Immersive VR-MoCap for Creative Motion Design in Character Animation Training: A Classroom-Based Comparative Study

Abstract

Although motion capture has become integral to contemporary animation pipelines, university teaching still asks students to learn motion largely through screen-based keyframing. To address this gap, this classroom-based comparative study evaluated one structured motion-design lesson within an immersive MoCap-supported training module. Sixty-eight undergraduates in a computer animation course completed the same task in either a Keyframe condition (n = 33) or a VR-MoCap condition (n = 35), with instructional delivery mode as the only difference. Creative performance was assessed in originality, fluency, aesthetic quality, clarity, and a composite score. MANOVA revealed a significant multivariate effect of condition (Pillai’s trace = 0.454, F(4, 63) = 13.12, p < 0.001). Relative to keyframe instruction, VR-MoCap produced significantly higher originality, fluency, clarity, and composite performance, whereas aesthetic quality did not differ significantly. Supplementary group-interview responses further indicated that students experienced the immersive condition as more engaging, more intuitive, and better suited to immediate feedback and embodied movement exploration. Immersive VR-MoCap appears most useful in the early phases of motion design and is better understood as complementing, rather than replacing, conventional keyframe training.

1. Introduction

Computer animation education is a long-term and highly integrated process that combines artistic sensibility, technical competence, and performance-based understanding. In the field of character animation in particular, students are expected not only to master form, timing, motion rhythm, acting intention, and narrative expression, but also to translate these capacities into executable animation workflows within digital production environments [1,2,3]. Traditional character animation training has long been grounded in observation, drawing, body mechanics, and performance awareness, with the ultimate aim not merely of making a character move, but of making that movement appear intentional, expressive, and believable [1,2,4]. However, as computer graphics technologies and digital pipelines have increasingly entered university classrooms, animation education has also encountered a new tension: although motion capture (MoCap) and virtual reality (VR) are now widely used across animation, games, movement analysis, and performance-oriented digital media, their pedagogical integration into animation curricula remains uneven and often insufficiently theorized [5,6,7].

One persistent problem in animation education is the gap between foundational animation training and software-based instruction. Although the Twelve Principles of Animation remain central to 3D animation because they address timing, motion structure, and visual credibility [1,8], students often learn software operation without fully understanding movement as expression, performance, and embodied communication [4,9]. Character animation is therefore inherently interdisciplinary, involving not only technical skills but also acting, anatomy, psychology, physics, and observation of human behaviour [10,11]. As Hooks [2] and Kennedy [12] suggest, animated movement is not merely technical output or raw data, but a performance shaped by intention, perception, and context. These perspectives indicate that character animation education should move beyond tool training alone and integrate movement understanding, bodily perception, spatial judgment, and performance logic into a more coherent pedagogical structure. This paper therefore focuses on the use of immersive VR and motion capture as instructional supports for computer animation practice and training.

However, the educational value of emerging technologies is often overstated at the moment of their introduction. Tools such as MoCap and VR are frequently valued for their novelty and technical capacity, while media attention and marketing may inflate expectations before their actual learning benefits are verified [13,14]. In this context, motion capture has increasingly entered animation education as a possible bridge between classroom learning and industry practice. Originally developed for medicine, sports, law, and movement analysis, MoCap later became an important tool in computer animation and games [5], and advances in hardware and software have made it more accessible in educational settings [15,16]. As a result, the first objective of this study was to design and develop an immersive MoCap learning environment for animation training.

Another issue is whether MoCap can genuinely support creativity in character animation learning. Captured motion does not automatically become expressive animation, because performance depends not only on motion accuracy but also on intention, emotion, and role understanding [2,12]. Mou [17], further suggests that keyframing and motion capture may support different dimensions of creativity: keyframing may better support unusual movement ideas, whereas motion capture may better support the generation of more valid motion options. This means that the value of MoCap cannot be judged only by efficiency or realism, but by how it supports students’ exploration and creation of movement. As a result, the second objective of this study was to examine whether an immersive MoCap learning environment could better support creativity in animation motion design than conventional keyframe-based instruction.

The contribution of this study is threefold. First, it provides a classroom-based comparison between immersive VR-MoCap-supported instruction and conventional keyframe-based instruction in an authentic undergraduate animation course rather than in a laboratory or prototype-only setting. Second, it operationalizes creative motion-design performance as a multidimensional construct comprising originality, fluency, clarity, and aesthetic quality, enabling a more differentiated analysis than single-score creativity comparisons. Third, it shows that the pedagogical value of VR-MoCap is selective rather than uniform, thereby moving beyond a simple assumption of general technology superiority. Unlike studies that discuss VR or MoCap mainly in terms of realism, engagement, or workflow efficiency, the present framework examines how immersive, embodied instruction may support different components of creativity in motion design in different ways.

To sum up, whereas the first research objective concerns the design of an immersive MoCap learning environment for animation training, the second is concerned with examining whether this approach can better support creativity in animation motion design than conventional keyframe-based instruction. The study therefore addresses the following research questions:

RQ1: 

Does immersive MoCap-supported training enhance creativity in computer animation motion design more than keyframe-based training?

RQ2: 

How do immersive MoCap-supported training and keyframe-based training differ across key dimensions of creativity, including originality, fluency, clarity, and aesthetic quality?

2. Literature Review

2.1. Traditional Character Animation Training

Traditional character animation training remains the conceptual foundation of animation education. Earlier scholarship emphasizes that believable animation depends not only on technical execution but on drawing, anatomy, body mechanics, timing, rhythm, exaggeration, and performance sensitivity [1,2,3]. These principles continue to matter in digital production: Lasseter [8] argues that they remain directly applicable to 3D animation, and Musa et al. [9] likewise show that weak computer animation often reflects inadequate grounding in these fundamentals. At the same time, scholars caution that traditional principles are sometimes taught as formulaic rules rather than as perceptual and embodied understanding of action [4]. Figure 1: Animation training through traditional hand-drawn methods [3].

Recent work therefore frames character animation as a hybrid discipline requiring the integration of technical skill, acting, emotion, spatial judgment, and performance logic [2,12]. This view also aligns with STEAM-oriented models that connect art, technology, and design thinking [10,11]. In contemporary animation education, the challenge is not simply to preserve traditional principles, but to reconnect them with lived movement experience within digital production contexts [18]. To address these issues, the present study designs and evaluates a user-centred MoCap training environment and compares it with conventional training methods.

2.2. Motion Capture Technology in Education

Motion capture has developed from a specialist technology in medicine, sports, and movement analysis into an important tool in digital animation and game production [5,7]. As a result, it has increasingly entered educational settings as a way to connect classroom learning with industry practice [15,16,19]. Existing studies suggest that MoCap can support animation training by making movement more concrete, improving students’ understanding of production workflows, and enabling more direct engagement with bodily motion as a basis for design [15,16]. Recent work also frames MoCap as an immersive and interdisciplinary learning medium rather than merely a technical tool [19,20].

At the same time, the literature shows that MoCap is not automatically effective and does not simply replace conventional methods. Its educational value depends on how it is integrated into teaching, what learning goals are targeted, and how captured motion is interpreted, edited, and transformed into expressive animation [5,17,21]. Scholars also stress that believable animated movement depends not only on motion accuracy, but on intention, acting, and performance logic [2,12,22]. Beyond animation, studies in other educational fields further suggest that motion capture can support cognitive, psychomotor, and affective learning when properly contextualized [23,24,25]. In this sense, MoCap contains several key affordances that may enhance learning and animation training outcomes, including embodied movement input, real-time feedback, workflow visibility, and performance-based exploration.

2.3. Virtual Reality Technology in Education

Virtual reality has become an important educational technology because it can create presence, interactivity, and embodied engagement in spatially organized virtual environments [13,26,27]. Compared with conventional screen-based learning, VR allows learners to experience and explore information from within a virtual space, which makes it especially relevant for tasks involving spatial and bodily understanding [28,29,30]. This is particularly significant for animation education, where movement must be understood as a spatial-temporal and embodied event rather than as a purely visual sequence.

Research on embodied cognition further supports the educational relevance of VR. Cognition is closely tied to bodily action and environmental interaction [31], and effective immersive learning depends not only on visual immersion but also on meaningful body-based interaction [27,32]. In art and design education, immersive environments may deepen learners’ bodily relationship to content and support more direct understanding of spatial performance [33]. At the same time, the literature cautions that VR is not automatically effective; its value depends on alignment with learning goals and instructional design [13,29]. For animation education, the central question is whether VR can better support the perception, evaluation, and design of movement in space.

2.4. Creativity in Motion Design

Creativity is a central goal of animation education, but in motion design it needs to be defined in an operational way. In creative research, creative outcomes are usually understood in terms of both novelty and appropriateness [34]. For character animation, this means that movement should not only be original, but also readable, workable, and appropriate to the character and scene. Creativity in motion design is therefore multidimensional rather than vague or purely subjective. Figure 2: Four Dimensions of Creativity Assessment.

Existing creativity studies particularly emphasize originality and fluency [34,35], while design research further shows that quality and communicative value must also be considered [36]. This is especially relevant to animation, where a movement may be novel but unclear, or rich in ideas but weak in expressive coherence. For this reason, the present study defines creativity in motion design through four dimensions: originality, fluency, clarity, and aesthetic quality. This definition is also consistent with research suggesting that movement and embodied exploration can shape creative thinking [37], and with Mou’s [17] finding that keyframe and motion capture may support different creativity dimensions in animation learning.

This framework directly supports the research questions of the study. It makes it possible to examine not only whether immersive MoCap-supported instruction enhances creativity overall, but also how it may influence different aspects of creative motion-design performance when compared with conventional keyframe-based teaching.

2.5. Current Study

This study describes a classroom-based comparative study situated in a virtual character animation course. It examines whether immersive VR-MoCap-supported instruction is more effective than conventional keyframe-based instruction in supporting students’ creative motion-design performance, and whether any difference appears across specific creativity dimensions. To address these aims, the study uses a primarily quantitative comparative design, evaluating between-group differences in originality, fluency, aesthetic quality, clarity, and a composite creativity score, because the central research question concerns measurable differences in creative outcomes under comparable classroom conditions. This quantitative comparison is supplemented by a group interview with selected VR-MoCap learners, which is appropriate for examining a shared immersive learning experience and for capturing how students perceived embodied interaction, immediacy, engagement, and creative exploration, all of which are experiential aspects of animation learning that numerical ratings alone cannot fully explain.

3. Methods

3.1. Objectives

Creative motion design requires both novelty and appropriateness: animators must generate multiple plausible action options and then select those that most effectively communicate character, intention, and dramatic meaning. Consistent with divergent-thinking traditions, creativity in motion can be operationalized through observable indicators such as originality, fluency, and quality-oriented criteria including clarity/readability and aesthetic appeal. However, although immersive technologies are increasingly used in animation education, computer animation courses still rarely offer a validated instructional workflow showing how embodied or immersive learning experiences translate into measurable gains in creative performance. To address this theoretical and practical gap, the present study developed an immersive MoCap-oriented training module for use in undergraduate animation instruction and conducted an initial evaluation of its effectiveness, with particular attention to the immersive delivery of motion-design learning.

3.2. Materials and Methods

This study adopted a design-and-development approach [38]. Design-and-development research is a systematic approach concerned with the creation, refinement, and evaluation of educational products, instructional models, or learning environments. An immersive training module was created (As shown in Figure 3 and

Table 1. Design framework and instructional phases of training module.

Steep	Topics	Activities
1	Introduction MoCap	Overview of motion capture technology, its role in animation, types of systems, and its relevance to student learning in character animation.
2	System Setup	Introduction to the system, camera calibration, software configuration, and basic movement capture (e.g., walking and running), followed by initial processing in Maya.
3	Basic Practice	Review of the 12 principles of animation and their application in character animation.
4	Animation Principles and Performance	Applying MoCap data to character rigs in Maya while maintaining animation principles. Importing MoCap data into Unreal Engine for VR applications.
5	Design Practice	Retargeting captured motion data to character rigs in Maya, importing animation into Unreal Engine for VR applications, and developing short creative animation tasks.
6	Performance Recording and Final Animation Development	Recording students’ or actors’ movement data, transferring it to virtual characters, refining the animation, and completing a final MoCap-based animation project.

), tested, and iteratively refined for application in undergraduate computer animation courses. The module was implemented in different virtual lesson environments and was designed to align MoCap-informed practice with classic animation principles and motion-design objectives.

3.2.1. Sample

Participants were undergraduate students enrolled in a computer animation-related course at a comprehensive Asian university. Two instructional conditions were compared: a conventional Keyframe condition (n = 33) and an immersive VR-MoCap condition (n = 35). Most participants were between 19 and 22 years of age. Because all participants were native speakers of the instructional language, the experimental briefing and learning materials were presented in their native language. In addition, all participants had previously received training in Autodesk Maya, providing a shared baseline of 3D animation software experience across conditions.

Prior exposure to virtual reality was limited across the sample. Forty-three participants reported no previous VR use, 14 reported one prior use, and 11 reported two or more prior uses; no participant reported more than three prior VR experiences before the study. Demographic equivalence checks showed no significant difference between groups in sex distribution, χ²(1) = 0.92, p = 0.338, or in self-reported prior experience level, Mann–Whitney U = 545.00, p = 0.659. The VR-MoCap group was, however, slightly younger than the Keyframe group (VR-MoCap: M = 20.74, SD = 0.70; Keyframe: M = 21.21, SD = 0.78), Welch’s t(64.20) = −2.60, p = 0.011. Accordingly, age was retained as a control variable in robustness analyses, together with sex and prior experience.

3.2.2. Procedures

The study was conducted during a structured motion-design lesson. Both groups received the same instructional content and completed the same creative task. The critical difference between conditions concerned the presentation mode of the instructional motion references and learning environment. The Keyframe group viewed the lesson through a standard 2D display and engaged with the material through a conventional, non-immersive workflow. By contrast, the VR-MoCap group experienced the same instructional content through an immersive VR presentation combined with embodied MoCap-oriented interaction.

Following the lesson, participants completed the creative motion-design task individually. Their task outcomes were then assessed on four creativity dimensions: Originality, Fluency, Aesthetic, and Clarity. For analytic purposes, a composite total creativity score was also computed by summing the four-dimension scores. Figure 4 presents an overview of the training procedure.

Creative performance was evaluated by three independent domain experts: two animation instructors from the host university, each with more than five years of teaching experience, and one senior professional animator from industry. Each student outcome was rated independently on a 7-point Likert scale (1 = minimally met the criterion; 7 = fully met the criterion) for originality, fluency, clarity, and aesthetic quality. Before formal scoring, the raters reviewed the dimension definitions and practiced on sample works not included in the main analysis in order to align their interpretations. Inter-rater agreement was then examined using an intraclass correlation coefficient (ICC = 0.80), and the mean of the three raters’ scores was used for analysis.

The instructional session examined in this study was situated within a broader MoCap-related teaching sequence, but the present analysis focused specifically on the comparative effects of the two lesson-delivery conditions.

3.2.3. Materials and Devices

The core instructional material consisted of a virtual lesson on the principles of character animation and motion design developed in Unreal Engine 5.4 (see Figure 5). All digital assets used in the lesson, including character animations, models, and environmental elements, were sourced from freely available assets on the Unreal Marketplace and other open platforms in order to support reproducibility and appropriate licensing for research use.

The lesson began with a tutor-led introduction to the 12 principles of animation. Participants were then encouraged to explore movement patterns associated with these principles, such as anticipation and exaggeration, within the virtual lesson environment. To support embodied observation and experimentation, the environment included 3D character models and supplementary 3D scenes intended to increase perceptual richness and immersion.

Participants were assigned to one of two training modalities. In the immersive modality, users wore VR and MoCap equipment and could move and perform freely within the virtual environment, resulting in a high level of interactivity. In the conventional modality, users completed the lesson through a keyframing-based training approach presented on a standard display, which provided a comparatively lower level of interactivity. This design allowed the study to isolate the contribution of immersive, MoCap-oriented delivery while holding instructional content constant across groups.

Within the VR-MoCap lesson, motion-to-character mapping was handled through a standardized automated workflow supervised by the instructor. After system calibration (see Figure 6), performer movements were aligned to the target character rig using the software’s built-in retargeting tools, including skeleton alignment and overall body-scale normalization, so that major joint motions could be transferred consistently to the virtual character. Students did not manually edit the captured data during the lesson and were not asked to perform advanced post-processing such as cleanup, smoothing, or IK-based correction, because the purpose of the session was introductory motion-design learning rather than professional MoCap finishing.

3.2.4. Data Analysis

Descriptive statistics were computed for all study variables. Group equivalence on demographic variables was examined using Pearson’s chi-square test for sex distribution, Welch’s independent-samples t test for age, and the Mann–Whitney U test for prior experience level [39].

Because Originality, Fluency, Aesthetic, and Clarity represent related but non-identical dimensions of creative performance, the primary between-group analysis used a one-way multivariate analysis of variance (MANOVA), with instructional condition (Keyframe vs. VR-MoCap) as the independent variable and the four creativity dimensions as dependent variables. Pillai’s trace was used as the principal multivariate statistic because of its robustness. When the omnibus multivariate effect was significant, follow-up independent-samples t tests were conducted for each creativity dimension. To control family-wise Type I error across the four subscale comparisons, the p values were reported. The composite total creativity score was analyzed separately using a Welch’s t test because it was derived directly from the four subscales. Effect sizes are reported as Hedges’ g. As a robustness check, covariance-adjusted models controlling for age, sex, and prior experience were also estimated.

4. Results

4.1. Demographic Group Differences

The two groups were largely comparable with baseline characteristics. No significant between-group difference was found in sex distribution, χ²(1) = 0.92, p = 0.338. Likewise, prior experience level did not differ significantly between the Keyframe and VR-MoCap groups, Mann–Whitney U = 545.00, p = 0.659. A modest age difference was observed, with the VR-MoCap group being slightly younger on average than the Keyframe group, Welch’s t(64.20) = −2.60, p = 0.011. Given this result, age was included in subsequent robustness analyses alongside sex and prior experience.

4.2. Group Differences in Creative Training Outcomes

The four creativity indicators showed good internal consistency in the present sample (Cronbach’s α = 0.84), supporting their interpretation as related aspects of a broader creative performance construct. The omnibus MANOVA revealed a significant multivariate effect of instructional condition on the combined creativity outcomes, Pillai’s trace = 0.454, F(4, 63) = 13.12, p < 0.001. This result indicates that the overall profile of creative performance differed significantly between the Keyframe and VR-MoCap conditions. Follow-up Welch’s t tests with Holm adjustment showed that participants in the VR-MoCap condition scored significantly higher than those in the Keyframe condition on Originality, Fluency, and Clarity. The between-group difference in Aesthetic was in the expected direction but did not reach statistical significance. The composite total score was also significantly higher in the VR-MoCap condition. As shown in

Table 2. Between-group comparisons of creative training outcomes.

Outcome	Keyframe (n = 33) M +/− SD	VR-MoCap (n = 35) M +/− SD	Mean Difference [95% CI]	Welch’s t (df)	p	Holm-Adjusted p	Hedges’ g
Originality	3.55 +/− 1.23	4.43 +/− 1.17	0.88 [0.30, 1.46]	3.03 (65.25)	0.003	0.007	0.73
Fluency	2.91 +/− 0.84	4.14 +/− 1.09	1.23 [0.76, 1.70]	5.24 (63.64)	<0.001	<0.001	1.25
Aesthetic	3.15 +/− 1.50	3.74 +/− 1.58	0.59 [−0.15, 1.34]	1.58 (65.99)	0.118	0.118	0.38
Clarity	3.45 +/− 1.18	4.97 +/− 0.95	1.52 [1.00, 2.04]	5.82 (61.72)	<0.001	<0.001	1.41
Composite total score	13.06 +/− 3.60	17.29 +/− 4.07	4.23 [2.37, 6.08]	4.54 (65.73)	<0.001	n/a	1.09

Note. Mean difference = VR-MoCap minus Keyframe. Holm adjustment was applied to the four subscale comparisons only. The omnibus MANOVA for Originality, Fluency, Aesthetic, and Clarity was significant: Pillai’s trace = 0.454, F(4, 63) = 13.12, p < 0.001.

, the largest effects were observed for Clarity and Fluency, followed by the composite total score and Originality. The robustness analysis yielded the same substantive pattern. When age, sex, and prior experience were entered as covariates, the multivariate effect of instructional condition remained significant, Pillai’s trace = 0.451, F(4, 60) = 12.34, p < 0.001, indicating that the observed group differences were not explained by baseline demographic variation.

4.3. Participants’ Attitudes Towards MoCap Design

To supplement the quantitative findings, the study also explored participants’ attitudes toward MoCap-based design learning. A group interview was conducted with selected students from the VR-MoCap condition because this format was appropriate for examining a shared immersive learning experience. It allowed participants to reflect on specific moments during training, respond to each other’s comments, and express both common and differing views on how MoCap supported animation motion design. Overall, most participants described the MoCap experience as engaging, intuitive, and different from conventional keyframe-based instruction. They frequently referred to immediate feedback, stronger bodily involvement, greater creative flexibility, and improved collaboration as major advantages of the MoCap condition.

One of the most frequently mentioned features was the intuitive and embodied nature of MoCap. Compared with keyframe animation, which usually depends on manipulating controllers and manually adjusting poses, MoCap allowed students to use their own bodies to generate motion directly. Participants felt that this made motion design more immediate and easier to understand, especially for beginners or students who found software operation difficult. Several students also noted that the real-time mapping between body movement and virtual character performance increased their sense of presence and made the learning process more vivid.

• Participant A: “Compared with MAYA, MoCap is more direct and faster. It pays more attention to bodily performance rather than keyboard-and-mouse operation.”
• Participant B: “Because I had to perform the action myself, I became fully involved. It gave me a stronger sense of immersion.”
• Participant C: “I felt that my own actions were transferred from reality to the virtual character. It was a very interesting and even magical experience.”

Another positive point raised by participants was efficiency. Many students contrasted MoCap with the repetitive and time-consuming process of setting keyframes manually. They emphasized that MoCap reduced the burden of technical operation and allowed them to focus more on designing and refining motion itself. This perception is consistent with the quantitative result that the VR-MoCap group performed particularly well on fluency and clarity. Participants often described MoCap as a faster way to test ideas, preview actions, and generate continuous motion in a shorter period.

• Participant D: “Motion capture avoided repeated keyframing and improved production efficiency.”
• Participant E: “It is more convenient and efficient than Maya keyframing. It is easier to get started.”
• Participant F: “Using motion capture improved the fluency of motion design, because I could quickly design continuous movements.”

Participants also associated MoCap with stronger creative stimulation. A recurring response was that MoCap enabled them to improvise, try exaggerated movements, and discover unexpected ideas during performance. Unlike conventional keyframe work, which many students felt required them to think first and then construct movement step by step, MoCap encouraged a more exploratory process in which action ideas emerged through bodily experimentation. Some students explained that the system helped them notice motion details, amplitudes, and emotional nuances more clearly, which in turn supported originality and expressive performance.

• Participant G: “MoCap allowed me to improvise, and unexpected effects could appear during performance.”
• Participant H: “In keyframe animation, my ideas are often restricted. MoCap allows me to imitate and create more quickly.”
• Participant I: “When using MoCap, I was more willing to try exaggerated actions, which improved creativity and performance.”

A further theme concerned motivation, emotional engagement, and role empathy. Many students reported that MoCap made them feel more involved in the creative process because they were not simply operating a character from outside, but temporarily “becoming” the character through performance. This embodied experience appeared to strengthen emotional resonance with the animated role. Some students described the virtual character as their “shadow,” while others said that seeing their own movements transferred onto the character helped them understand the role’s personality, emotion, or dramatic state more clearly.

• Participant J: “When I was recording with motion capture, the virtual character felt like my shadow.”
• Participant K: “I do what the character does, and the character does what I do. I felt that I was inside the script and the character setting.”
• Participant L: “When I watched the animation afterward, it felt as if the character and I were connected emotionally.”

Team collaboration was another advantage repeatedly mentioned by the interviewees. Because MoCap-based production involved actors, directors, scene preparation, technical setup, and repeated action adjustment, students experienced the design process as more collective than conventional individual computer-based animation work. Participants explained that real-time feedback encouraged discussion, negotiation, and immediate revision among team members. In this sense, MoCap was not only a tool for producing movement, but also a framework that promoted communication and joint problem-solving during animation design.

• Participant M: “Motion capture promoted team communication. Every part is connected, and each step requires coordination.”
• Participant N: “This production method gives group members more chances to collaborate with one another.”
• Participant O: “When recording an emotional action, we discussed different bodily expressions together before choosing the final performance.”

Despite the generally positive responses, several participants also pointed out some limitations of MoCap. These included the inconvenience of wearing the capture suit, embarrassment when performing in front of others, the need for repeated calibration, limited accuracy in complex actions, and the lack of props or sufficiently large space. Some students also noted that although MoCap was efficient and expressive, certain exaggerated, dangerous, or highly stylized actions still required later adjustment or could not be fully achieved through live performance alone.

• Participant P: “The most challenging part was wearing the motion capture suit. At first it felt embarrassing, but I gradually adapted.”
• Participant Q: “When recording complex actions, the accuracy was sometimes not high, and recalibration was needed.”
• Participant R: “Some dangerous or highly exaggerated actions could not be completed directly and had to be simulated or transformed.”

The interview findings suggest that students perceived immersive MoCap-supported instruction as more engaging, efficient, and creative-supportive than conventional keyframe-based learning. Participants associated MoCap with stronger bodily immersion, faster idea generation, clearer motion feedback, and more active collaboration. To some extent, these responses can be understood in relation to embodied cognition, which emphasizes that understanding is closely connected to bodily action and sensorimotor experience rather than abstract operation alone [31], as well as experiential learning, in which knowledge develops through action, feedback, and revision [40]. Students’ repeated references to immersion and direct involvement also correspond to the idea of presence in immersive virtual environments, especially the feeling of “being there” during interaction [41]. At the same time, the interview responses suggest that creativity under the MoCap condition was not only an individual matter, but was often supported by discussion, coordination, and shared adjustment within the group, which is also consistent with the view that creative practice can emerge through collaboration [42]. However, participants also recognized practical and technical constraints that should be addressed in future instructional design, including better equipment, more action guidance, richer props, and more flexible production support. These qualitative responses provide useful contextual support for the quantitative results, especially the observed advantages of the VR-MoCap condition in originality, fluency, clarity, and overall creative performance.

5. Discussion

This study examined whether immersive MoCap-supported instruction could better support creativity in animation motion design than conventional keyframe-based instruction in a computer animation learning context. The findings showed that the immersive VR-MoCap condition was associated with higher overall creative outcomes, with significant advantages in originality, fluency, and clarity. By contrast, no significant group difference was found for aesthetic quality. These results suggest that immersive instructional delivery may selectively support some dimensions of creative motion-design learning more effectively than others.

The results reported in Section 4.2 show that the advantage of the VR-MoCap condition was not marginal but patterned. The omnibus multivariate effect was statistically significant (Pillai’s trace = 0.454, F(4, 63) = 13.12, p < 0.001), and the same substantive pattern remained after controlling for age, sex, and prior experience (Pillai’s trace = 0.451, F(4, 60) = 12.34, p < 0.001), which strengthens the interpretation that the observed differences were linked to the lesson-delivery condition rather than to baseline variation. More importantly, the effects were selective rather than uniform. The largest gains appeared in clarity (Hedges’ g = 1.41, Holm-adjusted p < 0.001) and fluency (g = 1.25, Holm-adjusted p < 0.001), followed by the composite score (g = 1.09, p < 0.001) and originality (g = 0.73, Holm-adjusted p = 0.007), whereas aesthetic quality showed only a small, non-significant difference (g = 0.38, p = 0.118). This distribution matters because it argues against a simple claim that immersive technology enhances all aspects of animation creativity equally. A more plausible interpretation is that immersive VR-MoCap mainly supported the generation of multiple motion possibilities and the legibility of action in space.

This pattern partly converges with earlier research, but it also extends it in an important way. Mou [17] reported that motion capture supported fluency, whereas keyframe work offered slightly stronger support for originality, and explained that result partly by students’ ability to manipulate more extreme poses manually. The present findings agree with that earlier study on fluency, but they do not reproduce the originality disadvantage. Instead, originality was significantly higher in the VR-MoCap condition. This difference suggests that originality is not inherently advantaged by keyframing and weakened by motion capture; rather, it appears to depend on how motion capture practice is designed. When MoCap is treated mainly as a recording tool, students may remain close to habitual or natural movement patterns. When it is embedded in an immersive lesson that encourages bodily improvisation, immediate retakes, and spatial evaluation, it can also stimulate novelty. This interpretation is reinforced by the group interview, in which students described the immersive condition as more direct, efficient, and supportive of improvisation and unexpected ideas. It is also compatible with earlier research (Wang et al. 2011) [16] suggesting that MoCap can function as a more efficient teaching workflow than traditional keyframing when students learn motion through practice and feedback.

5.1. Theoretical Implications

Moreno and Mayer [43], Makransky and Petersen [29], and Petersen et al. [30] argue that the educational effects of immersive media are better understood in terms of specific affordances and mediating learning processes than as simple media-superiority effects. From this perspective, the present findings support a feature-based interpretation: immersive VR-MoCap instruction appears to strengthen particular components of creative motion-design performance rather than improving all dimensions equally. In character animation, the generation and evaluation of movement depend heavily on the perception of timing, spatial trajectory, body orientation, and performance intent, and embodied or immersive systems can make such information more directly available through sensorimotor engagement and action-centred perception [27,31,44]. This helps explain why the immersive condition showed stronger effects on originality, fluency, and clarity than on aesthetic quality. Existing research on MoCap in animation education likewise suggests that motion-based and immersive workflows can support embodied exploration, creative experimentation, and more intuitive understanding of movement, while different instructional media may support different dimensions of animation creativity [16,17,20]. At the same time, the non-significant difference in aesthetic quality is theoretically in-formative rather than contradictory. From a design-learning perspective, aesthetic quality typically develops through iterative critique, timing adjustment, stylistic calibration, and repeated revision cycles rather than through a single short-term lesson [45,46,47]. This suggests that immersive instruction may be especially effective in the earlier phases of creative motion design, when students are generating alternatives, testing motion ideas, and evaluating whether an action reads clearly, whereas aesthetic refinement may require longer instructional exposure and repeated feedback. In this sense, immersive MoCap-supported instruction should be understood not merely as a technical enhancement, but as a pedagogical bridge linking animation principles, embodied observation, and creative motion design [20,48].

5.2. Practical Implications

The present findings also have practical implications for computer character animation education. A recurring problem in animation education is the gap between foundational character animation training and software-based instruction, such that students may learn digital procedures without fully understanding movement as embodied performance. In this context, the results suggest that immersive VR-MoCap should be positioned not as a replacement for conventional keyframe instruction, but as a complementary pedagogical layer that reconnects animation principles with bodily action, spatial perception, and performance intent. Because the immersive condition showed stronger outcomes in originality, fluency, and clarity, but not in aesthetic quality, immersive MoCap appears especially useful in the earlier phases of motion-design learning, when students need to generate alternatives, test movement ideas, and judge whether actions read clearly before later stages of stylistic refinement. A practical curriculum implication is therefore to adopt a staged or hybrid workflow in which immersive MoCap activities are introduced early for embodied observation, motion exploration, and performance testing, and are then followed by keyframe editing, critique, and refinement tasks that target timing control, aesthetic finish, and character-specific nuance [15,16,17,19,20].

The study also offers guidance for the design and development of immersive computer animation systems. The interview responses indicated that students valued immediate feedback, bodily involvement, creative flexibility, and collaboration, suggesting that effective systems should support rapid cycles of action, preview, reflection, and revision rather than relying on immersion as a display feature alone. From a design perspective, immersive animation platforms would benefit from real-time character preview, simple replay and comparison of motion takes, audience-perspective and performer-perspective viewing options, and smooth transitions between capture, VR visualization, and downstream animation software. They should also support collaborative review, peer observation, and instructor walkthroughs, because animation learning is shaped not only by individual practice but also by social critique and co-construction. At the same time, developers should pay close attention to usability, calibration demands, ergonomic comfort, and cognitive load, since embodied systems are most educationally effective when body-based interaction is meaningful and technological complexity does not overshadow instructional goals [13,27,28,30,49,50]. From this perspective, the practical goal is not simply to build more immersive systems, but to develop human-centred animation learning environments in which immersion, interactivity, feedback, collaboration, and workflow integration are deliberately aligned with the specific demands of creative motion design. The scalability of immersive VR-MoCap in higher education depends on institutional feasibility. Implementation requires access to VR headsets and motion capture equipment, adequate physical space, calibration time, and staff or technical support, which may limit routine use in resource-constrained settings. For this reason, immersive VR-MoCap may be most feasible as a targeted instructional component within a staged or hybrid workflow, using shared immersive facilities for early-stage motion exploration and conventional keyframing for later refinement and finishing.

5.3. Comparative Analysis with Related Studies

The findings of the present study can be further clarified by comparing them with related studies reviewed in Section 2. Earlier work by Bennett and Denton [15] and Wang et al. [16] emphasized the practical value of motion capture in animation education, particularly its ability to connect classroom learning with production-oriented workflows and to help students understand motion through direct practice. The present study supports this general view, as students in the VR-MoCap condition achieved stronger outcomes in fluency and clarity, and interview responses also suggested that MoCap made motion design feel more immediate, efficient, and intuitive. However, the present study extends these earlier contributions by providing a controlled classroom-based comparison with conventional keyframe instruction and by evaluating creative performance through multiple dimensions rather than discussing MoCap mainly as a workflow or technical training tool.

The closest comparison is Mou’s study on keyframe and motion capture in character animation education [17]. Mou found that motion capture could support fluency, whereas keyframe animation could offer advantages for originality because students could manually construct more unusual or exaggerated poses. The present study partly confirms and partly extends this finding. Consistent with Mou’s work, the VR-MoCap condition showed a strong advantage in fluency. However, unlike Mou’s findings, originality was also significantly higher in the VR-MoCap condition. This difference suggests that originality is not determined by the use of motion capture alone, but by how motion capture is embedded within the instructional environment. In the present study, the combination of immersive VR presentation, embodied movement exploration, immediate feedback, and repeated performance testing may have encouraged students to improvise and generate more novel motion ideas.

The results also contribute to broader research on immersive learning. Prior studies on VR education have argued that the learning value of immersive environments depends on specific affordances such as interaction, embodiment, presence, and feedback, rather than on immersion alone [13,29,30]. The present findings support this affordance-based interpretation. VR-MoCap did not improve all creativity dimensions equally; instead, its strongest effects appeared in originality, fluency, and clarity, while aesthetic quality did not differ significantly between groups. This selective pattern indicates that immersive MoCap is especially useful for early-stage motion ideation, embodied exploration, and action readability, but that aesthetic refinement may still require extended critique, timing adjustment, and post-production work.

Compared with recent work that discusses MoCap integration into animation, visual effects, and game-design curricula [19,20], the present study offers a more specific empirical contribution. It evaluates one structured immersive MoCap-supported lesson in an authentic undergraduate animation course and compares it directly with a conventional keyframe-based condition. Therefore, the novelty of this study lies not only in using VR and MoCap in animation education, but in demonstrating how an immersive MoCap-supported learning environment affects different dimensions of creative motion-design performance. The findings suggest that VR-MoCap should be understood as a complementary pedagogical approach for early-stage creative exploration rather than as a full replacement for conventional keyframe animation training.

5.4. Limitations and Suggestions for Future Research

Several limitations should be considered when interpreting the present findings. The study evaluated one structured motion-design lesson drawn from a broader MoCap-related teaching sequence and compared two lesson-delivery conditions at the whole-lesson level. The present design therefore cannot determine whether the advantage of the immersive condition was produced by immersion itself, by VR affordances such as head-tracked viewing and depth cues, or by the coupling of immersive presentation with embodied engagement. This distinction matters because recent immersive learning research suggests that educational effects are affordance-specific and strongly shaped by instructional design rather than by VR as a uniform medium [28,29,30].

The generalizability of the findings is also constrained by the sample and instructional context. Participants were drawn from a single undergraduate course at one university, the VR-MoCap group was slightly younger than the Keyframe group, and prior VR experience was limited across the sample, although robustness analyses suggested that the main pattern of results was not explained by these differences. Variability in learner characteristics, technological familiarity, and educational context has been shown to shape learning with immersive media [13,29,32]. Broader sampling across institutions, stronger randomization or stratified assignment, and more explicit controls for prior VR and MoCap experience would therefore strengthen future evaluations.

A further limitation concerns the assessment of creative outcomes. Creativity was operationalized through four rated dimensions: originality, fluency, aesthetic quality, and clarity, as well as a composite total score, while the qualitative component took the form of a supplementary group interview. Although this multidimensional approach is appropriate for motion-design tasks, creativity in design-oriented learning is difficult to capture through any single rating format, and stronger interpretation usually requires triangulation across products, processes, and learner accounts [14,35,36]. Future research should therefore combine dimensional ratings with expert assessment of finished animation artifacts, process-based indicators such as iteration and exploration breadth, and more systematic mixed-method data collected from both instructional conditions.

Ultimately, the non-significant difference in aesthetic quality should also be interpreted in relation to the duration and scope of the intervention. The present study examined an early-stage motion-design lesson, whereas aesthetic refinement in animation often depends on repeated critique, stylistic calibration, and extended cycles of revision. Prior work in animation education likewise suggests that MoCap becomes most pedagogically valuable when it is integrated into longer instructional sequences and linked to retargeting, critique, and production-oriented refinement [16,17,19,20]. Future work should therefore evaluate the full immersive MoCap training module over a longer period and examine whether early gains in originality, fluency, and clarity translate into stronger downstream animation quality and curriculum-level development.

6. Conclusions

This classroom-based comparative study provides an initial evaluation of one structured lesson within a broader immersive MoCap-supported training module for computer character animation. Accordingly, the observed advantages of the VR-MoCap condition should be interpreted primarily as short-term lesson-level effects rather than as evidence of curriculum-wide impact. The broader curriculum implications are promising, but they remain provisional and require longer-term evaluation of the full training module. Compared with conventional keyframe-based instruction, immersive VR-MoCap-supported delivery yielded higher overall creative performance, with significant advantages in originality, fluency, and clarity, whereas aesthetic quality did not differ significantly. These findings suggest that immersive instruction may be particularly valuable in the early phases of motion design, when students need to generate alternatives, explore movement possibilities, and judge whether actions read clearly in space. Supplementary group-interview data further indicated that students experienced the immersive condition as more engaging, intuitive, and supportive of immediate feedback, bodily involvement, and creative flexibility. Taken together, the study suggests that immersive MoCap is best understood not as a replacement for conventional keyframe training, but as a complementary pedagogical approach that can reconnect animation principles, embodied performance, and digital production practice. It therefore offers both an initial empirical basis for curriculum innovation and a foundation for longer-term evaluation of the full training module.