Brain-Music Duet: MEG signal complexity and auditory
perception in musicians and nonmusicians
by
Sarah M. Carpentier
A thesis submitted in conformity with the requirements
for the degree of Masters of Arts
Department of Psychology
University of Toronto
© Copyright by Sarah M. Carpentier 2011
ii
ii
Brain-Music Duet: MEG signal complexity and auditory perception
in musicians and nonmusicians
Sarah M. Carpentier
Degree of Masters of Arts
Department of Psychology
University of Toronto
2011
Abstract
Music training has been suggested to lead to an enhancement in the neural activity associated
with music processing. It has been proposed that brain signal complexity is a reflection of the
functional capacity of that neural system. The present study tested the hypothesis that musicians
have a larger repertoire of brain activity associated with musical perception then nonmusicians.
We used multiscale entropy to capture the complexity of the MEG signal while musicians and
nonmusicians listened to different melodies. We observed that initial melody presentation
elicited higher complexity in musicians compared to nonmusicians. Brain signal complexity
decreased in both groups as a function of stimulus repetition. We propose that the neural
networks that underlie auditory processing have a more diverse range of functioning in
musicians, as compared to nonmusicians. Repetition reduces the amount of information
processing and corresponding brain signal complexity.
iii
iii
Acknowledgements
First and foremost, I have to express my deep gratitude to my supervisor Randy. His
enrichment of my academic career has exceeded well beyond my expectations, and his scientific
innovation and strength of character are inspirational. By setting such high standards for himself
he encourages all of his students to strive past their upmost limits to the best of their abilities.
Thank you, Randy for your guidance, encouragement, and faith in my potential. I consider
myself to be extremely privileged not only to be one of your students, but to know you.
Secondly, I would like to thank the other members of the McIntosh Lab who have
provided various types of feedback and support: Marc Berman, Kelly Shen, Anjali Beharelle,
Zainab Fatima, Erin Gibson, Michele Korostil, Grigori Yourganov, Maria Karacchalios and
Tanya Brown. Special appreciation goes to the following people: Jennifer Heisz – for being a
stupendous and patient sounding board, for providing intellectual and personal support, and for
giving the world a strong female role model; Vasily Vakorin – for helping me make my silly
little ideas a computational reality and for always surpassing the definition of ‘complex’; Natasa
Kovacevic – for making labs across the world wish human cloning was a reality; Bratislav Misic
– for always supplying the best English translation for scientific jargon; Gleb Bezgin – for his
effervescent personality that lights up the office, and for being the person who could probably
answer my ideas better than me; and Hongye Wang – for helping me with every red error
message, and for doing the jobs of three research assistants.
Also, I need to thank Takako Fujioka and Bernhard Ross for being open to new ideas and
generously sharing their data for this project. I admire Takako’s dedication to the study of music,
both in and outside the lab. She has made significant contributions to science, has ameliorated
patients’ lives, and impressively still manages to find spare time to enriche peoples’ lives with
iv
iv
music. I thank Bernhard for his patience, for sharing his MEG expertise, and for invaluable help
with data pre-processing for this project.
Importantly, I also acknowledge the efforts put forth by subsidiary advisor, Glenn
Schellenberg, toward reading my drafts and providing helpful feedback. I thank Glenn for
tolerating the relatively micro examinations of music provided by neuroscience, and for the
reminders that music is not an experimental entity that exists in isolation from the rest of the
world. He reminds me that only through collaboration from multiple different approaches can
human behaviour and cognition be fully explained. I hope that someday we can put our different
pieces together and take a look at our completed puzzle. Also, to the other member of my
committee, Jennifer Ryan, thank you for taking your time to read this work, listen to me ramble
on about it, and provide helpful feedback.
Finally, I need to thank all my family and friends who helped me through this process.
My mother is the best teacher I have had the pleasure of knowing, and I thank her for the
countless hours she has spent editing every academic paper of my career. After all the help she
has given my sister and I over the years, in my family we say that my mother has Bachelor’s
degrees from Barnard College, McMaster University, and Queen’s University, a law degree from
the University of Ottawa, and Master’s degrees in Education from OISE, International Relations
from Carleton University, and now Psychology from the University of Toronto. Also, I thank my
father, for never stopping trying to be a better version of himself professionally and personally,
even though he is perfect now, and for trying now to fit as much neuroscience as possible into
his life. Lastly, Jonathan, my partner in adventure, who sets the bar high, maybe, sometimes,
possibly beats me at trivial pursuit, and never fails to supply me with the two most valuable
commodities, joy and relaxation.
v
v
Table of Contents
Abstract ...................................................................................................................................... ii
Acknowledgements ................................................................................................................... iii
Table of Contents ....................................................................................................................... v
List of Figures ........................................................................................................................... vi
Introduction ................................................................................................................................ 1
Materials and Methods ..............................................................................................................11
Participants ....................................................................................................................11
Stimuli ...........................................................................................................................12
Contour ..............................................................................................................12
Interval ...............................................................................................................12
Procedure.......................................................................................................................13
Audio Presentation .............................................................................................13
MEG Overview ..................................................................................................13
MEG Acquisition ...............................................................................................13
Data Analysis ................................................................................................................14
Multiscale Entropy .............................................................................................14
Partial Least Squares ..........................................................................................16
Results ......................................................................................................................................18
Discussion .................................................................................................................................19
Musicians versus nonmusicians .....................................................................................19
Unconscious, automatic processing ................................................................................20
Melodic Contour versus Intervals...................................................................................22
Entropy and ERP ...........................................................................................................24
Conclusion ................................................................................................................................25
Figures ......................................................................................................................................26
References .................................................................................................................................31
vi
vi
List of Figures
Figure 1. Musical stimuli ...........................................................................................................26
Figure 2. Results for standard melodies in the contour condition ...............................................27
Figure 3. Results for standard melodies in the interval condition ...............................................28
Figure 4. Results for deviant melodies in the contour condition .................................................29
Figure 5. Results for deviant melodies in the interval condition .................................................30
1
Brain-Music Duet: MEG signal complexity and auditory perception
in musicians and nonmusicians
Musical abilities are arguably some of the most complex of human accomplishments. A
successful concert pianist can bimanually coordinate the production of over 1,500 notes per
minute, and listening to a symphony can involve perception of over 30 different instruments.
Because music is an auditory stimulus that unfolds over time, music perception relies on the
listener’s ability to appreciate the contribution of different instruments separately and together as
they develop. Investigations of musical capabilities typically have divided a whole piece into
parts, and revolve around identifying the neural substrate of each particular component, such as
pitch (Patterson, Uppenkamp, Johnsrude, & Griffiths, 2002; Zatorre, Evans, & Meyer, 1994;
Zatorre & Samson, 1991), timbre (Menon et al., 2002; Samson, 2003), or rhythm (Chen,
Penhune, & Zatorre, 2008; Fujioka, Trainor, Large, & Ross, 2009; Snyder & Large, 2005). When
put back together, the neuroimaging evidence seems to suggest that music listening activates
distributed brain regions. How these regions interact to produce a unified perception remains to
be elucidated.
Each individual has his or her own unique musical experiences, and these experiences
influence perceptions, listening tastes, and abilities to play an instrument. These ‘musical
memories’ guide the processing of new, incoming music (Krumhansl & Castellano, 1983). For
example, a large experience-dependent effect is seen when people are better able to process and
remember new music from their own culture (Hannon & Trehub, 2005; Morrison, Demorest,
Aylward, Cramer, & Maravilla, 2003). The neural basis for these ‘memories’ has been the
subject of ample investigation. Musicians provide an opportunity to investigate how the brain
can be altered with experience to support behaviour and cognition. Previous work has
2
demonstrated that music training is associated with widespread structural (Gaser & Schlaug,
2003a, 2003b; Hutchinson, Lee, Gaab, & Schlaug, 2003; Hyde et al., 2009b; Schlaug, Forgeard,
Zhu, Norton, & Winner, 2009; Schmithorst & Wilke, 2002) and functional (Bhattacharya &
Petsche, 2005; Jentschke & Koelsch, 2009; Koelsch, Schroger, & Tervaniemi, 1999; Trainor,
Shahin, & Roberts, 2009) changes. However, similar to work with nonmusicians, these studies
typically single out a single component of music, and localize the difference in mean activation
between musicians and nonmusicians. While informative, this procedure fails to capture the full
neural dynamics that are responsible for processing intricate pieces of music over time. We
suggest that full processing requires neural activity that fluctuates with the changes in the
developing musical piece. It has been previously suggested that signal complexity serves as a
measure of the functional repertoire of the system: higher complexity indicates a wider range of
configurations (Ghosh, Rho, McIntosh, Kotter, & Jirsa, 2008; McIntosh, Kovacevic, & Itier,
2008). Therefore, higher complexity in the human brain might promote cognition (McIntosh et
al., 2010; Protzner, Valiante, Kovacevic, McCormick, & McAndrews, 2010; Tononi, Edelman,
& Sporns, 1998). The present study proposes that processing the intricate acoustic events in
music requires complex neural network activity that operates at multiple temporal and spatial
scales. The more information that is available in a musical piece, the greater the variations in
brain network activity necessary to capture that information. Accordingly, reflecting their
advanced musical capabilities, we hypothesize that musicians should exhibit higher brain signal
complexity during music perception.
Nonlinear, dynamical systems provide a useful framework for understanding the
information processing capacity of brain networks (Bullmore & Sporns, 2009; Deco, Jirsa, &
McIntosh, 2011). Unlike linear structures, a nonlinear network system is capable of parallel
processing of information at multiple spatial and temporal scales (Honey, Kotter, Breakspear, &
3
Sporns, 2007; Kelso, 1995). These systems have highly complex activity and process diverse
information simultaneously within segregated local regions, and merge it interactively between
spatially distributed neural populations (Tononi, Sporns, & Edelman, 1994). Anatomical and
functional reports support the brain as a complex, scale-invariant system (Bullmore & Sporns,
2009) that gives rise to consciousness (Tononi, Sporns & Edelman, 1998; Tononi & Edelman,
1998) and cognition (McIntosh et al., 2010). Complexity is supported through small-world type
organizational structure, including dense local connections and sparse long-range connections
(Watts & Strogatz, 1998). A system that has sparse-reciprocal connections is able to integrate
more information than a system that is completely interconnected or one in which regions are
arranged hierarchically (Tononi, Sporns & Edelman, 1994; McIntosh, 2000). Small-world
organization is also believed to be an economical option for biological system organization
because longer axonal projections require more materials and energy (Cherniak, 1994).
Anatomical examinations of invertebrates (Watts & Strogatz, 1998), non-human primates
(Passingham, Stephan, & Kotter, 2002), and humans (Bullmore & Sporns, 2009; Hagmann et al.,
2008; Honey et al., 2009) agree that nervous systems of all levels have the structural
organization of small-world networks. Segregated functional specialization is fundamental to
brain organization (Bendor & Wang, 2005; Dobelle, Mladejovsky, & Girvin, 1974; Penfield &
Boldrey, 1937), and lesion studies have improved our understanding of neural functioning
(Goodale, Milner, Jakobson, & Carey, 1991; Scoville & Milner, 1957). However, while
localization can inform about the critical elements for a particular function, the full realization of
a unified consciousness requires coordination of interactions between different parts of the
network (McIntosh, 2000; Tononi & Edelman, 1998). Early connectivity estimates in the non-
human primate visual cortex suggested that only 30-40% of all possible connections are evident
among the different cortical areas (Felleman & Van Essen, 1991). More recently, thousands of
4
tracer studies of the macaque monkey cortex have been compiled in the CoCoMac database
(http://www.cocomac.org; (Kotter, 2004). Meta-analysis of these studies support the proposal of
a balance between segregation and integration, indicative of high complexity and information
processing capacity (Passingham et al., 2002). In humans, a recent DSI study agreed that the
cortex is comprised of structurally segregated and functionally specialized regions that are
interconnected by a network of cortico-cortical axonal pathways (Hagmann et al., 2008).
Functional connectivity analysis, defined as temporal relation of activity between neural
populations (Friston, 1994), confirms the multiscale network interactions expected from small-
world architecture and simulations (Honey et al., 2007; Tononi et al., 1994). Cortical regions
dynamically couple to one another forming transient functional networks associated with
perception, cognition, and action (Bressler, 1995; Buzaski & Draguhn, 2004; Nyberg et al.,
2000; Singer & Gray, 1995; Varela, Lachaux, Rodriguez, & Martinerie, 2001). For example, it
was demonstrated that functional connectivity between the fusiform face area (FFA), dorsolateral
and ventrolateral prefrontal cortex (PFC), premotor cortex, intraparietal sulcus, caudate nucleus,
thalamus, hippocampus, and occipital regions is modulated by maintenance of a face image in
working memory (Gazzaley, Rissman, & D'Esposito, 2004). Spontaneous activity in the default
and resting states, where there is no experimental input to the system, also displays distributed
network dynamics (Biswal, Yetkin, Haughton, & Hyde, 1995; Fox et al., 2005; Greicius,
Krasnow, Reiss, & Menon, 2003; Raichle et al., 2001). For example, Fox and colleagues (2005)
reported reliable functional connectivity across participants between the posterior cingulate
cortex (PCC), medial PFC, left dorsolateral PFC, bilateral inferoparietal cortex, left inferolateral
temporal cortex, and left parahippocampal gyrus when participants were awake but with their
eyes closed. When participants switched to listening to narrative text, the ‘default’ PCC activity
(Raichle et al., 2001) became inversely correlated with activity in Broca’s area, which became
5
positively connected to Wernicke’s area and the left premotor cortex (Hampson, Peterson,
Skudlarski, Gatenby, & Gore, 2002). These studies demonstrate that the anatomical architecture
of the cerebral cortex enables a shift between the transient functional networks necessary for
neurocognitive functioning (McIntosh, 2000).
It has been suggested that brain signal variability, or more formally complexity, can
measure the diversity of functional networks in a given brain (Ghosh et al., 2008; McIntosh et
al., 2008; McIntosh et al., 2010), which in turn serve as an index of the individual’s cognitive
capabilities. In line with this view, increases in complexity are observed from infancy to
adulthood (Anokhin, Birbaumer, Lutzenberger, Nikolaev, & Vogel, 1996; Lippé, Kovacevic, &
McIntosh, 2009; McIntosh et al., 2008; Meyer-Lindenberg, 1996), which likely coincides with
concomitant maturational increases in behavioural repertoire. Furthermore, increased complexity
has been shown to correlate positively with performance accuracy (Garrett, Kovacevic,
McIntosh, & Grady, 2010; McIntosh et al., 2008; Misic, Mills, Taylor, & McIntosh, 2010), and
is also negatively correlated with response time variability (Garrett et al., 2010; Garrett,
Kovacevic, McIntosh, & Grady, 2011; McIntosh et al., 2008) suggesting that greater stability of
cognitive responses is a result of increased neural network complexity. One recent study
demonstrated that EEG complexity increases as a function of previous knowledge of the stimulus
(Heisz, Shedden, & McIntosh, under review). The results revealed that famous faces elicited
higher signal complexity than novel faces as a function of personal familiarity with the famous
face (e.g., familiarity ratings, naming accuracy). The same study also showed that complexity
increases with acquired familiarity in an exposure training paradigm. Taken together, these
results support the hypothesis that a more complex functional neural architecture reflects the
enhanced dynamic repertoire of the system and supports a greater capacity for information
processing (Ghosh et al., 2008; McIntosh et al., 2010).
6
As already noted, music listening involves widespread neural activity, including the
prefrontal and temporal cortices (Patterson et al., 2002; Peretz & Zatorre, 2005; Zatorre, 1988;
Zatorre et al., 1994; Zatorre & Samson, 1991), premotor and supplementary motor area (SMA)
(Halpern, Zatorre, Bouffard, & Johnson, 2004; Halsband, Tanji, & Freund, 1993) cerebellum
(Ivry & Keele, 1989; Janata & Grafton, 2003), and parahippocampal and paralimbic regions
(Green et al., 2008; Menon & Levitin, 2005; Mizuno & Sugishita, 2007; Salimpoor, Benovoy,
Larcher, Dagher, & Zatorre, 2011). Initially, different components of music, such as pitch,
rhythm, and dynamics (loudness), are processed separately (De Santis, Clarke, & Murray, 2007;
De Santis, Spierer, Clarke, & Murray, 2007; Peretz, 1990), and integrated later to give the
impression of a unified musical piece (Munte, Altenmuller, & Jancke, 2002; Peretz & Kolinsky,
1993). For example, fixed pitches are processed bilaterally in Heschl’s gyrus (Griffiths, 2003;
Patterson et al., 2002), whereas posterior regions of the secondary auditory cortex process pitch
height, and anterior regions process pitch chroma (Warren, Uppenkamp, Patterson, & Griffiths,
2003). The superior temporal gyrus and planume polare are activated by intervals, contour, and
melody (Patterson et al., 2002). Rhythm processing invokes activity in the cerebellum, basal
ganglia (Ivry & Keele, 1989; Janata & Grafton, 2003), premotor cortex, and SMA (Halpern et
al., 2004; Halsband et al., 1993).
Anatomical studies have observed increased grey and white matter volumes in musicians
relative to nonmusicians (Gaser & Schlaug, 2003a; Schlaug, 2001; Schlaug, Jancke, Huang, &
Steinmetz, 1995). In order to distinguish between predisposition and training specific effects,
investigators in a recent study gave different levels of musical instruction to children for 15
months (Hyde et al., 2009a). Consistent with structural differences found between adult
musicians and nonmusicians (Bermudez & Zatorre, 2005; Lee, Chen, & Schlaug, 2003; Schlaug,
7
Jancke, Huang, & Steinmetz, 1995; Schmithorst & Wilke, 2002; Schneider et al., 2002), the
authors observed increases in size of the right pre-central gyrus, corpus callosum, and Heschl’s
gyrus in the musically trained children compared to the nonmusically trained children. Increased
cerebellar volume is also reported in musicians and has been shown to correlate positively with
lifelong intensity of practice (Hutchinson et al., 2003).
Musicians show additional motor cortex activation compared to nonmusicians during
music perception (Grahn & Brett, 2007). Similarly, fMRI research has found reduced
asymmetrical activity in the motor cortices of pianists, corresponding to the bimanual
coordination required in piano performance (Jancke, Schlaug, & Steinmetz, 1997). The bulk of
neuroimaging studies concerned with the functional differences between musicians and
nonmusicians employ techniques such as electroencephalography (EEG) or
magnetoencephalography (MEG), which forego the spatial accuracy of fMRI in favour of
temporal precision. Evoked event-related potentials (ERP) are time-locked to a stimulus at rates
of hundreds of milliseconds, and reflect specific physical features of the acoustic environment
(Näätanen & Picton, 1987). One of the first indications of functional reorganization in musicians
was presented by Elbert and colleagues (1995). They found that somatosensory-evoked magnetic
fields were larger for the left-hand fingers of violinists. Pantev and colleagues (1998) found that
musicians show approximately 25% larger amplitude evoked N1 magnetic responses for piano
sounds compared to pure tones and nonmusicians, and that this enlargement was correlated with
the age at which music training began. Continued investigation revealed that this effect was most
pronounced for tones from the musicians’ own type of instrument (Pantev, Roberts, Schulz,
Engelien, & Ross, 2001). Similar enhanced electrophysiological responses have been observed
for spatial selectivity of sound sources in conductors compared to pianists and nonmusicians
(Munte, Nager, Beiss, Schroeder, & Altenmuller, 2003).
8
Studies of violations in musical expectancy yield characteristic ERP responses that are
heightened in musicians. The early right anterior negativity (ERAN) is observed in response to
violations of acoustical grammar or syntax (Koelsch, Jentschke, Sammler, & Mietchen, 2007;
Koelsch, Schroger, & Tervaniemi, 2000; Maess, Koelsch, Gunter, & Friederici, 2001) and is
generally sensitive to the amount of harmonic appropriateness (Koelsch et al., 2001). Typically
this means that a negative spike in amplitude is recorded approximately 500 ms after an
individual is presented with violation in the regularities of the musical environment they were
raised in. For instance, the dominant-tonic progression is known as the authentic cadence and is
the most common ending of a harmonic progression in Western tonal music. Endings other than
the authentic cadence, considered acoustic ‘odd-balls’, elicit an ERAN response (Koelsch &
Friederici, 2003). Both musicians and nonmusicians display ERAN responses, though it is
enhanced by musical training (Jentschke & Koelsch, 2009), and it appears to develop by 5 years
of age (Jentschke, Koelsch, Sallat, & Friederici, 2008). MEG analysis suggests that the ERAN is
generated by the inferior frontal cortex (Koelsch & Friederici, 2003). The mismatch negativity
(MMN) is somewhat similar to the ERAN; however, it can be recorded after any change occurs
in a repeated auditory stimulus, even in the absence of attention (Näätanen & Alho, 1995;
Näätanen & Picton, 1987; Picton, Alain, Leun, Ritter, & Achim, 2000), and can therefore be
thought of as more sensitive to the immediate acoustic context. The MMN has been localized
mainly to the supratemporal plane (Alho, 1995; Levanen, Ahonen, Jari, McEvoy, & Sams,
1996). Relative to nonmusicians, musicians display a more pronounced MMN response to
deviations in harmonic chord progressions (Koelsch et al., 1999), pitch sequences (Brattico,
Winkler, Naatanen, Paavilainen, & Tervaniemi, 2002), and complex temporal patterns (Lopez et
al., 2003; Tervaniemi, Ilvonen, Karma, Alho, & Naatanen, 1997). Specifically, professional
musicians show MMN for timing deviations as small as 20 ms, whereas nonmusicians require a
9
deviation of at least 50 ms in order to exhibit an MMN response (Russeler, Altenmuller, Nager,
Kohlmetz, & Munte, 2001). Furthermore, when compared to woodwind players and
nonmusicians, drummers show a heighted MMN response to manipulated drum beat sequences
(Munte et al., 2003).
The data analyzed in the present study were previously collected and analyzed for
MMNm (magnetic mismatch negativity, as the data were acquired with MEG) responses in
musicians and nonmusicians (Fujioka, Trainor, Ross, Kakigi, & Pantev, 2004). Fujioka and
colleagues observed that musicians had consistent MMNm responses to deviant tones in two
different conditions, in which melodies were learned over the course of the experiment.
Nonmusicians showed MMNm only in one of the two conditions and the amplitude of their
response was significantly attenuated compared to musicians. These results paralleled those of a
behavioural measure. Musicians’ were at ceiling on a task that required them to detect deviant
tones (96.50% & and 95.83%, respectively for the two conditions), and significantly more
accurate than nonmusicians (63.0% and 86.17%). Importantly, both musicians and nonmusicians
exhibited a significant MMNm response to deviance from a repeated control pure tone,
indicating that musicians’ superior performance relied on consideration of the entire melody and
not an acoustical advantage.
Despite the collection of evidence that music perception involves distributed brain
regions, direct examinations of the network interactions between regions are limited. Two lines
of evidence stand out: increased corpus callosum size in musicians, and enhanced long-range
oscillatory synchronization. Interhemispheric interaction is believed to be crucial for the
integration necessary for conscious perception (Tononi, 2010; Tononi & Edelman, 1998).
Increased size of the anterior midline corpus callosum is observed consistently in musicians (Lee
et al., 2003; Schlaug, Jancke, Huang, Staiger, & Steinmetz, 1995; Schmithorst & Wilke, 2002).
10
These corpus callosum differences emerged in children after approximately 29 months of
practicing music, and total weekly music exposure predicted degree of change as well as
improvement (Schlaug et al., 2009). These results loosely support the suggestion that musicians
have the enhanced network integration to support their abilities.
Synchronization of neural oscillations at a specific frequency between different neuronal
populations is frequently proposed as a mechanism for information transfer in the brain
(Bressler, Coppola, & Nakamura, 1993; Buzaski, 2006; Varela et al., 2001). Transient periods of
oscillation synchronization in the gamma frequency band (25-80Hz), measured as phase-locking
between different neural recording sites, have been theorized to integrate distributed neuronal
sets together into a coherent ensemble that underlies a cognitive act (Canolty et al., 2007; Hipp,
Engel, & Siegel, 2011; Tallon-Baudry & Bertrand, 1999). Rodriguez and colleagues (1999)
observed that face perception, but not meaningless shapes, induced long-distance patterns of
gamma synchronization in EEG scalp recording corresponding to the moment of perception and
an ensuing motor response. In music perception, long-range oscillatory activity is proposed to
bind segregated musical features and to match acoustic information to learned templates in
memory for music (Fujioka et al., 2009). Gamma and beta band synchronization between the
auditory and motor cortices is modulated by beat processing (Fujioka et al., 2009; Snyder &
Large, 2005), and induced gamma activity is enhanced when participants listen to meaningful
stimuli compared with pure tones during a discrimination task (Crone, Boatman, Gordon, & Hao,
2001). Compared to nonmusicians, musicians show increased phase synchrony over distributed
cortical areas, predominately in the gamma band (Bhattacharya & Petsche, 2005; Snyder &
Large, 2005; Sokolov, Pavlova, Lutzenberger, & Birbaumer, 2004). This heightened effect
appears to be music specific because it is reduced to nonmusician levels when musicians listen to
text (Bhattacharya & Petsche, 2005). In an effort to delineate training specific effects, Trainor
11
and colleagues (2009), showed that induced oscillatory gamma band activity is enhanced in
musicians relative to nonmusicians specific to instrument of practice, and it develops in children
after one year of musical training, but not in children without training. Taken together, these
results suggest that music training leads to increased integration of information in neural
networks.
The presented evidence converges with the understanding that the dynamics that emerge
from neural networks are crucial for the manifestation of complex behaviours and cognitions
(McIntosh, 2000). Previous investigations have uncovered widespread brain activity associated
with music perception yet discussion of interactions among regions remains minimal. The
temporal coordination of activity flow through various music networks can be expected to
influence perception of a complex piece of music, and music training may result in distinct
patterns of functional dynamics. It has been suggested that a brain with higher signal complexity
has a more varied set of functional connections and the ability to perform more operations
(Ghosh et al., 2008; McIntosh et al., 2008; McIntosh et al., 2010; Tononi et al., 1994). The goal
of the present study was to investigate whether increased diversity of neural activity underlies
musicians’ superior perception of musical melodies. Specifically, we hypothesized that
compared to nonmusicians, musicians should exhibit higher brain signal complexity when
listening to music.
Materials and Methods
Participants
Participants were 12 musicians (8 female, ages 19-33 years) and 12 nonmusicians (9
female, ages 19-40 years). The musicians had studied and played two or more instruments
12
regularly for at least 10 years (M = 14.3 years, range: 10 to 23) with formal education including
musical schools or private lessons. The nonmusicians had almost no formal music training (3 of
12 had 2 years of lessons but quit playing more than 10 years ago) except for what they learned
in school. None of the individuals in either group had absolute pitch. All participants were right
handed (assessed by the Edinburgh handedness test) with normal hearing (range of 250 to 8000
Hz as tested by clinical audiometry), and they were screened for any history of medical,
neurological, psychiatric, and substance-abuse problems. After being informed about the nature
of the study, they provided written consent to participate and the experiment was approved by
The Ethics Commission of the Baycrest Centre for Geriatric Care in accordance with the
Declaration of Helsinki.
Stimuli
In both conditions, stimuli comprised CD-quality standard and deviant melodies, both
with five tones in a digitally recorded piano timbre. Standard melodies were presented 80% of
the time and the duration of each tone was 300 ms.
Contour.
In the contour condition, eight different five-tone ascending melodies were used (see
Figure 1). Each comprised different tones and intervals from the C-major diatonic scale, and
each started on one of five different tones between C5 and G5 (American notation). In the
corresponding deviant melodies, the contour was altered such that the fifth tone was lower in
pitch than the fourth tone. The interval size deviations in the contour melodies ranged from a
minor second to a major third (2 to 4 semitones, respectively). The mean value was a major
second (2 semitones) and this was the same value of deviation in the interval conditions.
Interval.
13
Standard stimuli in the interval condition consisted of a single five-tone major-key
melody with an up-up-down-up contour (do-re-fa-mi-sol), which was transposed to eight keys
with starting notes in the same range as those of the contour stimuli. For the deviant melodies,
the final note was raised by one whole tone (2 semitones, from sol to la). The altered tone did not
change the contour and it remained within the key of the melody.
Procedure
Audio Presentation.
Hearing thresholds for each participant were determined for the left and right ears for the
B5 piano sound. Stimuli were presented at 60 dB above threshold. Participants were presented
with three successive blocks of 300 trials of each condition, and condition presentation order was
counterbalanced. The order of melody variations was pseudo-random. Each melody was
separated by a 900ms silent interval.
MEG Overview.
MEG is a passive, non-invasive neurophysiological technique that measures the magnetic
fields generated by neuronal activity of the brain. Different areas of the brain communicate with
electrochemical impulses that generate electric and magnetic fields. MEG detects patterns of
brain activity as a number of sources of extremely miniscule electro-magnetic fields, such that
MEG is a ‘direct’ measure of brain activity, compared to fMRI and PET, which measure brain
metabolism. In the MEG equipment, magnetic field sensors rest in a helmet placed on the
individual’s head. Fiducials are placed on the nasion and pre-auicular points for head
localization.
MEG Acquisition.
14
Magnetic field responses were recorded with a 151-channel whole-cortex magnetometer
system (OMEGA, CTF Systems Inc, Port Coquitlam, Canada) at the Rotman Research Institute
at Baycrest Centre for Geriatric Care. Data were collected at a sampling rate of 312.5 samples/s.
Pre-processing included low-pass filtering at 55 Hz, and eye-blink and movement artifact
rejection at a threshold of 2 pT. Duration of recording epoch was 2.099 s, including a 0.4-s pre-
stimulus period. Onset of the first tone of each stimulus synchronized the stimulus presentation
and the data acquisition.
The recordings were performed while participants sat upright in an adjustable chair in a
magnetically shielded room. Prior to MEG acquisition, each participant was fitted with three
fiducial localization coils in order to localize the position of the individual’s head relative to the
MEG sensors. Participants watched a soundless movie of their choice and were instructed not to
pay attention to the sound stimuli. The movie was projected onto a screen placed in front of the
MEG chair. Compliance was verified by video monitoring. Participants were instructed not to
pay attention to the auditory stimuli and no explanation about the stimuli was provided.
Data Analysis
Multiscale Entropy.
Because interactions due to both local dense interconnectivity and sparse long-range
projections give rise to the outputs of neuronal networks (Kelso, 1995; Tononi et al., 1994), the
resulting dynamics could be expected to operate at multiple time scales. MSE calculates sample
entropy of a signal at different time scales (Costa, Goldberger, & Peng, 2002, 2005) and was
used to measure the complexity of the brain signal and music sequences. Sample entropy is the
negative of the logarithmic conditional probability that two sequences of m consecutive data
points that are similar to each other (within a given tolerance r) will remain similar at the next
point (m+1) in the data set (N), where N is the length of the time series (Richman & Moorman,
15
2000). MSE was calculated using the algorithm available at www.physionet.org/physiotools/mse
with parameter values pattern length, m, equal to 5, and the similarity criterion, r, equal to 1
(Vakorin et al., 2010). The value r is defined as a proportion of the standard deviation of the
original data (Costa, Goldberger, & Peng, 2004).
MSE calculation involves two procedures: (1) coarse-graining of the time series, and (2)
calculating sample entropy for each coarse-grained time series. For scale t, the time series is
constructed by averaging the data points with non-overlapping windows of length t. The number
of scales is determined by the reliability of the entropy estimation of the series and is a function
of the number of data points in the signal. A signal with fewer data points has fewer time scales.
MSE requires a time series of approximately 500 data points to get a stable measure of entropy
by scale, and the MEG epochs of 2.099 s with a sampling rate of 312.5Hz yielded 656 samples
and fulfilled this requirement. On an MSE curve plot (see Figure 2c), scale 1 is the sample
entropy for the non-averaged, original signal, scale 2 is the entropy from the signal average of 2
adjacent points, scale 3 is the average of 3 adjacent points, and so on. Unlike traditional single-
scale entropy measures, signals with low complexity, such as random noise or completely
deterministic signals, exhibit a steep decline in MSE curve with increasing scale, and signals
with greater temporal interdependencies, such as 1/f noise or cardiovascular inter-beat intervals,
display a more gradual shift in the curve (Costa et al., 2002, 2004, 2005; Costa & Healey, 2003;
Lippé et al., 2009; Nikulin & Brismar, 2004). This is because 1/f signals contain information
about dependences within and between timescales (Costa et al., 2005). For each subject in the
present analysis, a channel specific MSE estimate was obtained as a mean across single trial
entropy measures for timescales 1-10. MSE comparisons are done on the shape of the MSE
curve (when entropy is plotted by the coarse-grain level) or as an area-under-the-curve.
16
It is worth noting that a strong correlation between spectral power distribution and MSE
values has been observed (McIntosh, Kovacevic & Itier, 2008). Specifically, simulated data
demonstrated that alterations to Fourier coefficients changed the resulting MSE curve. It was
also demonstrated, however, that although random phase jitter of the time series significantly
impacted MSE, it had no effect on spectral power. Thus, it was concluded from these simulations
that MSE is sensitive to non-linearities that are not captured by spectral power. Other results
confirm that MSE delineates brain patterns not observable from spectral power (Heisz et al.,
under review; Misic et al., 2010).
Partial Least Squares Analysis.
Partial Least Squares (PLS) analysis is a multivariate statistical technique that has been
used in neuroimaging to extract commonalities between brain activity and experimental design
(McIntosh, Bookstein, Haxby, & Grady, 1996; McIntosh & Lobaugh, 2004). Although this
approach is similar to canonical correlation, it maximizes the covariance between two data sets
instead of the correlation. PLS has been validated for analysis of data from PET (McIntosh et al.,
1996), fMRI (Martinez-Montes, Valdes-Sosa, Miwakeichi, Goldman, & Cohen, 2004; McIntosh
& Lobaugh, 2004), ERP (Lobaugh, West, & McIntosh, 2001), and MEG (Misic et al., 2010).
Task PLS (McIntosh & Lobaugh, 2004) specifically analyzes the association between brain
activity (data set X) and experimental design (data set Y).
In the first step, MSE values were arranged into matrix X, with subject measures, by
group and condition in rows, and mean MSE values for each MEG channel at each time scale in
columns. To assess the relationship of group and condition with MSE value, matrix Y is a matrix
of dummy coding that codes for the experimental groups or conditions. In the second stage of
analysis, the average for each condition was then computed and stored as M. Each column of M
was then mean-centered by subtracting the mean of the column from each value of that column.
17
This mean-centered matrix was subjected to singular value decomposition (SVD) to compute an
optimal least-squares fit to the covariance between the original X and Y variables (i.e. MSE
across all channels and group/condition). The decomposition yields a set of orthogonal latent
variables (LV) each containing three matrices: (1) U, weights for the rows, indicating a contrast
that characterizes the differences between groups and or/tasks; (2) V, weights for the columns,
indicating the linear combination of channels maximally related to the contrast; and (3) the
singular value, which is the covariance between the contrast and the MSE weights. In this
analysis, each LV represented one contrast between experimental groups and/or conditions in
relation to a particular pattern of channels and temporal scales. SVD is similar to principle
components’ analysis (PCA), whereby an LV accounts for a proportion of the total variance in
the data matrix. The SVD LV is similar to loading on a factor in PCA. Also, it has been noted
that task PLS is somewhat akin to discriminant analysis (Abdi & Williams, 2010).
The statistical significance of each LV as a whole was determined using permutation tests
(McIntosh et al., 1996; McIntosh & Lobaugh, 2004). The purpose of conducting a permutation
test is to evaluate whether an LV is significantly different from random signal. Permutation tests
consist of randomly reordering the rows (i.e. subject observations) of matrix X, while leaving
matrix Y unchanged. PLS is recomputed on the permuted matrix to obtain a new matrix of
singular values. After repetition of this procedure (500 permutation), the set of all singular values
provides a sampling distribution from which the null hypothesis can be tested. The number of
permutations performed is proportional to the desired precision of the alpha critical value, thus,
500 permutations allow for precision to the third decimal point. P-values are determined by
calculating the proportion of permuted singular values that are equal to or exceed the original
singular value.
18
The stability of each statistical effect is assessed through bootstrap estimation of standard
error confidence intervals of the weights in U and V. Bootstrap samples are created by sampling
with replacement the observations in X and Y (Efron & Tibshirani, 1986). The standard error
was subsequently estimated from 500 bootstrap samples, and the singular vector weights for each
MSE coefficient were divided by this standard error to yield a bootstrap ratio. The bootstrap ratio
is similar to a z-score if the distribution of singular vector weights is Gaussian (McIntosh &
Lobaugh, 2004). Peak channels with a weight/standard error ratio > 3.5 (99% confidence
interval) were considered to be reliable (Sampson, Streissguth, Barr, & Bookstein, 1989) .
Results
In both conditions (contour, interval), standard melodies elicited greater sample entropy
in musicians compared to nonmusicians (p < .001, see Figures 2 and 3). In the contour condition,
both groups showed a consistent pattern for both standard (Figure 2) and deviant (Figure 4)
melodies. Musicians had higher MSE than nonmusicians for the first block, followed by a rapid
decrease in the second block, and a return to an intermediate level in the third block.
Nonmusicians displayed a gradual decrease in MSE across all three blocks (p = .001; the
difference between blocks 2 and 3 was non-significant, p = .906).
A similar general trend of decreasing MSE as a function of trial was observed in the
interval condition. Figure 3 illustrates that standard melodies elicited a rapid decrease in MSE
from block 1 to block 2 in musicians, with no significant change between blocks 2 and 3, and a
more gradual decline in MSE across all three blocks in nonmusicians. Deviant melodies in the
interval condition resulted in stable MSE across all three blocks in musicians (i.e., no significant
differences, see Figure 5). Nonmusicians exhibited greater MSE than musicians during the first
block, p < .001, and rapidly declined to musician levels for the second and third blocks (musician
19
vs. nonmusicians’ block 2 and block 3 were insignificant). Bootstrap ratios (Figure 2-5B)
demonstrate that all effects were reliable across most channels and observed at coarse scales.
Discussion
Musicians versus Nonmusicians
As predicted, musical training is associated with increases in brain signal complexity.
Compared to nonmusicians, musicians displayed higher multiscale entropy during initial melody
presentation. Increased signal complexity is indicative of increased diversity of transient
functional networks (McIntosh et al., 2010; Tononi et al., 1998). This highly variable activity
may be ideally suited to capture changes in elaborate harmonies, rhythms, and dynamics of
pieces of music, and may develop through experience to support a more accurate and wider
range of musical behaviours.
In addition to enhanced music performance, musicians also display enhanced
performance on tests of music perception (Peretz & Babai, 1992; Tervaniemi, Just, Koelsch,
Widmann, & Schroger, 2005). In an earlier study with the same sample as the present study
musicians outperformed nonmusicians in a deviant tone-detection task (Fujioka et al., 2004).
Unfortunately, the behavioural task was not performed at the time of MEG acquisition, which
precludes examination of correlations between brain activity and performance. However, recent
evidence has demonstrated that optimal neural complexity is positively correlated with
performance on tests of face recognition (Heisz et al., under review; McIntosh et al., 2008; Misic
et al., 2010), as well as with performance on tests of perceptual matching, attentional cueing, and
delayed match-to-sample (Garrett et al., 2010). Considered jointly with simulations
demonstrating that high complexity indicates rapid processing of high levels of information that
can flexibly adapt to changes in external input (Tononi, Sporns & Edelman, 1994), our results
20
provide support for the theory that increased neural complexity supports augmented music
perception. Further investigation involving simultaneous neuroimaging and behavioural testing is
required before this claim is definitive.
Maturational increases in brain signal complexity have been observed in infants exposed
to an auditory stimulus (Lippé et al., 2009), and this trend continues from childhood to adulthood
(McIntosh, Kovacevic & Itier, 2008). Because both groups in the present study were similar in
age, our results suggest that biological maturation cannot explain increased neural complexity
seen among musician. This account of experience dependent changes in transient neural activity
is further supported by a previous report of a positive relationship between MSE and stimulus
familiarity (Heisz et al., under review), in which famous faces elicited greater EEG complexity
than novel faces, an effect that was a function of personal familiarity measured with familiarity
ratings and naming. Heisz and colleagues also found that MSE correlated with acquired
familiarity in a multi-day training paradigm. Because the effect was distributed across brain
regions, the authors concluded that the increase in signal complexity indexed high integration of
specialized, segregated functional regions. Similarly, we propose that the heightened signal
complexity recorded in musicians during melody perception developed as a result of training in
order to facilitate efficient processing of elaborate music. Further investigation of the
relationship between the extent of music experience and brain signal complexity is required to
substantiate this claim.
Unconscious, automatic processing
A general decrease of entropy across blocks was observed in both groups but this effect
was more pronounced in musicians. The present experiment was originally designed to
investigate the association between extensive music training and automatic melodic processing
(Fujioka et al., 2004). Recall that participants watched a silent movie during melody presentation
21
and they were instructed not to ignore the sound stimuli. Reduced complexity was also reported
in a similar experimental paradigm that drew attention away from the stimulus (Vakorin et al.,
2010), and during repetition of a familiar stimulus (Heisz et al., under review). Initial
presentation of a novel stimulus attracts attention (Daffner, Mesulam, et al., 2000; Daffner,
Scinto, et al., 2000; Escera, Alho, Winkler, & Näätanen, 1998; Tiitinen et al., 1993), and
transient neural network activity resolves the input information (Honey et al., 2007; McIntosh,
2000). Auditory sensory processing involves integration of information in the input with stored
representations of preceding auditory events (Alain, Woods, & Knight, 1998). Thus, with further
repetition, listeners habituate to innocuous stimuli and minimal sensory processing descends to
an unconscious level (Fantz, 1964; Trainor, McDonald, & Alain, 2002). Decreases in signal
complexity are associated with diminished conscious awareness (Protzner et al., 2010; Shen,
Olbrich, Achermann, & Meier, 2003); therefore, while we did not assess levels of stimulus
awareness, we believe that the observed decreases in MSE may reflect a transition from
conscious processing of informative and novel information to automatic processing of repetitive
information.
Automatic processing is expected to coincide with decreased functional integration
between regions of the network and with decreased of the neural signal complexity (Tononi,
2010). The reductions in sample entropy in our analysis were observed reliably across coarse
time scales, indicative of long-range temporal correlations. The lack of source analysis in this
investigation makes spatial conclusions difficult. Recent analyses of EEG signals demonstrated,
however, that local information is typically represented at finer timescales whereas conduction of
distributed information conduction is expressed at coarser timescales (Vakorin, Lippé, &
McIntosh, 2011). Furthermore, maturational increases in coarse-scale entropy (McIntosh et al.,
2008; Vakorin et al., 2011) parallel developmental increases in integrated long-range
22
connectivity relative to local activity (Fair et al., 2009). Segregated, localized activity is
suggested to be associated with automatic or unconscious processes, and integration from
distributed neuronal groups is necessary for conscious, unified perception (Tononi & Edelman,
1998).
The importance of a balance between segregation and integration can be seen in
comparison of different conscious and unconscious states. Massive brain synchrony and hyper-
integration result in seizures, diminished consciousness, and decreased complexity (Protzner et
al., 2010). Conversely, slow-wave sleep (Massimini et al., 2007; Massimini et al., 2005) and
anaesthetic states (Ferrarelli et al., 2010) are associated with decreased connectivity between
brain regions and decreased complexity. During REM sleep, when conscious-like dreams occur
(Hobson, 2006; Stickgold, Hobson, Fosse, & Fosse, 2001), wakefulness-like EEG patterns are
associated with increased effective connectivity, integration (Ferrarelli et al., 2010), and signal
complexity (Shen et al., 2003). Simulations have shown that long-range integration appears
when a novel task is introduced, which decreases during routinization (Dehaene, Kerszberg, &
Changeux, 1998). Additionally, specialized, local processing corresponds to lower complexity
when compared to collaboration between distributed regions (Tononi et al., 1994). Animal
recordings also show that activity evoked by a habituated stimulus is restricted to local sensory
pathways (Horel et al., 1967), and fMRI has shown that evoked activity is restricted to primary
and secondary auditory cortex (BA 41 and 42) following auditory tone habituation (Celsis et al.,
1999).
Melodic Contour versus Intervals
Compared with nonmusicians, deviant tones in the interval condition elicited minimal
MSE among musicians during block 1 (Figure 5). We suggest that this effect is because the
intervallic change contained nominal information relative to the changes in the contour
23
condition, and consequently may be more easily and rapidly processed by trained musicians.
This explanation is in line with previous results that have observed lower brain signal entropy as
a function of stimulus information content (Heisz et al., under review; Misic et al., 2010). Firstly,
the deviant melodies were dispersed infrequently (20% of trials) among the standard melodies.
As a result, consequences of habituation on neural signal complexity were expected to contribute
to a reduction in recorded entropy. Secondly, neural network activity that accompanies deviant
melodies in the interval condition may have been lower than in the contour condition because
these melodies contain less inherent information than the contour melodies, simply by virtue of
their design. In the contour condition, all eight melodies were composed of different intervals.
The deviant tones changed contour as well as the final interval of the melody. For example, in
contour melody 2 (Figure 1), the fifth (last) tone is one semitone higher than the fourth tone,
whereas the deviant tone is four semitones lower. By contrast, melodies in the interval condition
are transpositions of the same sequence, and all deviant tones increased the size of the final
interval by one whole tone. Consequently, the melodies in the interval condition are much more
statistically stable and carry less novel information than the melodies in the contour condition
(Shannon & Weaver, 1949). In a probabilistic sense, the contour melodies involved a wider
range of possible outcomes and hence a higher level of uncertainty then interval melodies.
It has been suggested that the brain operates in a Bayesian probabilistic sense to generate
predictions about the likely network activity configuration that would be optimal for a given
input (McIntosh et al., 2010). Indeed, there is ample evidence that there are neural mechanisms
that operate in a predictive and regular sequence even in the omission of an expected stimuli
(Rankin, Large, & Fink, 2009; Ritter, Sussman, Deacon, Cowan, & Vaughan, 1999; Russeler et
al., 2001; Snyder & Large, 2005; Zanto, Large, Fuchs, & Kelso, 2005). Specific to music, Large
and colleagues (Zanto, Snyder, & Large, 2006) observed induced gamma-band activity that
24
reflected temporally precise rhythm predictions, which coincided with participants’ anticipation-
based synchronized responses to changing rhythms (Large & Jones, 1999; Rankin et al., 2009).
In a continually changing environment, activity that flows through variable network
configurations enables adaptability of response (Manoel & Connolly, 1995). Conversely, when
the temporal regularity of previous successive events decreases the number of possible
outcomes, as is the case of the interval melodies in the present study, network activity stabilizes
because highly complex and variable network activity is no longer necessary, in a probabilistic
sense, to ensure a correct response. Although this explanation is plausible, further exploration of
the effect of stimulus information on brain signal complexity is required before a definite
conclusion can be reached.
Entropy and ERP
Fujioka and colleagues (2004) observed larger amplitude MMNm in musicians for
interval compared to contour deviants, and only for interval deviants in nonmusicians. Further
analyses of the evolution of this ERP over the course of the experiment were not conducted in
the present study, making it difficult to delineate the specific relationship between signal entropy
and event-related potentials. Indeed, signal-to-noise ratio difficulties make determining the
evolution of ERP problematic. MSE is computed on neural activity containing both induced and
evoked signal components. Importantly, previous analyses observed no change in task associated
MSE values after subtraction of the average evoked response (Misic et al., 2010). Because
MMNm evoked potentials were more pronounced for the more statistically regular interval
melodies, it seems that ERP responses were more sensitive to the impact of a deviant in the
overall regularity of the sequence. In other words, the ‘deviance’ value assigned to a particular
tone is taken relative to the variability of the standard sequences. By contrast, our measure of
25
neural complexity that appears to be sensitive to the amount of consciously processed
information.
Conclusion
When presented with a new melody, musicians display more brain signal complexity
compared to age-matched nonmusicians. Such increases in the variability of their brain activity
suggest that music training increases neural resources that are available for music processing.
Highly variable brain activity may be better able to capture the subtle intricacies and the large
amount of information in a dynamic piece of music, and thus be responsible for the improved
behavioural and cognitive skills of musicians. The present study represents an important step in
testing this topic hypothesis.
26
27
28
29
30
31
References
Abdi, H., & Williams, L. (2010). Barycentric discriminant analysis. In N. Salkind, D. Dougherty
& B. Frey (Eds.), Encyclopedia of Research Design (pp. 64-75). Thousand Oaks, CA:
Sage.
Alain, C., Woods, D., & Knight, R. (1998). A distributed cortical network for auditory sensory
memory in humans. [Article]. Brain Research, 812(1-2), 23-37.
Alho, K. (1995). Cerebral generators of mismatch negativity (MMN) and its magnetic
counterpart (MMNm) elicited by sound changes. Ear and Hearing, 16, 38-51.
Anokhin, A., Birbaumer, N., Lutzenberger, W., Nikolaev, A., & Vogel, F. (1996). Age increases
brain complexity. [Article]. Electroencephalography and Clinical Neurophysiology,
99(1), 63-68.
Bendor, D., & Wang, X. (2005). The neuronal representation of pitch in primate auditory cortex.
[Article]. Nature, 436(7054), 1161-1165. doi: DOI 10.1038/nature03867
Bermudez, P., & Zatorre, R. (2005). Differences in gray matter between musicians and
nonmusicians. [Proceedings Paper]. Neurosciences and Music Ii: From Perception To
Performance, 1060, 395-399. doi: DOI 10.1196/annals.1360.057
Bhattacharya, J., & Petsche, H. (2005). Phase synchrony analysis of EEG during music
perception reveals changes in functional connectivity due to musical expertise. [Article].
Signal Processing, 85(11), 2161-2177. doi: DOI 10.1016/j.sigpro.2005.07.007
Biswal, B., Yetkin, F., Haughton, V., & Hyde, J. (1995). Functional Connectivity in the Motor
Cortex of Resting Human Brain Using Echo-Planar MRI. [Article]. Magnetic Resonance
in Medicine, 34(4), 537-541.
32
Brattico, E., Winkler, I., Naatanen, R., Paavilainen, P., & Tervaniemi, M. (2002). Simultaneous
storage of two complex temporal sound patterns in auditory sensory memory. [Article].
Neuroreport, 13(14), 1747-1751.
Bressler, S. (1995). Large-scale cortical networks and cognition. Brain Research Reviews, 20,
288-304.
Bressler, S., Coppola, R., & Nakamura, R. (1993). Episodic multiregional cortical coherence at
multiple frequencies during visual task performance. Nature, 366, 153-156.
Bullmore, E., & Sporns, O. (2009). Complex brain networks: graph and theoretical analysis of
structural and functional systems. Nature Reviews Neuroscience, 10, 186-198.
Buzaski, G. (2006). Rhythms of the Brain. New York, NY: Oxford University Press.
Buzaski, G., & Draguhn, A. (2004). Neuronal oscillations in cortical networks. Science, 304,
1926-1929.
Canolty, R., Soltani, M., Dalal, S., Edwards, E., Dronkers, N., Nagarajan, S., . . . Knight, R.
(2007). Spatiotemporal dynamics of work processing the human brain. Frontiers in
Human Neuroscience, 1, 185-196.
Celsis, P., Boulanouar, K., Doyon, B., Ranjeva, J., Berry, I., Nespoulous, J., & Chollet, F.
(1999). Differential fMRI responses in the left posterior superior temporal gyrus and left
supramarginal gyrus to habituation and change detection in syllables and tones. [Article].
Neuroimage, 9(1), 135-144.
Chen, J., Penhune, V., & Zatorre, R. (2008). Listening to Musical Rhythms Recruits Motor
Regions of the Brain. [Article]. Cerebral Cortex, 18(12), 2844-2854. doi: DOI
10.1093/cercor/bhn042
Cherniak, C. (1994). Component Placement Optimization in the Brain. [Article]. Journal of
Neuroscience, 14(4), 2418-2427.
33
Costa, M., Goldberger, A., & Peng, C. (2002). Multiscale entropy analysis of complex
physiologic time series. [Article]. Physical Review Letters, 89(6), -. doi: ARTN 068102
DOI 10.1103/PhysRevLett.89.068102
Costa, M., Goldberger, A., & Peng, C. (2004). Comment on "Multiscale entropy analysis of
complex physiologic time series" - Reply. [Editorial Material]. Physical Review Letters,
92(8), -. doi: ARTN 089804 DOI 10.1103/PhysRevLett.92.089804
Costa, M., Goldberger, A., & Peng, C. (2005). Multiscale entropy analysis of biological signals.
[Article]. Physical Review E, 71(2), -. doi: ARTN 021906 DOI
10.1103/PhysRevE.71.021906
Costa, M., & Healey, J. (2003). Multiscale entropy analysis of complex heart rate dynamics:
Discrimination of age and heart failure effects. [Proceedings Paper]. Computers in
Cardiology 2003, Vol 30, 30, 705-708.
Crone, N., Boatman, D., Gordon, B., & Hao, L. (2001). Induced electrocorticographic gamma
activity during auditory perception. [Article]. Clinical Neurophysiology, 112(4), 565-582.
Daffner, K., Mesulam, M., Scinto, L., Acar, D., Calvo, V., Faust, R., . . . Holcomb, P. (2000).
The central role of the prefrontal cortex in directing attention to novel events. [Article].
Brain, 123, 927-939.
Daffner, K., Scinto, L., Calvo, V., Faust, R., Mesulam, M., West, W., & Holcomb, P. (2000).
The influence of stimulus deviance on electrophysiologic and behavioral responses to
novel events. [Article]. Journal of Cognitive Neuroscience, 12(3), 393-406.
De Santis, L., Clarke, S., & Murray, M. (2007). Automatic and intrinsic auditory "what" and
"where" processing in humans revealed by electrical neuroimaging. [Article]. Cerebral
Cortex, 17(1), 9-17. doi: DOI 10.1093/cercor/bhj119
34
De Santis, L., Spierer, L., Clarke, S., & Murray, M. (2007). Getting in touch: Segregated
somatosensory what and where pathways in humans revealed by electrical neuroimaging.
[Article]. Neuroimage, 37(3), 890-903. doi: DOI 10.1016/j.neuroimage.2007.05.052
Deco, G., Jirsa, V., & McIntosh, A. (2011). Emerging concepts for the dynamical organization of
resting-state activity in the brain. Nature Reviews Neuroscience, 12, 43-56.
Dehaene, S., Kerszberg, M., & Changeux, J. (1998). A neuronal model of a global workspace in
effortful cognitive tasks. [Article]. Proceedings of the National Academy of Sciences of
the United States of America, 95(24), 14529-14534.
Dobelle, W., Mladejovsky, M., & Girvin, J. (1974). Artificial vision for blind - electrical-
stimulation of visual-cortex offers hope for a functional prothesis. [Article]. Science,
183(4123), 440-444.
Efron, B., & Tibshirani, R. (1986). Bootstrap Methods for Standard Errors, Confidence Intervals,
and Other Measures of Statistical Accuracy. Statistical Science, 1(1), 54-75.
Elbert, T., Pantev, C., Wienbruch, C., Rockstroh, B., & Taub, E. (1995). Increased cortical
representation of the fingers of the left hand in string players. [Article]. Science,
270(5234), 305-307.
Escera, C., Alho, K., Winkler, I., & Näätanen, R. (1998). Neural mechanisms of involuntary
attention to acoustic novelty and change. [Article]. Journal of Cognitive Neuroscience,
10(5), 590-604.
Fair, D., Cohen, A., Power, J., Dosenbach, N., Church, J., Miezin, F., . . . Petersen, S. (2009).
Functional Brain Networks Develop from a "Local to Distributed" Organization.
[Article]. Plos Computational Biology, 5(5), -. doi: ARTN e1000381 DOI
10.1371/journal.pcbi.1000381
35
Fantz, R. (1964). Visual experience in infants - Decreased attention to familiar patterns relative
to novel ones. [Article]. Science, 146(364), 668-&.
Felleman, D., & Van Essen, D. (1991). Distributed Hierarchical Processing in the Primate
Cerebral Cortex. Cerebral Cortex, 1, 1-47.
Ferrarelli, F., Massimini, M., Sarasso, S., Casali, A., Riedner, B., Angelini, G., . . . Pearce, R.
(2010). Breakdown in cortical effective connectivity during midazolam-induced loss of
consciousness. [Article]. Proceedings of the National Academy of Sciences of the United
States of America, 107(6), 2681-2686. doi: DOI 10.1073/pnas.0913008107
Fox, M., Snyder, A., Vincent, J., Corbetta, M., Van Essen, D., & Raichle, M. (2005). The human
brain is intrinsically organized into dynamic, anticorrelated functional networks.
[Article]. Proceedings of the National Academy of Sciences of the United States of
America, 102(27), 9673-9678. doi: DOI 10.1073/pnas.0504136102
Friston, K. (1994). Functional and effective connectivity: A synthesis. Human Brain Mapping,
2(1/2), 56-78.
Fujioka, T., Trainor, L., Large, E., & Ross, B. (2009). Beta and Gamma Rhythms in Human
Auditory Cortex during Musical Beat Processing. [Proceedings Paper]. Neurosciences
and Music Iii: Disorders and Plasticity, 1169, 89-92. doi: DOI 10.1111/j.1749-
6632.2009.04779.x
Fujioka, T., Trainor, L., Ross, B., Kakigi, R., & Pantev, C. (2004). Musical training enhances
automatic encoding of melodic contour and interval structure. [Article]. Journal of
Cognitive Neuroscience, 16(6), 1010-1021.
Garrett, D., Kovacevic, N., McIntosh, A., & Grady, C. (2010). Blood Oxygen Level-Dependent
Signal Variability Is More than Just Noise. [Article]. Journal of Neuroscience, 30(14),
4914-4921. doi: Doi 10.1523/jneurosci.5166-09.2010
36
Garrett, D., Kovacevic, N., McIntosh, A., & Grady, C. (2011). The Importance of Being
Variable. [Article]. Journal of Neuroscience, 31(12), 4496-4503. doi: Doi
10.1523/jneurosci.5641-10.2011
Gaser, C., & Schlaug, G. (2003a). Brain structures differ between musicians and non-musicians.
[Article]. Journal of Neuroscience, 23(27), 9240-9245.
Gaser, C., & Schlaug, G. (2003b). Gray matter differences between musicians and nonmusicians.
[Proceedings Paper]. Neurosciences and Music, 999, 514-517. doi: DOI
10.1196/annals.1284.062
Gazzaley, A., Rissman, J., & D'Esposito, M. (2004). Functional connectivity during working
memory maintenance. Cognitive, Affective & Behavioral Neuroscience, 4(4), 580-599.
Ghosh, A., Rho, Y., McIntosh, A., Kotter, R., & Jirsa, V. (2008). Noise during Rest Enables the
Exploration of the Brain's Dynamic Repertoire. [Article]. Plos Computational Biology,
4(10), -. doi: ARTN e1000196 DOI 10.1371/journal.pcbi.1000196
Goodale, M., Milner, A., Jakobson, L., & Carey, D. (1991). A Neurological Dissociation
Between Perceiving Objects and Grasping Them. [Article]. Nature, 349(6305), 154-156.
Grahn, J., & Brett, M. (2007). Rhythm and beat perception in motor areas of the brain. [Article].
Journal of Cognitive Neuroscience, 19(5), 893-906.
Green, A., Baerensten, K., Stodkilde-Jorgensen, H., Wallentin, M., Roepstorff, A., & Vuust, P.
(2008). Music in minor activates limbic structures: a relationship with dissonance?
Neuroreport, 19, 711-715.
Greicius, M., Krasnow, B., Reiss, A., & Menon, V. (2003). Functional connectivity in the resting
brain: A network analysis of the default mode hypothesis. [Article]. Proceedings of the
National Academy of Sciences of the United States of America, 100(1), 253-258. doi: DOI
10.1073/pnas.0135058100
37
Griffiths, T. (2003). Functional Imaging of Pitch Analysis. Annals of the New York Academy of
Sciences, 999, 40-49.
Hagmann, P., Cammoun, L., Gigandet, X., Meuli, R., Honey, C., Wedeen, V., & Sporns, O.
(2008). Mapping the structural core of human cerebral cortex. [Article]. Plos Biology,
6(7), 1479-1493. doi: ARTN e159 DOI 10.1371/journal.pbio.0060159
Halpern, A., Zatorre, R., Bouffard, M., & Johnson, J. (2004). Behavioral and neural correlates of
perceived and imagined musical timbre. Neuropsychologia, 42, 1281-1292.
Halsband, U., Tanji, J., & Freund, H.-J. (1993). The role of premotor cortex and the
supplementary motor area in the temporal control of movement in man. Brain, 116, 243-
246.
Hampson, M., Peterson, B., Skudlarski, P., Gatenby, J., & Gore, J. (2002). Detection of
functional connectivity using temporal correlations in MR images. [Article]. Human
Brain Mapping, 15(4), 247-262. doi: DOI 10.1002/hbm.10022
Hannon, E., & Trehub, S. (2005). Metrical categories in infancy and adulthood. [Article].
Psychological Science, 16(1), 48-55.
Heisz, J., Shedden, J., & McIntosh, A. (under review). Relating brain signal variability to
knowledge representation. The Journal of Neuroscience.
Hipp, J., Engel, A., & Siegel, M. (2011). Oscillatory synchronization in large-scale cortical
networks predicts perception. Neuron, 69, 387-396.
Hobson, J. A. (2006). Sleep and Dreaming Encyclopedia of Cognitive Science: John Wiley &
Sons, Ltd.
Honey, C., Kotter, R., Breakspear, M., & Sporns, O. (2007). Network structure of cerebral cortex
shapes functional connectivity on multiple time scales. [Article]. Proceedings of the
38
National Academy of Sciences of the United States of America, 104(24), 10240-10245.
doi: DOI 10.1073/pnas.0701519104
Honey, C., Sporns, O., Cammoun, L., Gigandet, X., Thiran, J., Meuli, R., & Hagmann, P. (2009).
Predicting human resting-state functional connectivity from structural connectivity.
[Article]. Proceedings of the National Academy of Sciences of the United States of
America, 106(6), 2035-2040. doi: DOI 10.1073/pnas.0811168106
Horel, J., Vierck, C., Pribram, K., Spinelli, D., John, E., & Ruchkin, D. (1967). Average Evoked
Responses and Learning. [Article]. Science, 158(3799), 394-&.
Hutchinson, S., Lee, L., Gaab, N., & Schlaug, G. (2003). Cerebellar volume of musicians.
[Article]. Cerebral Cortex, 13(9), 943-949.
Hyde, K., Lerch, J., Norton, A., Forgeard, M., Winner, E., Evans, A., & Schlaug, G. (2009a).
The Effects of Musical Training on Structural Brain Development A Longitudinal Study.
[Proceedings Paper]. Neurosciences and Music Iii: Disorders and Plasticity, 1169, 182-
186. doi: DOI 10.1111/j.1749-6632.2009.04852.x
Hyde, K., Lerch, J., Norton, A., Forgeard, M., Winner, E., Evans, A., & Schlaug, G. (2009b).
Musical Training Shapes Structural Brain Development. [Article]. Journal of
Neuroscience, 29(10), 3019-3025. doi: Doi 10.1523/jneurosci.5118-08.2009
Ivry, R., & Keele, S. (1989). Timing functions of the cerebellum. Journal of Cognitive
Neuroscience, 1, 136-152.
Janata, P., & Grafton, S. (2003). Swining in the brain: shared neural substrates for behaviors
related to sequencing and music. Nature Neuroscience, 6, 682-687.
Jancke, L., Schlaug, G., & Steinmetz, H. (1997). Hand skill asymmetry in professional
musicians. [Article]. Brain and Cognition, 34(3), 424-432.
39
Jentschke, S., & Koelsch, S. (2009). Musical training modulates the development of syntax
processing in children. [Article]. Neuroimage, 47(2), 735-744. doi: DOI
10.1016/j.neuroimage.2009.04.090
Jentschke, S., Koelsch, S., Sallat, S., & Friederici, A. (2008). Children with Specific Language
Impairment Also Show Impairment of Music-syntactic Processing. [Article]. Journal of
Cognitive Neuroscience, 20(11), 1940-1951.
Kelso, J. (1995). Dynamic Patterns: The Self-Organization of Brain and Behavior. Cambridge,
Massachusetts
London, England: The MIT Press.
Koelsch, S., & Friederici, A. (2003). Toward the neural basis of processing structure in music -
Comparative results of different neurophysiological investigation methods. [Proceedings
Paper]. Neurosciences and Music, 999, 15-28.
Koelsch, S., Gunter, T., Schroger, E., Tervaniemi, M., Sammler, D., & Friederici, A. (2001).
Differentiating ERAN and MMN: An ERP study. [Article]. Neuroreport, 12(7), 1385-
1389.
Koelsch, S., Jentschke, S., Sammler, D., & Mietchen, D. (2007). Untangling syntactic and
sensory processing: An ERP study of music perception. [Article]. Psychophysiology,
44(3), 476-490. doi: DOI 10.1111/j.1469-8986.2007.00517.x
Koelsch, S., Schroger, E., & Tervaniemi, M. (1999). Superior pre-attentive auditory processing
in musicians. [Article]. Neuroreport, 10(6), 1309-1313.
Koelsch, S., Schroger, E., & Tervaniemi, M. (2000). Superior pre-attentive and attentive
processing of auditory information in musicians: an MMN study. [Meeting Abstract].
Journal of Psychophysiology, 14(1), 64-65.
40
Kotter, R. (2004). Online retrieval, processing, and visualization of primate connSectivity data
from the CoCoMac database. [Proceedings Paper]. Neuroinformatics, 2(2), 127-144.
Krumhansl, C., & Castellano, M. (1983). DYNAMIC PROCESSES IN MUSIC PERCEPTION.
[Article]. Memory & Cognition, 11(4), 325-334.
Large, E., & Jones, M. (1999). The dynamics of attending: How people track time-varying
events. [Review]. Psychological Review, 106(1), 119-159.
Lee, D., Chen, Y., & Schlaug, G. (2003). Corpus callosum: musician and gender effects.
[Article]. Neuroreport, 14(2), 205-209. doi: DOI 10.1097/01.wnr.0000053761.76853.41
Levanen, S., Ahonen, A., Jari, R., McEvoy, L., & Sams, M. (1996). Deviant auditory stimuli
activate human left and right auditory cortex differently. Cerebral Cortex, 6, 288-296.
Lippé, S., Kovacevic, N., & McIntosh, A. (2009). Differential maturation of brain signal
complexity in the human auditory and visual system. [Article]. Frontiers in Human
Neuroscience, 3, -. doi: ARTN 48 DOI 10.3389/neuro.09.048.2009
Lobaugh, N., West, R., & McIntosh, A. (2001). Spatiotemporal analysis of experimental
differences in event-related potential data with partial least squares. [Article].
Psychophysiology, 38(3), 517-530.
Lopez, L., Jurgens, R., Diekmann, V., Becker, W., Ried, S., Grozinger, B., & Erne, S. (2003).
Musicians versus nonmusicians - A neurophysiological approach. [Proceedings Paper].
Neurosciences and Music, 999, 124-130.
Maess, B., Koelsch, S., Gunter, T., & Friederici, A. (2001). Musical syntax is processed in
Broca's area: an MEG study. Nature Neuroscience, 4(5), 540-545.
Manoel, E., & Connolly, K. (1995). Variability and the development of skilled actions. [Article].
International Journal of Psychophysiology, 19(2), 129-147.
41
Martinez-Montes, E., Valdes-Sosa, P., Miwakeichi, F., Goldman, R., & Cohen, M. (2004).
Concurrent EEG/fMRI analysis by multiway Partial Least Squares. [Article].
Neuroimage, 22(3), 1023-1034.
Massimini, M., Ferrarelli, F., Esser, S., Riedner, B., Huber, R., Murphy, M., . . . Tononi, G.
(2007). Triggering sleep slow waves by transcranial magnetic stimulation. [Article].
Proceedings of the National Academy of Sciences of the United States of America,
104(20), 8496-8501. doi: DOI 10.1073/pnas.0702495104
Massimini, M., Ferrarelli, F., Huber, R., Esser, S., Singh, H., & Tononi, G. (2005). Breakdown
of cortical effective connectivity during sleep. [Article]. Science, 309(5744), 2228-2232.
doi: DOI 10.1126/science.1117256
McIntosh, A. (2000). Towards a network theory of cognition. [Article]. Neural Networks, 13(8-
9), 861-870.
McIntosh, A., Bookstein, F., Haxby, J., & Grady, C. (1996). Spatial pattern analysis of functional
brain images using partial least squares. [Article]. Neuroimage, 3(3), 143-157.
McIntosh, A., Kovacevic, N., & Itier, R. (2008). Increased Brain Signal Variability Accompanies
Lower Behavioral Variability in Development. [Article]. Plos Computational Biology,
4(7), -. doi: ARTN e1000106 DOI 10.1371/journal.pcbi.1000106
McIntosh, A., Kovacevic, N., Lippé, S., Garrett, D., Grady, C., & Jirsa, V. (2010). The
development of a noisy brain. [Review]. Archives Italiennes De Biologie, 148(3), 323-
337.
McIntosh, A., & Lobaugh, N. (2004). Partial least squares analysis of neuroimaging data:
applications and advances. [Proceedings Paper]. Neuroimage, 23, S250-S263. doi: DOI
10.1016/j.neuroimage.2004.07.020
42
Menon, V., & Levitin, D. (2005). The rewards of music listening: Response and physiological
connectivity of the mesolimbic system. [Article]. Neuroimage, 28(1), 175-184. doi: DOI
10.1016/j.neuroimage.2005.05.053
Menon, V., Levitin, D., Smith, B., Lembke, A., Krasnow, B., Glazer, D., . . . McAdams, S.
(2002). Neural correlates of timbre change in harmonic sounds. [Article]. Neuroimage,
17(4), 1742-1754. doi: DOI 10.1006/nimg.2002.1295
Meyer-Lindenberg, A. (1996). The evolution of complexity in human brain development: an
EEG study. Electroencephalography and Clinical Neurophysiology, 99(405-411).
Misic, B., Mills, T., Taylor, M., & McIntosh, A. (2010). Brain Noise Is Task Dependent and
Region Specific. [Article]. Journal of Neurophysiology, 104(5), 2667-2676. doi: DOI
10.1152/jn.00648.2010
Mizuno, T., & Sugishita, M. (2007). Neural correlates underlying perception of tonality-related
emotional contents. Neuroreport, 18, 1651-1655.
Morrison, S., Demorest, S., Aylward, E., Cramer, S., & Maravilla, K. (2003). FMRI
investigation of cross-cultural music comprehension. [Article]. Neuroimage, 20(1), 378-
384. doi: Doi 10.1016/s1053-8119(03)00300-8
Munte, T., Altenmuller, E., & Jancke, L. (2002). The musician's brain as a model of
neuroplasticity. [Review]. Nature Reviews Neuroscience, 3(6), 473-478. doi: DOI
10.1038/nrn843
Munte, T., Nager, W., Beiss, T., Schroeder, C., & Altenmuller, E. (2003). Specialization of the
specialized: Electrophysiological investigations in professional musicians. [Proceedings
Paper]. Neurosciences and Music, 999, 131-139.
Näätanen, R., & Alho, K. (1995). Mismatch negativity - a unique measure of sensory processing
in audition. International Journal of Neuroscience, 80, 317-337.
43
Näätanen, R., & Picton, T. (1987). The N1 wave of the human electric and magnetic response to
sound: a review and analysis of the component structure. Psychophysiology, 38, 283-299.
Nikulin, V., & Brismar, T. (2004). Comment on "Multiscale entropy analysis of complex
physiologic time series". [Editorial Material]. Physical Review Letters, 92(8), -. doi:
ARTN 089803 DOI 10.1103/PhysRevLett.92.089803
Nyberg, L., Persson, J., Habib, R., Tulving, E., McIntosh, A., Cabeza, R., & Houle, S. (2000).
Large scale neurocognitive networks underlying episodic memory. Journal of Cognitive
Neuroscience, 12(163-173).
Pantev, C., Oostenveld, R., Engelien, A., Ross, B., Roberts, L., & Hoke, M. (1998). Increased
auditory cortical representation in musicians. [Article]. Nature, 392(6678), 811-814.
Pantev, C., Roberts, L., Schulz, M., Engelien, A., & Ross, B. (2001). Timbre-specific
enhancement of auditory cortical representations in musicians. [Article]. Neuroreport,
12(1), 169-174.
Passingham, R., Stephan, K., & Kotter, R. (2002). The anatomical basis of functional
localization in the cortex. [Review]. Nature Reviews Neuroscience, 3(8), 606-616. doi:
DOI 10.1038/nrn893
Patterson, R., Uppenkamp, S., Johnsrude, I., & Griffiths, T. (2002). The processing of temporal
pitch and melody information in auditory cortex. [Article]. Neuron, 36(4), 767-776.
Penfield, W., & Boldrey, E. (1937). Somatic motor and sensory representation in the cerebral
cortex of man as studied by electrical stimulation. [Article]. Brain, 60, 389-443.
Peretz, I. (1990). Processing of Local and Global Musical Information by Unilateral Brain-
Damaged Patients. [Article]. Brain, 113, 1185-1205.
Peretz, I., & Babai, M. (1992). The role of contour and intervals it the recognition of melody
parts: Evidence from cerebral asymmetries in musicians. Neuropsychologia, 30, 277-292.
44
Peretz, I., & Kolinsky, R. (1993). Boundaries of separability between melody and rhythm in
music discrimination - A neuropsychological perspective. [Article]. Quarterly Journal of
Experimental Psychology Section a-Human Experimental Psychology, 46(2), 301-325.
Peretz, I., & Zatorre, R. (2005). Brain organization for music processing. Annual Reviews of
Psychology, 56, 89-114.
Picton, T., Alain, C., Leun, O., Ritter, W., & Achim, A. (2000). Mismatch Negativity: Different
Water in the Same River. Audiology & Neurotology, 5, 111-139. doi: 10.1159/000013875
Protzner, A., Valiante, T., Kovacevic, N., McCormick, C., & McAndrews, M. (2010).
Hippocampal signal complexity in mesial temporal lobe epilepsy: a noisy brain is a
healthy brain. Archives Italiennes De Biologie, 148, 289-297.
Raichle, M., MacLeod, A., Snyder, A., Powers, W., Gusnard, D., & Shulman, G. (2001). A
default mode of brain function. [Article]. Proceedings of the National Academy of
Sciences of the United States of America, 98(2), 676-682.
Rankin, S., Large, E., & Fink, P. (2009). Fractal Tempo Fluctuation and Pulse Prediction.
[Article]. Music Perception, 26(5), 401-413. doi: Doi 10.1525/mp.2009.26.5.401
Richman, J., & Moorman, J. (2000). Physiological time-series analysis using approximate
entropy and sample entropy. [Article]. American Journal of Physiology-Heart and
Circulatory Physiology, 278(6), H2039-H2049.
Ritter, W., Sussman, E., Deacon, D., Cowan, N., & Vaughan, H. (1999). Two cognitive systems
simultaneously prepared for opposite events. [Article]. Psychophysiology, 36(6), 835-
838.
Rodriguez, E., George, N., Lachaux, J., Martinerie, J., Renault, B., & Varela, F. (1999).
Perception’s shadow: long-distance synchronization of human brain activity. Nature, 397,
430-433.
45
Russeler, J., Altenmuller, E., Nager, W., Kohlmetz, C., & Munte, T. (2001). Event-related brain
potentials to sound omissions differ in musicians and non-musicians. [Article].
Neuroscience Letters, 308(1), 33-36.
Salimpoor, V., Benovoy, M., Larcher, K., Dagher, A., & Zatorre, R. (2011). Anatomically
distinct dopamine release during anticipation and experience of peak emotion to music.
[Article]. Nature Neuroscience, 14(2), 257-U355. doi: DOI 10.1038/nn.2726
Sampson, P., Streissguth, A., Barr, H., & Bookstein, F. (1989). Neuro-Bahavioral Effects of
Prenatal Alcohol .2. Partial Least-Squares Analysis. [Article]. Neurotoxicology and
Teratology, 11(5), 477-491.
Samson, S. (2003). Neuropsychological studies of musical timbre. [Proceedings Paper].
Neurosciences and Music, 999, 144-151.
Schlaug, G. (2001). The Brain of Musicians. Annals of the New York Academy of Sciences, 930,
281-299.
Schlaug, G., Forgeard, M., Zhu, L., Norton, A., & Winner, E. (2009). Training-induced
Neuroplasticity in Young Children. [Proceedings Paper]. Neurosciences and Music Iii:
Disorders and Plasticity, 1169, 205-208. doi: DOI 10.1111/j.1749-6632.2009.04842.x
Schlaug, G., Jancke, L., Huang, Y., Staiger, J., & Steinmetz, H. (1995). Increased Corpus-
Callosum Size in Musicians. [Article]. Neuropsychologia, 33(8), 1047-&.
Schlaug, G., Jancke, L., Huang, Y., & Steinmetz, H. (1995). In-vivo evidence of structural brain
asymmetry in musicians. Science, 267(5198), 699-701.
Schmithorst, V., & Wilke, M. (2002). Differences in white matter architecture between
musicians and non-musicians: a diffusion tensor imaging study. [Article]. Neuroscience
Letters, 321(1-2), 57-60. doi: Pii s0304-3940(02)00054-x
46
Schneider, P., Scherg, M., Dosch, H., Specht, H., Gutschalk, A., & Rupp, A. (2002).
Morphology of Heschl's gyrus reflects enhanced activation in the auditory cortex of
musicians. [Article]. Nature Neuroscience, 5(7), 688-694. doi: DOI 10.1038/nn871
Scoville, W., & Milner, B. (1957). Loss of rencent memory after bilateral hippocampal lesions.
[Article]. Journal of Neurology Neurosurgery and Psychiatry, 20(1), 11-21.
Shannon, C., & Weaver, W. (1949). The Mathematical Theory of Information. Urbana, IL:
University of Illinois Press.
Shen, Y., Olbrich, E., Achermann, P., & Meier, P. (2003). Dimensional complexity and spectral
properties of the human sleep EEG. [Article]. Clinical Neurophysiology, 114(2), 199-
209. doi: Doi 10.1016/s1388-2457(02)00338-3
Singer, W., & Gray, C. (1995). Visual feature integration and the temporal correlation
hypothesis. Annual Reviews of Neuroscience, 18(555-586).
Snyder, J., & Large, E. (2005). Gamma-band activity reflects the metric structure of rhythmic
tone sequences. [Article]. Cognitive Brain Research, 24(1), 117-126. doi: DOI
10.1016/j.cogbrainres.2004.12.014
Sokolov, A., Pavlova, M., Lutzenberger, W., & Birbaumer, N. (2004). Reciprocal modulation of
neuromagnetic induced gamma activity by attention in the human visual and auditory
cortex. [Article]. Neuroimage, 22(2), 521-529. doi: DOI
10.1016/j.neuroimage.2004.01.01.045
Stickgold, R., Hobson, J. A., Fosse, R., & Fosse, M. (2001). Sleep, Learning, and Dreams: Off-
line Memory Reprocessing. Science, 294(5544), 1052-1057. doi:
10.1126/science.1063530
Tallon-Baudry, C., & Bertrand, O. (1999). Oscillatory gamma activity in humans and its role in
object representation. [Editorial Material]. Trends in Cognitive Sciences, 3(4), 151-162.
47
Tervaniemi, M., Ilvonen, T., Karma, K., Alho, K., & Naatanen, R. (1997). The musical brain:
Brain waves reveal the neurophysiological basis of musicality in human subjects.
[Article]. Neuroscience Letters, 226(1), 1-4.
Tervaniemi, M., Just, V., Koelsch, S., Widmann, A., & Schroger, E. (2005). Pitch discrimination
accuracy in musicians vs. nonmusicians: an event-related potential and behavioural study.
Experimental Brain Research, 161, 1-10.
Tiitinen, H., Alho, K., Huotilainen, M., Ilmoniemi, R., Simola, J., & Näätanen, R. (1993).
Tonotopic Auditory-Cortex and the Magnetoencephalographic (MEG) Equivalent of the
Mismatch Negativity. [Note]. Psychophysiology, 30(5), 537-540.
Tononi, G. (2010). Information integration: its relevance to brain function and consciousness.
[Article]. Archives Italiennes De Biologie, 148(3), 299-322.
Tononi, G., & Edelman, G. (1998). Neuroscience - Consciousness and complexity. [Review].
Science, 282(5395), 1846-1851.
Tononi, G., Edelman, G., & Sporns, O. (1998). Complexity and coherency: integrating
information in the brain. [Review]. Trends in Cognitive Sciences, 2(12), 474-484.
Tononi, G., Sporns, O., & Edelman, G. (1994). A measure for brain complexity - relating
functional segregation and integration in the nervous system. [Article]. Proceedings of
the National Academy of Sciences of the United States of America, 91(11), 5033-5037.
Trainor, L., McDonald, K., & Alain, C. (2002). Automatic and controlled processing of melodic
contour and interval information measured by electrical brain activity. [Article]. Journal
of Cognitive Neuroscience, 14(3), 430-442.
Trainor, L., Shahin, A., & Roberts, L. (2009). Understanding the Benefits of Musical Training
Effects on Oscillatory Brain Activity. [Proceedings Paper]. Neurosciences and Music Iii:
Disorders and Plasticity, 1169, 133-142. doi: DOI 10.1111/j.1749-6632.2009.04589.x
48
Vakorin, V., Lippé, S., & McIntosh, A. (2011). Variability of Brain Signals Processed Locally
Transforms into Higher Connectivity with Brain Development. [Article]. Journal of
Neuroscience, 31(17), 6405-6413. doi: Doi 10.1523/jneurosci.3153-10.2011
Vakorin, V., Ross, B., Krakovska, O., Bardouille, T., Cheyne, D., & McIntosh, A. (2010).
Complexity analysis of source activity underlying the neuromagnetic somatosensory
steady-state response. [Article]. Neuroimage, 51(1), 83-90. doi: DOI
10.1016/j.neuroimage.2010.01.100
Varela, F., Lachaux, J., Rodriguez, E., & Martinerie, J. (2001). The brainweb: Phase
synchronization and large-scale integration. [Review]. Nature Reviews Neuroscience,
2(4), 229-239.
Warren, J., Uppenkamp, S., Patterson, R., & Griffiths, T. (2003). Separating pitch chroma and
pitch height in the human brain. [Article]. Proceedings of the National Academy of
Sciences of the United States of America, 100(17), 10038-10042. doi: DOI
10.1073/pnas.1730682100
Watts, D., & Strogatz, S. (1998). Collective dynamics of 'small-world' networks. Nature, 393,
440-442.
Zanto, T., Large, E., Fuchs, A., & Kelso, J. (2005). Gamma-band responses to perturbed auditory
sequences: Evidence for synchronization of perceptual processes. [Proceedings Paper].
Music Perception, 22(3), 531-547.
Zanto, T., Snyder, J., & Large, E. (2006). Neural correlates of rhythm expectancy. Advances in
Cognitive Psychology, 2(2), 221-231.
Zatorre, R. (1988). Pitch perception of complex tones and human temporal-lobe function.
Journal of the Acoustical Society of America, 84, 566-572.
49
Zatorre, R., Evans, A., & Meyer, E. (1994). Neural Mechanisms Underlying Melodic Perception
and Memory for Pitch. [Article]. Journal of Neuroscience, 14(4), 1908-1919.
Zatorre, R., & Samson, S. (1991). Role of the Right Temporal Neocortex in Retention of Pitch in
Auditory Short-Term Memory. [Article]. Brain, 114, 2403-2417.