examine these far-transfer effects, and only with more experimental results can we start to understand the whole picture of effects of early music training on general cognition.
Randomized Controlled Studies
As the results from cross-sectional and longitudinal stud- ies start to converge, it becomes critical to address one outstanding issue in these studies, genetic predisposition. Indeed, musicians could be genetically predisposed to have better auditory skills, which might have prompted them to seek more musical activity and training early in life. The key to addressing this issue is by conducting ran- domized controlled intervention studies. By randomly assigning participants to receive either music training or a control activity for a period of time in a controlled manner, we can measure with confidence whether any differences in outcome measures are due to the music training rather than genetic predispositions.
Here, we review the few existing randomized controlled experiments in more detail. A few studies have focused on school-age children. Moreno and colleagues (2009) randomly assigned nonmusically trained 8-year-old chil- dren to attend 24 weeks of either a music or a painting class. Both before and after the training period, a series of
Fall 2020 • Acoustics Today 65