Enhancing dance learning by optimising instructional perspective: A preregistered study on the effect of dancers’ viewing perspective, cognitive load and expertise

Elisabetta Versace; Manuela Angioi; Kristin Hadfield; Vincent Gallo; Lars Juhl Jensen; Juncal Roman; Karen Sheriff; Emma Redding; Frances Clarke; Julian Deering; Dylan Morrissey; Stephen Buckingham

doi:10.1371/journal.pone.0355860

Abstract

While movement sequences in choreography are often taught through instructor demonstrations, the influence of visual perspective on learning outcomes has received limited attention. Drawing on cognitive load theory, we investigated how the requests imposed by different instructional perspectives affect learning and performance. We compared the effects of Back view (first-person), Front mirror view (third-person mirrored), and Front opposite view (third-person) between dancers. We hypothesized lower cognitive load and better performance in the Back view, particularly for less experienced dancers. After preregistering the study, 32 dancers with varying levels of experience learned a moderately complex choreography from pre-recorded videos, in the three perspectives. Five dance experts assessed movement accuracy, spatial skills, and rhythm/dynamics using a validated scoring system and a new open-source digital tool. The Back view consistently yielded superior outcomes, especially among less trained dancers. No significant differences emerged between the two third-person views, possibly due to sample size or dancer background. These findings suggest that first-person perspectives may support motor learning, with performance patterns consistent with reduced cognitive load. Practically, camera perspective should be considered in online instruction, while first-person views may benefit early in-person training. Our results highlight the role of visual perspective in learning choreography and complex motor routines, offering both theoretical and applied contributions, including validation of a novel digital assessment tool.

Citation: Versace E, Angioi M, Hadfield K, Gallo V, Jensen LJ, Roman J, et al. (2026) Enhancing dance learning by optimising instructional perspective: A preregistered study on the effect of dancers’ viewing perspective, cognitive load and expertise. PLoS One 21(9): e0355860. https://doi.org/10.1371/journal.pone.0355860

Editor: Emiliano Cè, Università degli Studi di Milano: Universita degli Studi di Milano, ITALY

Received: January 15, 2026; Accepted: July 27, 2026; Published: September 16, 2026

Copyright: © 2026 Versace et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files.

Funding: This study was funded by the Economic and Social Research Council research grant ES/V015354/1 granted to EV, MA and DM. EV was also funded by the Royal Society Leverhulme Trust fellowship SRF\R1\21000155. DM acknowledges the support of the National Institute for Health Research Barts Biomedical Research Centre (NIHR203330).

Competing interests: The authors have declared that no competing interests exist.

Introduction

Millions of people worldwide learn some form of dance, either recreationally or professionally, from a teacher present in the same room or from screens [1–3]. For all these different approaches it is still not clear what the most effective ways are to teach a piece of choreography, or any sophisticated motor sequence. This study investigates how the different visual perspectives in which a dancer sees the instructor do influence choreography learning and performance. While we focus on dance, similar issues apply to other sports, aesthetic disciplines in particular [4].

In sports settings, training load management has been implemented to optimise performance and reduce the risk of injury [5,6]. While the role of physiological and mechanical load has been emphasized [5,7–9], the impact of cognitive load on learning motor sequences is only starting to be understood. Cognitive load refers to the amount memory and other cognitive resources involved in task execution and learning [10].

A high cognitive load can negatively impact learning and performance, as observed in the laboratory [11–13] and in different sports settings. For instance, rowers engaged in solving arithmetic problems reduce movement complexity [14,15], team ball athletes show lower agility when engaged in cognitively demanding tests [16], and dancers reduce postural control when engaged in concurrent mental tasks [17]. The impact of cognitive load on the acquisition of complex motor sequences, such as dance choreographies, is less understood, although we know that expert dancers can chunk movement sequences with lower effort [18] and potentially engage with lower attentional effort than novices [19].

In dance, perceptual, cognitive, motor, emotional and social demands are tightly integrated [20–22], as learners must simultaneously process multimodal information (visual, auditory and proprioceptive) movement sequences, spatial orientation and timing, often under time constraints, artistic demands and social pressure. Thus, understanding conditions that enhance dance learning is relevant not only for the wider field of dance but also to understand the mechanisms underlying complex motor learning more broadly [23]. One of the important factors in play is cognitive load.

Cognitive load builds from different components, including intrinsic and extraneous load. First, the intrinsic load refers to the inherent difficulty of a task, which is influenced by the expertise of the learner [24,25]. For instance, choreography that is cognitively demanding for a novice may impose little cognitive load on an expert. We used the current amount of training as a proxy for expertise, as previous studies showed that dancers with more experience are better in chunking and recalling dance sequences [26], and that current practice is strongly associated with dance performance [27]. According to the well-established cognitive load theory of the expertise reversal effect, task demands differentially affect learners based on their existing schemas [24].

Second, the extraneous load refers to how the instructional material is presented (rather than to the intrinsic characteristics of the learned material), with the effect of hindering or supporting learning [28]. For instance, slower instructions enhance learning in tasks with medium and high complexity [29]. In this work, we study the effect of the extraneous load imposed by the instructional point of view. An efficient instructional design should reduce the extraneous cognitive load [30,31]. We used a between-subjects design to investigate the impact of three different teacher viewing perspectives on dance learning (see Fig 1): (1) first-person view from behind the teacher (Back view, hypothesised to impose the lowest extraneous cognitive load, Fig 1a), (2) third-person view mirrored, namely the viewpoint of someone facing the mirror image of the instructor (Front mirror view; expected to impose a moderate extraneous cognitive load, Fig 1b), and (3) third-person view without a mirror image (Front opposite: expected to impose the highest cognitive load, Fig 1c). We focused on a choreography of medium complexity, that required a minimum of three years of ballet training to be learnt in two teaching sessions of approximately 60 minutes.

Download:

Fig 1. The three different perspectives of the teacher.

(a) First person: Back view, where participant could simply follow the choreographer. (b) Third person: Front mirror view: where participants could move the limb on the same right/left side of the choreographer. (c) Third person: opposite view: where participants could move the limb on the same right/left side of the choreographer.

https://doi.org/10.1371/journal.pone.0355860.g001

Based on the cognitive load theory, our predictions were as follows: that the extraneous cognitive load is lower (and performance better) in the first-person point of view, intermediate in the third person mirrored and higher in the third person non-mirrored condition. Cognitive studies on mental rotation and mental imagery have shown longer processing times (higher cognitive load) for objects and bodies that must be “mentally rotated” at greater angular distances to perform matching tasks [32–34]. Recognition of mirror images, such as those of a teacher demonstrating a sequence in front of the learner, depends on the point of view of the observer [35]. Moreover, different studies have found a better accuracy of execution after participants had viewed an instructional video that demonstrated simple step-by-step procedures from a first-person perspective, compared to a third-person perspective [36–39]. However, a first-person perspective can also hinder dance learning, as the face and gestures of the instructor are less visible, and this is a non-canonical view, to which practitioners are less exposed.

To investigate whether learning a choreography is more effective when taught from a first-person perspective, a mirrored third-person view, or a standard third-person perspective, we used pre-recorded videos of one choreographer explaining and demonstrating a dance sequence from different points of view, as shown in Fig 1.

Validity and applicability of the experimental setting

In this study, we investigated the cognitive load involved in acquiring a complex motor sequence by examining dancers as they learned a new solo piece of choreography. The task is directly relevant to real-life scenarios, as both in training and professional settings dancers learn new sequences to be recalled and performed later, independently. This exposes dancers to significant cognitive demands. Moreover, dancers are described as “performing” or “aesthetic athletes” [20], due to the simultaneous physical and artistic challenges of their discipline. This unique combination of demands makes dancers an ideal model for studying cognitive load and learning from instructions, with potential relevance to other contexts.

The setting based on teaching from different perspectives is ecologically valid, as in dance training and professional practice, exercises and choreographies are taught by teachers located in front of the learner, performing the same sequence from their own point of view, reversing teachers’ instructions between the right and left side, and from the third to first person points of view. For instance, from early stages of ballet training, exercises at the barre are usually taught from an ego-centric frame of reference, with all pupils starting with the left hand on the barre, irrespective of their position with respect to the teacher. Moreover, not only do dancers often learn choreography from video recordings taken from the point of view of the audience (in third person), but online instruction has become ubiquitous in professional, vocational and recreational settings since the expansion of digital platforms and social media (e.g., TikTok, Instagram, YouTube) and the transformation of teaching imposed by the COVID-19 pandemic, [2,3]. In the video formats, dancers, demonstrators and teachers are typically viewed from a third-person perspective.

Despite most dance teaching happening in the third person perspective, the work by Fiorella et al. [36] suggests that learning simple motor sequences in the first perspective from instructional videos is more effective than learning the same task in third person videos. Moreover, Koning et al. [38] showed that learning the manipulative-procedural task of knot-tying from video animations is more effective from the first-person compared to the third person perspective. Whether the same principle applies to sports and aesthetic disciplines such as dance, remains unknown.

Aims and hypotheses

The main aim of this study was to clarify the role of the viewing perspective as a source of extraneous cognitive load. We tested the hypothesis that dance learning is enhanced when the teacher is seen in first-person perspective (Back view), compared to when they are seen in third-person perspective (Front view), as predicted by the reduced cognitive load of the first-person view. We also predicted that the third-person mirror view (Front mirror) produced a better performance compared to the third-person non-mirror view (Front opposite), as mentally rotating an image produces a higher cognitive load.

The second aim of this study was to test the effect of the expertise (intrinsic cognitive load) and its interaction with the point of view (extraneous load), in enhancing dance learning. Based on the expertise reversal effect theory, less expert practitioners were expected to benefit more from the reduction of cognitive load compared to more expert practitioners.

We also had the applied goal of measuring the reliability of our open-source digital tool MotionPerfection [40], implemented at www.wemovetogether.uk, to support the objective assessment of dance performance.

Materials and methods

Preregistration

The main aim of this study was preregistered on AsPredicted.org prior to data collection (Registration number: 66,583, 22 May 2021). The preregistration included hypotheses, study design, and planned analyses.

Participants

After ethical approval from the Queen Mary University Research Ethics Committee (QMREC2020/34), we recruited voluntary participants among active dancers aged 18 years or older (range 18–43 years old, M = 23, SD = 5.15), with at least 3 years of ballet training. Dancers were recruited using a convenience sampling approach via advertisements at three London-based vocational dance schools and through affiliated dance teachers. Recruitment started on June 2021 and ended on January 2023. We recruited a total of 32 participants (27 females and 5 males). One participant was excluded due to a procedural error (incorrect point of view presentation), resulting in a final analysed sample of 31 participants (26 females, 5 males). Participants were drawn from three different backgrounds: vocational ballet schools (n = 4), vocational contemporary dance schools (n = 8), and independent dancers not enrolled in vocational training (n = 19). Participants were stratified by sex and randomly assigned to different experimental conditions.

The pre-registered design targeted a relatively homogeneous sample of highly trained dancers, with an expected small effect size (f = 0.26; α = 0.05, power = 0.8; target N ≤ 147). As shortly before data collection lockdown restrictions were lifted and in-person teaching resumed, recruitment opportunities were substantially reduced for the vocational students initially targeted. To address this challenge, we broadened recruitment to include dancers with a minimum experience of three years of ballet, which was needed to match the requirements of the choreography. While this introduced greater variability in participants’ expertise, it increased the expected effect size and allowed us to explore a key aspect of cognitive load theory: the differences in learning between different levels of expertise.

Experimental design

In this confirmatory pre-registered study, we investigated the effect of cognitive load imposed by observing the dance teacher from different points of view, testing participants in three experimental conditions (see Fig 1): first-person view: Back (lowest expected cognitive load), third person view: Front matched (intermediate expected cognitive load), and third person view: Front opposite (highest expected cognitive load). Cognitive load was not directly measured; instead, we interpreted differences in performance across conditions as reflecting variation in processing demands induced by the instructional perspective. The use of primary task performance to infer cognitive load dynamics is validated methodology [31,41,42].

To assess dancers’ performance, dance experts evaluated: Control of movements; Accuracy of movements; Technique; Spatial Skills; Dynamics and Rhythmical accuracy; Performance qualities. Assessment criteria, adapted from Angioi et al. [20] are summarised in S1 Table. Dance experts also assessed the proportion of errors in body segment positions (matching of joint positions between a model and individual dancers, in 13 poses). As this variable did not allow discrimination between participants, and did not capture the complexity of dance dynamics and artistry, we focused on the other measures of performance. Because dancers came from different training backgrounds, this was accounted for by using as covariate the current training expertise of dancers, based on the weekly hours of dance and physical training at the time of the experiment.

Apparatus

The dancers watched the recording of the teacher from a monitor connected to a laptop, during a recorded Zoom call. The dance floor was a square, 2 m side, delimited by coloured tape. Experimenters helped the dancer to setup the square making sure that dancers were fully visible throughout the recording. To support and systematise the assessment, scoring and annotation, the dance experts assessed dancers’ performance using the digital platform www.wemovetogether.org, that implements our open access framework [40].

Stimuli and test

Participants were randomly assigned to three conditions: either Back (for participants to copy), Front mirror (for participants to match the image seen, as in a mirror) or Front opposite (for participants to flip the steps, as in an interaction with a teacher located in front), see Fig 1.

Conditions differed only in the point of view of the teaching recording. To make sure that participants assigned to the different conditions had the same instructional information, we video recorded the choreographer simultaneously from the front and back points of view, while he provided verbal and visual instructions. The teacher (Mr Erico Montes) was an expert ballet master, ballet teacher and choreographer, former professional dancer at the Royal Ballet, with hundreds of hours of teaching experience. To obtain the Front mirror video, we flipped the left/right side of the Front opposite image. The same audio file was used for all three videos. Videos were recorded using a GoPro Hero 7 camera.

The choreography lasted 1’15” overall. In the first session, the entire choreography was taught. In the second and last session, the instructor rehearsed the choreography, to prepare dancers for the performance. To reduce the risk of injury, the choreographer included no high impact jumps and only a single turn [43,44]. In line with this, no injuries or physical issues have been recorded during the experiment.

Procedure

All participants took part in two sessions, on two consecutive days. After recruitment, participants were given an information sheet, had the opportunity to ask questions, and then signed the consent form. We recorded demographic information before Session 1, and feedback on the project at the end of Session 2.

In Session 1, participants completed the demographic questionnaires and then proceeded to the video-recorded session. The video-recorded session included a 20-minute warm-up, followed by a 60-minute choreographic session led by the choreographer. To familiarise participants with the performance format, Session 1 ended with two solo attempts, where participants performed the choreography without any external feedback. In Session 2, after the warm-up and a 45-minute rehearsal session, participants performed the choreography twice, without any external feedback. We assessed performance using the second recording.

Video analysis

Objective assessment is a key challenge for aesthetic disciplines such as dance [45]. To support and standardise this task, dance experts conducted structured video-based assessments on the freely available online platform https://www.wemovetogether.uk/, based on the open source software MotionPerfection [40]. The software is designed for the visualisation, synchronisation and annotation of videos that display motor sequences with an acoustic component to anchor the execution time.

Dance experts assessed the execution using the Aesthetic Competence Tool [20], which is based on audition criteria of international dance schools and dance companies across the United Kingdom, Australia and USA. Our analyses focused on the three criteria of the Aesthetic Competence Tool that correspond to our pre-registered measures of interest: Accuracy of movement, Spatial skills, Dynamics/timing/rhythmical accuracy. Dance experts scored dancers for all the seven criteria of the scale, and we observed (as an exploratory analysis) that all the measures (including also Control of movement, Technique; Performance qualities and Overall performance) exhibited the same pattern of the pre-registered variable of interest.

Participants’ performance was assessed by five dance experts: four with direct experience in vocational training and dance performance, one with dance training. Dance experts were blind to the experimental condition and level of training of each participant. Each participant was assessed by at least two experts, except for Participant 16, 18, 31, who were assessed by one dance expert only, due to limited availability of the experts.

Each performance variable was rated on a scale from 10 to 0, where 10 was the best possible execution, 1 an extremely poor execution and 0 no execution. All dancers executed the choreography. Experts also assessed the individual pose accuracy of body segment position, focusing on 13 core poses compared to the position held by the choreographer. As this approach was not sensitive enough to participant differences, due to the low number of errors, we focused the analyses on the established artistic elements of choreography.

Dance experts further quantified the number of errors looking at 13 static poses during the choreography, for which the teacher explicitly trained the participants, comparing the static executed pose, with that of a “ground-truth” model. Due to the low number of errors in the static poses and to the dynamic nature of dance, this measure was not sensitive enough. Moreover, all dance experts commented that the evaluation of poses was not adequate to capture the core of dance performance. Based on this, data analyses focused on the Aesthetic Competence Tool results only.

Data analysis

Analyses and plots were performed conducted in R (R version 4.4.2 (2024-10-31) via R-studio 2023.12.1.402). Data are available in the S1 File. For all analyses we set a significance level of p ≤ 0.05.

We initially assessed the inter-rater reliability using Intraclass Correlation Coefficients (ICC) [46,47], as visualised in Fig 2. We first examined a subset of three participants (P12, P14, P15) who had been rated by all five experts. This subset analysis revealed substantial agreement among individual raters (ICC[2]=0.66) and excellent reliability when averaging across raters (ICC[2k]=0.91). While based on a small sample (n = 3), this initial result demonstrated strong consistency when complete rater coverage was present. To validate these findings more broadly, we then assessed inter-rater reliability across the full sample of participants, including those with some missing ratings. The ICC analysis showed moderate to substantial agreement among individual raters (ICC[1]=0.58, ICC[2]=0.59, ICC[3]=0.69), and excellent reliability when averaging scores (ICC[1k]=0.87, ICC[2k]=0.88, ICC[3k]=0.92). These results support the use of averaged expert ratings as a reliable measure of movement accuracy in this context.

Download:

Fig 2. Heatmaps showing expert ratings for the participants across three performance variables: (A) Accuracy of movements, (B) Spatial skills, and (C) Dynamics and Rhythmical accuracy.

Each row corresponds to a participant, each column represents one of five expert raters (A–E). Ratings range from 1 (poor) to 10 (high), with darker shades indicating higher accuracy. Grey cells indicate missing ratings. The use of a consistent colour scale across all three plots enables direct comparison between variables. Three participants (P12, P14, P15) have been rated by all experts.

https://doi.org/10.1371/journal.pone.0355860.g002

Before inferential analyses, we examined the interrelationships among the dependent variables Accuracy of movements, Spatial skills, Dynamics and Rhythmical accuracy. Pearson correlation analysis showed high correlations among all three variables (r ≈ 0.90). This strong multicollinearity likely reflects a common underlying construct in performance. For completeness, we report results of all three variables, as they are potentially independent (e.g., accurate movements can be performed in the wrong spatial location, at the wrong time). As these measures are often used together in dance assessments, it will be informative to provide full reporting of all metrics to facilitate future meta-analyses and establish standardized benchmarks in dance science.

To examine whether dancers’ viewing perspective of the teacher (Point of View: Back, Front mirror, Front opposite), expertise (measured as weekly hours of training), and their interaction influenced dance performance, we fitted a linear model with the average expert ratings as a continuous dependent variable (Accuracy of movements, Spatial skills, Dynamics and rhythmical accuracy). To further characterise the interaction between Training and Point of View, we used post hoc analyses using the emmeans package. Simple slopes of Training (weekly hours) were estimated separately for each Point of View condition. Differences between slopes were assessed using pairwise comparisons with Tukey adjustment for multiple testing. This approach allowed us to determine whether the relationship between Training and performance differed across viewing perspectives.

For all dependent variables, model diagnostics indicated good fit of the linear regression model. Residual plots showed no obvious deviations from homoscedasticity or normality, and the Shapiro-Wilk test confirmed normality of residuals. No influential points were detected via Cook’s distance. To visualise the results, we computed predicted values using the ggeffects package [48], which calculates marginal effects (predicted means) from the fitted model, while accounting for interactions. Specifically, we used the function ggpredict to generate predicted values for the dependent variables across a range of current training hours, for each video condition. Predicted values, including 95% confidence intervals, were plotted alongside the raw participant data using the ggplot2 package [49].

Deviations from pre-registration

Sample size and sample composition.

This study was originally planned and funded as an acute response to the lockdown training experienced during the Covid-19 pandemic, focusing on isolated individual vocational dance students (16–18 years), with a similar level of training. However, as the lockdown was lifted, vocational dancers were extremely busy with in-person group classes, so that only 4 vocational dancers could enrol in this experiment. Hence, we have been forced to reduce the sample size and to open-up registration to non-vocational dancers. As inclusion criteria, we required participants to be currently active in dance and to have at least 3 years of experience in ballet. Because we included dancers with levels and type of experience, we used training practice as a proxy for expertise. We measured training as the number of hours spent in physical training every week. Accounting for expertise is essential, as it can dramatically affect the cognitive load (see Introduction).

Dependent variables.

As per preregistration protocol, dance experts scored individual dancers’ performance. In the preregistration we planned to analyse the proportion of errors (number of errors/overall visible positions or segments) for each dancer at predefined times as dependent variable, focusing on (a) errors in the position of body segments with respect to the own body, (b) errors in direction in space and (c) timing errors at 20 specific time points. However, due to the need to use a short choreography that could fit a 2-day project, and to the small number of errors detected with the analysis of static poses, this approach was not sensitive enough. Hence, we asked dance experts to assess the dancers’ overall performance, focusing on the established measures of Accuracy of movements, Spatial skills, Dynamics and Rhythmic accuracy [20].

Results

Performance

Accuracy of movements.

To examine the effect of Point of View and Training on accuracy of movements, we fitted a linear model including main effects and their interaction (Fig 3a). The overall model was significant, F_5,25 = 4.33, p = 0.006, explaining 36% of the variance (adjusted R² = 0.36). At the effect level, we observed a significant main effect of Training (F_1,25 = 11.463, p = 0.002), Point of View (F_2,25 = 6.093, p = 0.007) and their interaction (F_2,25 = 3.459, p = 0.047). At the coefficient level (Table 1), relative to the Front opposite condition, the Back condition showed higher accuracy (β = 2.48, p = 0.003), whereas the Front mirror condition showed lower accuracy (β = −1.92, p = 0.026). Training level was positively associated with accuracy (β = 0.085, p = 0.002). These coefficients are consistent with the observed interaction.

Download:

Table 1. Summary of the linear regression model examining the effects of Training (weekly hours), Point of View (Back, Front mirror; reference category: Front opposite), and their interaction on Accuracy of movements. The table reports model estimates, standard error of the mean, t-values, and p-values for each predictor.

https://doi.org/10.1371/journal.pone.0355860.t001

Download:

Fig 3. Effect of Point of View and Training on the performance, focusing on (a) Accuracy of movements, (b) Spatial skills, (c) Dynamics and rhythmical accuracy.

Solid lines and error bars indicate predicted values and 95% confidence intervals respectively.

https://doi.org/10.1371/journal.pone.0355860.g003

To further characterise the interaction between Training and Point of View, we examined simple slopes of Training for each condition. Training was not reliably associated with accuracy in the Back condition (β = −0.003, SE = 0.044, 95% CI [−0.094, 0.088]), whereas it was positively associated with accuracy in both the Front mirror condition (β = 0.161, SE = 0.045, 95% CI [0.069, 0.252]) and the Front opposite condition (β = 0.097, SE = 0.042, 95% CI [0.011, 0.183]). Pairwise comparisons of these slopes indicated that the effect of Training was significantly stronger in the Front mirror condition than in the Back condition (Δβ = 0.164, SE = 0.063, p = 0.039, Tukey-adjusted). Differences between the Back and Front opposite conditions (p = 0.246) and between the two Front conditions (p = 0.558) were not statistically significant. The full table of coefficients is shown in Table 1.

Spatial skills.

To examine the effect of Point of View and Training on spatial skills, we fitted a linear model including main effects and their interaction (Fig 3b). The overall model was significant, F_5,25 = 2.95, p = 0.032, explaining 24% of the variance (adjusted R² = 0.24). At the effect level, we observed a significant main effect of Training (F_1,25 = 7.82, p = 0.010) and Point of View (F_2,25 = 3.97, p = 0.032), but their interaction was not significant (F_2,25 = 1.99, p = 0.158).

At the coefficient level (Table 2), relative to the Front opposite condition, the Back condition showed higher spatial skills (β = 2.04, p = 0.010), whereas the Front mirror condition did not differ significantly (β = −1.41, p = 0.092). Training level was positively associated with spatial skills (β = 0.070, p = 0.010). The interaction terms were not significant (although showed the same trends observed for accuracy of movement), indicating that the effect of Training did not differ reliably across perspectives. The full table of coefficients is shown in Table 2.

Download:

Table 2. Summary of the linear regression model examining the effects of Training (weekly hours), Point of View (Back, Front mirror; reference category: Front opposite), and their interaction on Spatial skills. The table reports model estimates, standard error of the mean, t-values, and p-values for each predictor.

https://doi.org/10.1371/journal.pone.0355860.t002

The main effects indicate that greater levels of training were associated with higher spatial skill scores, and that performance varied depending on the viewing perspective. Estimated marginal means showed that participants in the Back view condition had the highest spatial skills scores, while those in the Front views (Mirror and Opposite) scored lower on average. Although the pairwise comparisons were not significant, effect size estimates suggested a moderate advantage for the Back view over both Front perspectives: Cohen’s d ≈ 0.75–0.79 for the Back vs Front conditions, whereas Cohen’s d ≈ 0.040 for the Front mirror vs Front opposite difference. Given that these estimates were associated with non-significant comparisons, they should be interpreted cautiously, as do not provide conclusive evidence for differences between conditions.

Dynamics and rhythmical accuracy.

To examine the effect of Point of View and Training on dynamics and rhythmic accuracy, we fitted a linear model including main effects and their interaction (Fig 3c). The overall model was significant, F_5,25 = 5.05, p = 0.003, explaining 38% of the variance (adjusted R² = 0.38). At the effect level, we observed a significant main effect of Training (F_1,25 = 10.17, p = 0.004) and Point of View (F_2,25 = 4.99, p = 0.015). There was also a trend toward an Interaction between Training and Point of View (F_2,25 = 2.65, p = 0.091). At the coefficient level (Table 3), relative to the Front opposite condition, the Back condition showed higher dynamics and rhythmical accuracy (β = 2.54, p = 0.005), whereas the Front mirror condition did not differ significantly (β = −1.79, p = 0.057). Training level was positively associated with performance (β = 0.088, p = 0.004). The full table of coefficients is shown in Table 3.

Download:

Table 3. Summary of the linear regression model examining the effects of Training (weekly hours), Point of View (Back, Front mirror; reference category: Front opposite), and their interaction on Dynamics and rhythmical accuracy. The table reports model estimates, standard error of the mean, t-values, and p-values for each predictor.

https://doi.org/10.1371/journal.pone.0355860.t003

The trend in the interaction suggests that the relationship between training and performance may vary by viewing perspective. Participants in the Back view condition showed relatively high rhythmic accuracy even at lower levels of training, whereas participants in the Front views appeared to benefit more clearly from additional training. Effect size estimates supported these differences, with moderate to large advantages for the Back view over both Front perspectives (Cohen’s d ≈ 0.73–0.82). Although confidence intervals for these effect sizes included zero, indicating these differences were not statistically significant at the 0.05 level, the pattern of results suggests a practical advantage for the Back view condition in dynamics and rhythmical accuracy. Conversely, the effect size of difference between the two Front views was negligible (Cohen’s d = −0.089).

Discussion

We investigated how different visual perspectives of a dance instructor (first-person Back view, third-person Front mirror, and third-person Front opposite) influence dance learning and performance, as moderated by participants’ training experience. This addresses the issue of extraneous cognitive load (cognitive load dependent on the instructional layout) in motor learning. Our pre-registered hypothesis predicted that the Back view would support better learning outcomes by imposing the lowest extraneous cognitive load. The Back view in fact corresponds to a first-person visual alignment of the dancer with the instructor and the exercise. As more expert dancers have lower cognitive load for a given level of choreographic complexity, the largest advantages were expected in less experienced dancers. We also predicted an intermediate load and performance for the Front mirror view, namely a third person perspective in which the learners move on the same direction of the demonstrator, and a heavier load for the Front opposite view, in which the learner moved in opposite direction of the demonstrator.

Overall, this pattern is consistent with the hypothesis of the lower cognitive load for the Back view, as we observed an enhanced performance either as a significant effect or as a trend across all variables: accuracy of movements, spatial skills, and dynamics/rhythm. Front views posed greater challenges, particularly for dancers with lower training levels. The data did not provide full support for the predicted differentiation between the two front conditions, as performance was not significantly different between Front mirror and Front opposite views. This suggests that the primary distinction captured in this study may be between body-aligned (Back) and front-facing perspectives, rather than between graded levels of front-view difficulty. However, the lack of significant differences between the two front views could also be attributed to the reduced sample size, which was impacted by the recruitment challenges following the ease of the COVID-19 restrictions, and that forced us to reduce the recruitment targets. Indeed, a limitation of our study is the reduced sample size, as recruiting dancers specifically trained in ballet for individual sessions across multiple days has been a challenge. Hence, one possibility is that limited statistical power or participants’ prior exposure to both front-facing perspectives reduced sensitivity to detect differences between these conditions. Alternatively, or additionally, the lack of significant differences between mirror and opposite front views can reflect the current exposure of dancers to both perspectives in their in-person dance training, or to the widespread exposure to third-perspective videos via social media and digital tools widely available. Future studies should clarify how different types of training affect cognitive load longitudinally.

The observed advantage for the Back view aligns with evidence that “mental rotation” and spatial transformations increase cognitive load and hinder performance [32–35]. In our case, the Back view appears to mitigate this load, by allowing participants to map observed movements directly onto their own experience, producing a better performance. Our findings are also consistent with studies in basic procedural learning via video instruction, where first-person demonstrations enhance learning [36,38]. Our findings suggest that the lack of direct experience with the face and gestures of the instructor as seen mostly from the back, does not substantially hinder learning under these conditions.

Confirming other research (e.g., 27), our findings showed that training experience significantly affected performance, with more experienced dancers performing better than less experienced ones. The performance gap due to experience was especially evident in dancers with fewer than 25 hours of weekly training, the only group to receive scores in the lower range.

The effect of Point of View and Training is better understood when considering their interaction, at least for the movement accuracy. An interaction between Point of View and Training was observed for movement accuracy, whereas this pattern was weaker or not statistically robust for the other outcomes. The modulation by expertise was clearest for movement accuracy. In the contest of movement accuracy, the effect of the point of view was strongly affected by the expertise of dancers, as shown by the significant interactions on Point of View and Training. This suggests that, in front of the same choreography, different strategies might be needed for dancers with different levels of expertise. While the fact that the Back view appeared to particularly facilitate less experienced dancers suggests that the benefit of a first-person perspective is higher in less experienced dancers, only few experienced dancers have been tested on the Back point of view. Taken together, these findings provide some partial support for an expertise-reversal account.

These results pave the way to further studies on the effectiveness of the first-person perspective for experienced dancers with challenging choreographies, and whether the benefit of first-person point of view recapitulates the expertise reversal effect [24]. Future studies should clarify to what extent expert professional dancers, who are often requested to learn a choreography from videos, could benefit from a first-person rather than third person perspective. Taken together, our results suggest that first-person perspectives may represent a useful strategy to adapt instructions to learner expertise and facilitate the task. The first-person view is often under-utilised in real-world dance education, in particular in ballet classes, where dancers training at the barre are often asked to start the exercise on the same egocentric position (e.g., usually with the left hand on the barre), irrespectively of whether it is the same or the different side of the demonstrator/teacher in front of them [50]. Our study opens the way for further investigation on the efficacy of first-person perspective in learning a choreography or motor sequences with different levels of expertise and developmental stages.

The potential impact of dance and motor sequence learning enhancement is huge, as dance is a behaviour observed in all human cultures [51,52]. For instance, according to the official data of the UK [53], in this country there are approximately 7,000 dancers and choreographers who undergo daily training and are engaged in the teaching and learning choreographies through in person or video instruction. In England, 27% of children and young adults take part at least once a week to recreational dance classes in person or online, a format increasingly popular since the COVID-19. Other disciplines directly rely on choreographies and complex motion sequences from artistic and rhythmic gymnastics to figure skating, synchronised swimming, but the teaching and learning of motor routines is a broader issue across sports.

Overall, our findings indicate that a first-person perspective may support motor sequence learning, when high cognitive load is involved. By confirm high inter-rater reliability in performance, we validated a digital tool for the analysis and assessment of motor sequences. These outcomes may have theoretical and practical implications for the advancement of instructional design, the practice and teaching of dance. Future studies should assess the generalisability to other disciplines in performing arts and sports that require complex sequence learning, both in video/online and in-person settings. On the one hand, as remote learning and video-based teaching become increasingly common, instructional design should consider the cognitive impact of camera angle and perspective. Conversely, as dance training typically engages practitioners of different expertise with potentially challenging third-person perspectives, teachers and choreographers should consider shifting their position, or that of their students, to overcome difficulties of less experienced athletes. This may support motor sequence learning in dance and beyond.

Supporting information

S1 Table. Assessment criteria and scoring scale.

https://doi.org/10.1371/journal.pone.0355860.s001

(DOCX)

S1 File. Dataset: Support dataset.

https://doi.org/10.1371/journal.pone.0355860.s002

(TXT)

Acknowledgments

We thank the artistic director and choreographer Erico Montes; the dancers, who have so enthusiastically participated in this experiment and a previous pilot project (summarised at https://youtu.be/m7gOGwZ1zTY); the dance masters/teachers: Raymond Chai, Hubert Essakow, Nina Thilas-Mohs, Sandra Corsini; Diane Abdhalla, Theano Vikatou and Simran Nadkarni for their support with data collection; the support staff at The English National Ballet School, The Royal Ballet School and Trinity Laban Conservatoire for Music and Dance. We used AI-assisted technologies (chat-GPT, release June 2025) as a search engine and to improve readability of the work.

References

1. L Biernacki J, Stracciolini A, Fraser J, J Micheli L, Sugimoto D. Risk factors for lower-extremity injuries in female ballet dancers: a systematic review. Clin J Sport Med. 2021;31(2):e64–79. pmid:30589745
- View Article
- PubMed/NCBI
- Google Scholar
2. Li J, Ahmad MA. Evolution and trends in online dance instruction: a comprehensive literature analysis. Front Educ. 2025;10:1523766.
- View Article
- Google Scholar
3. Parrish M. Toward transformation: Digital tools for online dance pedagogy. Arts Education Policy Review. 2016;117(3):168–82.
- View Article
- Google Scholar
4. Donti O, Donti A, Theodorakou K. A review on the changes of the evaluation system affecting artistic gymnasts’ basic preparation: the aspect of choreography preparation. SGJ. 2014;6(2):63–72.
- View Article
- Google Scholar
5. Bourdon PC, Cardinale M, Murray A, Gastin P, Kellmann M, Varley MC, et al. Monitoring athlete training loads: consensus statement. Int J Sports Physiol Perform. 2017;12(Suppl 2):S2161–70. pmid:28463642
- View Article
- PubMed/NCBI
- Google Scholar
6. McLaren SJ, Macpherson TW, Coutts AJ, Hurst C, Spears IR, Weston M. The relationships between internal and external measures of training load and intensity in team sports: a meta-analysis. Sports Med. 2018;48(3):641–58. pmid:29288436
- View Article
- PubMed/NCBI
- Google Scholar
7. Foster C, Florhaug JA, Franklin J, Gottschall L, Hrovatin LA, Parker S, et al. A new approach to monitoring exercise training. J Strength Cond Res. 2001;15(1):109–15. pmid:11708692
- View Article
- PubMed/NCBI
- Google Scholar
8. Gabbett TJ. The training-injury prevention paradox: should athletes be training smarter and harder?. Br J Sports Med. 2016;50(5):273–80. pmid:26758673
- View Article
- PubMed/NCBI
- Google Scholar
9. Halson SL. Monitoring training load to understand fatigue in athletes. Sports Med. 2014;44 Suppl 2(Suppl 2):S139-47. pmid:25200666
- View Article
- PubMed/NCBI
- Google Scholar
10. de Jong T. Cognitive load theory, educational research, and instructional design: some food for thought. Instr Sci. 2009;38(2):105–34.
- View Article
- Google Scholar
11. Brattan VC, Baker DH, Tipper SP. Spatiotemporal judgments of observed actions: contrasts between first- and third-person perspectives after motor priming. J Exp Psychol Hum Percept Perform. 2015;41(5):1236–46. pmid:26076174
- View Article
- PubMed/NCBI
- Google Scholar
12. Jackson PL, Meltzoff AN, Decety J. Neural circuits involved in imitation and perspective-taking. Neuroimage. 2006;31(1):429–39. pmid:16406257
- View Article
- PubMed/NCBI
- Google Scholar
13. Watanabe R, Higuchi T. Behavioral advantages of the first-person perspective model for imitation. Front Psychol. 2016;7:701. pmid:27242614
- View Article
- PubMed/NCBI
- Google Scholar
14. Herrebrøden H, Jensenius AR, Espeseth T, Bishop L, Vuoskoski JK. Cognitive load causes kinematic changes in both elite and non-elite rowers. Hum Mov Sci. 2023;90:103113. pmid:37331066
- View Article
- PubMed/NCBI
- Google Scholar
15. Meguerdichian M, Walker K, Bajaj K. Working memory is limited: improving knowledge transfer by optimising simulation through cognitive load theory. BMJ Simul Technol Enhanc Learn. 2016;2(4):131–8. pmid:35514866
- View Article
- PubMed/NCBI
- Google Scholar
16. Büchel D, Gokeler A, Heuvelmans P, Baumeister J. Increased cognitive demands affect agility performance in female athletes - implications for testing and training of agility in team ball sports. Percept Mot Skills. 2022;129(4):1074–88. pmid:35703458
- View Article
- PubMed/NCBI
- Google Scholar
17. Kuczyński M, Szymańska M, Bieć E. Dual-task effect on postural control in high-level competitive dancers. J Sports Sci. 2011;29(5):539–45. pmid:21294035
- View Article
- PubMed/NCBI
- Google Scholar
18. Bläsing BE. Segmentation of dance movement: effects of expertise, visual familiarity, motor experience and music. Front Psychol. 2015;5:1500. pmid:25610409
- View Article
- PubMed/NCBI
- Google Scholar
19. Carey K, Moran A, Rooney B. Learning choreography: an investigation of motor imagery, attentional effort, and expertise in modern dance. Front Psychol. 2019;10:422. pmid:30881331
- View Article
- PubMed/NCBI
- Google Scholar
20. Angioi M, Metsios GS, Twitchett E, Koutedakis Y, Wyon M. Association between selected physical fitness parameters and esthetic competence in contemporary dancers. J Dance Med Sci. 2009;13(4):115–23. pmid:19930813
- View Article
- PubMed/NCBI
- Google Scholar
21. Bläsing B, Calvo-Merino B, Cross ES, Jola C, Honisch J, Stevens CJ. Neurocognitive control in dance perception and performance. Acta Psychol (Amst). 2012;139(2):300–8. pmid:22305351
- View Article
- PubMed/NCBI
- Google Scholar
22. Hackney ME, Burzynska AZ, Ting LH. The cognitive neuroscience and neurocognitive rehabilitation of dance. BMC Neurosci. 2024;25(1):58. pmid:39506634
- View Article
- PubMed/NCBI
- Google Scholar
23. Cross ES, Acquah D, Ramsey R. A review and critical analysis of how cognitive neuroscientific investigations using dance can contribute to sport psychology. International Review of Sport and Exercise Psychology. 2013;7(1):42–71.
- View Article
- Google Scholar
24. Kalyuga S, Ayres P, Chandler P, Sweller J. The expertise reversal effect. Educational Psychologist. 2003;38(1):23–31.
- View Article
- Google Scholar
25. Mann DTY, Williams AM, Ward P, Janelle CM. Perceptual-cognitive expertise in sport: a meta-analysis. J Sport Exerc Psychol. 2007;29(4):457–78. pmid:17968048
- View Article
- PubMed/NCBI
- Google Scholar
26. Starkes JL, Deakin JM, Lindley S, Crisp F. Motor versus verbal recall of ballet sequences by young expert dancers. Journal of Sport Psychology. 1987;9(3):222–30.
- View Article
- Google Scholar
27. Burzynska AZ, Finc K, Taylor BK, Knecht AM, Kramer AF. The dancing brain: structural and functional signatures of expert dance training. Front Hum Neurosci. 2017;11:566. pmid:29230170
- View Article
- PubMed/NCBI
- Google Scholar
28. Sweller J, Chandler P, Tierney P, Cooper M. Cognitive load as a factor in the structuring of technical material. Journal of Experimental Psychology: General. 1990;119(2):176–92.
- View Article
- Google Scholar
29. Jarraya M, Rekik G, Belkhir Y, Chtourou H, Nikolaidis PT, Rosemann T, et al. Which presentation speed is better for learning basketball tactical actions through video modeling examples? the influence of content complexity. Front Psychol. 2019;10:2356. pmid:31695646
- View Article
- PubMed/NCBI
- Google Scholar
30. Sweller J. Cognitive architecture and instructional design.
31. Orru G, Longo L. The evolution of cognitive load theory and the measurement of its intrinsic, extraneous and germane loads: a review. Longo L, Leva MC. Human mental workload: Models and applications. Cham: Springer International Publishing. 2019. 23–48.
- View Article
- Google Scholar
32. Munzert J, Lorey B, Zentgraf K. Cognitive motor processes: the role of motor imagery in the study of motor representations. Brain Res Rev. 2009;60(2):306–26. pmid:19167426
- View Article
- PubMed/NCBI
- Google Scholar
33. Parsons LM. Imagined spatial transformation of one’s body. J Exp Psychol Gen. 1987;116(2):172–91. pmid:2955072
- View Article
- PubMed/NCBI
- Google Scholar
34. Shepard RN, Metzler J. Mental rotation of three-dimensional objects abstract. Science. 1971;171(February):701–3. pmid:5540314
- View Article
- PubMed/NCBI
- Google Scholar
35. Vanrie J, Willems B, Wagemans J. Multiple routes to object matching from different viewpoints: mental rotation versus invariant features. Perception. 2001;30(9):1047–56. pmid:11694082
- View Article
- PubMed/NCBI
- Google Scholar
36. Fiorella L, van Gog T, Hoogerheide V, Mayer RE. It’s all a matter of perspective: viewing first-person video modeling examples promotes learning of an assembly task. Journal of Educational Psychology. 2017;109(5):653–65.
- View Article
- Google Scholar
37. Garland TB, Sanchez CA. Rotational perspective and learning procedural tasks from dynamic media. Computers & Education. 2013;69:31–7.
- View Article
- Google Scholar
38. Koning BB, Mok K, Marcus N, Ayres P. Investigating the role of hand perspective in learning from procedural animations. Br J Educ Psychol. 2023;93 Suppl 2:251–69. pmid:36047932
- View Article
- PubMed/NCBI
- Google Scholar
39. Leng X, Zhu W, Mayer RE, Wang F. The viewing perspective effect in learning from instructional videos: a replication and neuroimaging extension. Learning and Instruction. 2024;94:102004.
- View Article
- Google Scholar
40. Gallo V, Zhou X, Abdallah D, Angioi M, Redding E, Ackroyd B, et al. MotionPerfection: an agile tool for the visualisation, analysis, annotation, and record of motor practice. SoftwareX. 2022;20:101247.
- View Article
- Google Scholar
41. Sweller J, Ayres P, Kalyuga S. Cognitive Load Theory. New York, NY: Springer New York. 2011.
- View Article
- Google Scholar
42. Paas F, Tuovinen JE, Tabbers H, Van Gerven PWM. Cognitive load measurement as a means to advance cognitive load theory. Educational Psychologist. 2003;38(1):63–71.
- View Article
- Google Scholar
43. Allen N, Nevill A, Brooks J, Koutedakis Y, Wyon M. Ballet injuries: injury incidence and severity over 1 year. J Orthop Sports Phys Ther. 2012;42(9):781–90. pmid:22814244
- View Article
- PubMed/NCBI
- Google Scholar
44. Mattiussi AM, Shaw JW, Williams S, Price PD, Brown DD, Cohen DD, et al. Injury epidemiology in professional ballet: a five-season prospective study of 1596 medical attention injuries and 543 time-loss injuries. Br J Sports Med. 2021;55(15):843–50. pmid:33837002
- View Article
- PubMed/NCBI
- Google Scholar
45. Fischer C, Fagundes AG, Poton R. Analysis of current tests for assessing dance aesthetic performance: a systematic review. International Journal of Art, Culture, Design, and Technology. 2022;11(2):1–9.
- View Article
- Google Scholar
46. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15(2):155–63. pmid:27330520
- View Article
- PubMed/NCBI
- Google Scholar
47. McGraw KO, Wong SP. Forming inferences about some intraclass correlation coefficients. Psychological Methods. 1996;1(1):30–46.
- View Article
- Google Scholar
48. Lüdecke D. Ggeffects: tidy data frames of marginal effects from regression models. JOSS. 2018;3(26):772.
- View Article
- Google Scholar
49. Wickham H. Ggplot2: elegant graphics for data analysis. Springer. 2009.
50. Kimmerle ML. Lateral bias, functional asymmetry, dance training and dance injuries. Journal of Dance Medicine & Science. 2010;14(2):58–66.
- View Article
- Google Scholar
51. Fink B, Bläsing B, Ravignani A, Shackelford TK. Evolution and functions of human dance. Evolution and Human Behavior. 2021;42(4):351–60.
- View Article
- Google Scholar
52. Laland K, Wilkins C, Clayton N. The evolution of dance. Curr Biol. 2016;26(1):R5-9. pmid:26766235
- View Article
- PubMed/NCBI
- Google Scholar
53. One Dance UK. Written evidence submitted by One Dance UK Author: Institution: Date: 2022. [cited 2025 Jul 9]. https://committees.parliament.uk/writtenevidence/133982/pdf/

[ref1] 1. L Biernacki J, Stracciolini A, Fraser J, J Micheli L, Sugimoto D. Risk factors for lower-extremity injuries in female ballet dancers: a systematic review. Clin J Sport Med. 2021;31(2):e64–79. pmid:30589745
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. Li J, Ahmad MA. Evolution and trends in online dance instruction: a comprehensive literature analysis. Front Educ. 2025;10:1523766.
View Article
Google Scholar

[6] View Article

[7] Google Scholar

[ref3] 3. Parrish M. Toward transformation: Digital tools for online dance pedagogy. Arts Education Policy Review. 2016;117(3):168–82.
View Article
Google Scholar

[9] View Article

[10] Google Scholar

[ref4] 4. Donti O, Donti A, Theodorakou K. A review on the changes of the evaluation system affecting artistic gymnasts’ basic preparation: the aspect of choreography preparation. SGJ. 2014;6(2):63–72.
View Article
Google Scholar

[12] View Article

[13] Google Scholar

[ref5] 5. Bourdon PC, Cardinale M, Murray A, Gastin P, Kellmann M, Varley MC, et al. Monitoring athlete training loads: consensus statement. Int J Sports Physiol Perform. 2017;12(Suppl 2):S2161–70. pmid:28463642
View Article
PubMed/NCBI
Google Scholar

[15] View Article

[16] PubMed/NCBI

[17] Google Scholar

[ref6] 6. McLaren SJ, Macpherson TW, Coutts AJ, Hurst C, Spears IR, Weston M. The relationships between internal and external measures of training load and intensity in team sports: a meta-analysis. Sports Med. 2018;48(3):641–58. pmid:29288436
View Article
PubMed/NCBI
Google Scholar

[19] View Article

[20] PubMed/NCBI

[21] Google Scholar

[ref7] 7. Foster C, Florhaug JA, Franklin J, Gottschall L, Hrovatin LA, Parker S, et al. A new approach to monitoring exercise training. J Strength Cond Res. 2001;15(1):109–15. pmid:11708692
View Article
PubMed/NCBI
Google Scholar

[23] View Article

[24] PubMed/NCBI

[25] Google Scholar

[ref8] 8. Gabbett TJ. The training-injury prevention paradox: should athletes be training smarter and harder?. Br J Sports Med. 2016;50(5):273–80. pmid:26758673
View Article
PubMed/NCBI
Google Scholar

[27] View Article

[28] PubMed/NCBI

[29] Google Scholar

[ref9] 9. Halson SL. Monitoring training load to understand fatigue in athletes. Sports Med. 2014;44 Suppl 2(Suppl 2):S139-47. pmid:25200666
View Article
PubMed/NCBI
Google Scholar

[31] View Article

[32] PubMed/NCBI

[33] Google Scholar

[ref10] 10. de Jong T. Cognitive load theory, educational research, and instructional design: some food for thought. Instr Sci. 2009;38(2):105–34.
View Article
Google Scholar

[35] View Article

[36] Google Scholar

[ref11] 11. Brattan VC, Baker DH, Tipper SP. Spatiotemporal judgments of observed actions: contrasts between first- and third-person perspectives after motor priming. J Exp Psychol Hum Percept Perform. 2015;41(5):1236–46. pmid:26076174
View Article
PubMed/NCBI
Google Scholar

[38] View Article

[39] PubMed/NCBI

[40] Google Scholar

[ref12] 12. Jackson PL, Meltzoff AN, Decety J. Neural circuits involved in imitation and perspective-taking. Neuroimage. 2006;31(1):429–39. pmid:16406257
View Article
PubMed/NCBI
Google Scholar

[42] View Article

[43] PubMed/NCBI

[44] Google Scholar

[ref13] 13. Watanabe R, Higuchi T. Behavioral advantages of the first-person perspective model for imitation. Front Psychol. 2016;7:701. pmid:27242614
View Article
PubMed/NCBI
Google Scholar

[46] View Article

[47] PubMed/NCBI

[48] Google Scholar

[ref14] 14. Herrebrøden H, Jensenius AR, Espeseth T, Bishop L, Vuoskoski JK. Cognitive load causes kinematic changes in both elite and non-elite rowers. Hum Mov Sci. 2023;90:103113. pmid:37331066
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref15] 15. Meguerdichian M, Walker K, Bajaj K. Working memory is limited: improving knowledge transfer by optimising simulation through cognitive load theory. BMJ Simul Technol Enhanc Learn. 2016;2(4):131–8. pmid:35514866
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref16] 16. Büchel D, Gokeler A, Heuvelmans P, Baumeister J. Increased cognitive demands affect agility performance in female athletes - implications for testing and training of agility in team ball sports. Percept Mot Skills. 2022;129(4):1074–88. pmid:35703458
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref17] 17. Kuczyński M, Szymańska M, Bieć E. Dual-task effect on postural control in high-level competitive dancers. J Sports Sci. 2011;29(5):539–45. pmid:21294035
View Article
PubMed/NCBI
Google Scholar

[62] View Article

[63] PubMed/NCBI

[64] Google Scholar

[ref18] 18. Bläsing BE. Segmentation of dance movement: effects of expertise, visual familiarity, motor experience and music. Front Psychol. 2015;5:1500. pmid:25610409
View Article
PubMed/NCBI
Google Scholar

[66] View Article

[67] PubMed/NCBI

[68] Google Scholar

[ref19] 19. Carey K, Moran A, Rooney B. Learning choreography: an investigation of motor imagery, attentional effort, and expertise in modern dance. Front Psychol. 2019;10:422. pmid:30881331
View Article
PubMed/NCBI
Google Scholar

[70] View Article

[71] PubMed/NCBI

[72] Google Scholar

[ref20] 20. Angioi M, Metsios GS, Twitchett E, Koutedakis Y, Wyon M. Association between selected physical fitness parameters and esthetic competence in contemporary dancers. J Dance Med Sci. 2009;13(4):115–23. pmid:19930813
View Article
PubMed/NCBI
Google Scholar

[74] View Article

[75] PubMed/NCBI

[76] Google Scholar

[ref21] 21. Bläsing B, Calvo-Merino B, Cross ES, Jola C, Honisch J, Stevens CJ. Neurocognitive control in dance perception and performance. Acta Psychol (Amst). 2012;139(2):300–8. pmid:22305351
View Article
PubMed/NCBI
Google Scholar

[78] View Article

[79] PubMed/NCBI

[80] Google Scholar

[ref22] 22. Hackney ME, Burzynska AZ, Ting LH. The cognitive neuroscience and neurocognitive rehabilitation of dance. BMC Neurosci. 2024;25(1):58. pmid:39506634
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref23] 23. Cross ES, Acquah D, Ramsey R. A review and critical analysis of how cognitive neuroscientific investigations using dance can contribute to sport psychology. International Review of Sport and Exercise Psychology. 2013;7(1):42–71.
View Article
Google Scholar

[86] View Article

[87] Google Scholar

[ref24] 24. Kalyuga S, Ayres P, Chandler P, Sweller J. The expertise reversal effect. Educational Psychologist. 2003;38(1):23–31.
View Article
Google Scholar

[89] View Article

[90] Google Scholar

[ref25] 25. Mann DTY, Williams AM, Ward P, Janelle CM. Perceptual-cognitive expertise in sport: a meta-analysis. J Sport Exerc Psychol. 2007;29(4):457–78. pmid:17968048
View Article
PubMed/NCBI
Google Scholar

[92] View Article

[93] PubMed/NCBI

[94] Google Scholar

[ref26] 26. Starkes JL, Deakin JM, Lindley S, Crisp F. Motor versus verbal recall of ballet sequences by young expert dancers. Journal of Sport Psychology. 1987;9(3):222–30.
View Article
Google Scholar

[96] View Article

[97] Google Scholar

[ref27] 27. Burzynska AZ, Finc K, Taylor BK, Knecht AM, Kramer AF. The dancing brain: structural and functional signatures of expert dance training. Front Hum Neurosci. 2017;11:566. pmid:29230170
View Article
PubMed/NCBI
Google Scholar

[99] View Article

[100] PubMed/NCBI

[101] Google Scholar

[ref28] 28. Sweller J, Chandler P, Tierney P, Cooper M. Cognitive load as a factor in the structuring of technical material. Journal of Experimental Psychology: General. 1990;119(2):176–92.
View Article
Google Scholar

[103] View Article

[104] Google Scholar

[ref29] 29. Jarraya M, Rekik G, Belkhir Y, Chtourou H, Nikolaidis PT, Rosemann T, et al. Which presentation speed is better for learning basketball tactical actions through video modeling examples? the influence of content complexity. Front Psychol. 2019;10:2356. pmid:31695646
View Article
PubMed/NCBI
Google Scholar

[106] View Article

[107] PubMed/NCBI

[108] Google Scholar

[ref30] 30. Sweller J. Cognitive architecture and instructional design.

[ref31] 31. Orru G, Longo L. The evolution of cognitive load theory and the measurement of its intrinsic, extraneous and germane loads: a review. Longo L, Leva MC. Human mental workload: Models and applications. Cham: Springer International Publishing. 2019. 23–48.
View Article
Google Scholar

[111] View Article

[112] Google Scholar

[ref32] 32. Munzert J, Lorey B, Zentgraf K. Cognitive motor processes: the role of motor imagery in the study of motor representations. Brain Res Rev. 2009;60(2):306–26. pmid:19167426
View Article
PubMed/NCBI
Google Scholar

[114] View Article

[115] PubMed/NCBI

[116] Google Scholar

[ref33] 33. Parsons LM. Imagined spatial transformation of one’s body. J Exp Psychol Gen. 1987;116(2):172–91. pmid:2955072
View Article
PubMed/NCBI
Google Scholar

[118] View Article

[119] PubMed/NCBI

[120] Google Scholar

[ref34] 34. Shepard RN, Metzler J. Mental rotation of three-dimensional objects abstract. Science. 1971;171(February):701–3. pmid:5540314
View Article
PubMed/NCBI
Google Scholar

[122] View Article

[123] PubMed/NCBI

[124] Google Scholar

[ref35] 35. Vanrie J, Willems B, Wagemans J. Multiple routes to object matching from different viewpoints: mental rotation versus invariant features. Perception. 2001;30(9):1047–56. pmid:11694082
View Article
PubMed/NCBI
Google Scholar

[126] View Article

[127] PubMed/NCBI

[128] Google Scholar

[ref36] 36. Fiorella L, van Gog T, Hoogerheide V, Mayer RE. It’s all a matter of perspective: viewing first-person video modeling examples promotes learning of an assembly task. Journal of Educational Psychology. 2017;109(5):653–65.
View Article
Google Scholar

[130] View Article

[131] Google Scholar

[ref37] 37. Garland TB, Sanchez CA. Rotational perspective and learning procedural tasks from dynamic media. Computers & Education. 2013;69:31–7.
View Article
Google Scholar

[133] View Article

[134] Google Scholar

[ref38] 38. Koning BB, Mok K, Marcus N, Ayres P. Investigating the role of hand perspective in learning from procedural animations. Br J Educ Psychol. 2023;93 Suppl 2:251–69. pmid:36047932
View Article
PubMed/NCBI
Google Scholar

[136] View Article

[137] PubMed/NCBI

[138] Google Scholar

[ref39] 39. Leng X, Zhu W, Mayer RE, Wang F. The viewing perspective effect in learning from instructional videos: a replication and neuroimaging extension. Learning and Instruction. 2024;94:102004.
View Article
Google Scholar

[140] View Article

[141] Google Scholar

[ref40] 40. Gallo V, Zhou X, Abdallah D, Angioi M, Redding E, Ackroyd B, et al. MotionPerfection: an agile tool for the visualisation, analysis, annotation, and record of motor practice. SoftwareX. 2022;20:101247.
View Article
Google Scholar

[143] View Article

[144] Google Scholar

[ref41] 41. Sweller J, Ayres P, Kalyuga S. Cognitive Load Theory. New York, NY: Springer New York. 2011.
View Article
Google Scholar

[146] View Article

[147] Google Scholar

[ref42] 42. Paas F, Tuovinen JE, Tabbers H, Van Gerven PWM. Cognitive load measurement as a means to advance cognitive load theory. Educational Psychologist. 2003;38(1):63–71.
View Article
Google Scholar

[149] View Article

[150] Google Scholar

[ref43] 43. Allen N, Nevill A, Brooks J, Koutedakis Y, Wyon M. Ballet injuries: injury incidence and severity over 1 year. J Orthop Sports Phys Ther. 2012;42(9):781–90. pmid:22814244
View Article
PubMed/NCBI
Google Scholar

[152] View Article

[153] PubMed/NCBI

[154] Google Scholar

[ref44] 44. Mattiussi AM, Shaw JW, Williams S, Price PD, Brown DD, Cohen DD, et al. Injury epidemiology in professional ballet: a five-season prospective study of 1596 medical attention injuries and 543 time-loss injuries. Br J Sports Med. 2021;55(15):843–50. pmid:33837002
View Article
PubMed/NCBI
Google Scholar

[156] View Article

[157] PubMed/NCBI

[158] Google Scholar

[ref45] 45. Fischer C, Fagundes AG, Poton R. Analysis of current tests for assessing dance aesthetic performance: a systematic review. International Journal of Art, Culture, Design, and Technology. 2022;11(2):1–9.
View Article
Google Scholar

[160] View Article

[161] Google Scholar

[ref46] 46. Koo TK, Li MY. A guideline of selecting and reporting intraclass correlation coefficients for reliability research. J Chiropr Med. 2016;15(2):155–63. pmid:27330520
View Article
PubMed/NCBI
Google Scholar

[163] View Article

[164] PubMed/NCBI

[165] Google Scholar

[ref47] 47. McGraw KO, Wong SP. Forming inferences about some intraclass correlation coefficients. Psychological Methods. 1996;1(1):30–46.
View Article
Google Scholar

[167] View Article

[168] Google Scholar

[ref48] 48. Lüdecke D. Ggeffects: tidy data frames of marginal effects from regression models. JOSS. 2018;3(26):772.
View Article
Google Scholar

[170] View Article

[171] Google Scholar

[ref49] 49. Wickham H. Ggplot2: elegant graphics for data analysis. Springer. 2009.

[ref50] 50. Kimmerle ML. Lateral bias, functional asymmetry, dance training and dance injuries. Journal of Dance Medicine & Science. 2010;14(2):58–66.
View Article
Google Scholar

[174] View Article

[175] Google Scholar

[ref51] 51. Fink B, Bläsing B, Ravignani A, Shackelford TK. Evolution and functions of human dance. Evolution and Human Behavior. 2021;42(4):351–60.
View Article
Google Scholar

[177] View Article

[178] Google Scholar

[ref52] 52. Laland K, Wilkins C, Clayton N. The evolution of dance. Curr Biol. 2016;26(1):R5-9. pmid:26766235
View Article
PubMed/NCBI
Google Scholar

[180] View Article

[181] PubMed/NCBI

[182] Google Scholar

[ref53] 53. One Dance UK. Written evidence submitted by One Dance UK Author: Institution: Date: 2022. [cited 2025 Jul 9]. https://committees.parliament.uk/writtenevidence/133982/pdf/

Figures

Abstract

Introduction

Validity and applicability of the experimental setting

Aims and hypotheses

Materials and methods

Preregistration

Participants

Experimental design

Apparatus

Stimuli and test

Procedure

Video analysis

Data analysis

Deviations from pre-registration

Sample size and sample composition.

Dependent variables.

Results

Performance

Accuracy of movements.

Spatial skills.

Dynamics and rhythmical accuracy.

Discussion

Supporting information

S1 Table. Assessment criteria and scoring scale.

S1 File. Dataset: Support dataset.

Acknowledgments

References