Neuropsychologia 119 (2018) 118–127

Contents lists available at ScienceDirect

Neuropsychologia
journal homepage: www.elsevier.com/locate/neuropsychologia

Roles of posterior parietal and dorsal premotor cortices in relative pitch T

processing: Comparing musical intervals to lexical tones
⁎
Chen-Gia Tsaia,b, Tai-Li Choub,c,d,e, Chia-Wei Lif,
a
Graduate Institute of Musicology, National Taiwan University, Taipei, Taiwan
b
Neurobiology and Cognitive Science Center, National Taiwan University, Taipei, Taiwan
c
Department of Psychology, National Taiwan University, Taipei, Taiwan
d
Graduate Institute of Brain and Mind Sciences, National Taiwan University, Taipei, Taiwan
e
Graduate Institute of Linguistics, National Taiwan University, Taipei, Taiwan
f
Department of Radiology, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan

A R T I C LE I N FO A B S T R A C T

Keywords: Humans use time-varying pitch patterns to convey information in music and speech. Recognition of musical
Attention melodies and lexical tones relies on relative pitch (RP), the ability to identify intervals between two pitches. RP
Pitch processing in music is usually more fine-grained than that in tonal languages. In Western music, there are twelve
Lexical tone pitch categories within an octave, whereas there are only three level (non-glide) lexical tones in Taiwanese (or
Music
Taiwanese Hokkien, a tonal language). The present study aimed at comparing the neural substrates underlying
FMRI
RP processing of musical melodic intervals with that of level lexical tones in Taiwanese. Functional magnetic
resonance imaging data from fourteen participants with good RP were analyzed. The results showed that ima-
gining the sounds of visually presented musical intervals was associated with enhanced activity in the central
subregion of the right dorsal premotor cortex (dPMC), right posterior parietal cortex (PPC), and right dorsal
precuneus compared to auditory imagery of visually presented Taiwanese bi-character words with level lexical
tones. During the sound-congruence-judgement task (auditory imagery of musical intervals or bi-character
words, and subsequently judging if the imagined sounds were melodically congruent with heard sounds), the
contrast of the musical minus linguistic conditions yielded activity in the bilateral dPMC-PPC network and dorsal
precuneus, with the dPMC activated in the rostral subregion. The central dPMC and PPC may mediate the
attention-based maintenance of pitch intervals, whereas the dorsal precuneus may support attention control and
the spatial/sensorimotor processing of the fine-grained pitch structures of music. When judging the congruence
between the imagined and heard musical intervals, the bilateral rostral dPMC may play a role in attention
control, working memory, evaluation of motor activities, and monitoring mechanisms. Based on the findings of
this study and recent studies of amusia, we suggest that higher order cognitive operations are critical to the more
fine-grained pitch processing of musical melodies compared to lexical tones.

1. Introduction
One prominent feature shared between music and language is the
deliberate use of pitch variation in conveying information. Parallels of
pitch perception in speech and music can be noted in the global and
local pitch patterns (Nan and Friederici, 2013; Xie and Myers, 2015).
Sentence intonation is analogous to the global pitch contour of a mu-
sical melody, which tracks the direction of musical pitch movement
over time (Stalinski et al., 2008). At the lexical level, local pitch var-
iations determine word meanings in tonal languages. The latter form of
pitch information for lexical tone categorization may be comparable to
short musical melodies (Nan and Friederici, 2013) or melodic intervals
in music (Bradley, 2016).
While the global pitch perception is essential for both tonal and non-
tonal languages, the local pitch perception for lexical tone categoriza-
tion seems to play a more important role for tonal languages relative to
non-tonal languages. This raises a question about how experience with
a tone language affects the local pitch perception for music. Deutsch
et al. (2006) hypothesized that an acquired tonal language might have a
considerable impact on the development of absolute pitch (AP), a rare
ability to identify musical notes without a reference pitch. This hy-
pothesis, however, was not supported by the experimental results that

⁎
Corresponding author.
E-mail address: 799032@w.tmu.edu.tw (C.-W. Li).

https://doi.org/10.1016/j.neuropsychologia.2018.07.028
Received 22 March 2018; Received in revised form 24 July 2018; Accepted 25 July 2018
Available online 26 July 2018
0028-3932/ © 2018 Elsevier Ltd. All rights reserved.
C.-G. Tsai et al. Neuropsychologia 119 (2018) 118–127

Table 1
The five conditions in the fMRI study.
Task Stimuli Cognitive processes

Visual presentation Auditory Auditory imagery via Comparing the imagined and Fine-grained pitch
presentation subvocal rehearsal heard pitch intervals processing

Auditory-imagery task Number-pseudoword* – – – –

Bi-character word – O – –
Musical score – O – O
Sound-congruence- Bi-character word Tone pair O O –
judgement task Musical score Tone pair O O O

Note: The auditory-imagery task was to ask the participant to imagine the sound of a written number-pseudoword combination, a written bi-character Mandarin
word (in Taiwanese), or written musical scores (with numerical notations) via covertly speaking/singing. The participant was instructed to judge whether the visual
stimulus could be spoken/sung. Moreover, the sound-congruence-judgment task was to ask the participant to imagine the sounds of visually-presented musical
intervals or bi-character words. The participant was instructed to judge if the imagined sounds were melodically congruent with the auditory presentation of a tone
pair.
* Baseline condition. O The involved cognitive process(es).

Fig. 1. Examples of the stimuli for the lin-

guistic conditions, music conditions, and
baseline. There were three types of visual
presentation: Taiwanese bi-character words,
tone pairs notated using numerical notation,
and number-pseudoword combinations. In this
example of numerical notation, “5” and “1”
correspond to sol and do, respectively. A vi-
sually presented bi-character word was fol-
lowed by the sound of a tone pair, which was
either congruent or incongruent with the lex-
ical tones of the word with regard of melody. A
visually presented number pair was followed
by the sound of a tone pair, which was either
congruent or incongruent with the numerical
musical notation. We used the incongruent
tone pair with the pitch interval of an adjacent
category of the congruent tone pair. For the
interval category of Mandarin words, the pitch
interval between F#4 and B3 (seven semi-
tones) was adjacent to the pitch interval be-
tween F#4 and E4 (two semitones). For the
interval category of Western music, the pitch
interval between F#4 and B3 was adjacent to
the pitch interval between G4 and B3 (eight semitones). The fact that there was a smaller difference between the congruent and incongruent musical pitch intervals
than the difference between the congruent and incongruent linguistic pitch intervals reflects the more fine-grained pitch structures of music compared to language.

no correlation between AP and lexical tone identification was observed
in musicians (Lee and Lee, 2010; Lee et al., 2011). AP should not be
confused with another important musical ability, relative pitch (RP),
which allows to identify musical pitch intervals or relations between
several pitches. RP possessors are able to categorize pitch intervals
while singing and perceiving pitch intervals. When a familiar melody is
transposed, RP possessors are able to recognize it even though the
fundamental frequencies of this melody are changed. Since lexical tones
are mainly determined by contextual pitch cues, it is believed that RP is
more essential than AP for comprehension of a tonal language (Zhang
et al., 2012, 2017; Zhang and Chen, 2016).
Evidence has accumulated showing that the RP processing of lexical
tones and musical melodic intervals rely on overlapping neural cir-
cuitry. Several regions in the posterior parietal cortex (PPC) were in-
volved in the processing of lexical tones (Klein et al., 2001; Gandour
et al., 2004) and musical melodies in RP possessors (Schulze et al.,
2009), including the superior parietal lobule (SPL), intraparietal sulcus
(IPS), and inferior parietal lobule (IPL). Li et al. (2010) showed in-
creased activation in the right IPL for the processing of lexical tones
compared to the processing of consonants and rhymes. Foster and
Zatorre (2010) found that activity within the right IPS predicted task
performance in recognition of transposed melody, a task requiring lis-
teners to use pitch interval information. In a study on musicians’
neuroanatomical features, musicians with RP exhibited greater cortical
thickness or gray matter concentration in the IPL, IPS, SPL, and dorsal
premotor cortex (dPMC) than musicians with AP (Bermudez et al.,
2009). This may be linked to RP musicians’ superior ability of tonal
working memory than AP musicians. It has been proposed that the
processing of lexical tones was closely related to RP and working
memory (Nan et al., 2010; Wong et al., 2012; Bidelman et al., 2013).
During RP tasks, working memory may be involved with (1) retrieving
pitch intervals from long-term memory, (2) maintaining and/or at-
tending to the retrieved and/or heard pitch intervals, and (3) com-
paring these pitch intervals. The PPC and dPMC may be part of a do-
main-general network subserving RP processing in music and speech.
Although there are shared neural substrates for the RP processing of
lexical tones and musical melodic intervals, very few neuroimaging
studies directly compared them. This is partially due to the fact that the
majority of lexical tones are characterized by distinct patterns of gliding
pitch, whereas musical pitches are usually steady. Specifically, three of
the four Mandarin tones are characterized by pitch glides and thus not
exactly analogous to the scale tones in Western music (Lee and Lee,
2010). To circumvent this problem, Zhang et al. (2017) used Cantonese
words to examined the neural substrates of RP processing of lexical
tones and musical melodic intervals. The reason to use Cantonese in-
stead of Mandarin was that Cantonese has three level tones with steady

119
C.-G. Tsai et al. Neuropsychologia 119 (2018) 118–127

Fig. 2. Schematic description of the functional scanning runs. Each trial began with a silence and warning tone. Then, numbers or characters were presented on the
screen. For the baseline condition, musical auditory-imagery condition, and linguistic auditory-imagery condition, the participants were instructed to imagine sounds
via subvocal rehearsal according to the visual presentation. At the end of a trial the participant answered whether the visual presentation could be sung or spoken by
pressing a button. For the musical and linguistic sound-congruence-judgement conditions, the participant was instructed to covertly sing/speak the visual pre-
sentation with the heard tone pair, which was consecutively presented for three times within 1.8 s. If the visually presented musical interval or bi-character word
could be correctly sung/spoken with the heard pitch interval, the visually presented pitch interval and the heard pitch interval were judged as melodically congruent.
Otherwise, they were judged as melodically incongruent. At the end of a trial the participant answered whether the pitch interval of the visual presentation was
melodically congruent with the heard tone pair by pressing a button.

Fig. 3. Behavioral results of the correct response rate of the two sound-congruence-judgment conditions. (a) Bar graph with mean and standard error of the mean
(SEM). (b) Stripcharts of linked observations. (c) Stripcharts. The correct rate for the musical sound-congruence-judgment condition was significant lower than that of
the linguistic sound-congruence-judgment condition (* p < 0.05).

120
C.-G. Tsai et al. Neuropsychologia 119 (2018) 118–127

Table 2 patterns of bi-character words with level tones are called linguistic pitch
Contrast results for the auditory-imagery task. The MNI coordinates identify the intervals.
locations of cluster maxima and local maxima (> 4 mm apart within each Compared to the RP processing for tonal languages, the RP pro-
cluster). cessing for music requires more fine-grained pitch resolution. In
X Y Z Volume Information t-value Cluster Western music, for example, there are twelve pitch categories within an
(voxels) octave, whereas there are only three level lexical tones in Taiwanese.
The perfect fourth (five semitones), augmented fourth (six semitones),
Linguistic auditory-imagery condition > Baseline
12 − 84 −2 Lingual gyrus, paraippocampal 10.24 5840
and perfect fifth (seven semiones) can be used to speak the same
gyrus Taiwanese bi-charater word. In music domain, however, they are three
10 − 76 − 14 Cerebellar lobule V-VI 8.41 different pitch intervals. The aim of this study was to identify the neural
36 − 72 − 20 Fusiform gyrus 8.02 regions showing increased activity for the more fine-grained RP pro-
18 − 92 20 Superior/middle occipital gyrus 7.47
cessing of musical pitch intervals compared to linguistic pitch intervals.
6 − 44 −2 Cerebellar lobule IV 7.05
12 − 84 34 Cuneus 6.99 While undergoing fMRI, participants were asked to imagine the sounds
12 − 72 56 Precuneus 6.32 of visually presented musical intervals or Taiwanese words. Sounds of
34 − 76 48 Superior parietal lobule, 5.25 tone pairs were presented in half of the experimental trials, in which the
angular gyrus
participants were additionally asked to judge whether the imagined
− 20 12 −2 Putamen 9.79 364
4 − 16 10 Thalamus 6.07 86
sounds were melodically congruent with the heard sounds (sound-
12 16 2 Caudate 5.21 137 congruence-judgement task). As mentioned above, the PPC-dPMC net-
6 −6 16 Lentiform nucleus 4.94 192 work may be part of a domain-general working memory network for
− 42 4 − 12 Superior temporal gyrus 8.48 804 pitch processing. It seems plausible to predict that the PPC-dPMC net-
− 54 8 14 Rolandic operculum 5.93
work may show enhanced activity in both musical and linguistic RP-
− 44 12 −8 Anterior insula 5.48
− 64 − 40 22 Supramarginal gyrus, superior 4.37 50 related tasks relative to the baseline. Moreover, the PPC-dPMC network
temporal gyrus (left) may be observed in the contrast of the musical minus linguistic con-
58 − 40 22 Supramarginal gyrus, superior 3.96 52 ditions due to its putative role in the more fine-grained pitch processing
temporal gyrus (right) of music.
38 58 20 Middle frontal gyrus (right) 6.44 514
30 64 14 Superior frontal gyrus (right) 5.47
When comparing the musical to linguistic sound-congruence-jud-
− 30 40 28 Middle frontal gyrus (left) 5.59 215 gement conditions, brain regions related to attention, evaluation of
− 32 58 24 Superior frontal gyrus (left) 3.60 subvocal rehearsal and monitoring mechanisms may be activated. The
12 − 72 56 Precuneus 6.32 155 dorsolateral prefrontal cortex (dlPFC) has been found to play a role in
− 10 − 90 − 32 Cerebellar lobule Crus I-II 5.73 89
error-monitoring (Bengtsson et al., 2009), attention control (Vohn
58 14 − 10 Superior temporal pole 5.59 416
38 6 6 Insula, inferior frontal gyrus 5.48 et al., 2007; Natale et al., 2009; Wang et al., 2010; Braga et al., 2013),
64 −4 −8 Superior/middle temporal gyrus 4.80 and working memory (Veltman et al., 2003; Manoach et al., 2004;
4 2 62 Supplementary motor area 5.08 412 Mitchell, 2007; Takahama et al., 2010; Harding et al., 2016), and is
−6 24 30 Anterior cingulate gyrus 4.95 therefore a candidate region that supports attention and working
−2 20 38 Middle cingulate gyrus 3.76
2 16 42 Superior/medial frontal gyrus 3.56
memory in the fine-grained processing of musical pitch intervals during
− 28 − 100 10 Middle occipital gyrus 4.71 194 the sound-congruence-judgement task. Recent studies have indicated
− 12 − 94 14 Cuneus 3.97 that higher order cognitive operations may be critical to the perfor-
50 6 48 Precentral gyrus 4.62 102 mance of musical RP processing tasks. Albouy et al. (2013) reported a
48 6 54 Middle frontal gyrus 4.45
correlation between the impairment in maintaining musical melodic
(dorsolateral prefrontal cortex)
54 0 48 Premotor cortex (right) 3.63 information and decreased gamma oscillations within the right dlPFC.
− 50 − 2 52 Premotor cortex (left) 4.22 51 Zhang et al. (2017) observed that impaired attention to repeated pitch
8 − 36 −8 Cerebellar lobule III 4.61 33 stimuli or effortful encoding of repeated pitch stimuli may be reflected
Musical auditory-imagery condition > Linguistic auditory-imagery condition in abnormal activation in the middle frontal gyrus and precuneus. In
30 − 52 50 Inferior parietal lobule 7.84 303 the current study, we investigated the role of higher order cognitive
50 − 32 46 Supramarginal gyrus 7.15 operations in RP processing by comparing the musical versus linguistic
48 − 48 60 Intraparietal sulcus 5.46
conditions. Compared to the RP processing for lexical tones, the RP
28 −2 54 Dorsal premotor cortex 7.53 120
28 0 66 Superior frontal gyrus 5.78 processing for music may engage higher order cognitive operations
18 − 66 52 Superior parietal lobule, dorsal 4.19 33 such as working memory and attention control for more fine-grained
precuneus pitch discrimination.

2. Materials and methods

pitches. The stimuli in Zhang et al.’s (2017) study were spoken Can-
tonese words contrasting the three level tones and musical stimuli that
were matched in pitch with these spoken Cantonese words. Although
Zhang et al. (2017) demonstrated the common and differential activa-
tion maps yielded by listening to spoken words and melodically mat-
ched musical stimuli, the participants were not instructed to explicitly
detect changes in the heard pitch intervals. Therefore, the neural me-
chanisms underlying explicit RP processing still remain unclear. To
characterize the explicit RP processing for musical melodic intervals
and lexical tones, the present study used pairs of musical tones and
Taiwanese bi-character words with level tones as stimuli. Similar to
Cantonese, Taiwanese (a tonal language, also known as Taiwanese
Hokkien) has three level tones (Wu, 2000). The pitch variation patterns
of bi-character words with level tones bear resemblance to musical
melodic intervals. In the remaining part of this paper, pitch variation
2.1. Participants
Fifty-two individuals responded to an online advertisement of the
present study and completed an online RP pre-test, which was a shor-
tened version of the sound-congruence-judgement task in the fMRI
experiment (see below). Eighteen of them were allowed to enter the
fMRI experiment because their correct response rates for the four ex-
perimental conditions were equal to or higher than 80%, but one was
not willing to participate in the fMRI experiment. The remaining 17
participants completed the fMRI experiment. The data of one partici-
pant were discarded because of hardware glitches during the fMRI
scans. The data of the other two participants were discarded because
their correct rates were lower than 70% during the fMRI scans. The
final sample consisted of 14 participants (11 female, age ranged 20–27

121
C.-G. Tsai et al. Neuropsychologia 119 (2018) 118–127

Fig. 4. Contrast results for the auditory-imagery task. The threshold was set at FDR-corrected p < 0.05 with a minimum cluster size of 30 voxels. (A) The linguistic
auditory-imagery condition was associated with significant greater activation in the bilateral anterior insula, bilateral dorsal premotor cortices, bilateral caudate,
bilateral middle frontal gyrus, bilateral superior temporal gyrus, precuneus, supplementary motor area, and right angular gyrus relative to the baseline condition. (B)
Significantly increased activation in the right intraparietal sulcus, right angular gyrus, right superior parietal lobule, right precuneus, and right dorsal premotor
cortex (central subregion) was observed for the contrast of the musical minus linguistic auditory-imagery conditions.

years) with correct rates higher than 70% during the fMRI scans. None
of the participants reported neurological, mental, or hearing problems.
This fMRI study was conducted in accordance with the Declaration of
Helsinki, and all procedures were approved by the Internal Review
Board of National Taiwan University, Taiwan. Written informed con-
sent was obtained from all participants, and written instructions were
given to the participants before the fMRI experiment.
2.2. Tasks
There were two tasks during the fMRI scans, and the participants
performed one of them for each trial. The first task, auditory-imagery
task, was to attempt to imagine the sound of a visually presented
number-pseudoword combination (a number and a one-character
pseudoword), a bi-character Mandarin word (in Taiwanese), or musical
scores (with numerical notations) via covertly speaking/singing, and to
judge whether the visual presentation could be spoken/sung. The
second task, sound-congruence-judgement task, was to imagine the
sounds of visually presented musical pitch intervals or bi-character
words, and subsequently to judge if the imagined sounds were melo-
dically congruent with the auditory presentation of a tone pair. In this
task, the participant attempted to covertly sing/speak the visual pre-
sentation with the heard pitches. If the visually presented musical in-
terval or bi-character word could be correctly sung/spoken with the
heard pitch interval, the visually presented pitch interval and the heard
pitch interval were judged as melodically congruent. Otherwise, they
were judged as melodically incongruent. A mixed visual-auditory de-
sign was used for the sound-congruence-judgement task because we
thought that if bi-character words were auditorily presented, the ma-
jority of participants may use fine-grained pitch processing to judge
whether the pitch interval of the auditorily presented word was con-
gruent with the auditorily presented tone pair.
There were four experimental conditions and a baseline condition in
the present study. The four experimental conditions were linguistic
auditory-imagery, musical auditory-imagery, linguistic sound-con-
gruence-judgement, and musical sound-congruence-judgement
conditions. The stimuli for the baseline condition were visual pre-
sentations of number-pseudoword, which were not associated with any
pitch interval because the lexical tones of pseudowords could not be
determined. These five conditions involve different cognitive processes,
as shown in Table 1. For the linguistic and musical auditory-imagery
condition, working memory was involved in auditory imagery via
subvocal rehearsal. For the linguistic and musical sound-congruence-
judgement conditions, working memory was involved in auditory
imagery via subvocal rehearsaland comparing the imagined pitch in-
terval with the heard pitch interval. In addition, fine-grained pitch
processing was involved in the musical auditory-imagery condition and
musical sound-congruence-judgement condition.
2.3. Stimuli and procedure
Example stimuli are shown in Fig. 1. For the auditory-imagery task,
a visual stimulus was presented in a trial. For the sound-congruence-
judgement task, a visual stimulus and an auditory stimulus were pre-
sented in a trial. There were three types of visual stimuli: bi-character
words, musical scores with numerical notation, and number-pseudo-
word combinations. The auditory stimuli were tone pairs.
We used numerical notation for the visual presentation of musical
scores because numerical notation is commonplace in Taiwan. The
numbers 1, 2, 3, ……, and 7 correspond to the solfège pitch names do,
re, mi, ……, and si, respectively. A musical score had two numbers that
indicated a melodic interval. The auditory stimuli of tone pairs were
generated from the oboe instrument sound by Reason 7.0 (Reason
Software Company Inc.). All oboe tones were within the pitch range
from G#3 to A4 (fundamental frequency 207.7–440.0 Hz), which is
encompassed by the pitch range of human female adult speech.
For the sound-congruence-judgement task, there were congruent
and incongruent trials. A linguistic congruent trial included a visually
presented Taiwanese word and an auditorily presented tone pair with
the same linguistic pitch interval category. A musical congruent trial
included visually presented musical scores and an auditorily presented
tone pair with the same musical pitch interval category. For the

122
C.-G. Tsai et al. Neuropsychologia 119 (2018) 118–127

Table 3 interval differed by one semitone.

Contrast results for the sound-congruence-judgment task. The MNI coordinates The experimental procedure is schematically illustrated in Fig. 2.
identify the locations of cluster maxima and local maxima (> 4 mm apart There were four runs of fMRI scanning. Each run consisted of 30 trials.
within each cluster). The total 120 trials (24 trials per condition) were presented in a unique
X Y Z Volume Information t-value Cluster random order for each participant. Each trial began with a silence (2 s)
(voxels) and warning tone (2.8 kHz, 0.7 s). Then, numbers or characters were
presented on the screen. For the baseline condition (number-pseudo-
Linguistic sound-congruence-judgment condition > Baseline
56 0 − 10 Superior temporal gyrus, 14.08 67237
word), the musical auditory-imagery condition, and the linguistic au-
supramarginal gyrus ditory-imagery condition, the participants were instructed to attempt to
− 38 18 0 Anterior insula 13.51 imagine the sounds via subvocal rehearsal three times according to the
− 58 − 26 2 Middle temporal gyrus 13.12 visual presentation. The participants answered whether the visual
52 4 − 12 Superior temporal pole 11.99
presentation could be sung or spoken by pressing a button (a left button
10 10 54 Supplementary motor area, 11.35
anterior/middle cingulate gyrus in response to “can” and a right button for “cannot”). For the musical
8 − 20 10 Thalamus 11.35 sound-congruence-judgement condition and the linguistic sound-con-
8 − 74 6 Middle/inferior occipital gyrus, 9.15 gruence-judgement condition, the participants were instructed to cov-
lingual gyrus, precuneus, cuneus
ertly sing/speak the visual presentation with the heard tone pair, which
− 34 − 60 − 30 Cerebellar lobule IV-Crus I 8.97
50 4 48 Precentral gyrus, dorsal premotor 8.85
was consecutively presented for three times within 1.8 s. This tone pair
cortex (right) was either melodically congruent or incongruent (50% probability for
− 48 16 20 Inferior frontal gyrus 8.83 each, randomly presented) with the pitch interval of the visual pre-
50 30 24 Dorsolateral prefrontal cortex 8.71 sentation. Then, the participants answered whether the pitch interval of
−6 26 46 Superior/middle/inferior frontal 8.71
the visual presentation was congruent with the pitch interval of the
gyrus
− 52 2 48 Precentral gyrus, dorsal premotor 7.66 auditory presentation by button pressing (a left button in response to
cortex (left) “no” and a right button for “yes”).
42 − 46 46 Inferior parietal lobule 7.19
− 18 − 28 − 12 Hippocampus, parahippocampal 7.17 2.4. MRI data acquisition
gyrus
24 4 6 Putamen, caudate 6.02
− 10 − 26 − 14 Pons, midbrain 6.02 For imaging data collection, participants were scanned using a 3 T
− 38 − 48 40 Inferior parietal lobule 5.68 344 MR system (MAGNETOM Prisma, Siemens, Erlangen, Germany) and a
− 46 − 40 38 Supramarginal gyrus 3.57 20-channels array head coil at the Imaging Center for Integrated Body,
− 54 − 32 38 Intraparietal sulcus 3.28
Mind, and Culture Research, National Taiwan University. In the func-
− 12 − 68 36 Precuneus 4.36 83
− 48 − 52 52 Inferior parietal lobule 3.93 62 tional scanning, about 2.5 mm slices of axial images were acquired
52 − 38 − 20 Inferior temporal gyrus 3.93 55 using a gradient echo planar imaging (EPI) with the following para-
Musical sound-congruence-judgment condition > Linguistic sound-congruence- meters: time to repetition = 2500 ms, echo time = 30 ms, flip angle
judgment condition = 87°, in-plane field of view = 192 × 192 mm, and acquisition matrix
36 − 54 54 Postcentral gyrus, superior/ 8.57 535 = 78 × 78 × 45 to cover whole cerebral areas. For spatial individual-
inferior parietal lobule (right) to-template normalization in preprocessing, a Magnetization Prepared
30 − 70 30 Middle occipital gyrus 6.44
Rapid Gradient Echo T1-weighted imaging with spatial resolution of
30 − 66 38 Superior occipital gyrus 6.12
30 − 58 42 Angular gyrus 5.15 0.9 mm isotropic was acquired for each participant.
24 − 60 30 Precuneus 4.71
− 32 − 54 52 Superior/inferior parietal lobule 5.29 82 2.5. Data analysis
(left)
12 − 68 56 Dorsal precuneus (right) 6.20 184
− 10 − 62 52 Dorsal precuneus (left) 5.40 32
The correct response rates for all five conditions during the fMRI
24 6 54 Dorsal premotor cortex (right) 6.09 131 scan session were computed. Participants with a correct rate lower than
− 24 12 54 Dorsal premotor cortex (left) 5.21 40 70% were excluded from the fMRI data analysis. We performed a one-
36 − 38 34 Intraparietal sulcus 5.44 58 tailed t-test on participants’ correct rates for the musical versus lin-
40 − 44 36 Supramarginal gyrus 5.39
guistic sound-congruence-judgement conditions because of the a priori
− 34 − 84 36 Middle occipital gyrus 5.37 37
hypothesis that the musical sound-congruence-judgement condition
required more fine-grained pitch discrimination and thus was more
linguistic sound-congruence-judgement task, the tone pairs had the difficult than the linguistic sound-congruence-judgement condition.
melodic intervals of 0, ± 2, and ± 7 semitones, which approximately For preprocessing and statistical analysis of the fMRI data, we used
correspond to the pitch intervals of Taiwanese words of three level SPM12 statistical parametric mapping software (Wellcome Trust Center
tones. Therefore, the linguistic pitch intervals had five categories. A for Neuroimaging; http://www.fil.ion.ucl.ac.uk/spm) and the Artifact
linguistic incongruent trial consisted of a visually presented Taiwanese Detection Tools (ART) toolkit. The first four volumes of each run were
word and an auditorily presented tone pair with the pitch interval ad- discarded to allow for magnetic saturation effects. The remaining
jacent to the linguistic interval category of the Taiwanese word. For the functional images were corrected for head movement artifact and
musical sound-congruence-judgement condition, the tone pairs had the timing differences in slice acquisitions. Preprocessed functional images
melodic intervals of 0, ± 1, ± 2, ± 3, ± 7, and ± 8 semitones. A were then coregistered to the individual anatomical image and nor-
musical incongruent trial consisted of a visually presented number pair malised to the standard MNI brain template and resampled to a 2-mm
and an auditorily presented tone pair, with their pitch intervals dif- isotropic voxel size. Normalised images were spatially smoothed with a
fering by one semitone. The degree of pitch-interval incongruence Gaussian kernel of 4-mm full width at half maximum (FWHM) to ac-
varied across the linguistic incongruent trials and musical incongruent commodate any anatomical variability across participants. ART toolkit
trials. For linguistic incongruent trials, the auditorily presented pitch was used to check the spikes and motion of each dataset. All functional
interval and the visually presented pitch interval differed by approxi- dataset showed head motion below 2.5 mm of maximal translation (in
mately two or five semitones. For musical incongruent trials, the au- any direction) and 1.0° of maximal rotation throughout the course of
ditorily presented pitch interval and the visually presented pitch scanning, and none outliers happened in all functional dataset.
The individual functional data analysis was performed in SPM12.

123
C.-G. Tsai et al.
Fig. 5. Contrast results for the sound-congruence-judgment task. The threshold was set at FDR-corrected p < 0.05 with a minimum cluster size of 30 voxels. (A) The
linguistic sound-congruence-judgment condition was associated with significant greater activation of the insula, striatum, superior/middle temporal gyrus, inferior
frontal gyrus, dorsal premotor cortex, supplementary motor area, caudate, and inferior parietal lobule relative to the baseline condition. (B) Significantly increased
activation in the superior/inferior parietal lobules, bilateral dorsal premotor cortex (rostral subregion), and right precuneus was observed for the contrast of the
musical minus linguistic sound-congruence-judgment conditions.
Statistical inference was based on a random effect approach at two
levels. At the single-subject level, the data of each participant were
analyzed using the general linear model by fitting the time series data
with the canonical hemodynamic response function (HRF) modeled at
events (Josephs and Henson, 1999). The events were modeled for the
response period in each of the five conditions. In the group-level ana-
lysis, paired t-tests were used to calculate the contrasts of the linguistic
auditory-imagery and linguistic sound-congruence-judgement condi-
tions minus the baseline condition, as well as the contrasts of the mu-
sical minus linguistic auditory-imagery conditions and the musical
minus linguistic sound-congruence-judgement conditions. The statis-
tical significant threshold was set at p < 0.05 (FDR-corrected) and a
minimum cluster size of 30 voxels in a whole brain analysis.
3. Results
Fig. 3 shows the results of the correct response rates for the two
sound-congruence-judgement conditions. Participants had significantly
lower correct rates in the musical sound-congruence-judgement con-
dition compared to the linguistic sound-congruence-judgement condi-
tion (p = 0.0317, one-tailed t-test). This result suggests that the musical
sound-congruence-judgement task was more difficult than the linguistic
sound-congruence-judgement task.
Table 2 and Fig. 4 show the significantly activated brain regions for
the contrast of the linguistic auditory-imagery condition minus baseline
and the contrast of the musical minus linguistic auditory-imagery
conditions. When comparing the linguistic auditory-imagery condition
to baseline condition, greater activation was observed within the su-
perior temporal gyrus, inferior/superior parietal lobules, precuneus,
inferior frontal gyrus (IFG), anterior insula, dPMC, dlPFC, supplemen-
tary motor cortex (SMA), lingual gyrus, cuneus, parahippocampal
gyrus, striatum, thalamus, and cerebellum. Moreover, the musical au-
ditory-imagery condition was associated with greater activation in the
right dPMC, right superior/inferior parietal lobules, and precuneus
compared to the linguistic auditory-imagery condition.
Table 3 and Fig. 5 show the significantly activated brain regions for
124
Neuropsychologia 119 (2018) 118–127
the contrast of the linguistic sound-congruence-judgement condition
minus baseline and the contrast of the musical minus linguistic sound-
congruence-judgement conditions. When comparing the linguistic
sound-congruence-judgement condition to baseline condition, greater
activation was observed within the right superior/middle temporal
gyri, superior/inferior parietal lobules, dlPFC, IFG, SMA, dPMC, ante-
rior insula, anterior cingulate cortex, precuneus, striatum, cerebellum,
and thalamus/midbrain. Moreover, the musical sound-congruence-
judgement condition was associated with greater activation in the
dPMC, superior/inferior parietal lobules, and dorsal precuneus com-
pared to the linguistic sound-congruence-judgement condition.
4. Discussion
Perception of lexical tones in a tonal language and perception of
musical melodies rely on parsing the relationship between pitches,
namely, RP processing. In the current study, we used fMRI and working
memory paradigms to compare the neural substrates underlying the
explicit RP processing of musical and linguistic pitch intervals. The
linguistic stimuli were Taiwanese bi-character words with level tones.
Their steady pitches allow for a direct comparison to the scale tones in
Western music. Pitch intervals of Taiwanese words are less fine-grained
compared to musical pitch intervals. Analysis of the behavioral data
confirmed that the musical sound-congruence-judgement task was more
difficult than the linguistic sound-congruence-judgement task. As ex-
pected, we found that several brain regions implicated in higher order
cognitive operations such as working memory and attention control
showed enhanced activity for the musical versus linguistic conditions.
All the four experimental conditions were associated with sig-
nificantly greater activity within the dPMC and PPC relative to the
baseline. This finding is concordant with the results from prior fMRI
studies showing the involvement of the dPMC-PPC network in both
linguistic and musical RP processing (Klein et al., 2001; Gandour et al.,
2004; Bermudez et al., 2009; Schulze et al., 2009; Foster and Zatorre,
2010; Li et al., 2010; Angenstein et al., 2012; Brown et al., 2013). The
dPMC-PPC network has been termed the dorsal attention network
C.-G. Tsai et al.
because it mediates the top-down guided voluntary allocation of at-
tention to stimuli features, and links stimuli to appropriate responses
(Corbetta et al., 2008). Recently, Genon et al. (2016) employed con-
nectivity-based parcellation to divide the right dPMC into the central,
ventral, dorsal, rostral, and caudal subregions. The central subregion of
the right dPMC tends to work in tandem with the PPC to support
working memory maintenance (Genon et al., 2016). In the present
study, the right PPC and the right central dPMC showed increased ac-
tivation for the musical versus linguistic auditory-imagery conditions.
This result is consistent with Schon et al.’s (2010) finding of increased
activity in the right central dPMC for a melody discrimination task than
a speech discrimination task. We suggest that the right dPMC-PPC
network may be responsible for the voluntary allocation of attention to
pitch features of the stimuli and the attention-based maintenance of
working memory (Berryhill et al., 2011) for the musical auditory-ima-
gery, musical sound-congruence-judgement, and linguistic sound-con-
gruence-judgement conditions. The activation clusters in the dPMC and
PPC for the linguistic auditory-imagery condition were more limited.
This may be due to the relatively lower attention to pitch features of the
imagined Taiwanese words.
Experimental and theoretical studies have suggested that the right
PPC contains the neural substrate of a general magnitude system for
space, time, and numbers (Walsh, 2003; Cohen Kadosh et al., 2007;
Bueti and Walsh, 2009). This general magnitude system may support
accurate representation and processing of pitch relations for the task of
judging the melodic congruence between an imagined musical pitch
interval and a heard interval. Notably, there are twelve pitch categories
within an octave, whereas there are only three level lexical tones in
Taiwanese. The significantly enhanced activity in the right PPC for the
musical versus linguistic auditory-imagery conditions may reflect the
increased cognitive load for the more fine-grained pitch processing.
Moreover, it was proposed that the right dPMC-PPC network may
contribute to systematically transforming and comparing pitch in-
formation in a manner similar to visuospatial information (Zarate and
Zatorre, 2008; Foster et al., 2013). Converging with this view, our
findings indicated that the visuospatial working memory function of the
right dPMC-PPC network plays a more important role in discrimination
of musical pitch intervals compared to linguistic pitch intervals.
One crucial question arising from the present study is the functional
specialization of the dPMC during RP processing. We observed that the
musical sound-congruence-judgement condition was associated with
significantly greater activity within the rostral (y > 3) dPMC than the
linguistic sound-congruence-judgement condition. Compared with the
right central dPMC, the right rostral dPMC typically supports higher
cognitive functions, including error-monitoring (Bengtsson et al.,
2009), evaluating hand motor capability (Hirose et al., 2010), response
selection/inhibition (Rowe et al., 2008; Batterink et al., 2010; Padmala
and Pessoa, 2010; Cai et al., 2014), attention control (Vohn et al., 2007;
Natale et al., 2009; Wang et al., 2010; Braga et al., 2013), mental cal-
culation (Fehr et al., 2008), and working memory (Veltman et al., 2003;
Manoach et al., 2004; Mitchell, 2007; Takahama et al., 2010; Harding
et al., 2016). It is also recognized that the left rostral dPMC may
mediate error-monitoring (Chen et al., 2006; Provost et al., 2010) and
top-down attention to memory retrieval (Ciaramelli et al., 2010). We
suggest that the bilateral rostral dPMC may support attention control,
working memory, evaluation of motor activities, and monitoring me-
chanisms during judging the congruence between the imagined and
heard musical pitch intervals.
In the present study, the contrast of musical minus linguistic sound-
congruence-judgement conditions yielded co-activation of the bilateral
rostral dPMC and dorsal precuneus. This result accords well with Genon
et al.’s (2016) finding that the right rostral dPMC connects with the
precuneus. In visual modality, the dorsal precuneus has been implicated
in higher order cognitive operations such as spatial working memory
and attention control (Lepsien et al., 2005; Natale et al., 2006). In au-
ditory modality, the dorsal precuneus plays a role in detecting/
125
Neuropsychologia 119 (2018) 118–127
processing salient sounds (Watkins et al., 2007), effortful auditory
imagery (Huijbers et al., 2011), and shifts of auditory attention
(Shomstein and Yantis, 2006). In an fMRI study of music score reading
and playing, Stewart et al. (2003) showed increased dorsal precuneus
activation when participants used the right hand to play melodies from
musical notation after training, suggesting the role of this area in
translating musical notation into sensorimotor codes. Moreover,
Meister et al. (2005) found that pianists’ dorsal precuneus exhibited
increased activation during silently playing musical scores with the
right hand, relative to passively reading the score of repetition of a
single note. In the present study, the musical conditions entailed the
transformation from musical notations into laryngeal sensorimotor
codes for subvocal rehearsal. During the auditory-imagery and sound-
congruence-judgement tasks, increased activation in the right dorsal
precuneus for the musical versus linguistic conditions may reflect its
role in attention control and the spatial/sensorimotor processing of the
more fine-grained pitch structures of music compared to lexical tones.
While the present fMRI study recruited individuals with good RP,
our findings also echo several studies on congenital amusia, a disorder
affecting the processing of musical pitch. Past studies have demon-
strated that the impaired fine-grained processing of pitch in congenital
amusics seems to be characterized by reduced activity in the right IFG
in response to musical melodies (Hyde et al., 2011), reduced white
matter concentration in the right IFG (Hyde et al., 2006), as well as the
reduced connectivity between the right IFG and auditory cortex (Loui
et al., 2009; Peretz, 2016). In the present study, both the contrast of the
musical minus linguistic auditory-imagery conditions and the contrast
of the musical minus linguistic sound-congruence-judgement conditions
showed subthreshold (uncorrected p < 0.003 with cluster size > 50
voxels for overall AlphaSim corrected p < 0.05) activation in the right
IFG (extending into the ventral premotor cortex). Based on this and
previous studies, we suggest that the right vPMC/IFG may contribute to
sensorimotor and auditory encoding of musical pitches.
Besides deficits in sensorimotor and auditory encoding, there is
emerging evidence linking amusia with higher order cognitive control
deficits. Zhang et al. (2017) recruited Cantonese-speaking amusics and
musically intact controls to investigate the processing of pitch intervals,
finding abnormally strong activation in amusics’ precuneus during the
repetition of pitch stimuli. This result resonates with our finding of
increased activation in the precuneus for the music minus linguistic
conditions. In addition to the precuneus, the dorsal prefrontal cortex
might also plays a role in cognitive control functions for the fine-
grained RP processing of music. Albouy et al. (2013) reported that the
impairment in maintaining musical melodic information in amusics
might be linked to decreased gamma oscillations within the right dlPFC
during the retention of musical melodic information. A follow-up study
by Schaal et al. (2015) showed that modulating the right dlPFC with
35 Hz transcranial alternating current stimulation in amusics selectively
improved pitch memory performance. The critical role of this region in
the processing of musical pitch intervals is supported by our finding
that the rostral dPMC, a transitional zone between the dPMC and dlPFC,
showed greater activity for the musical versus linguistic sound-con-
gruence-judgement conditions.
Several limitations of the study need to be pointed out. First, the
final sample of fMRI data consisted of 11 females and 3 males. Future
research should recruit a larger sample size with a good gender balance.
Second, the major screening criterion to be able to participate in the
fMRI experiment was the score of a RP pre-test, which was a shortened
version of the congruence-sound-judgement task in the fMRI experi-
ment. While this pre-test selected participants with good RP, our study
did not take into account the effect of music training. Moreover, future
research could explore whether RP-related musical training is beneficial
for sensorimotor/spatial processing, working memory, attention con-
trol, or error-monitoring.
To summarize, the present study has examined the neural substrates
underlying the explicit RP processing of musical and linguistic pitch
C.-G. Tsai et al.
intervals. Results confirm and extend previous findings of the involve-
ment of the dPMC-PPC network in RP-related tasks (Klein et al., 2001;
Gandour et al., 2004; Bermudez et al., 2009; Schulze et al., 2009; Foster
and Zatorre, 2010; Li et al., 2010; Schon et al., 2010), linking RP with
the voluntary allocation of attention to pitch features of the stimuli. The
contrast of musical minus linguistic conditions has identified activation
in the dorsal precuneus and dPMC. The dorsal precuneus may support
attention control and the spatial/sensorimotor processing of the more
fine-grained pitch structures of music. Enhanced activity in the rostral
dPMC for the musical versus linguistic sound-congruence-judgement
conditions may reflect its role in attention control, working memory,
evaluation of motor activities, and monitoring mechanisms during
judging the congruence between the imagined and heard musical pitch
intervals. This study highlights the importance of higher order cognitive
operations in the more fine-grained pitch processing of musical me-
lodies compared to lexical tones.
Acknowledgements
This study was supported by grant projects (MOST 104–2410-H-
002–114) from Ministry of Science and Technology, Taiwan.
References
Albouy, P., Mattout, J., Bouet, R., Maby, E., Sanchez, G., Aguera, P.E., Daligault, S.,
Delpuech, C., Bertrand, O., Caclin, A., Tillmann, B., 2013. Impaired pitch perception
and memory in congenital amusia: the deficit starts in the auditory cortex. Brain: J.
Neurol. 136, 1639–1661.
Angenstein, N., Scheich, H., Brechmann, A., 2012. Interaction between bottom-up and
top-down effects during the processing of pitch intervals in sequences of spoken and
sung syllables. NeuroImage 61, 715–722.
Batterink, L., Yokum, S., Stice, E., 2010. Body mass correlates inversely with inhibitory
control in response to food among adolescent girls: an fMRI study. NeuroImage 52,
1696–1703.
Bengtsson, S.L., Lau, H.C., Passingham, R.E., 2009. Motivation to do well enhances re-
sponses to errors and self-monitoring. Cereb. Cortex 19, 797–804.
Bermudez, P., Lerch, J.P., Evans, A.C., Zatorre, R.J., 2009. Neuroanatomical correlates of
musicianship as revealed by cortical thickness and voxel-based morphometry. Cereb.
Cortex 19, 1583–1596.
Berryhill, M.E., Chein, J., Olson, I.R., 2011. At the intersection of attention and memory:
the mechanistic role of the posterior parietal lobe in working memory.
Neuropsychologia 49, 1306–1315.
Bidelman, G.M., Hutka, S., Moreno, S., 2013. Tone language speakers and musicians share
enhanced perceptual and cognitive abilities for musical pitch: evidence for bidir-
ectionality between the domains of language and music. PLoS One 8, e60676.
Bradley, E.D., 2016. Phonetic dimensions of tone language effects on musical melody
perception. Psychomusicol.: Music Mind Brain 26 (4), 337–345.
Braga, R.M., Wilson, L.R., Sharp, D.J., Wise, R.J., Leech, R., 2013. Separable networks for
top-down attention to auditory non-spatial and visuospatial modalities. NeuroImage
74, 77–86.
Brown, R.M., Chen, J.L., Hollinger, A., Penhune, V.B., Palmer, C., Zatorre, R.J., 2013.
Repetition suppression in auditory-motor regions to pitch and temporal structure in
music. J. Cogn. Neurosci. 25, 313–328.
Bueti, D., Walsh, V., 2009. The parietal cortex and the representation of time, space,
number and other magnitudes. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci. 364,
1831–1840.
Cai, W., Cannistraci, C.J., Gore, J.C., Leung, H.C., 2014. Sensorimotor-independent pre-
frontal activity during response inhibition. Human. Brain Mapp. 35, 2119–2136.
Chen, J.L., Zatorre, R.J., Penhune, V.B., 2006. Interactions between auditory and dorsal
premotor cortex during synchronization to musical rhythms. NeuroImage 32,
1771–1781.
Ciaramelli, E., Grady, C., Levine, B., Ween, J., Moscovitch, M., 2010. Top-down and
bottom-up attention to memory are dissociated in posterior parietal cortex: neuroi-
magingand and neuropsychological evidence. J. Neurosci.: Off. J. Soc. Neurosci. 30,
4943–4956.
Cohen Kadosh, R., Cohen Kadosh, K., Linden, D.E., Gevers, W., Berger, A., Henik, A.,
2007. The brain locus of interaction between number and size: a combined functional
magnetic resonance imaging and event-related potential study. J. Cogn. Neurosci. 19,
957–970.
Corbetta, M., Patel, G., Shulman, G.L., 2008. The reorienting system of the human brain:
from environment to theory of mind. Neuron 58, 306–324.
Deutsch, D., Henthorn, T., Marvin, E., Xu, H.S., 2006. Absolute pitch among American
and Chinese conservatory students: prevalence differences, and evidence for a
speech-related critical period (L). J. Acoust. Soc. Am. 119, 719–722.
Fehr, T., Code, C., Herrmann, M., 2008. Auditory task presentation reveals predominantly
right hemispheric fMRI activation patterns during mental calculation. Neurosci. Lett.
431, 39–44.
Foster, N.E., Halpern, A.R., Zatorre, R.J., 2013. Common parietal activation in musical
126
Neuropsychologia 119 (2018) 118–127
mental transformations across pitch and time. NeuroImage 75, 27–35.
Foster, N.E., Zatorre, R.J., 2010. A role for the intraparietal sulcus in transforming mu-
sical pitch information. Cereb. Cortex 20, 1350–1359.
Gandour, J., Tong, Y., Wong, D., Talavage, T., Dzemidzic, M., Xu, Y., Li, X., Lowe, M.,
2004. Hemispheric roles in the perception of speech prosody. NeuroImage 23,
344–357.
Genon, S., Li, H., Fan, L., Muller, V.I., Cieslik, E.C., Hoffstaedter, F., Reid, A.T., Langner,
R., Grefkes, C., Fox, P.T., Moebus, S., Caspers, S., Amunts, K., Jiang, T., Eickhoff, S.B.,
2016. The Right Dorsal Premotor Mosaic: organization, functions, and connectivity.
Cereb. Cortex.
Harding, I.H., Harrison, B.J., Breakspear, M., Pantelis, C., Yucel, M., 2016. Cortical re-
presentations of cognitive control and working memory are dependent yet non-in-
teracting. Cereb. Cortex 26, 557–565.
Hirose, S., Hagura, N., Matsumura, M., Naito, E., 2010. Human rostral dorsal premotor
cortex mediates graspability judgment of external objects by evaluating hand motor
capability. Brain Res. 1313, 134–142.
Huijbers, W., Pennartz, C.M., Rubin, D.C., Daselaar, S.M., 2011. Imagery and retrieval of
auditory and visual information: neural correlates of successful and unsuccessful
performance. Neuropsychologia 49, 1730–1740.
Hyde, K.L., Zatorre, R.J., Griffiths, T.D., Lerch, J.P., Peretz, I., 2006. Morphometry of the
amusic brain: a two-site study. Brain: J. Neurol. 129, 2562–2570.
Hyde, K.L., Zatorre, R.J., Peretz, I., 2011. Functional MRI evidence of an abnormal neural
network for pitch processing in congenital amusia. Cereb. Cortex 21, 292–299.
Josephs, O., Henson, R.N., 1999. Event-related functional magnetic resonance imaging:
modelling, inference and optimization. Philos. Trans. R. Soc. Lond. Ser. B Biol. Sci.
354, 1215–1228.
Klein, D., Zatorre, R.J., Milner, B., Zhao, V., 2001. A cross-linguistic PET study of tone
perception in Mandarin Chinese and English speakers. NeuroImage 13, 646–653.
Lee, C.Y., Lee, Y.F., 2010. Perception of musical pitch and lexical tones by Mandarin-
speaking musicians. J. Acoust. Soc. Am. 127, 481–490.
Lee, C.Y., Lee, Y.F., Shr, C.L., 2011. Perception of musical and lexical tones by Taiwanese-
speaking musicians. J. Acoust. Soc. Am. 130, 526–535.
Lepsien, J., Griffin, I.C., Devlin, J.T., Nobre, A.C., 2005. Directing spatial attention in
mental representations: interactions between attentional orienting and working-
memory load. NeuroImage 26, 733–743.
Li, X., Gandour, J.T., Talavage, T., Wong, D., Hoffa, A., Lowe, M., Dzemidzic, M., 2010.
Hemispheric asymmetries in phonological processing of tones versus segmental units.
NeuroReport 21, 690–694.
Loui, P., Alsop, D., Schlaug, G., 2009. Tone deafness: a new disconnection syndrome? J.
Neurosci.: Off. J. Soc. Neurosci. 29, 10215–10220.
Manoach, D.S., White, N.S., Lindgren, K.A., Heckers, S., Coleman, M.J., Dubal, S.,
Holzman, P.S., 2004. Hemispheric specialization of the lateral prefrontal cortex for
strategic processing during spatial and shape working memory. NeuroImage 21,
894–903.
Meister, I., Krings, T., Foltys, H., Boroojerdi, B., Muller, M., Topper, R., Thron, A., 2005.
Effects of long-term practice and task complexity in musicians and nonmusicians
performing simple and complex motor tasks: implications for cortical motor orga-
nization. Human Brain Mapp. 25, 345–352.
Mitchell, R.L., 2007. fMRI delineation of working memory for emotional prosody in the
brain: commonalities with the lexico-semantic emotion network. NeuroImage 36,
1015–1025.
Nan, Y., Friederici, A.D., 2013. Differential roles of right temporal cortex and Broca's area
in pitch processing: evidence from music and Mandarin. Human Brain Mapp. 34,
2045–2054.
Nan, Y., Sun, Y., Peretz, I., 2010. Congenital amusia in speakers of a tone language:
association with lexical tone agnosia. Brain: J. Neurol. 133, 2635–2642.
Natale, E., Marzi, C.A., Girelli, M., Pavone, E.F., Pollmann, S., 2006. ERP and fMRI cor-
relates of endogenous and exogenous focusing of visual-spatial attention. Eur. J.
Neurosci. 23, 2511–2521.
Natale, E., Marzi, C.A., Macaluso, E., 2009. FMRI correlates of visuo-spatial reorienting
investigated with an attention shifting double-cue paradigm. Human Brain Mapp. 30,
2367–2381.
Padmala, S., Pessoa, L., 2010. Interactions between cognition and motivation during re-
sponse inhibition. Neuropsychologia 48, 558–565.
Peretz, I., 2016. Neurobiology of congenital amusia. Trends Cogn. Sci. 20, 857–867.
Provost, J.S., Petrides, M., Monchi, O., 2010. Dissociating the role of the caudate nucleus
and dorsolateral prefrontal cortex in the monitoring of events within human working
memory. Eur. J. Neurosci. 32, 873–880.
Rowe, J., Hughes, L., Eckstein, D., Owen, A.M., 2008. Rule-selection and action-selection
have a shared neuroanatomical basis in the human prefrontal and parietal cortex.
Cereb. Cortex 18, 2275–2285.
Schaal, N.K., Pfeifer, J., Krause, V., Pollok, B., 2015. From amusic to musical?–Improving
pitch memory in congenital amusia with transcranial alternating current stimulation.
Behav. Brain Res. 294, 141–148.
Schon, D., Gordon, R., Campagne, A., Magne, C., Astesano, C., Anton, J.L., Besson, M.,
2010. Similar cerebral networks in language, music and song perception. NeuroImage
51, 450–461.
Schulze, K., Gaab, N., Schlaug, G., 2009. Perceiving pitch absolutely: comparing absolute
and relative pitch possessors in a pitch memory task. BMC Neurosci. 10, 106.
Shomstein, S., Yantis, S., 2006. Parietal cortex mediates voluntary control of spatial and
nonspatial auditory attention. J. Neurosci.: Off. J. Soc. Neurosci. 26, 435–439.
Stalinski, S.M., Schellenberg, E.G., Trehub, S.E., 2008. Developmental changes in the
perception of pitch contour: distinguishing up from down. J. Acoust. Soc. Am. 124,
1759–1763.
Stewart, L., Henson, R., Kampe, K., Walsh, V., Turner, R., Frith, U., 2003. Brain changes
after learning to read and play music. NeuroImage 20, 71–83.
C.-G. Tsai et al.
Takahama, S., Miyauchi, S., Saiki, J., 2010. Neural basis for dynamic updating of object
representation in visual working memory. NeuroImage 49, 3394–3403.
Veltman, D.J., Rombouts, S.A., Dolan, R.J., 2003. Maintenance versus manipulation in
verbal working memory revisited: an fMRI study. NeuroImage 18, 247–256.
Vohn, R., Fimm, B., Weber, J., Schnitker, R., Thron, A., Spijkers, W., Willmes, K., Sturm,
W., 2007. Management of attentional resources in within-modal and cross-modal
divided attention tasks: an fMRI study. Human Brain Mapp. 28, 1267–1275.
Walsh, V., 2003. A theory of magnitude: common cortical metrics of time, space and
quantity. Trends Cogn. Sci. 7, 483–488.
Wang, L., Liu, X., Guise, K.G., Knight, R.T., Ghajar, J., Fan, J., 2010. Effective connectivity
of the fronto-parietal network during attentional control. J. Cogn. Neurosci. 22,
543–553.
Watkins, S., Dalton, P., Lavie, N., Rees, G., 2007. Brain mechanisms mediating auditory
attentional capture in humans. Cereb. Cortex 17, 1694–1700.
Wong, P.C., Ciocca, V., Chan, A.H., Ha, L.Y., Tan, L.H., Peretz, I., 2012. Effects of culture
127
Neuropsychologia 119 (2018) 118–127
on musical pitch perception. PLoS One 7, e33424.
Wu, S., 2000. Mandarin-Taiwanese Comparative Living Dictionary (in Mandarin and Min
Nan). Yuan-Liou, Taipei.
Xie, X., Myers, E., 2015. The impact of musical training and tone language experience on
talker identification. J. Acoust. Soc. Am. 137, 419–432.
Zarate, J.M., Zatorre, R.J., 2008. Experience-dependent neural substrates involved in
vocal pitch regulation during singing. NeuroImage 40, 1871–1887.
Zhang, C., Chen, S., 2016. Toward an integrative model of talker normalization. J. Exp.
Psychol. Human. Percept. Perform. 42, 1252–1268.
Zhang, C., Peng, G., Shao, J., Wang, W.S., 2017. Neural bases of congenital amusia in
tonal language speakers. Neuropsychologia 97, 18–28.
Zhang, C., Peng, G., Wang, W.S., 2012. Unequal effects of speech and nonspeech contexts
on the perceptual normalization of Cantonese level tones. J. Acoust. Soc. Am. 132,
1088–1099.