Edited by:
Reviewed by:
*Correspondence:
This article was submitted to the journal Frontiers in Human Neuroscience.
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Several studies comparing adult musicians and non-musicians have provided compelling evidence for functional and anatomical differences in the brain systems engaged by musical training. It is not known, however, whether those differences result from long-term musical training or from pre-existing traits favoring musicality. In an attempt to begin addressing this question, we have launched a longitudinal investigation of the effects of childhood music training on cognitive, social and neural development. We compared a group of 6- to 7-year old children at the start of intense after-school musical training, with two groups of children: one involved in high intensity sports training but not musical training, another not involved in any systematic training. All children were tested with a comprehensive battery of cognitive, motor, musical, emotional, and social assessments and underwent magnetic resonance imaging and electroencephalography. Our first objective was to determine whether children who participate in musical training were different, prior to training, from children in the control groups in terms of cognitive, motor, musical, emotional, and social behavior measures as well as in structural and functional brain measures. Our second objective was to determine whether musical skills, as measured by a music perception assessment prior to training, correlates with emotional and social outcome measures that have been shown to be associated with musical training. We found no neural, cognitive, motor, emotional, or social differences among the three groups. In addition, there was no correlation between music perception skills and any of the social or emotional measures. These results provide a baseline for an ongoing longitudinal investigation of the effects of music training.
Observing a young child in a choir, singing a melody with perfect pitch, or another at a recital translating musical notations into precisely timed finger movements on a small violin, it is natural to wonder whether traits that might precede training favor musical abilities and even motivate children to pursue music training instead of other activities. For example, are children with pre-existing superior listening skills more likely than others less gifted to seek participation in music training programs? Or is it the case that the better listening skills found in adult musicians can be fully explained by their long-term regular and intensive training in music? In either case, the difference between musically gifted or merely competent playing can be related, at least in part, to differences in measureable aspects of cognitive, motor, and emotional performance, as well as neural structure and function.
Playing music is a complex task that of necessity engages many different brain regions because it requires the concurrent recruitment of distinct sensory systems, including the auditory, somatosensory, and visual, as well as the interplay of these sensory systems with the motor, executive and affective systems. The mastering of this rich and demanding process requires regular and intense practice, and it involves coordinating both hands, and communicating emotionally with other players and listeners. The combination of such demands is likely to influence the development and maintenance of brain structures and their function.
Accordingly, over the past two decades, several investigators have reported differences in the brain and behavior of musicians compared to non-musicians (For comprehensive reviews please see
Music training has also been reported to have positive associations with cognitive domains in both adults and children that are only indirectly related to music, namely language skills including phonological awareness (
Turning to the brain itself, differences between musicians and non-musicians have been found, predictably, in auditory (
Despite the increasing interest in the benefits of music training and in the differences found in the brain and behavior of musicians and non-musicians, the interpretation of the findings remains unclear. For example, the differences reported in cross-sectional studies, which mostly employ quasi-experimental designs, might be due to long-term regular and intensive training or might result primarily from pre-existing biological factors that would predispose an individual to develop musical aptitude if exposed to music during a sensitive period of development. The differences also may result from contributions of both training and pre-existing factors, with certain differences possibly relating more primarily to training and others more primarily to pre-existing factors.
Learning and practicing new skills other than music have been previously shown to correlate with changes in the structure and function of the brain (
Another much-debated question in relation to music training is the extent to which benefits for non-musical abilities are strictly cognitive or extend to emotional and social skills. Music is intimately connected with emotion and feelings (
Performing in an ensemble, as opposed to private music tutoring, requires each musician to attend carefully to different aspects of the sound produced by the other musicians, including intonation, timing, and dynamics. It also requires each player to connect emotionally with the other musicians and to integrate the intended emotional impact with their own playing. Furthermore, it has been suggested that collective music making strengthens group cohesion by forcing participants to orient their attention to a shared temporal framework (
One way of addressing the uncertainties related to the effects of music training would consist of examining children prior to the onset of their group music training, and, once training begins, compare them to children involved in equally socially interactive but non-musical training, such as a group athletics program, and follow both groups longitudinally. This article describes reports on the first year results of such a study.
The first aim for the initial phase of the study was to investigate whether children who participate in intense after-school musical training, compared to a group of children participating in after-school athletic training, show differences in evaluations testing cognitive, motor, and musical behavior or in structural and functional brain measures prior to their training. The athletic training group – after-school soccer playing – was selected as to control for those aspects of musical training that would be shared with such sport activity – a motivating, sustained, and engaging sensorimotor group learning activity. Both music and sports programs were community-based programs offered free of charge in neighborhoods of downtown Los Angeles and Rampart district. Participants, in both programs, enrolled voluntarily in their respective programs. All the students who signed up for the sports program were admitted, while the final selection for the music program was by a lottery admission from the list of students who had signed up. The reason for the lottery enrolment is due to the limited number of available slots per year. A second comparison group of children from the same area was recruited. In this case the intention was to recruit children not involved in any particular systematic organized after-school activity.
The second aim was to determine whether the children participating in music training, compared to the control groups, showed different emotional and socials skills at the start of the study. Furthermore, we aimed to examine whether musical abilities as measured by a music perception assessment correlate (prior to training) with emotional and social benefits (including empathy and prosocial behavior) that have been previously associated with music training in adulthood.
The third aim was to establish the base level for a 5 year longitudinal study (currently in its second year) designed to investigate the effects of early music training on cognitive, social, emotional, and neural development.
Fifty 6- to 7-year-olds were recruited from public elementary schools and community music and sports programs in the greater Los Angeles area. Seventeen children (7 girls and 10 boys, mean age 79.9 months, SD = 6.8) were about to begin training with the Youth Orchestra of Los Angeles at Heart of Los Angeles, YOLA at HOLA, for short. The program is based on the Venezuelan approach known as “El Sistema” and offers free instruction 5 days a week to children from underserved areas of Los Angeles. The children are involved in a systematic and high intensity musical training that focuses on work on rhythm, melody, harmony, timbre, and emphasizes ensemble practice and group performances. Children enrolled in this program are selected, by lottery, up to a maximum of 20 per year, from a list of interested families. Seventeen children (5 girls and 12 boys, mean age 79.06 months, SD = 9.2) formed the first control group (control sports) who were about to begin training with a community-based soccer program and were not engaged in any musical training. This program allows all children whose parents enroll them in the program. The 17 enrolled in the study (to match the music target group) were those first 17 that showed interest in participating. The soccer program offers free and high intensity training (three times a week for 2 h each, with an addition of 1 h game each weekend) to children ages 6 and older. In addition, sixteen children (4 girls and 12 boys, mean age 81.5 months, SD = 6.47) formed the second control group (control non-sports). Children in this second control group were recruited from public schools in the same area of Los Angeles and were not involved in any systematic and intense after-school program of any kind. All three cohorts are from equally under-served minority communities including primarily Latino and Korean families, of downtown Los Angeles. All children were raised in bilingual households, but all attended English speaking schools and spoke English fluently.
One child from the music group and two children from the control sports group discontinued their participation in the program after the initial assessment and were not included in the final analysis. Additionally one child from the control non-sports group and one child from the music group relocated after completing the first part of the testing session, and could not continue to participate in the study; therefore they were not included in the final analysis. In total 45 children, 15 in each group, participated in the study (after attrition: music group
Study protocols were approved by the University of Southern California Institutional Review Board. Informed consent was obtained in writing, in preferred language, from the parents/guardians on behalf of the child participants and verbal assent was obtained from all individual children. Either the guardians or children could end their participation at any time. Participants (parents/guardians) received monetary compensation for their child’s participation and children were awarded small prizes (e.g., toys or stickers).
Children were tested individually, and in private, in two to three sessions, for a total of 5 h, completed over the course of 2–3 weeks. Testing sessions took place at the children’s music school or at our laboratory in the Brain and Creativity Institute at the University of Southern California. Magnetic resonance (MR) imaging and electroencephalography (EEG) sessions took place in the Dornsife Cognitive Neuroscience Imaging Center also at the University of Southern California. All structural MR images were handled according to the established policy of Dornsife Cognitive Neuroscience Imaging Center. They were sent to a neuroradiologist for review. If an incidental finding would be detected, the neuroradiologist would have contacted the physician designated specifically for that purpose by the family at the time they signed the informed consent and suggest further evaluation if needed.
The order of the children’s first behavioral test was counterbalanced across participants and subsequent tests were administered in a pseudorandom order as testing rooms and experimenters were available. Privacy was ensured during all the steps of the research project including recruitment, data collection, and analysis. All research procedures were conducted in person and in a private setting and data were labeled with a code that only the research team could link to personal identifying information.
Parents also answered an extensive structured interview on family income, education and ethnicity, perceptions of child’s academic achievement and school participation, the child’s current and previous participation in extracurricular activities including involvement in sports or music programs, and the presence of any professional artists currently living in the child’s home. The structured interview was conducted in the parents’ preferred language, English, Spanish, or Korean, by a research assistant who was a native speaker of that language.
Parents indicated their highest level of education and annual household income on a questionnaire. Responses to education level were scored on a 5-point scale: (1) elementary/middle school; (2) high school; (3) college education; (4) master’s degree (MA, MS, MBA); (5) professional degree (PhD, MD, JD). Responses to annual household income were scored on a five-point scale: (0) <$ 10,000 (1) $10,000–$19,999 (2) $20,000–29,999 (3) $30,000–39,999 (4) $40,000–49,999 and (5) >$50,000. A final socio-economic status (SES) score was calculated as the mean of each parent’s education score and annual income. Combination of education and average household income is considered to be a robust measure of SES.
Handedness was assessed as part of the Bruininks–Oseretsky Test of Motor Proficiency (BOT 2-brief). Children were asked to write their name, throw a ball to the experimenter and kick a ball to the experimenter. Children were classified as right or left-handers if they used either left or right hand/foot for all three tasks. They were classified as mixed-handers if they used either left or right hand for only one of the tasks. In the music group, there were one left-handed boy and one left-handed girl; in the sports control group, one left-handed girl; and in the non-sports control group, three left-handed boys (see
Summary of participants.
Music |
Sports control |
Non-sports control |
||||
---|---|---|---|---|---|---|
Male | Female | Male | Female | Male | Female | |
Right-handed | 7 | 6 | 10 | 4 | 8 | 4 |
Left-handed | 1 | 1 | 0 | 1 | 3 | 0 |
The Block Design, Vocabulary, Matrix Reasoning, and Similarities subtests from the Wechsler Abbreviated Scale of Intelligence (WASI-II) for children 6 years and older were administrated (
The brief form of the Bruininks–Oseretsky Test of Motor Proficiency, Second Edition (BOT-2-brief form) was used. The test uses 12 engaging, goal-directed subtests to measure a wide array of gross and fine motor skills for subjects 4–21 years.
To determine whether Gordon’s PMMA correlated with any of the social or emotional assessments, two-tailed bivariate Pearson correlation were performed and corrected for multiple comparisons on the age and SES matched samples collapsed across the three groups.
Children underwent structural, diffusion, and functional MR imaging of their brain. Data collection and analysis for the structural scan is described below; functional, diffusion imaging, and EEG components of this study will be reported separately.
We designed a child-friendly protocol that included a training session prior to the actual scanning session. Children learned about the scanner by watching a video and getting acquainted with a scanning session in a mock scanner while listening to the different types of sound made by the scanner. During the actual scanning session, if children wished, one of the investigators remained in the scanner room and held the child’s hand. To assist children to remain motionless during the structural scans, they watched a video of their choice. After the scanning session, children were shown an actual image of their brain on the computer. High-resolution T1-weighted structural MRI images were acquired using an MPRAGE sequence on a 3T scanner equipped with a 12-channel head coil, with the following parameters: 1 mm × 1 mm × 1 mm resolution over a 256 mm × 256 mm × 208 mm FOV; TI/TE/TR = 800/3.09/2530 ms; flip angle = 10∘∘; GRAPPA acceleration factor
To analyze our structural MR images, we used the BrainSuite software
Using a surface to volume registration technique (SVReg, please see
Each brain was examined after automated transfer and manual correction was applied whenever necessary due to edge mislabeling. We then determined gray matter volume, cortical thickness, and surface area, for each individual lobe, the insula and cingulate. Gray matter volume was computed using volumetric labels as well as the partial tissue fraction data generated by the BrainSuite and SVReg sequences which contains tissue fraction value corresponding to the percentages of gray matter, white matter and CSF in every voxel. Using these values, we computed the total amount of gray matter in each region of interest. Cortical thickness was defined as the linked distance between inner cortical surface (the gray/white separation line) and pial surface. This measure was computed at every vertex on the surface mesh defining the two surfaces. The vertexwise cortical thickness measure for all the vertices within a region of interest was subsequently averaged to produce the final cortical thickness value for that region of interest. Surface area was defined as the area of the pial cortical surface mesh for the particular region of interest. We also obtained the total volume (gray and white matter) for the cerebellum and total white matter volume for the corpus callosum.
Children from the three groups performed within the normal range reported for all the standard behavioral measures (
Mean and SD for behavioral assessments by group; univariate ANOVA results for each behavioral outcome by group.
Assessment | Music, sports control, and non-sports control | ANOVA results |
---|---|---|
FSIQ | 100.7 (11.5); 96.2 (8.45); 92.6 (10.5) | |
Digit span | 10.1 (2.1); 9.5 (2.8); 9.9 (2.0) | |
Auditory analysis | 10.6 (2.9); 9.5 (4); 10 (2.8) | |
BOT motor development | 57.1 (8); 57.4 (7.8); 57.5 (6.3) | |
Gordon’s PMMA | 61 (7.2); 60.9 (6.3); 60.4 (8) | |
Pro-social behavior | 2 (1.6); 2.3 (1.7); 2.53 (1.32) | |
Mind in the eye | 15.6 (4.2); 15.6 (4.7); 13.4 (4) | |
Index of empathy | 11.9 (4.1); 12.1 (2.2); 11.3 (2.7) | |
Emotion match | 0.48 (0.3); 0.49 (0.4); 0.6 (0.3) |
As for the correlation between Gordon’s PMMA and social tasks, including Pro-social Behavior, Emotion Match Assessment, Reading the Mind in the Eyes and Index of Empathy, the only significant correlation was between Gordon’s PMMA, and the Mind in the Eye’s test, however after Bonferroni correction for multiple comparisons, the correlation was no longer significant (
Intercorrelations among behavioral outcome variables for three groups (two tailed bivariate correlations).
FSIQ | Mind in the eye | Index of empathy | Emotion match | Pro-social behavior | |
---|---|---|---|---|---|
Gordon’s PMMA | 0.2067 | 0.3488* | 0.0535 | 0.1414 | -0.0303 |
FSIQ | -0.0849 | 0.1939 | -0.0329 | -0.2271 | |
Mind in the eye | -0.0473 | -0.1763 | -0.076 | ||
Index of empathy | -0.0015 | -0.0154 | |||
Emotion match | 0.1387 |
The values found in the three groups of children for gray matter volume, cortical thickness and surface area were within the range of previously reported measures for normal brain development in this age group (
Mean and SD for brain measures by groups.
Area | Music | Sports control | Non-sports control |
---|---|---|---|
R frontal lobe | 4.20 (0.16) | 4.27 (0.25) | 4.25 (0.16) |
L frontal lobe | 4.21 (0.16) | 4.26 (0.22) | 4.20 (0.20) |
R parietal lobe | 3.94 (0.21) | 3.99 (0.12) | 3.94 (0.17) |
L parietal lobe | 4.02 (0.25) | 4 (0.20) | 3.99 (0.21) |
R temporal lobe | 4.38 (0.16) | 4.45 (0.13) | 4.33 (0.23) |
L temporal lobe | 4.43 (0.18) | 4.45 (0.11) | 4.36 (0.17) |
R occipital LOBE | 3.50 (0.26) | 3.51 (0.25) | 3.41 (0.21) |
L occipital lobe | 3.52 (0.27) | 3.56 (0.27) | 3.43 (0.21) |
R cingulate | 4.10 (0.25) | 4.11 (0.25) | 4.07 (0.26) |
L cingulate | 4.12 (0.22) | 4.07 (0.18) | 4.10 (0.24) |
R insula | 5.16 (0.25) | 5.18 (0.39) | 5.12 (0.28) |
L Insula | 5.19 (0.25) | 5.10 (0.28) | 5.14 (0.29) |
R frontal lobe | 382.72 (24.19) | 384.64 (48.45) | 365.84 (28.90) |
L frontal lobe | 379.74 (24.30) | 380.39 (44.22) | 363.03 (26.40) |
R parietal lobe | 256.86 (27.84) | 269.66 (32.59) | 258.80 (21.27) |
L parietal lobe | 270.29 (20.80) | 287.14 (35.48) | 266.07 (21.18) |
R temporal lobe | 227.40 (18.54) | 233.27 (31.72) | 221.19 (17.70) |
L temporal lobe | 214.38 (15.99) | 224.11 (36.01) | 214.62 (18.35) |
R occipital lobe | 158.65 (13.73) | 154.64 (14.37) | 157.33 (17.74) |
L occipital lobe | 153.54 (11.76) | 149.13 (23.93) | 150.82 (20.74) |
R cingulate | 39.91 (6.41) | 39.86 (6.43) | 38.94 (4.71) |
L cingulate | 46.62 (3.92) | 46.9 (5.44) | 44.29 (5.45) |
R insula | 15.70 (1.38) | 16.16 (2.08) | 15.38 (2.36) |
L insula | 15.93 (1.42) | 16.63 (2.25) | 15.53 (2.30) |
R frontal lobe | 108.30 (8.72) | 110.92 (10.56) | 103.92 (9.40) |
L frontal lobe | 107.36 (8.35) | 110.04 (10.38) | 102.93 (9.43) |
R parietal lobe | 71.41 (8.41) | 76.19 (6.25) | 72.91 (6.69) |
L parietal lobe | 74.54 (7.12) | 79.47 (8.43) | 74.33 (6.23) |
R temporal lobe | 72.22 (5.75) | 75.12 (9.33) | 68.44 (7.95) |
L temporal lobe | 68.84 (4.84) | 72.33 (10.63) | 67.30 (7.08) |
R occipital lobe | 38.42 (3.46) | 37.86 (3.99) | 37.75 (4.07) |
L occipital lobe | 36.50 (3.77) | 35.3 (4.79) | 35.75 (5.64) |
R cingulate | 11.92 (1.93) | 11.8 (1.44) | 11.44 (1.19) |
L cingulate | 13.45 (1.43) | 13.14 (1.40) | 12.49 (1.51) |
R insula | 5.99 (0.50) | 6.15 (1.02) | 5.94 (0.96) |
L insula | 5.98 (0.64) | 5.84 (1.09) | 5.64 (1.20) |
Measures of total volume of cerebellum and corpus callosum in cc (mean and SD) by group – Univariate ANOVA results for cerebellum and corpus callosum by group.
Area | Music | Sports control | Non-sports control | ANOVA results |
---|---|---|---|---|
Corpus callosum | 6.41 (0.66) | 6.30 (1.03) | 6.14 (0.75) | |
Cerebellum | 122.59 (12.54) | 122.43 (13.77) | 115.71 (12.68) |
A power analysis with fixed effects model and alpha level of 0.05 showed that the current sample size yields, an 80% probability to find an effect size typical of what has previously been reported for comparisons of adult musicians and non-musicians (
The immediate objective of the study was to determine if children beginning music training were different from children of the same age beginning sports training, or not involved in any formal training program. Students in both music and sports group had enrolled voluntarily in their respective programs; all students who sign up for the sports program are admitted, while the final group in the music program is based on a lottery admission from the list of students who have signed up. We did not find any differences in cognitive, motor, or musical ability, nor were there anatomical differences in terms of gray matter volume, cortical thickness, or surface area in any regions of interest between the three groups of children at the onset of the study. The absence of pre-existing differences between the music group and the control groups provides a foundation for the future investigation of how musical training affects the brain and cognitive/emotional development, the goal of the longitudinal study in which we are engaged. It also addresses a classic question: are the differences which have been consistently found in brain and behavior measures between adult musicians and non-musicians in the general population, present prior to training thus representing traits favorable to music education?
The likelihood that the differences between musicians and non-musicians are accounted for by experience and training is supported by evidence of brain plasticity with acquiring and practicing new skills (
The absence of pre-existing differences in our subject groups does not allow us to conclude that adult musicians and non-musicians are not different as children. We also note that the children in our study were not randomly assigned to their groups. Such a design would not have been feasible, as noted by
The second aim of this phase of the study was to determine whether children participating in music training, compared to a control group of children not engaged in music training, differ in terms of emotional and social skills at the beginning of their training. We did not find any differences in emotional or social skills in the three groups. We further aimed to determine whether musical skills, as measured by a music perception instrument, correlate with emotional and social outcomes that have been proposed to be associated with music practice, including empathy and pro-social behavior. No correlation was found between Gordon’s PMAA and the results of the social/emotional assessments performed at this time, suggesting that, at least prior to training, there are no evident relationship between musical abilities as measured by a music perception task and emotional and social skills. Our results also give some indirect support for the idea that the kinds of social and emotional skills reported in children who have studied music may be a by-product of music training (
Few studies to date have examined the association between music training and the development of emotional abilities and social skills. Given that music is often considered to be intimately linked with emotions and feelings, it is surprising that available research shows mixed findings (
Our longitudinal study will allow us to examine the role of group training (in music and sports) on the development of emotional and social skills. The participants in the music group are training in a program inspired by the Venezuelan method known as
In summary, the analysis reported here lays the groundwork for the longitudinal study of the effects of music training on childhood development as assessed by cognitive, emotional, social, and neural measures. In our groups, we found no differences in the levels of intellectual, emotional, and social performance, nor did we find differences in brain structure. The socio-economical and family environments are comparable. We hope our findings will demonstrate the value of musical education at a time when music education programs are being eliminated from standard curricula.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
First, we wish to thank all of the children and their parents who have agreed to participate in this study. We also want to acknowledge the enthusiasm and collaboration of Deborah Borda, president of the Los Angeles Philharmonic, and Tony Brown, executive director of Heart of Los Angeles. For their help in recruitment of children, we wish to thank, Brotherhood Crusade and their Soccer for Success Program and the following elementary schools and community programs in the Los Angeles area: Vermont Avenue Elementary School, Saint Vincent School and Macarthur Park Recreation Center. We also thank Leslie Chinchilla, Mayra Mexco, and Theodosia Roussos for their assistance with recruitment and data collection.
This study was supported by the Brain and Creativity Institute and by a grant from an anonymous donor. David W. Shattuck, Anand A. Joshi, and Richard M. Leahy were supported in part by the National Institute of Neurological Disorders and Stroke of the National Institutes of Health under Award Number R01NS074980. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. This study was made possible by the close collaboration of the Los Angeles Philharmonic, the Youth Orchestra of Los Angeles, and Heart of Los Angeles.