Technology-driven initiating actions influence movement patterns in HMEAYC musical activities

Abstract

This study examines the integration of technology into the Holistic Music Educational Approach for Young Children (HMEAYC) by analyzing children’s movement sequences during musical rhythm activities. Specifically, it investigates the effects of technology-enhanced music on children’s initiating actions and movement patterns, comparing those with prior music experience to those without. A total of 75 children (43 boys and 32 girls), aged 3–6 years and enrolled in central Taiwan kindergartens, participated in the study. They were categorized into two clusters: Cluster 1 (n = 23) with minimal prior music exposure and Cluster 2 (n = 52) with previous musical experience. Researchers analyzed over 10 h of video recordings using a coding scheme that classified behaviors into six categories: Observation, Execution, Correct Performance, Abandonment, and Restart. Sequential behavior analysis revealed that children in Cluster 2 demonstrated significantly higher adaptability and reflective learning, showing more effective transitions, particularly from Observation to Execution and from Execution to Correct Performance. In contrast, Cluster 1 children exhibited limited behavioral flexibility, frequently repeating actions without evident improvements. Individual Performance (IP) did not manifest a statistically significant disparity among the clusters; nevertheless, a trend-level variation was detected. This indicates a plausible differentiation in children’s independent rhythmic execution, which necessitates additional inquiry. These findings suggest that technology-integrated HMEAYC enhances young learners’ engagement and adaptability. Future studies should explore targeted interventions to support children with minimal musical experience and conduct longitudinal research to assess the long-term cognitive and social impacts of technology-enhanced music education.

Introduction

In recent years, the integral role of music in early childhood development has gained increasing recognition. This investigation predominantly utilizes the theoretical construct of situated learning, which underscores the importance of acquiring knowledge through genuine, contextually relevant activities^1,2. Situated learning, as articulated by Lave and Wenger³, emphasizes learning through social participation and engagement in real-world tasks, making it an essential framework for understanding the developmental role of music in young children. There is substantial evidence indicating that music fosters creativity while supporting cognitive, emotional, and physical growth. Programs such as"Movement for Learning"have demonstrated that incorporating movement into daily activities enhances children’s physical development and academic outcomes, addressing concerns about the decline in physical abilities. Movement-based interventions also contribute to cognitive processing and behavioral control, particularly when aligned with educational objectives⁴. Rhythmic music, in particular, has been shown to foster self-regulation by enhancing impulse control, emotional management, and motor coordination—critical skills for managing cognitive and emotional responses^5,6. Further research, including investigations into rhythmic movement within cultural frameworks, further corroborates these assertions and underscores the extensive applicability of rhythmic learning methodologies across varied educational contexts^7,8. However, while these studies highlight the role of movement and rhythm in early learning, fewer studies have explicitly examined how technology-enhanced interventions mediate these effects. Given the increasing integration of digital tools in early childhood education, it is essential to explore how interactive music technology contributes to motor skill development and self-regulation. Prior research suggests that children’s engagement with digital music tools may support multisensory integration and cognitive flexibility, yet systematic investigations remain limited^9–11.

Advancements in technology have transformed traditional music activities into multisensory experiences, offering new ways to engage young children. Integrating technology within music education frameworks, such as the Holistic Music Educational Approach for Young Children (HMEAYC), has demonstrated significant benefits, particularly for children with developmental delays. By incorporating structured rhythm and movement activities, HMEAYC enhances self-control and social competence, helping children build skills in delayed gratification and effortful control, both of which are essential for social competence and academic success^11,12. Exposure to rhythmic music has been shown to promote motor coordination and synchronization, particularly within structured group settings that encourage collaboration and active participation¹³. This engagement leverages rhythm perception as a multisensory experience, where body movement influences auditory rhythm perception through interconnected auditory and vestibular systems^14,15. Despite the growing body of research on multisensory learning, gaps remain regarding how young children internalize rhythm-based learning experiences when technology is incorporated. Studies have shown that digital learning environments can facilitate kinesthetic and visual-auditory synchronization, yet their role in early childhood musical engagement is underexplored^16,17. In social learning contexts, such as group music activities, synchronized movements improve rhythmic coordination and promote social bonds, prosocial behaviors, and emotional regulation¹⁸. Thus, interactive, technology-based music interventions offer powerful tools for supporting physical and social development. Music technology plays a crucial role in enhancing learning behaviors by promoting active engagement and social interaction. Within the HMEAYC framework, combining technology with multisensory musical activities has been shown to improve children’s learning habits, communication skills, and academic engagement^11,19. However, it remains unclear how varying levels of prior musical experience shape children’s responses to digital rhythm interventions. While some research suggests that digital tools can scaffold musical skill development, further empirical evidence is needed to understand their role in fostering adaptability in rhythm-based learning^20,21.

These activities also cultivate early and digital literacy skills as children engage with digital interfaces, learning to navigate digital texts and images through touch, thereby enhancing cognitive engagement²². The collaborative nature of technology in musical play fosters creativity and exploration, enabling children to dynamically engage with musical concepts²³. Consequently, the HMEAYC framework, which draws from situated learning and rhythmic learning theories, offers a comprehensive approach to supporting young children’s development across physical, cognitive, and emotional domains. By integrating music education with cultural, cognitive, and emotional learning, HMEAYC adapts flexibly to meet children’s individual developmental needs, making it highly suitable for early intervention programs in Taiwan^24,25. Incorporating technology, such as a ceiling projector with musical rhythm, further enhances the HMEAYC’s ability to address diverse developmental needs by improving accessibility and adaptability. Building on this context, the present study examines the effects of integrating technology-enhanced musical rhythm activities on young children’s movement behaviors and social competencies within the HMEAYC framework. Specifically, this study investigates the sequential behavior patterns exhibited by children during movement activities in a musical rhythm context. By conducting a behavioral sequence analysis, this research aims to enhance understanding of how technology-enhanced music education supports essential movement skills, adaptability, and social competencies in early childhood development.

Methods

Participants

Participants in this study were 75 children (43 boys and 32 girls) aged 3–6 years, recruited from preschools in central Taiwan based on their abilities and prior experience in music-related activities. Among them, 52 children had previous exposure to HMEAYC music-related courses (Cluster 2), while 23 had no such experience (Cluster 1). The study investigated their initiation and movement behaviors within the HMEAYC framework in the context of music rhythm. The principal researcher conducted the instructional research, assisted by three trained observers who had undergone researcher-led training and three teaching assistants, all of whom had received training in HMEAYC. An independent t-test showed no significant difference in age distribution between Cluster 1 (M = 5.09, SD = 0.67) and Cluster 2 (M = 5.06, SD = 0.24), t(73) = 0.204, p = 0.840. Furthermore, Cohen’s d was calculated to measure the effect size of age differences between groups, yielding a value of 0.07, indicating a negligible effect³¹. This confirms that age differences between clusters were minimal and unlikely to influence the study outcomes. Similarly, a chi-square test showed no significant difference in gender distribution between the two clusters (χ² = 0.0, p = 1.0). These results confirm that age and gender were not confounding factors in participants. For the reference, Cluster 1 comprised 14 male participants and 9 female participants, whereas Cluster 2 encompassed 29 male participants and 23 female participants. These demographic particulars are presented for thoroughness, although statistical analyses indicated the absence of any significant disparities between the groups. Study considered the classroom environment, noise levels, and peer interactions. These variables were documented throughout the study to assess their possible effects on participant behavior. Prior research has highlighted the importance of structured learning environments in optimizing young children’s engagement and self-regulation, particularly when integrating interactive music technology²⁶. Thus, we ensured consistency in spatial arrangement, noise control, and interaction opportunities across all sessions. Prior exposure to digital interactive tools may also impact children’s responsiveness to technology-enhanced activities. Research indicates that children familiar with touchscreen music applications or motion-based rhythm activities engage differently than those encountering such technology for the first time. To address this, we recorded participants’past experiences with digital music tools or interactive applications as a potential factor in their adaptability^11,19,24. This study was conducted in accordance with relevant ethical guidelines and regulations, with ethical approval obtained from the Central Regional Research Ethics Committee at China Medical University, Taichung, Taiwan (Approval No.: CRREC-110-117). All experimental protocols were reviewed and approved by the IRB (Institutional Review Board). Before participation, written informed consent was obtained from the legal guardians of all participating children. To ensure confidentiality, all collected data were anonymized, and no personally identifiable information was recorded.

Materials and instruments

In examining children’s social competence and movement behaviors within the context of the Holistic Music Educational Approach for Young Children (HMEAYC), the researchers grounded their methodology in the principles of situated learning instruction (e.g²⁷.,). This theoretical framework informed both the study design and the selection of instruments used to analyze children’s interactions and movement patterns within the HMEAYC framework, which was enhanced by a ceiling projector displaying rhythmic visual cues (Fig. 1). The ceiling projector served as a critical tool, providing visual prompts that complemented the musical tasks and facilitated children’s understanding and engagement with rhythm. To ensure a rigorous analytical approach, additional statistical methods were applied to validate the coding scheme and behavioral categorization. This included examining inter-category transitions and behavioral interactions to assess the relationships between different engagement phases. Furthermore, potential external influences, such as classroom environment and teacher involvement, were considered when interpreting behavior patterns, ensuring a comprehensive understanding of the factors affecting children’s responses within the HMEAYC framework.

Fig. 1.

HMEAYC activity space featuring multisensory instruments for musical rhythm interaction.

The researchers initiated the study by randomly sampling video recordings of participants to conduct preliminary qualitative observations. These observations offered initial insights into how children interacted with HMEAYC activities and responded to the rhythmic cues from the ceiling projector. Building upon these observations, the researchers developed a comprehensive coding scheme to systematically explore and categorize children’s movement behaviors during the activities. The coding scheme encompassed six behavior categories—Observation, Execution, Correct Performance, Individual Performance, Abandonment, and Restart—each meticulously defined with descriptions and examples to ensure clarity and consistency in its application across observers (see Table 1).

Table 1.

Coding for children’s movement behaviors in the HMEAYC model.

Code (phase)	Behavior definition	Illustrative example
O (observation)	The child observes the teacher’s instructions and demonstrations	The child watches the teacher and imitates the teacher’s demonstration of the musical rhythm
E (execution)	The child follows the teacher’s instructions and initiates movements accordingly	The child imitates the teacher’s movements, follows the beat of the music, and performs the next instructed action
CP (correct performance)	The child performs movements accurately and in alignment with the learning content	The child accurately performs the movements in sync with the music rhythm as instructed
IP (individual performance)	The child independently interprets and performs movements, sometimes leading to varied or creative outcomes	The child creatively interprets and performs the actions
A (abandonment)	The child stops engaging due to difficulty or failure to complete the movement task	The child stops performing due to movements not being in sync with the rhythm of the music
R (restart)	The child abandons the current task but later reinitiates an attempt, demonstrating persistence or adaptability	After an incorrect performance, the child decides to restart by imitating new movements in sync with the music rhythm

Each behavior category represented a specific phase of engagement in the HMEAYC course. Transitions between behavioral modalities were systematically coded in a sequential manner, predicated upon observable shifts in actions. In instances where a child transitioned directly from one behavioral state to another (e.g., from Correct Performance to Individual Performance), the transition was recorded without the inclusion of intermediary codes such as’Abandonment’or’Restart’, unless such specific disengagement or re-engagement behaviors were distinctly observed. For instance, if a child creatively modified a task during performance without cessation, the coding would reflect CP → IP. Conversely, should a pause or pronounced interruption transpire, an Abandonment code would be integrated prior to the subsequent behavior. This elucidated logical framework has been incorporated into Table 1 and the Methods section to augment both transparency and replicability. Each behavior was logged as a single dominant action per unit of observation, based on the child’s most salient movement at that time. Transitions were recorded in temporal order, and intermediary codes were included only when clearly observed. A detailed explanation of the analytical methodology, including the use of Lag Sequential Analysis (LSA), is provided in the “Data analysis” section.

Each behavioral code was consequently regarded as an independent entity within the behavioral continuum, facilitating the development of sequential configurations that encapsulate engagement, disengagement, and adaptability. In addition to task-aligned performance, the coding scheme accounted for independent variations in behavior. The Individual Performance category identified instances where children engaged in movements autonomously, which could produce varied outcomes based on their interpretation of the task. If a child decided to stop attempting the task due to difficulties, this was recorded under Abandonment, indicating a break in their engagement. Conversely, the Restart category captured situations where a child, after abandoning an activity, chose to reinitiate or shift focus to a new task, demonstrating resilience or a renewed attempt to participate. By systematically analyzing these behaviors, sequential behavior transitions were examined to identify patterns of engagement and adaptability among participants. This allowed for a deeper investigation into how children progressed through different learning stages and responded to rhythmic stimuli within the HMEAYC framework. This detailed coding scheme enabled a structured analysis of children’s behavior, allowing researchers to identify patterns and correlations across different phases of engagement. Applying this scheme to video data provided a rich dataset to uncover underlying behavioral patterns within the HMEAYC context. Notably, the ceiling projector enhanced sensory engagement by reinforcing rhythmic learning, making it a valuable component of the HMEAYC framework. To further validate the behavioral categorization, statistical comparisons were conducted between participant clusters, assessing adaptability and learning progression differences. These analyses ensured that variations in movement behaviors were systematically examined rather than subjectively interpreted. To ensure the reliability of the coding process, the researchers calculated the kappa value of inter-rater reliability, which confirmed the consistency and accuracy of the observations. This methodological rigor underscored the value of the coding scheme in analyzing children’s movement behaviors within educational settings, particularly within the situated learning framework of HMEAYC courses, where multisensory engagement—supported by the ceiling projector—proved instrumental in children’s rhythmic and social development.

Procedure

The study was conducted over approximately 30–40 min within the context of the Holistic Music Educational Approach for Young Children (HMEAYC) courses. The researchers introduced participants to the methods of expressing musical rhythm activities, including the use of a ceiling projector displaying musical rhythms to enhance rhythmic elements and engage participants visually. A detailed explanation of the process was provided to ensure participants’understanding of the tasks and objectives.

Following this introduction, participants were given time to actively engage in the HMEAYC courses, during which they could explore various musical rhythm tasks and activities, guided by the visual cues projected from the ceiling. To account for potential external influences, we documented variations in classroom layout, lighting conditions, and technological familiarity, which could impact engagement and movement behaviors. Efforts were made to standardize these conditions across sessions to ensure consistency in the learning environment.

Communication and peer interaction were explicitly encouraged throughout the session, which is a core component of the HMEAYC pedagogical framework, where social learning is seen as essential for young children’s development. The researchers systematically recorded each participant’s actions and interactions during the process through video and observational data collection methods.

Given that prior exposure to digital music tools might influence engagement levels, we recorded participants’familiarity with interactive learning technologies. This allowed us to assess whether differences in engagement were related to technological adaptability rather than the intervention itself. These recordings provided a robust dataset for subsequent analysis, allowing researchers to assess individual movement behaviors, engagement with the activities, and the social dynamics that emerged throughout the course. This methodological approach ensured that the study captured both individual and collective learning experiences, with the ceiling projector enhancing sensory engagement. These data contributed valuable insights into the application of HMEAYC in promoting young children’s musical and social development.

Data analysis

This study utilized both real-time and video-based observations, applying a sequential analysis approach to examine children’s engagement in the Holistic Music Educational Approach for Young Children (HMEAYC). Video recordings captured over ten hours of footage from 75 participating children, documenting their music rhythm learning behaviors during the course.

The research engaged in a comprehensive array of statistical evaluations to ascertain the validity and robustness of our results. A sequence of statistical assessments was executed to verify the integrity of our findings. These assessments encompassed an independent t-test to analyze age discrepancies between clusters and a chi-square test to evaluate gender distribution. Thereafter, we executed an ANCOVA analysis, integrating age as a covariate, to assess the influence of the digital rhythm intervention while adjusting for age. In addition, methodologies involving bootstrapping were applied to validate the durability of the outcomes given the skewed sample sizes across clusters. These analytical procedures were integrated prior to the clustering and sequential behavior pattern analysis to guarantee that variations in sample size, age distribution, and gender did not introduce bias into the study outcomes.

These children’s behaviors were systematically coded using a predefined coding scheme, which included six behavior categories: Observation, Execution, Correct Performance, Individual Performance, Abandonment, and Restart (see Table 1). Given the large volume of video data, requiring detailed coding, three coders with expertise in early childhood education or development were recruited to ensure consistent and reliable coding. To establish inter-rater reliability, the coders collaboratively encoded the video data of a single child, achieving a Fleiss Kappa coefficient (k) of 0.86, which indicated high consistency among the coders.

Following this, each observed movement pattern was coded as a discrete unit according to the scheme, with codes organized chronologically to capture the sequential nature of the children’s behavior. To examine these encoded behaviors, we utilized Lag Sequential Analysis (LSA), a statistical technique specifically developed to identify temporal patterns within sequentially organized categorical data. In LSA, each observational unit is assigned a single, dominant behavior code, and transitions are recorded based on the natural temporal flow of actions. Consequently, intermediary behaviors (such as ‘Abandonment’ or ‘Restart’) are incorporated only when they are clearly observed. This approach facilitates the detection of statistically significant behavior sequences and their associated transition probabilities. Behavioral definitions and the coding framework that support this analysis are outlined in Table 1.

Results and discussion

Behavioral patterns and musical exposure in cluster analysis

The video data from the 75 participants revealed consistent occurrences of the six behavior codes across course sessions. Each participant’s total occurrences of each code were divided by the number of sessions attended to standardize behavior frequencies. To interpret the coded behaviors, a two-stage cluster analysis was conducted to identify distinct learner profiles based on behavior frequencies.

First, hierarchical clustering analysis determined the number of clusters. This was followed by a k-means clustering analysis applied to the identified cluster number, which allowed for a detailed examination of the behavioral characteristics within each cluster. After establishing the clusters, the study analyzed the encoded data sequentially to identify patterns in behavior sequences (e.g^28,29.,). This sequential analysis involved creating a transfer matrix of behavioral frequencies, a conditional probability matrix, and an expected value matrix. The results were further supported by generating an adjusted residuals table (z-score table) to facilitate data interpretation and provide a clearer understanding of the sequential patterns.

To explore how social competence influenced movement behaviors, participants were grouped based on their prior experience with HMEAYC courses. Additionally, children were categorized according to their Social Competence status to analyze behavioral differences across varying social competence levels. Using Kelly’s³⁰ recommendation for categorizing data by the top and bottom 27% cutoffs, students were classified into high, medium, and low Social Competence groups based on their scores from the Social Competence Assessment System for Preschool (SBASP). Those scoring in the top 27% formed the high Social Competence group, those in the bottom 27% constituted the low Social Competence group. At the same time, the remaining students were designated as the medium Social Competence group. A sequential behavior pattern analysis was then conducted across these social competence groups to elucidate behavioral differences, particularly how social competence influenced engagement and response patterns within the HMEAYC framework. This approach provided a nuanced view of how prior HMEAYC experience and social competence levels shaped children’s movement behaviors and engagement sequences, offering insights into how these factors contribute to learning outcomes in music-based educational settings (Dennis 2008).

The analysis of Table 2: Cluster-based Comparison of Movement Behavior Frequencies with ANOVA Results highlights significant variances in the average frequency of six core movement behaviors—Observation (O), Execution (E), Correct Performance (CP), Individual Performance (IP), Abandonment (A), and Restart (R)—between the two identified clusters of children. The data show substantial differences in several behaviors, as evidenced by their F-values. Specifically, Observation (O) had an F-value of 47.84 (p =.000), Execution (E) had 9.01 (p =.000), and Correct Performance (CP) had 34.32 (p =.000), indicating strong statistical significance. Abandonment (A) also showed a significant difference between clusters, with an F-value of 15.38 (p =.000). Restart (R) demonstrated the highest variance with an F-value of 58.90 (p =.000). Although the difference in Individual Performance (IP) between clusters did not reach conventional levels of statistical significance (F = 3.58, p =.063), the result indicated a trend-level distinction that may reflect underlying behavioral differences related to independent rhythm execution.

Table 2.

Cluster-based comparison of children’s movement behavior frequencies with ANOVA.

Indicators of cluster analysis	Cluster 1 (n = 23)	Cluster 2 (n = 52)	F	P
O (observation)	1	3	47.84	.000
E (execution)	3	4	9.01	.000
CP (correct performance)	1	2	34.32	.000
IP (individual performance)	2	2	3.58	.063
A (abandonment)	3	2	15.38	.000
R (restart)	1	3	58.90	.000

Frequencies reflect standardized average occurrences of each behavior per participant across sessions. F-values result from one-way ANOVAs comparing the two clusters. Observation (O): Watching teacher/instruction, Execution (E): Attempting a task, Correct Performance (CP): Accurate execution, Individual Performance (IP): Creative/independent attempt, Abandonment (A): Disengaging from task, Restart (R): Reinitiating task after pause.

In order to enhance the clarity of statistical transparency, precise p-values are now enumerated for all comparisons delineated in Table 2. The clusters encompass 23 children within Cluster 1 and 52 children within Cluster 2. Cluster 1 included 8 children with prior experience in music-related activities, accounting for 34.78% of that group, while Cluster 2 included 44 such children, representing 84.62% of that group. Across the entire sample, 52 out of 75 children (69.33%) had prior music experience. These percentages signify the distribution of musical experience both within and across the designated clusters. While IP did not achieve statistical significance, the near-threshold result indicates a possible trend worth further exploration. This observation may reflect developmental differences in self-regulatory strategies and independent learning behaviors that are not yet robust enough to produce significant group-level effects but suggest meaningful directions for future inquiry. Prior research has demonstrated that musical exposure enhances self-regulatory skills, impulse control, and attention span in early learners, which may explain why Cluster 2 learners exhibited greater adaptability in movement-based learning^9,11. This suggests that prior musical exposure may play a pivotal role in shaping learning behaviors, particularly by positively correlating with enhanced self-regulation. However, further research is needed to clarify potential causal relationships and consider other developmental factors, such as environmental influences, teacher scaffolding, and peer interactions, which may also contribute to these behavioral differences. Studies on digital learning environments indicate that technology-assisted interventions can serve as effective scaffolding tools, particularly for learners with limited prior experience, by enhancing multisensory engagement and structured rhythm-based learning^10,16. This aligns with previous research findings highlighting the benefits of music exposure on self-regulation and social skills^5,6.

Sequential analysis of task engagement patterns in HMEAYC activities

The sequential analysis of the six behavior codes revealed distinct patterns in each cluster’s engagement style, highlighting differences in how children approached tasks during HMEAYC activities (Table 3). In order to enhance the comparability across clusters, the sequence of behavioral categories (encompassing both rows and columns) has been uniformly established. Only statistically significant transitions (z > 1.96, Bonferroni-corrected) are included in Table 3. All z-scores were evaluated with Bonferroni correction for multiple comparisons (α = 0.05/number of comparisons). Only transitions with z > 1.96 (p < 0.05) are displayed.

Table 3.

Adjusted z-scores of behavioral transitions for two clusters (Bonferroni-corrected).

Cluster	Code (phase)	O	E	CP	IP	A	R
Cluster 1	O (observation)	− 2.65	8.53***	− 1.69	− 2.38	− 1.75	− 1.18
	E (execution)	− 3.35	− 4	4.86**	2.14*	0.7	− 0.08
	CP (correct performance)	− 1.55	− 1.82	− 1	5.16***	− 0.07	2.13*
	R (restart)	− 2.31	− 2.73	− 1.48	− 2.08	1.81	− 1.04
	A (abandonment)	− 1.61	− 1.9	− 1.04	− 1.46	− 1.08	− 0.74
	IP (individual performance)	− 1.06	− 1.24	− 0.69	2.2*	2.11*	− 0.49
Cluster 2	O (observation)	− 1.1	5.08***	− 1.04	− 1.15	0.06	− 1.1
	E (execution)	− 1.56	− 2.07	0.67	2.26*	2.4*	0.48
	R (restart)	− 1.03	− 1.36	− 0.98	0.81	− 0.92	− 1.03
	A (abandonment)	− 1.17	− 1.54	− 1.1	− 1.22	− 1.04	− 1.17
	CP (correct performance)	− 0.96	− 1.27	2.44*	− 1.01	− 0.86	3.28**
	IP (individual performance)	− 1.1	− 1.45	1.91	1.53	− 0.98	− 1.1

All z-scores were evaluated with Bonferroni correction for multiple comparisons (α =.05/number of comparisons). Only transitions with z > 1.96 (p <.05) are displayed. Transition directions are derived from Lag Sequential Analysis (LSA). Behavioral codes: O = Observation, E = Execution, CP = Correct Performance, IP = Individual Performance, A = Abandonment, R = Restart.

*p <.05, **p <.01, ***p <.001.

In Cluster 1 (Fig. 2), learners showed a linear, task-focused approach with limited reflective learning. The most prominent transition observed was from O to E, yielding a z-score of 8.53 (p < 0.001). This pattern suggests that learners in this cluster swiftly moved from observing the teacher to actively engaging in task execution, yet they exhibited fewer transitions to reflective behaviors, such as R or CP after an initial error. This engagement aligns with findings that movement-based activities can effectively stimulate participation by encouraging immediate execution⁴. However, although Cluster 1 learners actively engaged at the start, they faced challenges sustaining the quality of their performance and adapting to task-related difficulties, highlighting the need for reflective processes to enhance learning outcomes. Prior studies have suggested that learners with minimal prior exposure to structured rhythm-based activities may rely more on direct imitation rather than developing adaptive strategies for self-correction, which could explain the observed limitations in Cluster 1²⁰.

Fig. 2.

Behavioral transition diagram of Cluster 1.

To further investigate behavioral transitions, we analyzed sequential dependencies within each cluster. In Cluster 1, the transition from Execution (E) to Individual Performance (IP) (z = 2.14, p < 0.05) suggests that learners frequently engaged in self-directed task execution without necessarily aligning their performance with the intended learning objectives. Research on digital scaffolding tools suggests that structured feedback mechanisms can help learners with limited prior experience transition more effectively from execution to correct performance by reinforcing iterative learning strategies¹⁷. In contrast, Cluster 2 exhibited a more structured engagement pattern, with stronger transitions from Execution (E) to Correct Performance (CP) (z = 2.26, p < 0.05) and from Correct Performance (CP) to Restart (R) (z = 2.44, p < 0.05), indicating that learners were more likely to refine their performance iteratively rather than persist in ineffective execution patterns.

Lack of Statistical Comparisons and Their Inclusion in Analysis Additional statistical analyses, including ANCOVA and bootstrapping, confirmed that these observed patterns were not solely influenced by prior musical experience, reinforcing the role of technology-enhanced interventions in shaping learning behaviors. An ANCOVA controlling for prior music experience indicated that technology-enhanced interventions significantly affected adaptability (p < 0.05), independent of previous exposure. Furthermore, bootstrapping analysis validated the robustness of these findings, ensuring that the sample size discrepancy did not bias the results. Previous research highlights that technology-supported musical interventions can foster cognitive flexibility and problem-solving skills, which may contribute to the observed performance differences between the two clusters²¹.

In contrast, research suggests that technology-enhanced music activities, which promote a multisensory learning environment, improve adaptability and persistence through reflective iteration, thereby supporting children’s problem-solving and self-regulation skills^11,19. In Cluster 2 (Fig. 3), the strongest transition observed was between Observation (O) and Execution (E), with a z-score of 5.08 (p < 0.001), indicating that learners quickly engaged with tasks after observing the teacher’s instructions. However, learners in this cluster demonstrated notably advanced learning strategies. A significant transition from Execution (E) to Correct Performance (CP) (z = 2.26, p < 0.05) suggests that these learners not only engaged with the tasks but also consistently performed them correctly, reflecting a strong understanding of task requirements and an ability to apply their observations to achieve accurate performance effectively. Further statistical comparisons validated these differences, with Cluster 2 learners showing significantly higher adaptability in iterative learning processes (p < 0.05). Another notable transition within Cluster 2 was from Correct Performance (CP) to Restart (R) (z = 2.44, p < 0.05), suggesting that learners frequently revisited tasks even after successful completion. This pattern indicates a tendency toward reflective learning, where learners seek task completion, refinement, and improvement through repeated attempts. This aligns with studies indicating that structured musical interventions foster metacognitive awareness and enhance children’s ability to self-correct¹⁶.

Fig. 3.

Behavioral transition diagram of Cluster 2.

These findings align with previous studies demonstrating the impact of multisensory and technology-supported learning environments in fostering self-regulation and persistence^11,19. The sequential behavior patterns identified in both clusters reveal notable differences in how children engaged with movement tasks during Musical Rhythm activities. Children in Cluster 2, with a higher proportion of prior music experience, demonstrated advanced reflective behaviors and a propensity for iterative learning, especially when addressing mistakes. In contrast, children in Cluster 1 showed more limited sequences, often failing to adjust strategies after errors. These results support prior findings that technology-enhanced music interventions contribute to developing self-regulation and adaptability^13,18. Furthermore, these findings underscore the role of interactive and multisensory learning approaches in facilitating deeper engagement, reinforcing the educational value of integrating digital tools within early childhood music instruction.

Our findings also contribute to the broader discourse on digital interventions in early childhood education. Prior research has shown that structured, technology-supported music learning enhances cognitive flexibility and problem-solving skills²³. This study extends this understanding by demonstrating how sequential task engagement varies across learners with different levels of prior music experience. Future studies should examine how technology-mediated scaffolding techniques, such as adaptive rhythm guidance or real-time feedback mechanisms, could enhance self-regulated learning in early childhood education. In conclusion, this study highlights how technology-driven interventions enhance engagement and self-regulated learning, particularly for children with varying levels of musical experience. Future research should further investigate how different levels of prior exposure impact long-term learning trajectories and whether structured interventions can mitigate gaps between learners. These insights provide valuable implications for designing early childhood music education programs that incorporate technology to foster engagement, adaptability, and creative exploration.

Conclusion

The findings of this study reveal clear distinctions in how children with varying levels of prior musical experience engage in movement tasks within HMEAYC Musical Rhythm activities. Cluster 2 (children who had previous exposure to HMEAYC music-related courses), comprising children with more extensive prior music experience, demonstrated a more advanced approach characterized by reflective behaviors, adaptability, and iterative engagement with tasks. This group showed a stronger ability to revisit tasks, refine performance, and explore independent execution, suggesting that prior exposure to music education fosters essential skills like self-regulation, persistence, and strategic learning. In contrast, Cluster 1 learners (with no prior HMEAYC exposure) adopted a more task-focused approach with limited adaptability. Frequent task abandonment and repetitive actions without self-correction revealed difficulties with reflective learning and persistence. These findings underscore the value of incorporating targeted interventions that emphasize scaffolded reflection, structured self-assessment, and adaptive feedback mechanisms—particularly for children with limited musical experience.

Based on the aforementioned findings, it is advisable for educators to incorporate reflective and adaptive pedagogical strategies within the HMEAYC framework. Systematically organized rhythm-oriented activities that encompass components of self-assessment and progressive mastery can cultivate persistence and adaptability in young learners. Moreover, the integration of technology-enhanced instruments—such as interactive digital platforms or rhythm-based applications—can facilitate multisensory engagement and promote iterative practice as well as self-evaluation. Such instruments not only aid in the refinement of skills but also contribute to the development of metacognitive awareness and problem-solving methodologies. Educators might also contemplate the utilization of real-time feedback and structured task revisitation to foster reflective learning environments. These methodologies are congruent with self-directed behaviors observed in children possessing heightened musical experience and may assist all learners in gradually enhancing resilience and self-regulation. This viewpoint is further corroborated by empirical research demonstrating that adaptive feedback significantly improves children’s resilience and emotional regulation within rhythm-focused learning environments.

While this study provides valuable insights into the relationship between prior musical experience and sequential behavioral patterns, future research could explore how cultural and educational differences shape the implementation and outcomes of HMEAYC. Specifically, comparing early childhood environments in Taiwan with those in Western countries may reveal how pedagogical frameworks and child behavior interact within varied sociocultural contexts. These cross-cultural investigations could yield critical insights for global applications of the HMEAYC model.

Prior to the commencement of the data collection phase, a priori power analysis conducted following the methodological framework outlined by Cohen³¹ establishing the alpha level at.05 and aspiring for a power of 1 minus beta at.80, with a specific focus on a one-way ANOVA model incorporating an expected medium effect size represented by f = 0.25; the findings from this analysis demonstrated that a minimum of 66 participants would be requisite to attain the requisite statistical power for our investigation. As a result, our conclusive sample size of 75 participants not only fulfills but also exceeds this predetermined criterion, thereby furnishing a solid foundation for our analytical procedures. However, we acknowledge that including an even larger sample size would substantially enhance the generalizability of our results across various populations and contexts. The relatively limited sample and the controlled experimental context may not adequately encapsulate the intricacies inherent in real-world classroom dynamics. Future research endeavors could investigate the influence of peer interactions, instructional methodologies, and classroom environments on learning behaviors within technology-enhanced music activities. Furthermore, longitudinal research designs may more effectively capture developmental variations within the HMEAYC framework over time.

Furthermore, exploring the effects of various types of music technology could offer deeper insights into the most effective tools for fostering adaptive learning. Longitudinal studies focusing on the long-term retention of adaptive learning strategies could further validate the role of multisensory engagement in sustained cognitive and motor skill development. This could ultimately contribute to the refinement of music-based educational frameworks, supporting the development of self-regulation and persistence in early childhood education. By integrating iterative learning strategies with technology-enhanced interventions, future research can further expand on how young learners develop adaptability and problem-solving skills, ensuring a more comprehensive understanding of their developmental trajectories in music education.

In practical applications, educators specializing in early childhood development can utilize these insights by integrating systematic reflection prompts within rhythm-oriented activities. For instance, educators may motivate children to articulate their actions verbally, reassess prior attempts, or analyze various movement strategies. When integrated within the HMEAYC framework, these reflective approaches may cultivate metacognitive advancement and adaptive learning methodologies in young children, ultimately contributing to enhancing musical and socio-emotional development.

Author contributions

L.L. (Liza Lee) designed the instructional framework and implemented the teaching activities. L.L. also supervised the research process and integrated the essay writing. H.J.H. (Han-Ju Ho) processed and analyzed the data, conducted the statistical analysis and contributed to the manuscript drafting. All authors reviewed and approved the final version of the manuscript.

Data availability

The analyses during the current study are not publicly available but could be accessed at a reasonable request from the corresponding author. The data have been made available to the journal for review purposes. Please visit: https://docs.google.com/spreadsheets/d/1gJvlzYHF8jJyrYTNiX1Qu3WvllPG0CrT/edit?usp=sharing&ouid=103465328735442855476&rtpof=true&sd=true.

Declarations

Competing interests

The authors declare no competing interests. The first author, L.L. and corresponding author, H.J.H., affirms that this statement is made on behalf of all authors.