Learning to play a musical instrument is a complex task
that integrates multiple sensory modalities and higher-order cognitive
functions. Therefore, musical training is considered a useful framework
for the research on training-induced neuroplasticity. However, the
classical nature-or-nurture question remains, whether the differences
observed between musicians and non-musicians are due to predispositions
or result from the training itself. Here we present a review of recent
publications with strong focus on experimental designs to better
understand both brain reorganization and the neuronal markers of
predispositions when learning to play a musical instrument.
Cross-sectional studies identified structural and functional differences
between the brains of musicians and non-musicians, especially in
regions related to motor control and auditory processing. A few
longitudinal studies showed functional changes related to training while
listening to and producing music, in the motor network and its
connectivity with the auditory system, in line with the outcomes of
cross-sectional studies. Parallel changes within the motor system and
between the motor and auditory systems were revealed for structural
connectivity. In addition, potential predictors of musical learning
success were found including increased brain activation in the auditory
and motor systems during listening, the microstructure of the arcuate
fasciculus, and the functional connectivity between the auditory and the
motor systems. We show that “the musical brain” is a product of both
the natural human neurodiversity and the training practice.
Introduction: What is Neuroplasticity? Why is it so Important to Study it?
The constantly changing environment, the drive for new
knowledge and skills, all require behavioral flexibility. The brain, as
the source of behavior, adapts its architecture and functions to perform
new tasks through processes broadly defined as neuroplasticity. These
processes include, among others, dynamic reconfiguration of neural
connections, cell shape, size, myelination, synaptic strength and
neurogenesis, the last one limited to the olfactory bulb and the
hippocampus in adults (Tardif et al., 2016).
In human neuroimaging studies, it is possible to indirectly measure
macroscopic effects of the neuroplastic biological dynamics via
functional and structural modalities (for the overview of the
relationship between macroscopic measures and the underlying biology,
see Tardif et al., 2016).
Although usually measured separately, functional and structural
neuroplasticity reflect various aspects of the same neuroplastic
processes and are thus inherently intertwined in a complex manner.
We currently understand that the human brain is not
shaped exclusively during critical periods of development. Neuroplastic
changes occur in response to internal and external stimuli throughout
the entire lifetime (Draganski and May, 2008).
From a social perspective, neuroplasticity processes underlie such
phenomena as education, neurological rehabilitation, or healthy aging.
Musical Training as a Framework for Studying Brain Plasticity
Generally, in studies on neuroplasticity, two questions
arise: what are the structural and functional changes related to a
particular behavioral need, and how do they occur over time. To
effectively answer these questions, we first need to elicit a novel
behavior. There is a wide spectrum of learning protocols which were
employed so far to understand neuroplasticity. Simple ones engage only a
single sensory modality, like auditory (Zatorre et al., 2012) or tactile (Hodzic, 2004).
More complex ones utilize sensorimotor associations and higher-order
cognitive functions tasks, like the acquisition of foreign languages or
tactile reading (Li et al., 2014; Siuda-Krzywicka et al., 2016). The complexity of music performance requires a unique and multi-system involvement from the human brain (Münte et al., 2002; Herholz and Zatorre, 2012; Schlaug, 2015).
Playing a musical instrument requires sensorimotor adaptations, as with
the use of any tool, and more: a mapping of specific movements to the
auditorily perceived outcomes, which follow a set of more or less
intuitively understood rules of musical harmony, esthetics and pleasure.
It comprises both feed-forward and feedback interactions between the
integrated multisensory input (tactile, proprioceptive, auditory, and
visual) with motor output, as well as higher-order cognitive functions
such as memory, attention, emotion, and the processing of musical syntax
(Zatorre et al., 2007; Brown et al., 2015). Additionally, as rewarding stimuli are learned better than non-rewarding ones (Schultz, 2000), it is likely that the highly rewarding nature of musical performance promotes learning and drives brain plasticity (Penhune, 2019). Therefore, learning to play a musical instrument provides a useful framework to study multimodal brain plasticity.
Secondly, the changes in brain structure and function
have to be sampled frequently enough to capture the dynamics of the
neuroplastic processes. Brain volume changes do not relate to practice
in a monotonically increasing way (Lövdén et al., 2013; Wenger et al., 2017).
Yet, we observe continuous behavioral improvement and the extent of
behavioral and plastic changes correlate with training duration. The
proposed model of neuroplasticity includes a period of initial growth,
after which comes a renormalization phase, when the efficiency of brain
circuits increases while cortical volume does not (Wenger et al., 2017).
From a functional perspective, plastic changes can be reflected in
increased functional activation of a brain area related to a function,
its expansion on neighboring areas, or an involvement of novel, often
distant, areas. Interestingly, cortical map plasticity may also follow a
comparable pattern of expansion followed by retraction to pre-training
levels during learning as seen in structural changes (for review see Wenger et al., 2017).
Therefore, the functional (and structural) expansion temporarily
increases the available pool of circuits to be used “exploratively”
until the most efficient circuit to perform the task is determined. As
learning continues, the selected circuitry is further stabilized through
practice, the performance increasingly relies on that circuit and thus
the cortical map renormalizes (Wenger et al., 2017).
Two experimental approaches are typically employed in
cognitive neuroscience to understand brain reorganization following
training, namely the cross-sectional and the longitudinal design.
Comparing musically naive and proficient individuals in cross-sectional
studies can provide important insights into the neuroplasticity of the
human brain (Münte et al., 2002).
Musicians practice musical performance regularly for most of their
lives, often starting in early childhood and practising for many years.
Juxtaposing musicians and non-musicians can show changes associated with
very long training. However, while it might be tempting, the causal
relationship between musical training and the observed differences
cannot be inferred from correlational studies (Schellenberg, 2019).
The cross-sectional study design does not reveal the time course of the
plastic changes nor does it correct for any possible predispositions.
To infer causality, a theoretical model needs to be constructed and
validated against properly designed longitudinal studies. Longitudinal
studies can account for the interindividual variability pre-training,
but are costly, with costs increasing with the duration of the
experiments.
Finally, advances in non-invasive neuroimaging methods
gave scientists specific tools to non-invasively study brain plasticity
in living humans. Structural and functional neuroimaging techniques were
used to compare brain anatomy and function between groups of musicians
and non-musicians, and, more recently, to study the plastic changes
related to musical training in longitudinal studies.
This review aims to present the newest evidence for
experience-related neuroplasticity in the context of musical training in
adults, concentrating on neuroimaging and with an emphasis on
longitudinal studies. Since the scope of this review is limited, and the
focus is on musical training as a model for studying brain plasticity
in neurotypical adults, studies of complex developmental and
aging-related changes are not discussed. We particularly focus on
experimental designs in order to better understand both brain
reorganization and the neuronal markers of predispositions when learning
to play a musical instrument. Since we include studies which use a
multitude of functional as well as structural neuroimaging techniques,
we also provided a brief overview of such methods highlighting the
advantages and disadvantages of each method for neuroplasticity research
(Table 1).
More at link.
Alicja M. Olszewska
Maciej Gaca
Aleksandra M. Herman
Katarzyna Jednoróg2 and 
Artur Marchewka
