Cerebral Cortex September 2013;23:2038–2043

doi:10.1093/cercor/bhs180
Advance Access publication July 10, 2012

Music Training for the Development of Speech Segmentation

Clément François1, Julie Chobert2, Mireille Besson2,3 and Daniele Schön1
1
Institut de neurosciences des systèmes, INSERM and Aix-Marseille University, Marseille, France 2Laboratoire de Neurosciences
Cognitives, CNRS and Aix-Marseille University, Marseille, France and 3Cuban Neuroscience Center, La Habana, Cuba

Address correspondence to Clément François. Email: [email protected]

Downloaded from https://academic.oup.com/cercor/article/23/9/2038/594399 by HALLYM UNIVERSITY user on 12 March 2021

The role of music training in fostering brain plasticity and develop- findings have also been extended to children. Musically
ing high cognitive skills, notably linguistic abilities, is of great inter- trained children better detect pitch changes in speech (Magne
est from both a scientific and a societal perspective. Here, we et al. 2006; Moreno et al. 2009; Kraus and Chandrasekaran
report results of a longitudinal study over 2 years using both behav- 2010) and show increased verbal and reading abilities than
ioral and electrophysiological measures and a test-training-retest children who did not receive music training (Moreno et al.
procedure to examine the influence of music training on speech 2009, 2011), thereby providing evidence for music to
segmentation in 8-year-old children. Children were pseudo-randomly language transfer effects (Besson et al. 2011).
assigned to either music or painting training and were tested on Turning to speech segmentation, the ability to extract
their ability to extract meaningless words from a continuous flow words from continuous speech, there is evidence that infants,
of nonsense syllables. While no between-group differences were children, and adults can use the statistical properties of audi-
found before training, both behavioral and electrophysiological tory input to discover words and sound patterns (Saffran et al.
measures showed improved speech segmentation skills across 1996, 1999; Aslin et al. 1998; Kuhl 2004; Gervain et al. 2008;
testing sessions for the music group only. These results show that Teinonen et al. 2009). In speech, the conditional probability
music training directly causes facilitation in speech segmentation, of syllable Y happening given syllable X will be higher for
thereby pointing to the importance of music for speech perception syllables that follow one another within a word than for those
and more generally for children’s language development. Finally at word boundaries (e.g., in “pretty music”, the probability of
these results have strong implications for promoting the develop- “ty” given “pre” is higher than that of “mu” given “ty”). Thus,
ment of music-based remediation strategies for children with the statistical structure of a language seems to greatly contrib-
language-based learning impairments. ute to speech segmentation. Interestingly, we recently found
that musical expertise facilitated speech segmentation of an
Keywords: brain plasticity, language, longitudinal study, musical training, artificial language in adults (François and Schön 2011). Par-
N400, statistical learning ticipants were familiarized with a stream of 5 artificial trisylla-
bic sung pseudo-words and then presented with 2-alternative
forced choice tests with trisyllabic spoken pseudo-words or 3
Introduction tones melodies played on a piano timbre. On each trial, par-
Music engages a wide range of processing mechanisms, from ticipants had to choose which of 2 sequentially presented
sound encoding to higher cognitive functions such as sequen- items sounded more familiar. Musicians were more accurate
cing, attention, memory, and learning. These functions, which than nonmusicians in both the musical and linguistic tests,
are shared with several other human abilities (e.g., language), although this difference remained a trend (P = 0.11).
might in turn be shaped by music training. Thus, musicians However, analyses of the event-related potentials (ERPs) to
are a privileged population for studying brain plasticity as both musical and linguistic items revealed that a fronto-central
well as for investigating the intriguing possibility that musical negative component was significantly larger for unfamiliar
expertise transfers to other domains such as language. It is than for familiar items for musicians only. This result was
now well-established that music training induces functional taken as evidence that musical practice facilitated stream
and structural changes in the auditory and sensori-motor segmentation.
systems, making musicians more efficient and more sensitive However, cross-sectional studies comparing musicians and
in music-related tasks than nonmusicians. For instance, brain- nonmusicians demonstrate correlations but not causality
stem and primary auditory cortex responses to synthetic or in- (Schellenberg 2004). Here, we used the longitudinal approach
strumental sounds show more robust pitch encoding for to test for causality. We conducted a longitudinal study span-
musicians than for nonmusicians (Shahin et al. 2003, 2005; ning over 2 school years and we followed the developmental
Musacchia et al. 2007; Wong et al. 2007). Also, musicians dis- dynamics of music to speech transfer effects. We controlled
criminate deviant chords or detect omitted sounds better than for any pre-existing predispositions for music by using a
nonmusicians (Koelsch et al. 1999; Rüsseler et al. 2001; Bratti- test-training-retest procedure with 8-year-old children
co et al. 2009). These functional differences can be pseudo-randomly assigned to 1 of 2 training groups (music or
accompanied by morphological differences in terms of grey painting) without self-selection. In the first “test session”, 24
matter volume and density in the auditory cortex (Schlaug 8-year old children listened to 5 min of an artificial sung
et al. 1995; Keenan et al. 2001; Bermudez and Zatorre 2005). language (Schön et al. 2008) built by random concatenation
Moreover, there is growing evidence that music training of 4 trisyllabic meaningless pseudo-words. Syllables were
benefits linguistic skills such as dynamic acoustic analysis, always sung using a fixed syllable-pitch mapping (Fig. 1A).
pitch and lexical stress processing, phonological awareness, After this familiarization phase, children were presented with
reading, and second-language proficiency (e.g., Tallal and 2 spoken items and had to decide which item sounded more
Gaab 2006; Kraus and Chandrasekaran 2010). Some of these familiar (32 trials). Importantly, all items in the test were

© The Author 2012. Published by Oxford University Press. All rights reserved.
For Permissions, please e-mail: [email protected]
 music or painting lessons privately either before or during the
project. All children were French native speakers. Parental informed
consent was obtained for each child and the data were analyzed anon-
ymously. This study was approved by the CNRS and was conducted
in accordance with national norms and guidelines for the protection
of human subjects.

Longitudinal Study: Procedure

Children were tested before training (at T0), after approximately 1
year (at T1), and 2 years (at T2). At T0, T1, and T2, children were
tested individually in a quiet room of their school in 2 separate ses-

Downloaded from https://academic.oup.com/cercor/article/23/9/2038/594399 by HALLYM UNIVERSITY user on 12 March 2021

sions that included neuropsychological assessments and electro-
physiological tests, respectively. Each session lasted for 2 h and was
separated by 4 or 5 days. Results were used for the pseudo-random
assignment of children to the music or painting groups and as a base-
line at T0 to evaluate the impact of the training programs at T1 and
T2. Pseudo-random assignment to music or to painting training group
was based on results of several neuropsychological tests issued from
the WISC-IV and NEPSY batteries (Verbal Comprehension, Perceptual
Reasoning, Working Memory, and Attention) as well as on age,
school level, sex, and socio-economic background. This was done to
ensure that no significant differences existed between the 2 groups
before training. Children had similar socio-economic backgrounds
ranging from middle to low social class according to the criteria of
the National Institute of Statistics and Economic Studies. None of the
children and none of their parents had formal training in music or
painting. Moreover, children enrolled in the music group did not have
Figure 1. Effects of music and painting training on speech segmentation. (A) their instruments at home to prevent any additional practice outside
Experimental design used in the 3 test sessions. (B) Percentage of correct responses the music classes.
in the 3 test sessions (T0, T1, and T2) for the music (solid line) and painting (dashed Two teachers professionally trained in music or painting were
line) groups. Error bars indicate the standard error of the mean. Asterisks show specifically hired for this project from October to May in year 1 and in
significant differences between groups (*P = 0.05; **P < 0.01). year 2. Music training was based on a combination of Kodaly and Orff
approaches (http://www.iks.hu/; http://www.orff.de/en.html). Paint-
ing training was based on the approach developed by Arno Stern
spoken and not sung. Both EEG and behavioral responses (http://www.arnostern.com/). The teaching activity was coordinated
were recorded during the task. Children were then by the research group and care was taken to ascertain that both
pseudo-randomly assigned to 2 training groups (controlling groups were similarly motivated and stimulated.
for age, school level, sex, socio-economic background, and
musical expertise and for the level of performance in several Speech Segmentation Experiment: Design and Procedure
neuropsychological tests assessing reasoning, memory, and During the familiarization phase, children were asked to listen care-
attentional processing). One group of children took music fully to a continuous stream of sounds. During the following test, chil-
and the other painting classes for 45 min, twice a week in dren had to choose, by pressing 1 of 2 response buttons, which of 2
year 1 and once a week in year 2. “Test sessions” 2 and 3 (T1 items (first or second) most closely resembled what they just heard in
the stream. In the test, items were spoken (i.e., flat contour). In each
and T2) were identical to “test session” 1 (T0) and took place
test trial, one item was a pseudo-word from the artificial language
approximately after 1 and 2 years. We hypothesized that chil- (i.e., gimysy, pogysi, pymiso, sipygy) while the other was built by
dren in the music group would improve their speech segmen- merging the last syllable of one pseudo-word with the first 2 syllables
tation abilities across the test sessions more than children in of another (e.g., Sisipy and Sypymi) or the last 2 syllables of one
the painting group. Based on our previous findings with pseudo-word with the first syllable of another (e.g., Gysigi and
adults (François and Schön 2011) and considering that we Pygygi). Pseudo-words and partial pseudo-words did not have any
simplified the stimuli for children (4 pseudo-words rather meaning in French. The mean transitional probabilities (TP) were 0.8
for pseudo-words (ranging from 0.6 to 1) and 0.4 for partial pseudo-
than 5), we expected a behavioral facilitation together with words (ranging from 0.32 to 0.6). Each pseudo-word was presented
larger ERP differences between familiar and unfamiliar items with each partial pseudo-word, making up 16 pairs and repeated
over frontal regions in the musically trained compared with twice in a quasi-random order (32 trials). Stimuli were presented via
the painting-trained children. headphones. Learning phase and test lasted 5 min each.

Material
Materials and Methods The artificial language was built using 9 syllables combined to give
rise to 4 trisyllabic pseudo-words (gimysy, pogysi, pymiso, sipygy:
Participants i.e., non-lexical vocables respecting the phonotactic constraints of
A total of 37 8-year-old nonmusician children were enrolled in these French). Each of the 9 syllables was associated with a distinct tone.
experiments. Thirteen children were excluded from final analysis Therefore, each pseudo-word had a unique melodic contour (gimysy
either because they moved away during either the first (5) or the C3 D3 F3, pymiso B3 E4 F4, pogysi D4 C4 G3, sipygy G3 B3 C4;
second year (3) or due to inattentive behavior and impulsiveness (5), Fig. 1). The language stream was built by a random concatenation of
thus leading to a final group of 24 children (mean age = 8, standard the 4 pseudo-words (without repetition of the same item twice in a
deviation = 0.45, 19 right-handed, 14 boys, normal hearing, no known row) and synthesized using Mbrola (http://tcts.fpms.ac.be/synthesis/
neurological problems). None of the children had taken part to such mbrola.html). Each pseudo-word was repeated 100 times in the
an experiment before this project. Moreover, none of them took stream to give rise to a 5 min stream.

Cerebral Cortex September 2013, V 23 N 9 2039

Data Acquisition older at T1 and 2 years older at T2. Importantly, this improve-
As children were not given speed instructions, analyses were con- ment was of similar size in both the music and the painting
ducted on the percentage of correct responses. EEG was recorded training groups (main effect of group: F < 1; group by session
before training (at T0) and after 2 years (at T2) from 32 scalp electro-
des located at standard positions (International 10/20 system sites)
interaction: F < 1).
during the behavioral task. The data were then re-referenced offline
to the algebraic average of the mastoids. Trials containing artifacts Behavioral Data
were excluded. Artifacts were first detected by eye-balling and then
by using a 75 µV maximum amplitude criterion (less than 10% of the Figure 1B clearly shows that the level of performance steadily
trials). Two extra participants were discarded from ERP analyses only increased for the music group across testing sessions while it
due to major EEG artifacts. The EEG was amplified by Biosemi ampli- did not change for the painting group (group by session inter-
fiers with a band-pass of 0–102.4 Hz and was digitized at 512 Hz. action F2,44 = 3.4, P = 0.04). Most importantly, while the level

Downloaded from https://academic.oup.com/cercor/article/23/9/2038/594399 by HALLYM UNIVERSITY user on 12 March 2021

of performance in the music group was at chance at T0 (P =
0.40, one-sample t-test), it was higher than chance after 1 (T1)
Data Analyses
Neuropsychological data were analyzed using repeated-measures
and 2 years (T2) of music training (P = 0.02 and 0.004,
multivariate ANOVAs with group (music vs. painting) as a between- respectively). Moreover, the benefit due to the second period
subject factor, “Test Session” (T0 vs. T1 vs. T2) as a within-subject of training was similar to the benefit of the first period (P =
factor, and the score at the test as the dependent variable. Behavioral 0.80). In contrast, the level of performance in the painting
data in the 2-alternative force choice test were analyzed using group remained at chance level (0.5) at T0, T1, and T2
repeated-measures ANOVAs (RM-ANOVAs) to compare the percentage (P always >0.40).
of correct responses across groups and testing sessions. The Tukey
Results were further analyzed by taking into account the
tests were used for post hoc comparisons. Average performance was
also compared with chance level using 2-tailed one-sample t-tests. statistical structure of each of the familiar items. Familiarity
Finally, further analyses of the behavioral data modeled the effect of accuracy at T1 and T2 was higher in trials containing items
the items. This was done using a 2 × 4 RM-ANOVA including group with high TP than in trials containing items with low TP
(music and painting) a between-subject factor and Items (4 words) as (Fig. 2, main effect of Items at T1: F3,66 = 3.1; P = 0.03; main
within-subject factors, respectively, after the first and the second effect of Items at T2: F3,66 = 2.9; P = 0.04).
period of training.
This fine-grained analysis showed several important results
(Fig. 2). First, children in the music group performed better
ERP Data Analyses than children in the painting group on almost all items (main
ERP data for familiar items (averaged across the 4 familiar items and effect of group: F1,22 = 5.3; P = 0.03 at T1 and F1,22 = 5.4; P =
across children) and for unfamiliar items (averaged across the 4 unfa- 0.02 at T2). Secondly, while children in the music group did
miliar items and across children) were analyzed by computing the not succeed at recognizing the pseudo-word with the lowest
mean amplitudes in successive non-overlapping 50 ms windows from TP at T1 ( performance was at chance, P = 0.27, one-sample
0 to 1000 ms post-stimulus onset. RM-ANOVA was used for statistical
t-test), they did succeed at T2 (P = 0.01, one-sample t-test).
assessment that included group (music vs. painting) and Familiarity
(familiar vs. unfamiliar items). Moreover, to test for the distribution of Thirdly, although children in the painting group were still at
the effects, the model included the anterior–posterior (frontal, chance level with all items at T1 (P’s > 0.16, one-sample
central, and parietal) and hemisphere factors (left and right). P-values t-test), they were above chance level on the 2 pseudo-words
were adjusted using the Greenhouse–Geisser correction. Because of with the highest TP at T2, although this did not reach signifi-
the increased likelihood of type I errors, only effects that reached sig- cance (P’s > 0.14, one-sample t-test).
nificance (P < 0.05) in at least 2 consecutive 50 ms windows were con-
sidered significant.
Electrophysiological Data
The difference between ERPs to familiar and unfamiliar items
Results was tested with a 4-way RM ANOVA including group (music
vs. painting) as a between-subject factor and familiarity (fam-
Neuropsychological Data iliar vs. unfamiliar), antero-posterior (frontal, central and par-
Insofar as children were pseudo-randomly assigned to the 2 ietal) and hemispheres (right and left) as within-subject
different training groups taking into account the results at the factors. At T0, and in both groups, unfamiliar items elicited a
neuropsychological tests, the 2 groups did not differ before larger negativity than familiar items between 450 and 550 ms
training in any of the tests used (the two groups did not differ post-stimulus onset over frontal regions (familiarity by antero-
before training on: Reading age, P = 0.31; digit span direct, posterior interaction: F2,40 = 10.5, P < 0.001). Post hoc analyses
P = 0.79; digit span indirect, P = 0.79; digit span total, P = 0.95; revealed that ERPs were more negative for unfamiliar than for
similitudes, P = 0.26; symbols, P = 0.48; PM47, P = 0.75; visual familiar items over frontal regions only (−2.6 µV of effect size;
attention, P = 0.62; arrows, P = 0.68; auditory attention, P = 0.008). The hemisphere and group factors were not sig-
P = 0.23; orientation, P = 0.49; visuomotor precision, P = 0.42; nificant in the main effects or in the interactions with the
irregular words reading, P = 0.81; regular words reading, other factors (all P’s > 0.29).
P = 0.39; pseudo-words reading, P = 0.34; phoneme suppres- At T2, and in both groups, the familiarity effect was signifi-
sion, P = 0.65; phoneme fusion, P = 0.84; logatom repetition, cant over frontal regions in the 200–300 ms and in the 450–
P = 0.26). The level of performance in both groups improved 550 ms ranges (familiarity by antero-posterior interaction:
from T0 to T1 and T2. An analysis of neuropsychological data F2,40 = 4.8; P = 0.03 and F2,40 = 5.0, P = 0.03, respectively). Post
(WISC-IV and NEPSY, see the Materials and Methods section) hoc comparisons revealed that the familiarity effect between
across the 3 test sessions showed a main effect of session 200 and 300 ms was maximal over frontal regions, but this
(RM-MANOVA: F2,21 = 85, P < 0.001). This pattern of results difference did not reach significance (1.7 µV of effect size;
was expected given that children were approximately 1 year P = 0.09). In contrast, unfamiliar items elicited significantly

2040 Music Training for the Development of Speech Segmentation • François et al.
 Downloaded from https://academic.oup.com/cercor/article/23/9/2038/594399 by HALLYM UNIVERSITY user on 12 March 2021
Figure 2. Effects of music and painting training on the sensitivity to TP of each item (given along the x-axis). Percentage of correct responses is given for each of the 4 familiar
items after 1 year (T1) and after 2 years (T2) of training for the music (solid line) and painting (dashed line) groups.

more negative ERPs than familiar ones over frontal regions efficiency of general mechanisms involved in regularity ex-
between 450 and 550 ms (2.4 µV of effect size; P = 0.007). traction and sequence learning (Janata and Grafton 2003).
Most importantly, the familiarity effect in the 450–550 ms Music training thus fosters brain plasticity and facilitates
latency window was larger after 2 years in the music group speech segmentation. This facilitation may result from several
(2 µV of effect size; P = 0.002) than in the painting group (0.1 (but not exclusive) processes. Music training may improve
µV of effect size; P = 0.99; familiarity by group interaction: general auditory encoding abilities encompassing the brain-
F1,20 = 7.9, P = 0.01). In both time windows, the main effect of stem and auditory regions that, in turn, facilitate speech seg-
hemisphere or the interactions involving this factor were not mentation (Tallal and Gaab 2006; Kraus and Chandrasekaran
significant (all P’s > 0.14), except for a main effect of hemi- 2010). Alternatively, music training may facilitate the emer-
sphere in the 200–300 ms latency band (P = 0.06). gence of more stable memory traces via a more efficient
working memory and sequencing processes integrating pitch
and syllabic structures, through anatomical and/or functional
Discussion modifications going beyond the auditory regions. Finally,
The main findings of the present study can be summarized as music training may reduce the effect of interference of adja-
follows. Children with musical training improved their speech cent syllables/items (Pechmann and Mohr 1992; Berti et al.
segmentation abilities while children in the painting group 2006), possibly via more efficient temporal dynamic proces-
did not. Moreover, while the electrophysiological responses sing (Tallal and Gaab 2006), focusing of attention (Baumann
were different for familiar and unfamiliar words in both et al. 2008), or executive functions (Moreno et al. 2011).
groups, this difference was greater in the music group than in In this respect, 2 results are of particular interest. First,
the painting group. accuracy was significantly higher with items having high TP
The data reported here extend previous findings showing than with items having low TP in the music group at T1 and
that in adults, musical expertise facilitates speech segmenta- T2. In contrast, this difference did not reach significance in
tion (François and Schön 2011). In this previous experiment, the painting group, although there was a trend at T2 (Fig. 2).
behavioral results showed a trend for a musical practice Thus, children were sensitive to TP and not simply to differ-
advantage and electrophysiological data revealed a signifi- ences in the frequency of occurrence when choosing between
cantly larger fronto-central negative component for unfamiliar the “pseudo-word” and the “partial pseudo-word” in a given
than for familiar items in musicians only. Interestingly, behav- trial (“pseudo-words” being heard 3 times more often than
ioral data in children showed a clear advantage of the music “partial pseudo-words” during the learning phase).
training group. This slight discrepancy between adults and Secondly, the scalp distribution of the familiarity effect is
children behavior might be due to the fact that the stream and frontal, in line with previous ERPs and functional magnetic
test used with adults were more complex. An alternative resonance imaging data showing activity in the inferior and
explanation could be that both adult musicians and nonmusi- middle frontal gyri taken to index the implicit detection of
cians are already skilled enough at stream segmentation while word boundaries in adults and children (McNealy et al. 2006,
this ability is still developing in 8 year-old children, thus al- 2010, 2011; Cunillera et al. 2009). Interestingly, Slumming
lowing to observe training-related differences. Concerning et al. (2002) reported an increased grey matter density and
electrophysiological data, the morphology and topography of volume in the left inferior frontal gyrus of musicians (Slum-
the negative component were similar in adults and children. ming et al. 2002). Therefore, while music training certainly
Indeed, as it was the case in adult musicians, a fronto-central influences the functional organization of the auditory subcor-
negative component was sensitive to the degree of familiarity tical and cortical network, it seems that its impact on brain
of the items in the children music group only. This music plasticity goes beyond the auditory system tapping onto the
training advantage could emerge, possibly via increased dorsal and ventral pathways which seem to play an important

Cerebral Cortex September 2013, V 23 N 9 2041

role in language acquisition and higher order processes (Scott
and Wise 2004; Hickok and Poeppel 2007; Rodriguez-Fornells
et al. 2009).
Importantly, the longitudinal approach, coupled with
pseudo-random assignment to 1 of the 2 training groups, con-
trols for possible pre-existing predispositions for music and
ascertains that music is the cause of the observed changes
(e.g., Schellenberg 2004; Moreno et al. 2009). Note also that
the effect described here is not a general effect due to higher
motivation or arousal in the music class, since children in

Downloaded from https://academic.oup.com/cercor/article/23/9/2038/594399 by HALLYM UNIVERSITY user on 12 March 2021

both groups improved equally well in the neuropsychological
tests. Intriguingly, we did not replicate the results reported by
Schellenberg (2004) who showed significant improvement on
intelligence tests after 1 year of musical training. However,
one should note that Schellenberg used the IQ-full scale in-
cluding 5 subtests for each type of IQ. Because our longitudi-
nal study did include several experiments, we only used a
subset of performance and verbal IQ in order to keep the
testing session duration reasonable. Moreover, other impor-
tant differences are related to the age of the children and to
the sample sizes. Children enrolled in Schellenberg’s study
were younger (6 years old) than the ones enrolled in the
present study (8 years old). Musical training at this younger
age may act as the general school environment and improve
IQ as typically seen at the start of schooling (Schellenberg
2004). Also, the sample size was twice as large in Schellen-
berg’s study than in ours. This factor seems to play an impor-
tant role since Moreno et al. (2011) recently tested 64 children
and reported improved verbal ability after 20 days of musical
training but not of visual arts training. In contrast, in a pre-
vious study of our group, conducted with another sample of
37 8-year-old children (Moreno et al. 2009), the group by
session interaction for the full-scale IQ was not significant
(P > 0.20). Also Hyde et al. (2009) conducted a similar longi-
tudinal experiment with 31 children and failed to find any sig- Figure 3. (A) Grand-averages event-related potentials to familiar (thick) and
nificant differences in general intelligence measures while unfamiliar (dotted) items (averaged across items) recorded during the behavioral task
they observed structural changes in auditory and motor brain before training (T0) in the music group (top) and in the painting group (below). (B)
Map showing the distribution of the effect (unfamiliar–familiar) in the 450–550 ms
areas after 15 months of training. They suggested that larger range window (averaged across sessions and groups). (C) Grand-averages
groups may be better suited to confirm the Schellenberg event-related potentials to familiar (thick) and unfamiliar (dotted) items (averaged
results. Thus, the similar general improvement in the 2 across items) recorded during the behavioral task after 2 years (T2) in the music and
groups is most likely driven by maturation and repetition in the painting groups (F3 electrode, n = 12). Shaded area shows a significant larger
familiarity effect size in musically trained children compared with painting-trained
effects. In contrast, both the higher level of performance in children.
the speech segmentation task and the larger ERP familiarity
effect in the music group were driven by musical training
(Fig. 3). This strongly supports the view that musical training, by fos-
Our findings provide new evidence that music training can tering rhythm perception and production, may have an
play an important role in children’s language development by important role for the development of language skills in chil-
facilitating speech segmentation, a building block of language dren. Thus, taken together, these results favor the idea that by
acquisition. Importantly, speech segmentation abilities are developing both perceptual and cognitive functions, music
known to be closely linked to other speech abilities in typi- training shapes individual development (Patel 2010).
cally developing children and to be impaired in children with
speech disorders. Recent studies in typically developing
infants and school-age children point to a strong link between Funding
speech segmentation abilities and more general linguistic pro- This research was conducted at the INCM, CNRS & Université
ficiency such as expressive lexicon (Newman et al. 2006) and de la Méditerranée, Marseille, France, and was supported by a
foreign language proficiency (McNealy et al. 2011). Moreover, grant from the ANR-Neuro-07-02401 to M.B. and
children with language-based learning impairments not only ANR-09-BLAN-0310 to D.S.
have difficulties in speech segmentation tasks (Evans et al.
2009) and an impoverished perception of speech rhythms
(Abrams et al. 2009; Goswami et al. 2011), but also have a Notes
poorer performance than typically developing children in We wish to thank the children who participated in this long-lasting
tasks involving musical metrical structures (Huss et al. 2011). project as well as their parents, the teachers, and the schools

2042 Music Training for the Development of Speech Segmentation • François et al.
principals, Mrs Muriel Gaiarsa and Mr Jean-Jacques Gaubert, as well children: Behavioral and electrophysiological approaches. J Cogn
as Johannes Ziegler, Jennifer Coull, and Nia Cason for helpful com- Neurosci. 18(2):199–211.
ments on a previous version of this manuscript. Conflict of Interest: McNealy K, Mazziotta JC, Dapretto M. 2011. Age and experience
None declared. shape developmental changes in the neural basis of
language-related learning. Dev Sci. 14(6):1261–1282.
McNealy K, Mazziotta J, Dapretto M. 2006. Cracking the language
code: Neural mechanisms underlying speech parsing. J Neurosci.
References 26(29):7629–7639.
Abrams DA, Nicol T, Zecker S, Kraus N. 2009. Abnormal cortical pro- McNealy K, Mazziotta JC, Dapretto M. 2010. The neural basis of
cessing of the syllable rate of speech in poor readers. J Neurosci. speech parsing in children and adults. Dev Sci. 13(2):385–406.
29(24):7686–7693. Moreno S, Bialystok E, Barac R, Schellenberg EG, Cepeda NJ, Chau T.
Aslin RN, Saffran JR, Newport EL. 1998. Computation of conditional 2011. Short-term music training enhances verbal intelligence and

Downloaded from https://academic.oup.com/cercor/article/23/9/2038/594399 by HALLYM UNIVERSITY user on 12 March 2021

probability statistics by 8-month-old infants. Psychol Sci. 9 executive function. Psychol Sci. 22(11):1425–1433.
(4):321–324. Moreno S, Marques C, Santos A, Santos M, Castro SL, Besson M. 2009.
Baumann S, Meyer M, Jäncke L. 2008. Enhancement of Musical training influences linguistic abilities in 8-year-old chil-
auditory-evoked potentials in musicians reflects an influence of dren: More evidence for brain plasticity. Cerebral Cortex. 19
musical expertise but not selective attention. J Cogn Neurosci. 20 (3):712–723.
(12):2238–2249. Musacchia G, Sams M, Skoe E, Kraus N. 2007. Musicians have en-
Bermudez P, Zatorre RJ. 2005. Differences in gray matter between hanced subcortical auditory and audiovisual processing in speech
musicians and nonmusicians. Ann N Y Acad Sci. 1060:395–399. and music. Proc Natl Acad Sci USA. 104(40):15894–15898.
Berti S, Münzer S, Schröger E, Pechmann T. 2006. Different interfer- Newman R, Ratner NB, Jusczyk AM, Jusczyk PW, Dow KA. 2006.
ence effects in musicians and a control group. Exp Psychol. 53 Infant’s early ability to segment the conversational speech signal
(2):111–116. predicts later language development: A retrospective analysis.
Besson M, Chobert J, Marie C. 2011. Transfer of training between Develop Psychol. 42(4):643–655.
music and speech: Common processing, attention, and memory. Patel AD. 2010. In: Bailar M, editor. Music, biological evolution, and
Front Psychol. 2:94. the brain. Emerging disciplines. Houston: Rice University Press. p.
Brattico E, Pallesen KJ, Varyagina O, Bailey C, Anourova I, Järvenpää 91–144.
M, Eerola T, Tervaniemi M. 2009. Neural discrimination of nonpro- Pechmann T, Mohr G. 1992. Interference in memory for tonal pitch:
totypical chords in music experts and laymen: An MEG study. J Implications for a working-memory model. Mem Cognit. 20
Cogn Neurosci. 21(11):2230–2244. (3):314–320.
Cunillera T, Càmara E, Toro JM, Marco-Pallares J, Sebastian-Galles N, Rodriguez-Fornells A, Cunillera T, Mestress-Misse A, De Diego Bala-
Ortiz H, Pujol J, Rodriguez-Fornells A. 2009. Time course and guer R. 2009. Neurophysiological mechanisms involved in
functional neuroanatomy of speech segmentation in adults. Neuro- language learning in adults. Phil Trans Roy Soc B. 364:3711–3735.
image. 48(3):541–553. Rüsseler J, Altenmüller E, Nager W, Kohlmetz C, Münte TF. 2001.
Evans JL, Saffran JR, Robe-Torres K. 2009. Statistical learning in chil- Event-related brain potentials to sound omissions differ in musi-
dren with specific language impairment. J Speech Lang Hear Res. cians and non-musicians. Neurosci Lett. 308:33–36.
52(2):321–335. Saffran JR, Aslin RN, Newport EL. 1996. Statistical learning by
François C, Schön D. 2011. Musical expertise boosts implicit learning 8-month old infants. Science. 274:1926–1928.
of both musical and linguistic structures. Cerebral Cortex. 21 Saffran JR, Johnson E, Aslin RN, Newport EL. 1999. Statistical learning
(10):2357–2365. of tone sequences by human infants and adults. Cognition. 70
Gervain J, Macagno F, Cogoi S, Peña M, Mehler J. 2008. The neonate (1):27–52.
brain detects speech structure. Proc Natl Acad Sci USA. Schellenberg EG. 2004. Music lessons enhance IQ. Psychol Sci. 15
105:14222–14227. (8):511–514.
Goswami U, Wang HL, Cruz A, Fosker T, Mead N, Huss M. 2011. Schlaug G, Janckë L, Huang Y, Steinmetz H. 1995. In vivo evidence of
Language-universal sensory deficits in developmental dyslexia: structural brain asymetry in musicians. Science. 267
English, Spanish and Chinese. J Cogn Neurosci. 23(2):325–337. (5198):699–701.
Hickok G, Poeppel D. 2007. The cortical organization of speech pro- Schön D, Boyer M, Moreno S, Besson M, Peretz I, Kolinsky R. 2008.
cessing. Nat Rev Neurosci. 8:393–402. Song as an aid for language acquisition. Cognition. 106(2):975–983.
Huss M, Verney JP, Fosker T, Mead N, Goswami U. 2011. Music, Scott SK, Wise RJ. 2004. The functional neuroanatomy of prelexical
rhythm, rise time perception and developmental dyslexia: Percep- processing in speech perception. Cognition. 92(1-2):13–45.
tion of musical meter predicts reading and phonology. Cortex. 47 Shahin A, Bosnyak DJ, Trainor LJ, Roberts LE. 2003. Enhancement of
(6):674–689. neuroplastic P2 and N1c auditory evoked potentials in musicians.
Hyde KL, Lerch J, Norton A, Forgeard M, Winner E, Evans AC, J Neurosci. 23(13):5545–5552.
Schlaug G. 2009. Musical training shapes structural brain develop- Shahin A, Roberts LE, Pantev C, Trainor LJ. 2005. Modulation of P2
ment. J Neurosci. 29(10):3019–3025. auditory-evoked responses by the spectral complexity of musical
Janata P, Grafton ST. 2003. Swinging in the brain: Shared neural sub- sounds. Neuroreport. 16(16):1781–1785.
strates for behaviors related to sequencing and music. Nat Neuro- Slumming V, Barrick T, Howard M, Cezayirli E, Mayes A, Roberts N.
sci. 6:682–687. 2002. Voxel-based morphometry reveals increased gray matter
Keenan JP, Thangaraj V, Halpern AR, Schlaug G. 2001. Absolute pitch density in Broca’s area in male symphony orchestra musicians.
and planum temporale. Neuroimage. 14(6):1402–1408. Neuroimage. 17(3):1613–1622.
Koelsch S, Schröger E, Tervaniemi M. 1999. Superior pre-attentive Tallal P, Gaab N. 2006. Dynamic auditory processing, musical experi-
auditory processing in musicians. Neuroreport. 10(6):1309–1313. ence and language development. Trends Neurosci. 29(7):382–390.
Kraus N, Chandrasekaran B. 2010. Music training for the development Teinonen T, Fellman V, Näätänen R, Alku P, Huotilainen M. 2009.
of auditory skills. Nat Rev Neurosci. 11(8):599–605. Statistical language learning in neonates revealed by event-related
Kuhl PK. 2004. Early language acquisition: Cracking the speech code. brain potentials. BMC Neurosci. 13:10–21.
Nat Rev Neurosci. 207(4427):203–205. Wong PCM, Skoe E, Russo NM, Dees T, Kraus N. 2007. Musical
Magne C, Schön D, Besson M. 2006. Musician children detect pitch experience shapes human brainstem encoding of linguistic pitch
violations in both music and language better than nonmusician patterns. Nat Neurosci. 10:420–422.

Cerebral Cortex September 2013, V 23 N 9 2043