Abstract
In this scoping review, we investigate music skill acquisition from an embodiment perspective. We conceptualize the acquisition of musical expertise, not only as a technical process but as a gradual incorporation process, in which the instrument becomes experienced as an extension of the body, and the body itself becomes the vehicle of musical meaning and expression. To structure this process of embodiment, we draw on theoretical perspectives from neuroscience, philosophy, and music cognition that articulate how neurophysiological internal dynamics and lived experience emerge and interact. Building on this, we introduce the concept of embodiment indicators, defined as psychophysiological and biomechanical patterns that reflect how the instrument impacts the body throughout learning. Thus, to identify these indicators, we conducted a scoping review of empirical studies that use psychophysiological and biomechanical methods to investigate instrumental music learning. We followed the PRISMA-ScR guidelines across three databases, including PubMed, Web of Science, and Scopus. A total of 37 studies met the eligibility criteria. Analysis of these studies revealed three main categories of skill indicators: (1) neural markers of sensorimotor integration functions; (2) neuromuscular and movement markers of efficiency; and (3) markers of movement control. We examine and map these indicators onto three levels of embodiment—the morphological, functional, and phenomenological—to illustrate how learning to play an instrument transforms the body structures and functional capacities in support of musical expertise. Specifically, we examine how morphological changes (e.g., neural plasticity, musculoskeletal adaptations) and functional adaptations (e.g., internal models, cognitive flexibility) underpin the transition from effortful control to fluid, expressive performance. Finally, we discuss the limitations of the included articles, such as small sample sizes, limited use of natural music tasks, and methodological heterogeneity, and we propose directions for future research.
Keywords
Introduction
Imagine a violinist in the early stages of learning. At first, much of the coordination between the different body parts must be consciously controlled to perform the necessary movements to play the violin and convey musical understanding and expression. Even playing a single note requires precise finger placement, wrist position, and arm movement, all while maintaining proper posture (Konczak et al., 2009). As learning progresses, these movements and posture become increasingly integrated in the bodily coordination system. The gestures required for playing the instrument become smoother, breathing naturally aligns with musical phrasing, and the motor engagement in performance subtly and dynamically adapts to the musical demands, all with reduced conscious decision-making (Lotze et al., 2003; Shaffer, 1980). We can, for example, observe this in experienced violinists who instinctively adjust bow pressure and speed to create dynamic contrast and convey an emotion to their audience (Shaffer, 1980). This transformation, from effortful control to fluid and expressive execution, is a hallmark of music skill acquisition and expertise (Fitts, 1964; Ward et al., 2019).
The above-described adaptability and flexibility of the skilled musician are crucial for expressive performance (Reybrouck & Schiavio, 2024). Expressivity in music arises when the performer does not simply execute learned movements but dynamically responds to musical and emotional nuances in real time (Rojas, 2015). Such perceptual openness and performative flexibility, in contrast to an exact mechanical reproduction of a prepared plan, adds to the living quality of a performance and turns it into a captivating experience for the listener (Ericsson & Lehmann, 1996). This capacity for expressive performance is the result of a long and intensive skill acquisition process that requires effort and perseverance to acquire the necessary instrumental skills. In this learning process, the development of an optimal relationship between musician and musical instrument plays a pivotal role (Nijs et al., 2013). Such an optimal relationship has been described as the merging of the musician and instrument, through which the instrument is no longer experienced as an external, detached object to be controlled, but rather as if it was an organic component of the musician's body (Nosulenko et al., 2005). In other words, the instrument becomes incorporated as a natural extension of the musician.
As this incorporation unfolds, performers are able to transition from focusing on the technicalities of handling the instrument to immersing themselves in and engaging deeply with the music itself (Kim, 2020; Leman, 2007; Nijs, 2017; Rojas, 2015). In this process, instrument-related postures and gestures become integral to the functioning of the performer's body, and the knowledge of how to use the body and the instrument during performance becomes intuitive (Nijs, 2017; Rojas, 2015). As a result, the musician's body becomes the living vehicle of musical meaning, in which musical understanding and expression emerges not from intellectual decisions alone but from embodied intuitions cultivated through years of physical practice (Bremmer & Nijs, 2024). In this sense, mastering an instrument means developing a specialized form of bodily intelligence, where musical concepts, physical technique, and emotional expression become integrated in the performer's lived experience of music (Nijs, 2017).
This idea of incorporation has been addressed by different scholars in disciplines such as neuroscience (e.g., a stick in Maravita & Iriki, 2004), sport science (e.g., a rope in Thorndahl & Ravn, 2016), and philosophy of mind (De Preester & Tsakiris, 2009). Even though the musician–instrument relationship can be seen as an example par excellence (Thorndahl & Ravn, 2016), in music research this idea is often invoked but rarely elaborated in depth (e.g., Alerby & Ferm, 2005, p. 181; Bahn et al., 2001, p. 2; Bresler, 2005, p. 176; Leman, 2007, p. 161; Schiavio et al., 2019, p. 6). Here, we treat the experience of incorporation not as a disembodied mental construct or a romanticized narrative of virtuosity, but as the experiential outcome of a multilevel embodied process unfolding across morphological, functional, and phenomenological levels, in which the sense of ownership and agency in relation to the instrument emerge from learning-related changes in bodily morphology, sensorimotor integration, and predictive control mechanisms (Friston, 2012; Tsakiris et al., 2007).
Theoretical Framework on Embodiment in Music Skill Acquisition
Although the experiential dimension of embodiment in musical performance is well recognized, its definition remains challenging, partly due to the concept's multidisciplinary nature (Raoul & Grosbras, 2023). To bring some conceptual clarity and to situate the process of incorporation within a broader framework of embodiment, we begin with a working definition: Embodiment refers to the process by which external entities, like the musical instrument, are integrated into the body representation (Forster et al., 2022). However, because such a definition alone does not specify how bodily changes, functional processes, and subjective experience relate to one another, we adopt Nijs’ (2017) three-level account of embodiment, which adapts Metzinger’s (2014) framework to music performance.
Metzinger proposes that the experience of being a self is grounded within an embodied system operating across three interrelated levels: the morphological (first order), the functional (second order), and the phenomenological (third order). Nijs extends this model to describe how musical instruments are incorporated at each of these three levels, leading to the experience of having incorporated the instrument as part of the self. This framework provides an integrative perspective that connects physical, cognitive, and experiential dimensions of embodiment in music skill acquisition and specifically addresses the musician–instrument relationship in this process.
Understanding the mechanisms by which such incorporation occurs benefits from additional insights from philosophy and neuroscience. Particularly relevant are contemporary perspectives that articulate how neurophysiological dynamics and lived experience emerge and interact (Blanke, 2012; Candia-Rivera et al., 2024). These include theories on presence and bodily self-consciousness (Forster et al., 2022; Kilteni et al., 2015; Raoul & Grosbras, 2023), and neuroimaging studies on body ownership (Tsakiris et al., 2007, 2010) and the embodiment of external objects (Segil et al., 2022). Across these approaches, multisensory integration emerges as the central mechanism of embodiment and of the experiential properties arising from it (e.g., body ownership, agency, and presence) (Candia-Rivera et al., 2024; de Vignemont, 2011). These frameworks provide useful parallels for understanding how a musical instrument can become incorporated into the body and experienced as part of the self.
Levels of Embodiment in Music Skill Acquisition
First-Order Embodiment: Morphological Adaptations
Metzinger (2014) proposes that the experience of the self is embodied, meaning that action and cognition are grounded in the body's physical and neural architecture (Quadt, 2018). At the most basic level, embodiment concerns how the body's morphological and mechanical characteristics shape our interaction with the world (Haans & IJsselsteijn, 2012). In the context of learning to play a musical instrument, the physical manipulation of the instrument directly affects the musician's body morphology, impacting the neural and musculoskeletal physiology, postures, and movements (Tubiana & Chamagne, 2005). In this sense, embodiment at this level refers to how the body physically adjusts to the demands of musical performance, shaping how the musician interacts with the music itself (Gallagher, 2005; Nijs, 2017).
A key mechanism at this level is the development of sensorimotor couplings, which are bidirectional neural associations between motor commands and sensory expectations (Leman, 2007; Novembre & Keller, 2014; Segil et al., 2022; Tsakiris et al., 2010). In musicians, these couplings are particularly evident in the auditory–motor domain, as they continuously integrate the auditory perception of a sound with the motor actions required to produce that sound (Bangert & Altenmüller, 2003; Novembre & Keller, 2014). Empirical evidence supports this, showing that auditory and motor regions within the brain of musicians co-activate not only during actual performance but also during passive listening and silent motor tasks, indicating that sound and action become integrated through training (Bangert et al., 2006; De Manzano et al., 2020; D’Ausilio et al., 2006).
The development of such couplings relies on processes of multisensory integration (Segil et al., 2022). Research on body and limb ownership shows that the integration of sensory signals across modalities depends on temporal and spatial synchrony (Riemer et al., 2019). This means that auditory, somatosensory, and proprioceptive inputs need to co-occur within a limited temporal window to be perceived as arising from a single coherent source (Segil et al., 2022; Tsakiris et al., 2010). This synchrony is necessary for multisensory integration and plays a central role in generating the sense of ownership (Segil et al., 2022). A similar mechanism may occur when a musician develops sensorimotor couplings in relation to the instrument. Through repeated practice, temporally and spatially aligned sensorimotor and multisensory signals become associated and increasingly stable, leading to long-term structural and functional changes in the body, such as cortical reorganization and strengthened functional connectivity (Altenmüller et al., 2019; Dalla Bella, 2016; Mathias et al., 2015; Rosenkranz et al., 2007).
At this level, embodiment can be understood as the progressive embedding of instrument-related demands into the body's sensorimotor architecture. These adaptations do not yet concern conscious experience but provide the necessary physiological substrate upon which the later conscious experience of incorporation can emerge (Dosenbach et al., 2025; Martel et al., 2016; Schilling & Cruse, 2008; Segil et al., 2022).
Second-Order Embodiment: Functional Integration
Second-order embodiment concerns the development of predictive internal models that emerge from the continuous reinforcement of sensorimotor associations (Maes et al., 2014; Quadt, 2018). These internal models can be understood as dynamic neural representations or small-scale models of how the body works in its interaction with its surroundings (Gallagher, 2005; Pezzulo et al., 2024). They include inverse and forward models, which are neural mechanisms that allow individuals to make inferences about the causes of sensory outcomes, as well as predict the sensory outcomes of their actions, before even performing the actions themselves (Maes et al., 2014; Pezzulo et al., 2022, 2024).
Empirical evidence supports the existence of internal models in musicians, as they show that motor actions become automatically activated, at the behavioral and neural levels, as a result of perceiving the auditory consequences associated with that musical action (Kajihara et al., 2013; Maes et al., 2014). These studies show that musicians exhibit differences in gray and white matter architecture and cortical thickness, particularly in regions of the dorsal auditory pathway, which support auditory–motor integration (D’Ausilio et al., 2006; Kajihara et al., 2013; Maes et al., 2014). Functionally, these differences translate into an increased capacity to process auditory and somatosensory information, as well as improvements in higher-order cognitive functions such as attention regulation, motor planning, and executive control (Penhune, 2019; Zatorre et al., 2007). For instance, musicians tend to show greater sensitivity to subtle pitch and rhythm variations (Maidhof et al., 2009, 2010, 2013; Münte et al., 2002), enhanced kinesthetic and proprioceptive awareness (Hirano et al., 2020), improved attentional control (Altenmüller et al., 2019; Brown et al., 2015), and more efficient, automated, and fine control of subtle body movements (Hirano & Furuya, 2022), skills that are essential for expressive and skilled music performance.
Internal models are not limited to action–effect mappings but also include representations of the body's physical properties (e.g., its shape, size, weight, strength, and speed) and how it moves within the world (Longo, 2015; Segil et al., 2022). For the instrument to be incorporated within these models, it needs to achieve a similar sensorimotor representation to the body itself, such that tool-mediated actions are governed by the same predictive control mechanisms that regulate bodily action (D’Angelo et al., 2018; Martel et al., 2016). From this perspective, second-order embodiment can be understood as the stage at which body- and instrument-related sensorimotor associations give rise to a single predictive system guiding musical action (Maes et al., 2014; Nijs, 2017). As these internal models are refined with experience, they are hypothesized to support increasingly fluent, precise, and efficient performance (Altenmüller et al., 2019; Maes et al., 2014). In turn, this level of embodiment is also thought to provide the basis for a sense of control and agency to emerge, extending to the instrument, as it becomes functionally integrated into the performer's action system (Forster et al., 2022; Nijs, 2017; Quadt, 2018).
Third-Order Embodiment: Phenomenological Incorporation
Third-order embodiment refers to the subjective experience of incorporation, understood as the experiential manifestation of lower-level processes developed through skill acquisition. Within Metzinger's framework, phenomenological experience is grounded in representational processes (second-order embodiment), which themselves depend on underlying morphological bodily structures (first-order embodiment) (Quadt, 2018). Accordingly, third-order embodiment reflects how learning-related bodily and functional adaptations give rise to subjective experience (Tsakiris et al., 2007).
Neuroimaging studies of body ownership and external object embodiment support this idea. For instance, O’Kane et al. (2024) showed that body-part ownership arises from the perceptual binding of visual and somatosensory signals from specific body parts. Using a Bayesian causal inference framework, the authors propose that the brain infers ownership based on the probability that multisensory signals arise from a single coherent source (O’Kane et al., 2024). Similarly, studies on prosthetic limb embodiment (de Vignemont, 2011; Forster et al., 2022; Segil et al., 2022) argue that agency emerges in action when the sensory feedback of a movement is compared with the predicted sensory outcome, such that no prediction error leads to a sense of being in control (Haggard, 2017; Uhlmann et al., 2020). These studies emphasize that subconscious multisensory correlations provide the substrate for conscious experiences of embodiment, including a sense of ownership and agency (Abdulkarim et al., 2023; Blanke, 2012).
Building on these perspectives, the phenomenological experience of merging with the instrument can be understood as the experiential consequence of morphological and functional integration of the instrument. As sensorimotor couplings are stabilized and predictive internal models are refined through practice, the instrument is no longer experienced as an external tool but as part of the performer's bodily self or body representation (Martel et al., 2016). This experiential shift supports a sense of ownership, agency, and unity in relation to musical action, enabling expressive decisions to emerge from an embodied interaction with the music (Leman, 2007; Nijs, 2017).
To conclude, Metzinger's framework on embodiment levels offers a valuable lens for understanding the mechanisms through which musicians come to experience incorporation of the instrument (Leman, 2007; Segil et al., 2022). Nijs (2017) argues that when embodiment happens at this level, the boundary between body and instrument dissolves, and the instrument contributes to the musician's artistic identity, enabling expressive decisions to emerge from an embodied interaction with the music itself (Leman, 2007; Nijs, 2017). In this case, this sense of incorporation is accompanied by feelings of ownership—that is, that the instrument belongs to the body—and agency—that is, a sense of control over one’s actions, in relation to the instrument (Candia-Rivera et al., 2024; Kilteni et al., 2012; Owens & Duncan, 2025; Raoul & Grosbras, 2023; Schilling & Cruse, 2008; Tsakiris et al., 2010).
Rationale and Objectives
Learning to play an instrument is a complex, dynamic process that directly impacts the body (Leman, 2007). Through practice and experience, individuals develop and refine psychophysiological and biomechanical patterns that support skilled performance (Button et al., 2021). Over time, these adaptations can lead to the experience of incorporation of the instrument into the body, an experience that has a multisensory and neurophysiological basis (Segil et al., 2022).
In this context, such adaptive patterns can be captured through what we term embodiment indicators, defined as psychophysiological and biomechanical markers that reflect how the instrument impacts the body throughout the learning process and reveal how learning-related bodily and neural changes progressively reflect the incorporation of the instrument into the performer's body representation (Martel et al., 2016). Psychophysiological measures capture the internal dynamics of the brain and body, while biomechanics reveal how these changes are expressed in movement and goal-directed action. Integrating both perspectives allows us to link the phenomenological aspects of musical expertise with their underlying psychophysiological and biomechanical mechanisms, providing a multidimensional picture of embodiment across different stages of skill acquisition.
While there is extensive research on the neural and physiological basis of music perception and performance (Altenmüller et al., 2019; Brown et al., 2015; Hirano et al., 2020; Zatorre et al., 2007), as well as studies on musicians’ movements and their expressive intentions (Chi et al., 2020; López-Pineda et al., 2023; Moura et al., 2024, 2025; Nusseck et al., 2024; Vidal et al., 2024; Volta & Di Stefano, 2024), limited work has investigated how these indicators evolve with skill acquisition. To our knowledge, no studies have focused on identifying indicators across the three levels of embodiment or across multiple stages of expertise. A better understanding of these processes is essential for informing and designing effective pedagogical approaches that optimize learning trajectories, prevent maladaptive movement patterns and injuries, and facilitate the development of expressive, technically proficient, and musically meaningful performance across diverse musical contexts and developmental stages.
To begin addressing this gap, we conducted a scoping review to identify psychophysiological and biomechanical indicators associated with different skill levels in instrumental music learning. We focus primarily on first- and second-order embodiment indicators, which can be measured quantitatively. Third-order embodiment involves subjective experience and would require qualitative approaches, which would render the review too broad. Therefore, it is not directly measured in the present review but is discussed as an emergent outcome of lower-level changes. By focusing on quantifiable indicators, our aim is to identify concrete markers of bodily adaptation that capture the development of musical expertise.
A scoping review is well suited to this aim, given the methodological diversity and emerging nature of embodiment research in music performance (Arksey & O’Malley, 2005). The heterogeneity of empirical approaches and the lack of prior reviews on embodiment indicators call for a comprehensive mapping of how embodiment is conceptualized and measured in this field. Accordingly, this review addresses the following research questions.
What psychophysiological and biomechanical indicators are used to assess bodily adaptations in instrumental musicians during the process of music skill acquisition? How are these indicators identified (e.g., through which methods and measures)? Which indicators define different skill levels of musical expertise?
Methods
This review was conducted in November 2024, following the PRISMA-ScR guidelines (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension for Scoping Reviews; Page et al., 2021).
Eligibility Criteria: Inclusion and Exclusion Criteria
The inclusion and exclusion criteria were established based on the PICOS (Population, Intervention, Comparison, Outcomes, Study Design) framework (see Table 1; Amir-Behghadami & Janati, 2020), to guide the identification and selection of eligible studies.
Eligibility criteria for study selection based on the PICOS framework.
Note. The table outlines the inclusion and exclusion criteria for eligible studies, in this case, investigating instrumental musicians’ learning processes and skill-level comparisons using psychophysiological and biomechanical methods during music performance.
Information Sources and Search Strategy
A comprehensive literature search was conducted across three databases: PubMed (1967–present), Web of Science (1977–present), and Scopus (1976–present). The search strategy was not restricted by publication year. However, it was limited to empirical, peer-reviewed articles. Gray literature, including comments, editorials, datasets, retracted articles, notes, and newspaper articles, were excluded. In addition, nonempirical research such as reviews, meta-analyses, systematic reviews, conference papers, books and book chapters, and theoretical articles were not considered.
The search string was defined based on the methods used (e.g., biomechanical and psychophysiological) to assess our population of interest (e.g., performing instrumental musicians) while learning or acquiring music skills. The full search string can be found in Section 1.1 of the Supplementary Material.
Study Selection Process
The initial literature search was done by the first author and provided 3,193 records. After removing duplicates and retracted articles, 1,401 records remained for screening. The author did an initial title screening, excluding 1,104 records based on the inclusion and exclusion criteria, leaving 297 reports for title and abstract reading. Subsequently, two authors independently screened these reports, excluding an additional 225. This resulted in 72 reports being assessed for full-text eligibility. The same two authors conducted the full-text screening independently. When issues arose, a third reviewer was consulted to resolve any disagreements, which led to the exclusion of 35 more reports. Ultimately, a total of 37 studies were included in the review. A flow diagram of the process is shown in Figure 1, and a detailed description of the reasons for exclusion at each step are provided in Section 1.2 of the Supplementary Material.
PRISMA flow diagram.
The reference manager Zotero was used to identify and remove duplicates, and the software Rayyan.ai was used to screen the articles independently.
Data Charting Process and Data Items
The data extraction was led by the first author who manually extracted key information into an Excel table with the following categories. (1) Study metadata: study, title, publication year, authors, journal, country, academic affiliation, and funding information. (2) Participant characteristics: sample size, skill level, gender, mean age, age range, mean years of experience, and musical instrument. (3) Methods and methodology: methods characteristics, study design, experimental protocol, task performed, independent and dependent variables measured and data analysis. (4) Main results: findings, conclusions, and skill indicators. The extraction process was done in consultation with, and supervised by a third reviewer, who cross-checked that the information extracted from the studies was accurate. The tables are available in Section 2 of the Supplementary Material.
Synthesis of Results
To synthesize and analyze the extracted data, we constructed a coding matrix in which the presence or absence of specific features was recorded for each study. All variables were coded using a binary scheme (1 = reported/present, 0 = not reported/absent). This coding was applied to (2) participant characteristics (skill level and instrument) and (3) methodological characteristics (measurement tools, study design, and task performed). This was done to be able to perform statistical descriptive analysis. To identify the skill indicators, we summarized the results of each study and extracted the skill indicators that the authors identified as differentiating musicians at different levels. These were then categorized by (1) methodological approach, classified as either psychophysiological or biomechanical; and by (2) functional category, classified as either neural, muscular, or movement related. This process was led by the first author in consultation with the third reviewer involved in the data charting process. The synthesis is presented in the following section, supported by summary tables (Tables 2–4) and figures (Figures 2–5) to facilitate interpretation. The figures were generated using R Studio (v2024.12.1 + 563).
Overview of the included studies.
Note. The table presents an overview of the 37 included empirical studies that investigate music performance and skill acquisition using psychophysiological and biomechanical methods. It includes study title, publication year, authors, journal, and primary academic affiliation.
Sample characteristics by skill level.
Note. This table summarizes the sample characteristics across studies, grouped by music skill level. For each level, the table reports on the mean sample size, sample size range, mean age, age range, mean years of experience, and experience range per study group, calculated across all studies that include participants at that level. For instance, across all studies that included a group of beginners, the average sample size was 9.4 participants, with the average number of participants ranging from 1 to 16 individuals. These participants had a mean age of 23.7 years and an average of 2.2 years of musical experience. The “All studies” row shows the overall means and ranges across skill levels, providing a general profile of all the studies’ samples.
Skill-level groups included and compared across studies.
Number of studies per measurement tool used and their combination.
Psychophysiological skill indicators grouped by indicator category.
Results
A full description of the measurement tools, study design (e.g., cross-sectional or longitudinal designs), experimental protocol (e.g., including the task performed), independent and dependent variables measured, data analysis, results, and conclusions is provided in Sections 2.3–2.6 of the Supplementary Material. In the present section, we offer a concise summary of key findings most relevant to the following discussion.
Study Metadata
The first study included in this review dates to 1998. Since then, 37 studies have examined music learning and the acquisition of instrumental music skills using psychophysiological and biomechanical methods. See Table 2 for an overview of the study metadata of the included studies.
Sample Characteristics
Table 3 presents a summary of the descriptive statistics of the samples across all studies, categorized by skill level. The number of participants per study ranged from 6 to 68 (M = 20.5, SD = 10.5) participants. Participant ages ranged from 6.8 to 68.9 years (M = 28.5, SD = 8.8) years across studies. Across all participants, 37.65% of participants were female and 51.52% were male; the remainder were not reported. The full description of the participant characteristics per study is included in Section 2.2 of the Supplementary Material.
To enable synthesis across studies using diverse skill labels (e.g., novice, world-class, inexperienced), we standardized participant groups on a common set of expertise levels. Where possible, this remapping was guided by reported years of instrumental training, resulting in the following categories:
nonmusicians (<1 year) beginners (1–3 years) intermediate (3–10 years) proficient (11–14 years) experts (>14 years).
The threshold for expert musicianship was chosen to reflect typical conservatoire-based training trajectories, in which professional-level performance generally follows approximately 10 years of pretertiary instruction and a 4-year undergraduate degree in music performance. Accordingly, a minimum of 14 years of training was used as a criterion for classifying expert musicians. On this basis, individuals in undergraduate-level training (11–14 years of experience) were defined as proficient musicians, whereas those in pretertiary training (3–10 years of experience) were classified as intermediate. Finally, those with fewer than 3 years of experience were classified as beginners, and those with less than one year were labelled nonmusicians. When studies used labels that could not be reliably converted into years of training (e.g., “amateur” defined by nonprofessional status, irregular training histories, or recruitment from nonconservatoire settings), we retained the study's label. For transparency, the original study terminology and the resulting category are both reported in Supplementary Table 2.
As shown in Table 3 and Figure 2, participant skill levels varied across studies. Of the 37 studies, 30 (81%) included groups of musicians at different levels, and the rest mainly looked at nonmusicians (n = 5, 13.5%) or beginners (n = 2, 5.4%) as they learned to play an instrument. Among the 30 studies that included musicians at different levels, most of them investigated two skill levels, generally comparing nonmusicians or beginners with proficient or expert musicians (43.2%, n = 16). Comparisons involving intermediate levels (13.5%, n = 5) or more than two skill level groups (16.2%, n = 6) were less common.
Instruments and Music Tasks
The studies also varied in the musical instruments investigated. Most studies focused on piano performance (52.6%, n = 20), with the rest distributed across drums (18.4%, n = 7), violin (10.5%, n = 4), cello (13.2%, n = 5), and guitar (5.3%, n = 2). Note that one study looked at both drums and cello players (see Gonzalez-Sanchez et al., 2019). Regarding the music tasks, participants were most often asked to perform single strokes, either on the piano or the drums, and sustained notes or single note bowing exercises (54%, n = 20). Other tasks included five-note melodies (22%, n = 8) or simple songs (e.g., Twinkle, Twinkle) (14%, n = 5), classical repertoire pieces (5%, n = 2), and improvisation (5%, n = 2). For a detailed description of the tasks for each study see Section 2.3 of the Supplementary Material.
Psychophysiological and Biomechanical Measurement Tools
Across the 37 studies, 31 (83.78%) used psychophysiological measurement tools and methods such as functional magnetic resonance imaging (fMRI) (n = 6), electroencephalography (EEG) (n = 8), transcranial magnetic stimulation (TMS) (n = 2), and electromyography (EMG) (n = 15). In turn, 25 (67.57%) employed biomechanical measurement tools and methods such as 2-dimensional position sensors (n = 6), motion capture (MoCap) (n = 10), force transducers (n = 7), F-scan sensors (n = 1), and a cyberglove (n = 1). In total, 17 (45.94%) studies used two or more measurement tools (see Figure 3). For a detailed description of the measurement tools used for each study see Section 2.4 of the Supplementary Material.
Skill Indicators: Neural, Muscular, and Movement Indicators
To extract the indicators that characterize skill level, we categorized the findings from each study by (1) methodological approach, classified as either psychophysiological or biomechanical; and by (2) functional category, classified as either neural, muscular, or movement related.
Studies on Psychophysiology: Neural and Muscular Indicators
Psychophysiology studies generally examined neural and muscular adaptations in music learning. These studies focused on how, as a result of training, sensorimotor integration functions develop, supporting neural and muscular efficiency and, in turn, enhanced musical performance. Figure 4 shows a summary of the identified skill indicators and the measurement tools and methods with which they were identified.
Indicator Category 1: Sensorimotor Integration Functions
A recurring theme across studies was that musical training led to the development of sensorimotor associations, which referred to the coupling between sensory (e.g., auditory or visual) and motor systems (Bangert & Altenmüller, 2003; Lotze et al., 2003; Stewart et al., 2003). Several studies described that, with increasing expertise, these couplings became sufficiently robust that activation of one modality, such as listening to a melody or viewing the score, could trigger the entire sensorimotor network, even in the absence of movement (Lotze et al., 2003).
Skill Indicator 1.1: Auditory–motor and visuo–motor Couplings
Studies using direct current electroencephalography (DC-EEG) and functional magnetic resonance imaging (fMRI) showed that even brief periods of training (e.g., 20 min) on a simple piano key-to-pitch mapping resulted in an observable coactivation of auditory and motor regions. This coactivation was observed when listening to trained melodies without movement, or when performing movements without auditory feedback (Bangert & Altenmüller, 2003; Chen et al., 2012).
Wollman et al. (2018) also found evidence for the early development of auditory–motor couplings in nonmusicians as they learned to play the cello. Using fMRI, they observed a rapid engagement of the dorsal auditory–motor pathway in novice cellists after just one week of training. Activation was maintained after four weeks and correlated with greater proficiency when performing. Stronger coactivation was found in the right auditory cortex (AC), Premotor cortex (PMC), supplementary motor area (SMA), preSMA, M1, basal ganglia, and cerebellum, in better performers, suggesting that early and stable auditory–motor associations support improved performance. Similarly, Brown and Penhune (2018) showed that in proficient pianists, increased activation in left M1 and bilateral S1 during auditory recall of trained melodies correlated with enhanced pitch and timing accuracy during performance. Additionally, Lotze et al. (2003) found that while performing, expert violinists displayed greater activation in contralateral M1 and right auditory cortex, while amateurs showed more diffuse activation, particularly in prefrontal areas, overall indicating weaker auditory–motor associations and less efficient processing.
Stewart et al. (2003) also provided evidence for the existence of visuo–motor couplings. Using fMRI, they looked at differences in brain activity before and after nonmusicians learned to read and perform piano melodies. They found training-related activation in the bilateral superior parietal cortex and the left fusiform gyrus, an effect that was absent in the control group. The authors argued that these activation patterns reflected visuospatial sensorimotor mappings, referring to the coupling of spatial note information to motor responses (Stewart et al., 2003). However, they argued that the dorsal visual processing stream, in which the superior parietal cortex resides, is known to be involved for coding of spatial rather than featural aspects of visual information. This led them to interpret activation in this region as independent of skill level. In contrast, they proposed that it was activation in the fusiform gyrus, which was skill dependent. This region is part of the ventral visual stream and is typically associated with object identification and pattern discrimination. According to the authors, its engagement indicated the presence of pattern recognition mechanisms, which are indeed skill dependent. In other words, after training, the notes were no longer perceived as abstract spatial locations but instead became stable visual patterns with associated meaning and actions. Overall, the authors suggest that the activation of the left fusiform gyrus may be an early indicator of perceptual expertise in novice readers, reflecting increased training-dependent visual–motor coupling (Stewart et al., 2003).
Taken together, the evidence indicates that musical training results in integrated sensorimotor coactivation patterns, which are not found in nonmusicians (Bangert & Altenmüller, 2003). As a result, expert musicians show no neural distinction between motor-only and auditory-only conditions, indicating a fully integrated sensorimotor network that supports neural efficiency (Bangert & Altenmüller, 2003; Chen et al., 2012).
Skill Indicator 1.2: Brain network synchronization and functional connectivity
Sensorimotor integration functions were not only evident in localized brain activity, but in the synchronization across brain networks. Evidence came from Houdayer et al. (2016), who found plastic changes in interhemispheric circuits connecting the left and right motor and premotor areas, in nonmusicians, after learning to play the piano. Specifically, they found an increased ipsilateral silent period (ISP), measured with transcranial magnetic stimulation (TMS), which indicated enhanced interhemispheric communication. From these findings they suggested that interhemispheric interaction played a key role in motor learning, as it supports the coordination and implementation of new sensorimotor commands in music-naïve subjects. Similarly, Wollman et al. (2018) showed that nonmusicians who performed better exhibited enhanced functional connectivity between the presupplementary motor area (preSMA) and the bilateral auditory cortex (AC), emphasizing the importance of these regions in linking sound and action. This increased connectivity was associated with greater pitch and temporal accuracy, further highlighting the role of sensorimotor integration functions in supporting skilled performance.
Indicator Category 2: Neural Efficiency
A second central theme identified across studies was that musical training led to more focused and efficient brain activity during performance, while minimizing energy consumption. This neural efficiency was evident in motor and cognitive control processes, where experts often showed reduced activation in higher-order brain regions, compared with novices or nonmusicians. Across studies, a consistent finding was that the strengthening of sensorimotor associations facilitated this efficiency and the automation of skill, which in turn, contributed to enhanced performance.
Skill Indicator 2.1: Motor control
One key indicator of neural efficiency was the reduction of activity in high-order cortical areas involved in motor planning and control. These included the PMC, SMA, ipsilateral primary motor cortex (iM1), and dorsolateral prefrontal cortex (DLPFC) (Brown & Penhune, 2018; Lotze et al., 2003). For instance, Chen et al. (2012) observed that, during later stages of piano training, beginners exhibit decreased activation in the PMC, specifically the dorsal (PMd) and ventral (PMv) premotor cortices, as well as in the superior temporal gyrus (STG). Similarly, Houdayer et al. (2016), using EEG, observed a decrease in mu rhythm task-related desynchronization (TRD) over premotor regions in nonmusicians after they learned to play the piano. Additionally, Wright et al. (2012b) found that intermediate-level guitarists had smaller and later onset of movement-related cortical potentials (MRCPs), compared with nonmusicians, suggesting reduced cognitive effort during motor planning. These neural changes were accompanied by improvements in pitch and timing accuracy, reinforcing the link between reduced neural activity, efficient auditory–motor processing, and enhanced motor control.
Neural efficiency was also evident in subcortical structures. For instance, professional violinists showed an absent activation of the basal ganglia during performance (Lotze et al., 2003). Since the basal ganglia are associated with consolidating learned auditory–motor couplings into smooth, habitual actions, this finding suggests that well-practiced motor sequences reduce the need for neural resources and conscious control once they become automated.
These studies described decreased brain activity with expertise; however, other studies found that training-related reductions in some regions were accompanied by increased activity in others (Wright et al., 2012b). This underscores the high complexity of brain dynamics, which appear to depend on factors including the learning stage, the task, and the instrument played (Houdayer et al., 2016; Wright et al., 2012b). Brown and Penhune (2018), for example, observed that motor cortex (M1) activity remained stable across training, in both a group of proficient pianists and nonmusicians, as they learned to perform novel melodies on the piano. Similarly, Wollman et al. (2018) showed that the auditory cortex (A1; Heschl’s gyrus) was more strongly activated in experts, whereas amateurs relied more on the anterior STG. From these findings, the authors suggest that expert performers may rely on the consistent recruitment of motor execution and auditory feedback-processing regions to continuously support performance accuracy, and their activation might be partially independent of skill level (Wollman et al., 2018).
Skill Indicator 2.2: Neuromuscular control
Neural efficiency was also reflected in inhibitory processes, particularly through surround inhibition (SI). SI allows the brain to focus neural activity on the prime mover while suppressing adjacent, non-task-relevant muscles, preventing unwanted movements (Márquez et al., 2018). Using transcranial magnetic stimulation, Márquez et al. (2018) quantified SI, measured as the percentage reduction in motor-evoked potential (MEP) amplitude in nonactive muscles relative to baseline. They found that expert musicians exhibited high SI during simple, isolated finger movements (∼40% reduction) but lower SI during complex, multifinger tasks (∼15% reduction), suggesting flexible modulation of inhibitory control. In contrast, nonmusicians showed moderate, unchanging SI (∼25%) regardless of task demands. Based on these findings, the authors suggested that this flexibility supports the coordinated muscle synergies required in the complex movements involved in skilled performance. Notably, SI may reduce the cognitive cost of motor preparation by constraining neural activation to only the required effectors, aligning with broader principles of neural efficiency (Márquez et al., 2018).
Skill Indicator 2.3: Cognitive control
Another key indicator of neural efficiency was the reduction of activity in high-order cortical areas involved in cognitive and attentional control. For instance, Berkowitz and Ansari (2010) showed through fMRI that proficient musicians exhibited a significant deactivation of the right temporoparietal junction (rTPJ) while improvising, compared with nonmusicians. This region is usually involved in shifting or reorienting attention, and its deactivation has been described in response to top-down signals during goal-driven behavior, to inhibit attentional shifts toward task-irrelevant stimuli that could cause decrements in performance (Berkowitz & Ansari, 2010). Also, increased deactivation was found to correlate with more successful task performance. The authors interpreted these findings as indicating that the deactivation of the rTPJ region in experienced musicians reflected their ability to enter a more focused attentional state during the improvisation task.
Oscillatory dynamics further supported this view. Using EEG, Blanco and Ramirez (2019) asked participants to produce a stable and sustained violin sound on an open string and showed that intermediate-level violinists exhibited lower beta and gamma power at frontal sites during this activity, while non-violinists showed heightened activity, probably reflecting higher cognitive load and effort. However, as they improved across training sessions, their activity patterns gradually aligned with those of more experienced musicians. In turn, experienced musicians showed very clear localized gamma desynchronization at the right parietal and left temporal sites and beta synchronization in the right temporal region, which the authors associated not only with enhanced sensorimotor integration and motor control but also with attentional processes. They interpreted these findings using the temporal binding model, where gamma activity reflects the cognitive effort necessary to integrate distributed and multimodal information, which diminishes as tasks become automatic. Furthermore, they suggested that beta and gamma band power at frontal sites could be potential significant features to discriminate between musicians at different skill levels.
In contrast, Bugos et al. (2024) found that jazz improvisation training modulated neural activity in frontal theta bands during the improvisation task. In their study, a beginner jazz piano training group showed a maintenance in relative theta power over frontal areas, whereas the control group showed reductions in relative theta power over frontal areas. Theta power has been associated with cognitive control inhibition, working memory, and sustained attention, with a decrease in power over frontal areas usually interpreted as a decrease in cognitive control, attention, or periods of boredom (Katahira et al., 2018). In turn, there is also evidence in the literature of increased alpha activity associated with higher creativity and creative ideation (Benedek et al., 2018). Based on this, the authors attributed the change in theta activity between groups to greater recruitment of more resources for cognitive control after training.
Skill indicator 2.4: Auditory and error processing
Neural efficiency was also seen in increased neural activity related to auditory processing, and more specifically error processing. Supporting evidence came from Katahira et al. (2008) and Yasuhara et al. (2024), both of whom used EEG to investigate expertise-dependent event-related potential (ERP) responses to sensory perturbations (i.e., a note was shifted up by a semitone or delayed) during piano performance. Katahira et al. found that proficient musicians exhibited larger ERP responses to auditory feedback errors, specifically the N210 and the imagery mismatch negativity (iMMN). These responses were absent in nonmusicians, suggesting that trained musicians were more efficient in detecting errors, even without auditory feedback. Similarly, Yasuhara et al. found that expert pianists showed larger P300 amplitudes, compared with beginner pianists, in response to the auditory perturbation, indicating greater cognitive efficiency in processing errors. Additionally, these responses correlated with better timing stability and strike intensity adjustments, suggesting that cognitive processing was associated with enhanced motor adaptation in expert pianists.
Together, these findings point to a neurofunctional reorganization that underpins musical expertise. Training strengthens auditory–motor and visuo–motor couplings, facilitates network synchronization, and promotes inhibitory control, all of which support and reflect optimized neural efficiency related to music training. As a result, experts exhibit specific brain dynamics, such as reduced activity in high-level cognitive and motor areas, while other motor and auditory regions remain engaged to ensure precision and feedback integration. These training-related changes reflect a shift toward optimized, task-specific resource allocation, supporting skilled, expressive, and adaptive musical performance.
Indicator Category 3: Muscular Efficiency and Consistency
The last indicator category identified across psychophysiological studies was muscular efficiency and consistency, reflected in reduced muscle cocontraction, low variability in muscle activation timing, and early rise and decline of muscle activity. For instance, Fujii et al. (2009a) used surface electromyography (sEMG) to examine the world's fastest drummer (WFD), as he achieved high tapping speed (10 Hz). They observed that he relied on reciprocal wrist muscle activation rather than cocontraction, ensuring a smooth and efficient motion. In a follow-up study, Fujii et al. (2009b) compared trained drummers to nondrummers during fast rapid unimanual tapping, and found that trained drummers exhibited less muscle cocontraction, lower variability in muscle activation timing, and an earlier decline in wrist flexor activity. In a similar task, Fujii and Moritani (2012a) found that WFD exhibited a faster rise and earlier decline in electromyographic amplitude, as well as more stable timing in muscle activation, compared with both nondrummers and intermediate players. Extending this work, Fujii and Moritani (2012b) found that the WFD displayed higher motor unit discharge rates, increased recruitment, and enhanced synchronization
Similar muscular patterns were observed in pianists. Furuya and Kinoshita (2008a) found that expert pianists showed lower coactivation between agonist and antagonist in the forearm and upper arm during fast right-hand octave keystrokes. In contrast, beginners displayed higher coactivation, resulting in greater muscle stiffness and less precise movements. Supporting this, Furuya (2011) also found that expert pianists could maintain high performance at fast tempos with minimal finger muscle coactivation, whereas amateurs relied on increased coactivation, reducing overall efficiency in performing. Taken together, these findings suggest that long-term musical training leads to optimized muscle coordination, reducing unnecessary effort and enhancing precision. In this sense, muscular efficiency may mirror neural efficiency, both serving as physiological indicators of skilled performance.
Summary of the Studies on Psychophysiology
Summary of the psychophysiological skill indicators grouped by functional category.
Abbreviations. AC: auditory cortex; DLPFC: dorsolateral prefrontal cortex; EEG: electroencephalography; ERP: event-related potentials; ERD: event-related desynchronization; ERS: event-related synchronization; IFG: inferior frontal gyrus; IP: inferior parietal cortex; M1: primary motor cortex; LPC: late positive component; MRCPs: movement-related cortical potentials; PMv: premotor ventral cortex; PMd: premotor dorsal cortex; preSMA: presupplementary motor area; PSD: power spectral density; rTPJ: right temporoparietal junction; SMA: supplementary motor area; STG: superior temporal gyrus; TRD: task-related desynchronization.
Studies on Biomechanics: Kinematic and Kinetic Parameters
The studies that used biomechanical measurement tools and methods examined kinematic and kinetic parameters in instrumental learning. These studies primarily focused on movement control and efficiency. Figure 5 shows a summary of the identified indicators that characterize skill level and the measurement tools with which they were identified.
Biomechanical skill indicators grouped by indicator category.
Indicator Category 1: Movement Control
A central marker of musical expertise was enhanced movement control, characterized by enhanced movement coordination, greater consistency, and increased flexibility, accuracy, and smoothness. Together, these dimensions reflect how skilled performers optimize the organization, timing, and precision of their movements to meet the biomechanical demands of musical performance.
Skill Indicator 1.1: Movement coordination
Proximal-to-distal organization vs. selective freezing of degrees of freedom
Several studies examined how musicians coordinated their joints during instrumental performance. A central question emerging from this body of work was whether expertise consistently followed a proximal-to-distal sequencing (PDS) joint pattern, where movement control progressed from larger, proximal joints (e.g., shoulder) to smaller, distal ones (e.g., wrist and fingers), or whether expertise involved a selective freezing or freeing of degrees of freedom (DOFs) to enhance control and precision.
For instance, Furuya and Kinoshita (2007) analyzed upper-limb kinematics during a one-hand octave keystroke on the piano. They observed a clear PDS pattern in experts, with peak angular velocities occurring sequentially from the shoulder to the elbow and then to the wrist. In contrast, beginners exhibited a less clear temporal organization. A subsequent study by Furuya and Kinoshita (2008b) supported these findings. They found that expert pianists used a clear PDS strategy, in contrast to beginners, in which they used the shoulder as a leading joint to generate large interaction torques at the elbow and wrist. This strategy allowed for greater movement efficiency and reduced energy expenditure.
However, Furuya (2011) challenged the generalizability of these findings. In a continuous keystroke task, they found that experts exhibited simultaneous peak velocities at the elbow and fingers, indicating that the elbow played an active role in generating fingertip velocity, rather than merely generating interaction torques, as suggested in the previous studies. From these results, they suggested that joint coordination in experts might be task dependent, rather than following a universal PDS pattern.
Adding to this, research on string players provided further nuance. Verrel et al. (2013a, 2013b, 2014) examined expertise-related differences in DOF coordination during cello bowing. Their kinematic analyses showed that proficient cello players displayed reduced movement variability, greater amplitude of motion in the elbow and wrist, and enhanced coordination of distal joints (elbow, wrist, and fingers). There was also evidence of a proximal-to-distal progression in joint control. Notably, proficient players integrated both the shoulder and elbow for bow transport, while the wrist played a stabilizing role in maintaining bow angle and velocity. These features were absent in nonmusicians, who relied predominantly on the shoulder, demonstrating a lack of clear interjoint coordination. Overall, this coordination strategy, characteristic of proficient players, directly contributed to enhanced sound control, supporting the functional relevance of enhanced joint coordination in motor control.
In contrast, Konczak et al. (2009) observed a selective restriction of proximal joints in proficient and expert string players. They found that shoulder range of motion was reduced in proficient and expert musicians, while elbow coordination was maintained, suggesting a strategy of minimizing proximal movement to enhance precision. Rather than freeing all joints, expertise in this case involved stabilizing specific joints to reduce variability and fatigue.
A similar pattern emerged in studies investigating percussion instruments. Altenmüller et al. (2020) investigated expertise-related differences in drumming movements and found that experts adopted a whip-like strategy that prioritized wrist motion and minimized elbow involvement. This use of gravity and distal control resulted in higher temporal and spatial accuracy, contrasting with the irregular stick trajectories and reduced control observed in nondrummers. Like Furuya (2011), they argued that motor control strategies were task-dependent, with cyclic drumming relying on distal control rather than strict PDS.
Overall, the evidence indicates that motor coordination appears to be instrument- and task-specific. In some cases, expertise involves a PDS of movement, while in others, it requires stabilizing or restricting certain joints to improve accuracy and reduce fatigue. These variations highlight that motor expertise might be characterized by its adaptability in the use of motor control strategies, shaped by the specific biomechanical, technical, and expressive demands of each musical task and instrument.
Independent control of movements
Another indicator of movement coordination was individuation of finger movements. For instance, Parlitz et al. (1998) found that experts, while performing piano finger exercises, exerted less unintended force with nonplaying fingers, indicating better finger independence. Experts also showed brief, efficient finger touches, followed by immediate relaxation of playing fingers, while amateurs tended to maintain excessive tension, especially in nonplaying fingers. This reflected a higher degree of independent finger control over both playing and nonplaying fingers in experts. Further supporting these findings, Furuya et al. (2014) found that short-term piano practice with nonpianists reduced movement covariation across fingers, especially between less-independent pairs (e.g., ring and little fingers), indicating enhanced finger independence after training. Similarly, Winges and Furuya (2015), using a cyberglove to measure joint angle data, found that expert pianists exhibited greater temporal precision, dynamic control, and motor efficiency, characterized by more consistent, individuated joint motions. In contrast, amateurs showed more variable movements with increased joint coactivation, indicating less independent motor control.
Skill Indicator 1.2: Movement consistency
Movement control in musical performance was often characterized by decreased movement variability. For instance, Furuya and Kinoshita (2008b) and Oku and Furuya (2017) found lower variability in both peak key force and peak key descending velocity in expert pianists compared with novices when they were performing repeated keystrokes at varying loudness levels, which was associated with their more stable and controlled motor execution.
Similar results were found in stringed instruments. For instance, Konczak et al. (2009) found that learning to play the violin was associated with reduced variability in bow movements, with expert players showing the lowest variability in bowing angles. Using linear regression with the standard deviation of bow angle as the dependent variable and hours of lifetime practice as the predictor, they observed a negative correlation indicating that variability of bowing precision decreased as practice hours and bowing precision increased. Similarly, Verrel et al. (2013a, 2013b) found that nonmusicians exhibited greater variability in bow duration, bow angles, bow touch on the string, and bow amplitude compared with expert cellists. Building on these results, Verrel et al. (2014) examined bow velocity and bowing angle variability during bow movements, as well as acceleration amplitude at bow reversals. They found that expert cellists exhibited low variability, which was indicative of good performance. Specifically, low within-bow variability scores for bow velocity reflected near-constant bow velocity, which in turn was associated with more stable tone during performance. Similar results were found in drummers, where nondrummers showed high variability of movements, in comparison with experts, who demonstrated more consistent and controlled movements (Altenmüller et al., 2020).
Taken together, these findings support the notion that increased variability is generally associated with more complex, less controlled musical tasks or lower levels of expertise (Allingham et al., 2021).
Skill Indicator 1.3: Movement flexibility
In a study by Allingham et al. (2021), movement control in violin performance was characterized by freedom of movement. Using motion capture, they analyzed the bow and violin movements of novices and experienced string players. A key measure was “scroll sway,” defined as the standard deviation of the mediolateral (Y-axis) position of a marker placed on the violin scroll. This measure reflected the extent of instrumental and torso movement in the horizontal plane, providing an index of how freely the performer moved during playing. Their analysis showed that expert players exhibited greater overall “scroll sway” than novices, indicating more freedom of body movement. Moreover, experienced string players also demonstrated consistent and precise control of sound, bowing, and body movement, demonstrating greater movement control overall.
Skill Indicator 1.4: Movement accuracy
Another indicator of movement control was temporal and spatial accuracy. For instance, Altenmüller et al. (2020), who compared drummers at different skill levels using motion capture, observed that experts showed the lowest variability in stroke timing and regular, self-similar motion cycles, which were supported by the predominant use of distal joints (e.g., wrist and hand) and the efficient use of gravity for precise impact. This strategy differentiated experts from proficient players and nondrummers, allowing them to avoid unwanted rebounds and spatial deviations. Overall, consistency in both amplitude and trajectory across cycles was an indicator of expertise, indicating high temporal and spatial precision, and movement control.
Similarly, Beveridge et al. (2020), combined sEMG data with drumstick motion capture to analyze muscle activation patterns and timing control during drumming exercises. They calculated timing variability via the coefficient of variation of intertap intervals (CV-ITI), providing an objective metric of performance consistency, and observed that timing precision improved with cumulative lifetime practice. Overall, they found that the expert drummers, compared with the amateurs, were characterized by efficient, reciprocal muscle activation patterns and superior timing consistency.
Skill Indicator 1.5: Movement smoothness
Lastly, Gonzalez-Sanchez et al. (2019) introduced movement fluency as a key indicator of expert performance, defined as smooth, accurate, and efficient sound-producing motions. They conceptualized movement fluency as a combination of coordination and smoothness. Coordination, characterized by anticipation, segmentation, and coarticulation of motor elements, was measured with EMG, to capture patterns of muscle activation. Smoothness was measured with motion capture data using spectral arc length (SAL), a metric that analyses the frequency spectrum of the velocity profile. The study examined cellists and drummers at different skill levels as they performed an accelerando–decelerando task. Their results indicated that intermediate players exhibited smoother movement trajectories (higher SAL measures), optimized muscle activation patterns (lower EMG variance), and reduced variability of movement across tempos, when compared with beginners, all of which point to greater movement control. Based on these findings, the authors suggested that movement fluency, and particularly, movement smoothness, could serve as an objective indicator of musical skill.
Indicator Category 2: Movement Efficiency
Another indicator of expertise in musical performance was movement efficiency, reflected in the capacity to achieve music tasks with minimal muscular effort and optimized use of biomechanical forces. Efficiency was expressed through the economical recruitment of muscles, the exploitation of interaction and gravitational torques, and the reduction of unnecessary muscle cocontractions, all of which contributed to greater endurance, stability, and precision.
Skill Indicator 2.1: Use of motion-dependent interaction and gravitational torques
Expert musicians optimized their movements by leveraging motion-dependent interaction torques (INT), muscular torques (MUS), and gravity-assisted movements. For instance, Furuya and Kinoshita (2008b) and Furuya et al. (2009) found that expert pianists minimized direct muscular effort when performing repeated keystrokes at different loudness levels. In contrast, beginners relied on muscular force-driven arm downswing to produce keystrokes, resulting in less efficient, more variable movements with lower endurance. Similar findings were described by Oku and Furuya (2017), who reported that expert pianists, when compared with nonmusicians, could achieve loud tones at faster tempos with lower peak force, reflecting consistent, controlled, and energy-efficient movement.
Evidence from percussion instruments mirrors these findings. Altenmüller et al. (2020) showed that expert drummers optimized wrist motion and exploited stick rebound, minimizing muscular effort while maintaining highly regular, self-similar trajectories. In experts, the use of distal joints predominated, shoulder involvement was minimal, and gravity contributed to movement control, all of which supported higher temporal precision and consistency compared with proficient student drummers and nondrummers.
Skill Indicator 2.2: Efficiency and consistency in muscle activation patterns
Movement efficiency was also evident at the muscular level. Expert performers consistently exhibited lower levels of muscle cocontraction and more reciprocal muscle activation. For instance, Furuya and Kinoshita (2008b), when comparing novices to expert pianists performing right-hand octave keystrokes, showed that novices displayed stronger activation of shoulder and elbow extensors and higher forearm cocontraction, indicating higher distal muscular stiffness and effort. In contrast, professional pianists showed reduced cocontraction in finger muscles (EDC and FDS), particularly at higher tempi, indicating more economical and precise muscle activation. Further evidence came from Beveridge et al. (2020) in drumming, where experts demonstrated pronounced reciprocal activation, minimal cocontraction, and highly stable timing compared with amateurs, representing an efficient neuromuscular strategy that supports precise, consistent, and efficient performance.
Skill Indicator 2.3: Economical use of muscular force
Finally, efficiency was reflected in the economical use of muscular force. Parlitz et al. (1998) and Oku and Furuya (2017), using an F-scan sensor and a force transducer respectively, found that expert pianists could produce the same loudness levels with smaller force impulses and less exerted muscular force compared with amateurs or nonmusicians, indicating energetic efficiency. Among the pianists, those who could play faster at maximum tempo were the ones who showed smaller force impulse at loud tones. These results indicate that the ability to use muscular force economically was correlated with motor speed capacity, a characteristic of expertise.
Summary of the Studies on Biomechanics
Taken together, these findings highlight that expert musicians exhibit greater movement control, reflected in lower movement variability, higher temporal and spatial accuracy, higher movement smoothness and flexibility, and specific joint coordination patterns compared with novices. Additionally, muscle efficiency, reflected in optimized muscle activation and economical use of gravitational forces and interaction torques, also emerged as key indicators of expertise. These patterns reflect the effects of extensive training in both movement and muscular adaptations that underlie skilled music performance. In Table 5 and Figure 5 we summarize and visualize the identified skill indicators extracted from the results of the studies on biomechanics.
Summary of the biomechanical skill indicators grouped by functional category.
Abbreviations. CV: coefficient of variation; DOFs: degrees of freedom; dSI: smoothness measure – SPARC Index; MoCap: Motion Capture; PCA: principal component analysis; PD: procrustes distance; PDS: proximal to distal joint organization; sEMG: surface electromyography; SD: standard deviation.
Discussion
This scoping review examined empirical findings of studies on music skill acquisition through the lens of first-, second-, and third-order embodiment (Metzinger, 2014; Nijs, 2017). Our aim was 1) to map psychophysiological and biomechanical indicators of skill acquisition onto distinct levels of embodiment and 2) to examine how these indicators may contribute to the conditions for phenomenological incorporation of the instrument.
To do so, we defined a set of criteria to map these indicators onto the first and second levels of embodiment.
First order: indicators reflecting structural and neurophysiological activation patterns (e.g., neural and muscular activation patterns)
Second order: indicators reflecting cognitive and motor processes underlying predictive control (e.g., error monitoring, attentional control, movement preparation and coordination)
The third level of embodiment was beyond the empirical scope of this review. Notably, only one study combined psychophysiological or biomechanical measures with subjective reports, by asking participants what they had been thinking about after performing a task (Allingham et al., 2021), highlighting that the selected literature does not address the third level of embodiment through qualitative methods to probe subjective experience. Although some cognitive (e.g., attentional control), behavioral (e.g., movement smoothness), neural (e.g., coactivation of sensorimotor areas) or physiological (e.g., reciprocal muscle activation) measures may serve as indirect proxies for subjective experience, these processes are not necessarily consciously accessible or reportable by performers. Consequently, they remain open to interpretation and are insufficient to capture and understand the lived dimension of music performance.
Furthermore, it is important to note that the three levels are not discrete categories but rather interrelated dimensions that unfold through learning. Embodiment indicators might thus be better conceptualized as existing along a continuum, from neural and muscular adaptations through predictive control and efficiency to lived experience of incorporation. In the following sections we discuss how each level manifests empirically and how the indicators build upon one another.
First-Order (Morphological) Embodiment
First-order embodiment refers to structural and physiological changes that emerge from the musician's interaction with the instrument during the process of skill acquisition (Nijs, 2017). Empirical evidence from the included studies shows that musical training induces neuroplastic changes in auditory, motor, and multimodal brain regions, leading to the development of integrated sensorimotor networks with increasing expertise (Bangert & Altenmüller, 2003; Lotze et al., 2003; Stewart et al., 2003).
Embodiment Indicator 1.1: Integrated Sensorimotor Networks
A key indicator of morphological embodiment that differentiates beginners and experts was the progressive integration of auditory, visual, and motor systems (Bangert & Altenmüller, 2003; Lotze et al., 2003; Stewart et al., 2003). In novices, even short periods of piano or cello practice led to the coactivation between auditory and motor brain regions, marking the beginning of integrated auditory–motor processing (Bangert & Altenmüller, 2003; Chen et al., 2012; Li et al., 2018; Luciani et al., 2022; Wollman et al., 2018). Similar processes occurred when learners associated spatial information (visually presented notes) with motor actions, leading to visuo–motor integration networks that engaged both visual regions (e.g., superior parietal or fusiform gyrus) and motor areas (Stewart et al., 2003). With continued training, these connections become more robust and automatic. For instance, expert musicians showed integrated sensorimotor networks that became active not only during performance but also when listening to music (without performing movements) or when executing movements silently (Bangert & Altenmüller, 2003; Brown & Penhune, 2018; Lotze et al., 2003).
Furthermore, as novices learned to perform, training-related changes were observed in interhemispheric circuits connecting motor, premotor, and auditory areas (Houdayer et al., 2016), and in enhanced functional connectivity between the preSMA and bilateral auditory cortices (Wollman et al., 2018). The authors suggest that these adaptations support the implementation and coordination of new sensorimotor commands in music-naïve subjects and correlate with greater pitch and temporal accuracy and better performance in general (Wollman et al., 2018). Such findings converge with resting-state evidence that shows that musicians exhibit stronger multisensory–motor connectivity than nonmusicians (Alluri et al., 2017; Herholz & Zatorre, 2012; Papadaki et al., 2023).
Embodiment Indicator 1.2: Muscle Activation Patterns
Complementary to neural adaptations, we also identified changes in muscle activation patterns as indicators of morphological embodiment. For instance, expert drummers and pianists showed reduced levels of muscle cocontraction and greater reciprocal activation patterns in comparison with nonmusicians and musicians at lower levels (Allingham et al., 2021; Fujii et al., 2009a, 2009b; Furuya & Kinoshita, 2008a). In turn, these adaptations were associated with enhanced temporal precision, greater endurance and reduced fatigue, and the ability to maintain high performance at fast tempos, highlighting the role of muscular efficiency in movement control and performance (Beveridge et al., 2020; Fujii et al., 2009a, 2009b). We suggest that such changes reflect increased efficiency in handling the instrument, as the body's structural configuration becomes progressively aligned with the mechanical constraints of the instrument (Visentin et al., 2015).
Synthesis: Morphological Adaptations
Taken together, the neural and muscular findings reviewed here support the idea that first-order embodiment consists in the progressive embedding of instrument-specific sensorimotor demands into the body's basic action–perception architecture (Maes et al., 2014; Segil et al., 2022). Across studies, musical training was associated with coordinated reorganization of auditory, motor, and multimodal brain networks, as well as with changes in muscular activation patterns, reflecting increasingly efficient and stable sensorimotor organization. These adaptations converge in showing that sensory representations and motor commands become increasingly coupled, linking sound perception to the movements that generate musical output (Altenmüller et al., 2019; Herholz & Zatorre, 2012; Hyde et al., 2009; Novembre & Keller, 2014; Penhune, 2019). At the same time, changes in muscular coordination, such as reduced cocontraction and more reciprocal activation, suggest that the body's biomechanical organization becomes progressively aligned with the mechanical constraints of the instrument.
At this level, embodiment reflects a largely automatic process of sensorimotor integration driven by experience-dependent plasticity (Zatorre et al., 2007). These stabilized mappings do not yet imply predictive control or conscious experience, but they constitute the neurophysiological foundation upon which internal models and anticipatory action control can later develop, thereby enabling second-order (functional) embodiment (Novembre & Keller, 2014).
Second-Order (Functional) Embodiment
Second-order embodiment concerns the functional organization of sensorimotor control, that is, the acquisition and refinement of internal models that enable enhanced prediction and control (Metzinger, 2014; Nijs, 2017). In the context of musical skill acquisition and building on the framework of predictive processing outlined in the introduction, the reviewed evidence supports the view that second-order embodiment is reflected in the consolidation of predictive control mechanisms that incorporate the instrument (Friston, 2012; Keller & Koch, 2008; Maes et al., 2014; Pezzulo et al., 2024). These predictive models do not a priori distinguish between bodily and tool-mediated actions; rather, any reliably coupled sensorimotor contingencies can become integrated into the predictive model (Segil et al., 2022; Van Elk, 2021). In musical performance, stable auditory–motor couplings therefore enable the instrument to be controlled within the same predictive architecture that governs skilled action, reducing the need for explicit tool-focused monitoring (Martel et al., 2016; Segil et al., 2022; Van Elk, 2021).
Embodiment Indicator 2.1: Neural Efficiency
A key indicator of second-order embodiment that emerged from the included studies was neural efficiency, often reflected in reduced brain activation or energy expenditure in task-relevant regions while maintaining or improving task performance. These reductions are interpreted as signs of automation and optimized processing efficiency, reflecting a reduced need for cognitive effort to execute skilled actions. For instance, studies reported decreased activation in premotor regions, such as the dorsal and ventral premotor cortex (PMd, PMv), and in auditory regions like the STG during later stages of training or in experts compared with novices (Chen et al., 2012). Houdayer et al. (2016) also found a training- and task-related mu rhythm desynchronization in premotor regions, suggesting improved efficiency in sensorimotor integration and motor control. Similarly, expert guitarists exhibited reduced movement-related cortical potentials, suggesting enhanced efficiency in motor planning and execution (Wright et al., 2012a, 2012b), and professional violinists showed basal ganglia deactivation, suggesting the automatization of motor sequences (Lotze et al., 2003).
However, neural efficiency was not only limited to reductions in activity. Instead, it also manifested as more consistent and selective recruitment of specific regions, including the M1 (Brown & Penhune, 2018) and the A1 after training (Wollman et al., 2018). In turn, these increased activation patterns were associated with enhanced auditory feedback processing and motor precision. Furthermore, expert musicians, compared with nonmusicians, also showed evidence of surround inhibition (SI) of muscle synergies, a neural mechanism where the inhibition of nonactive muscles helps focus motor output and reduce the energetic cost of motor action (Márquez et al., 2018).
Additionally, neural efficiency was also evident in error processing. Expert musicians exhibited larger ERP responses (e.g., P300, N210, iMMN) to auditory feedback errors, indicating more effective detection of intention–action outcome mismatches (Katahira et al., 2008; Kleber et al., 2013, 2017; Maidhof et al., 2013; Yasuhara et al., 2024). The authors interpret the heightened neural responses to errors as evidence for the existence of internal models in expert musicians, from which they can monitor the discrepancy between the corollary discharge created in auditory area (the intended sound) and incoming auditory feedback, to then evoke the N210 in response to discrepancy (Katahira et al., 2008).
Embodiment Indicator 2.2: Cognitive Control
Second-order embodiment was also characterized by the refined use of executive and attentional resources, allowing musicians to sustain focus, adapt, and self-monitor with reduced cognitive effort. For instance, musicians, in comparison with nonmusicians, exhibited a significant deactivation of rTPJ during improvisation reflecting a shift toward internally focused attention, while suppressing irrelevant external cues (Berkowitz & Ansari, 2010). This is in line with literature showing reduced activation in higher-order cognitive control regions after training, including the caudate nucleus, cerebellum, inferior frontal gyrus, posterior hippocampus, and superior parietal lobule, which is generally correlated with increased melodic and rhythmic accuracy, suggesting the automation of executive control and enhanced attentional control (Olszewska et al., 2024). Additionally, expert musicians showed lower beta and gamma power at frontal sites, indicating lower cognitive load (Blanco & Ramirez, 2019). Yet, in an improvisational context, a study observed that trained musicians maintained greater recruitment of cognitive resources at frontal theta bands, while the control group showed reductions. The authors suggest that the allocation of cognitive resources is task dependent and argue that creative engagement might require sustained cognitive effort (Bugos et al., 2024).
Embodiment Indicator 2.3: Motor Control
Finally, we observed indicators of motor coordination and control in expert pianists and string players, which not only include joint coordination patterns such as PDS (Furuya & Kinoshita, 2007; Furuya & Kinoshita, 2008b; Verrel et al., 2013a, 2013b, 2014), but the ability to flexibly reconfigure their movement strategies depending on both task and environmental constraints (Altenmüller et al., 2020; Furuya, 2011; Konczak et al., 2009). We also observed indicators of movement consistency, efficiency, and precision in experts across instruments, including lower movement variability (Furuya & Kinoshita, 2008b; Konczak et al., 2009; Verrel et al., 2013a, 2014), lower variability in muscle activation timing and faster rise/decline of EMG amplitude (Fujii & Moritani, 2012a, 2012b), enhanced motor unit discharge rates and synchronization (Fujii & Moritani, 2012b), enhanced movement smoothness and fluency (Gonzalez-Sanchez et al., 2019), independent finger control and reduced unintended force (Furuya et al., 2014; Parlitz et al., 1998), and a reliance on the use of interaction and gravitational torques that also minimized muscular effort (Altenmüller et al., 2020; Furuya et al., 2009; Furuya & Kinoshita, 2008b; Oku & Furuya, 2017). Taken together, these markers illustrate how functional embodiment manifests through optimized motor control (Campo et al., 2023; Furuya and Kinoshita, 2008b; Gonzalez-Sanchez et al., 2019; Verrel et al., 2014).
Synthesis: Functional Adaptations
From the reviewed evidence we suggest that second-order embodiment can be interpreted as the consolidation of predictive control mechanisms indexed by neural efficiency and refined cognitive and motor control. Neural efficiency plausibly reflects the stabilization and automation of internal models, reducing the need for effortful monitoring of instrument-mediated actions (Friston, 2012; Maes et al., 2014). Changes in cognitive and attentional control may indicate a shift away from explicit, tool-directed processing toward goal-driven regulation of musical intentions (Allingham & Wollner, 2022; Zatorre et al., 2007), while optimized motor coordination and flexible movement reconfiguration indicate that instrument constraints are increasingly integrated into the performer's action repertoire, supporting anticipatory planning (Maes et al., 2014).
At this level, these indicators converge in suggesting that the instrument has become integrated into the performer's predictive models for action, such that it is no longer processed as an external object but is governed by the same anticipatory mechanisms that regulate the body itself (Martel et al., 2016; Segil et al., 2022; Van Elk, 2021). This integrated control architecture, in turn, provides the necessary substrate for the emergence of phenomenological experiences of agency, ownership, and transparency in musicians (Friston, 2012).
Third-Order (Phenomenal) Embodiment
Phenomenological aspects of embodiment were not directly assessed in the included studies. According to Metzinger (2004, 2014), third-order embodiment concerns the subjective dimension that arises when information about bodily states and action-related processes is integrated into a coherent phenomenal self-model (PSM) (Metzinger, 2004) and becomes globally available to the cognitive systems involved in perception and action. In this framework, incorporation of a musical instrument can be understood as self-extension: Instrument-mediated action and its sensory consequences become integrated into the performer's PSM, such that the instrument is experienced less as an external object and more as part of the self's organization of perception and action (Metzinger, 2004, 2014).
Across the reviewed evidence, several clusters of first- and second-order indicators plausibly support this experiential shift. First, strengthened sensorimotor integration and increasingly stable auditory–motor mappings may reduce the need for explicit monitoring of instrument mechanics, enabling attention to be directed toward musical intentions and thereby supporting instrument transparency in performance (Chen et al., 2012; Houdayer et al., 2016; Kim, 2020; Wollman et al., 2018). Second, markers of increased efficiency, including reduced cognitive load, refined inhibition, decreased cocontraction, and smoother kinematics, align with lived reports of effortlessness and fluent absorption characteristic of expert performance and flow (Blanco & Ramirez, 2019; Colombetti, 2014, 2023; Csikszentmihalyi, 2009; Gonzalez-Sanchez et al., 2019; Nijs, 2017). Third, improvements in prediction and error-based adaptation, indexed by enhanced neural responses to auditory perturbations and more stable corrective adjustments, may support stronger experiences of agency by increasing the reliability with which intended and perceived outcomes match (Blanke, 2012; Candia-Rivera et al., 2024; Forster et al., 2022; Leman, 2007; Raoul & Grosbras, 2023; Segil et al., 2022; Yasuhara et al., 2024). Finally, biomechanical evidence showing stable sound control through adaptable coordination and reduced variability suggests that instrument constraints become integrated into the motor repertoire, potentially supporting self-efficacy and perceived control during performance (Leman, 2007; Nijs, 2017).
Taken together, while phenomenological experience cannot be explicitly and directly derived from these measures, the evidence across neural, physiological, and biomechanical levels identifies plausible constitutive conditions for third-order embodiment to emerge, including instrument transparency, reduced effort and explicit monitoring, and enhanced agency and control. These interpretations are consistent with broader accounts of tool embodiment and bodily self-consciousness, which argue that when tools are embodied, their properties are processed similarly to bodily properties, with subjective consequences (de Vignemont, 2011). These phenomenological dimensions are not only central to understanding expertise but also connect embodiment to wellbeing by fostering experiences of agency, self-efficacy, presence, and ownership (Mojica & Di Paolo, 2025). For instance, Simoens and Tervaniemi (2013) found that musicians who experienced their instrument as an integral part of themselves reported greater professional wellbeing, lower performance anxiety, and reduced external pressure. Conversely, those who perceived their instrument as an obstacle were more likely to experience distress and career difficulties. These findings suggest that fostering a positive, embodied musician–instrument relationship may not only enhance performance but also contribute to musicians’ long-term wellbeing (Simoens & Tervaniemi, 2013).
Summary of the Embodiment Indicators
Summary of the embodiment indicators grouped by levels.
Limitations and Future Recommendations
The studies reviewed provide evidence for the embodied nature of music skill acquisition; however, several methodological limitations challenge the identification and interpretation of embodiment indicators across skill and embodiment levels.
Limitations Concerning the Included Studies
One major limitation concerns the widespread use of simplified motor tasks (e.g., tapping, keystrokes, single-note exercises). While these paradigms facilitate experimental control and isolate specific movement components, they fail to capture the complexity of real-world musical activities (Maselli et al., 2023). As a result, key cognitive–motor processes involved in skilled performance may be underrepresented (Maselli et al., 2023). Future studies would benefit from incorporating tasks drawn from the musicians’ repertoire performed in naturalistic performance settings that better reflect musicians’ everyday practice and performance contexts.
Another limitation concerns how expertise and evidence strength are represented across the reviewed studies. Many investigations primarily compare beginners or nonmusicians with proficient or expert musicians, with limited attention to intermediate stages of skill acquisition. In addition, expertise is defined inconsistently across studies, making comparisons difficult. Combined with relatively small sample sizes (on average around 20 participants per study), these factors limit the generalizability of findings and hinder the identification of robust embodiment indicators across the skill continuum. As a consequence, many proposed embodiment indicators are supported by only one or two empirical studies, often with small samples and specific tasks or instruments. Their robustness across musical contexts, instruments, and stages of skill acquisition remains uncertain. While these findings provide valuable initial insights, replication across independent samples, methodologies, and musical contexts are necessary before such indicators can be considered stable markers of embodied musical expertise.
An additional interpretative challenge concerns the distinction between short-term learning effects observed within single experimental sessions or over a few weeks and long-term skill development resulting from years of training. Short-term paradigms primarily capture early-stage adaptations, such as the initial formation of auditory–motor couplings, reflecting rapid plasticity (e.g., Bangert & Altenmüller, 2003). In contrast, long-term expertise involves the consolidation and optimization of these processes, leading to durable structural reorganization, predictive control, and neural efficiency. Within the present framework, short-term learning effects can be understood as revealing the early mechanisms of first-order embodiment, whereas long-term expertise reflects the cumulative development of both first- and second-order embodiment. Longitudinal research is needed to explicitly link short-term adaptation trajectories with long-term embodiment outcomes across different stages of musical expertise.
Limitations Concerning the Scoping Review
In this scoping review, we used the three levels of embodiment as a heuristic to organize the findings and to take initial steps toward an overarching theoretical framework for music skill acquisition. This framework provides conceptual clarity across studies by distinguishing between morphological, functional, and phenomenal embodiment, allowing researchers to situate specific findings within a broader theoretical landscape. At the same time, it remains provisional and would benefit from further theoretical integration.
Future work could strengthen this framework by connecting it to complementary perspectives in cognitive science and movement research. For example, the hierarchical levels of prediction proposed by Pezzulo et al. (2018, 2021) offer a process-oriented account of how sensorimotor predictions are organized and updated, which could enrich the functional dimension of embodiment. Likewise, conceptual work on body schema and body image by scholars such as Gallagher (2005) provides a valuable foundation for refining the distinction between first- and second-order embodiment, especially in relation to sensorimotor integration and body representation. In addition, integrating insights from dynamical systems theory could help account for the continuous and adaptive nature of skill acquisition, emphasizing the interaction between performer, task, and environment (Button et al., 2021).
At the empirical level, operationalizing these theoretical connections remains challenging. While morphological and functional embodiment can be investigated using psychophysiological and biomechanical measures, the phenomenal level is often underrepresented, reflecting both methodological tendencies and the inherent difficulty of capturing subjective experience in rigorous and reproducible ways. As a result, embodiment cannot be fully understood through biomechanics or psychophysiology alone, and experiential factors such as motivation, intentionality, and expressivity remain insufficiently explored. Addressing this gap will require mixed-methods approaches that integrate quantitative measures (e.g., neural, muscular, and kinematic data) with qualitative and first-person methodologies. This gap highlights the need to bridge theory and methodological innovation. The heterogeneity of study designs, outcomes, and measures in the reviewed literature precluded statistical synthesis, necessitating a narrative and thematic approach. This limitation highlights the need for greater methodological coherence across the field to support cumulative evidence building.
To advance in this direction, future research could take several key steps to address these limitations.
Incorporate naturalistic performance tasks and settings, including music pieces drawn from the musician's repertoire and experimental studies conducted in conservatoire settings. Integrate psychophysiological and biomechanical methods with qualitative methods that explore the subjective aspects of learning and performance. For instance, studies could include self-report questionnaire data on perceived effort, self-efficacy (McPherson & McCormick, 2006; Ritchie & Williamon, 2011), exertion and fatigue (McCrary et al., 2022), and phenomenological interviews (Høffding, 2018). Move beyond the binary classifications of novice vs. expert and investigate all stages of the skill continuum, such as novice, advanced beginner, competent, proficient, and expert (Dreyfus & Dreyfus, 1988). In addition, studies should use standardized criteria to define expertise, incorporating factors such as years of training, performance experience, and relevant physiological or biomechanical indicators. Collect data from larger and more diverse samples to improve the generalizability of findings, allow for more robust statistical analyses, and provide a broader understanding of music skill acquisition across diverse musicians’ groups and music genres.
In summary, while significant progress has been made in understanding the embodied nature of music skill acquisition, aligning conceptual frameworks with methodological improvements will be crucial for a richer, multilevel understanding of embodiment.
Conclusion
This scoping review highlights the conceptualization of music skill acquisition as an embodied process, where the musical instrument becomes an extension of the musician's body. In this process, neural, muscular, and motor adaptations coevolve, giving rise to embodiment at three levels: first-, second-, and third-order embodiment (Metzinger, 2014; Nijs, 2017). This hierarchical framework effectively captures the dynamic and evolving nature of the embodiment process throughout music skill acquisition. This review further identifies embodiment indicators, namely observable psychophysiological and biomechanical measures of these levels. Specifically, the indicators identified in this review are highlighted as markers of first- and second-order embodiment. Despite their value in offering insights into bodily adaptations as a result of learning, these indicators also reveal existing methodological challenges, particularly in identifying and tracking standardized embodiment indicators across skill levels. To address these challenges, future research could aim to create a more precise map of embodiment indicators across the whole skill acquisition continuum and integrate mixed-methods approaches to understand how the cognitive, motor, and affective process interact throughout the embodiment process and the development of expertise.
Supplemental Material
sj-docx-1-mns-10.1177_20592043261439571 - Supplemental material for Embodiment Indicators of Music Skill Acquisition: A Scoping Review of Psychophysiological and Biomechanical Studies
Supplemental material, sj-docx-1-mns-10.1177_20592043261439571 for Embodiment Indicators of Music Skill Acquisition: A Scoping Review of Psychophysiological and Biomechanical Studies by Laura Serra Marin, Hajer Gammoudi, Bahareh Behzadaval, Luis A. Leiva, Inès Chihi and Luc Nijs in Music & Science
Footnotes
Action Editor
Alexander Refsum Jensenius, University of Oslo, RITMO Centre for Interdisciplinary Studies in Rhythm, Time, and Motion & Department of Musicology.
Peer Review
Laura Bishop, University of Oslo, RITMO Centre for Interdisciplinary Studies in Rhythm, Time, and Motion. Helen Prior, University of Hull, Department of Music.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available in the following figshare repository: https://doi.org/10.6084/m9.figshare.30050827 (Serra Marin, 2026).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Ethical Approval
This research did not require ethics approval. It did not involve the use of personal data, fieldwork, or experiments involving human or animal participants. Instead, we conducted a scoping review of published literature. Therefore, no ethical concerns arise.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was funded by Institute for Advanced Studies at the University of Luxembourg. AUDACITY Funding Scheme, call 2023.
Supplementary material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
