← Study 1 · Study 2 of 3 · Next: Study 3 →
Aim
Participatory design is a design philosophy that involves the users in the design process to understand more about the user's needs, challenges/problems, and requirements. A participatory design study was conducted with two groups of participants: experts and beginner practitioners of Tai Chi. The result validates the beginner's need for feedback and invaluable practical input from the experts to be implemented through iterations. The developed system used the Tai Chi movements recommended by the experts in the participatory design. PoseNet was used to process the target video and the user's real-time video input, and cosine similarity results are shown in the interface as visual feedback. The result showed that feedback is helpful for the participants, although more improvement would be required. Design recommendations must be considered when designing movement-based interactions for mindfulness applications.
Participants
Initial recruitment was conducted through the Tai Chi Union of Great Britain, approached East and North London Buddhist Center, NHS Tai Chi group and Chinese Internal Arts Association. However, this approach resulted in few leads or responses.
The researcher attended taster sessions to approach the experts and enrolled in Wu Tai Chi Academy in Archway, Shadwell Yang-style Tai Chi class and Adult Tai Chi Class at Battersea to approach the experts. Snowballing recruitment was also used to recruit beginner learners. This is because there is a challenge in recruiting Tai Chi practitioners.
The participants were given £30 as compensation for their time.
Method
The study consisted of three elements: a semi-structured interview, a design exercise with two user groups and a design exercise by doing prototyping. The study followed Aryana et al.(). This study focused on eliciting design requirements for implementation and aimed to analyse the process using a participatory design approach. In order to effectively leverage the human-centred design approach, it is imperative to involve experts and end users in the design process. The reason is that even though this application is designed for beginner learners, they would need to learn the critical elements of Tai Chi. Experts know this and would be a valuable resource in designing the content. It is also important to see from both points of view, from experts and beginner users, to know the essential points, their challenges, and what part of Tai Chi both experts and beginner learners think would work in an online environment.
Analysis
Thematic analysis was conducted using NVivo on every participant's transcribed recording data. Thematic analysis, a widely used qualitative research method, was employed to analyse the interview data collected for this study. Utilising NVivo software facilitated the systematic organisation and exploration of the transcribed recordings from each participant. Thematic analysis involves a rigorous process of identifying, analysing, and interpreting patterns or themes within the data, allowing for the extraction of meaningful insights and understanding of participants' perspectives.
Findings
Challenges in offline teaching
The experts considered that the main challenge in teaching Tai Chi is the people or the students. What they meant by the students was understanding the students' condition to give them a suitable style of teaching. Some examples of this are when the students do not fully inform the experts about their medical conditions, exhibit antisocial behaviour, or have unrealistic expectations of Tai Chi (e.g. Tai Chi can make them levitate, such as the depiction in movies or able to cure all conditions).
Another challenge for experts is the difference in students' learning pace. Naturally, age and understanding of one's body awareness will affect the pace of learning. Tai Chi also involves learning one's own body and movement. The movements themselves are not complex, but getting the proper movement with all coordinated limbs and factors and the correct sequences takes time to master. For example, a student who is also a dancer will be able to memorise the movements faster because they have higher body awareness than those who do not have good body awareness. Also, younger students will likely memorise the movement faster than older students due to the possible movement restrictions in older students.
The other challenge is the limitation in what experts can share in a class. The usual practice takes about 2 hours. In this time span, most of the time is usually spent warming up, teaching forms, and doing the forms in groups. Sometimes, there is not enough time to teach the theories and philosophies, or the experts put some theories here and there during practice. However, there was a frustration noted in the interview.
Suggestions from the participants
Experts
- E1. Human interaction and feedback are important; therefore, a multi-method approach is needed both offline and online. Offline as pre-recorded with 3D and 360 video. Another app for theory. A body suit for movement practice. The remote control and sensor in hand can detect movement and return feedback with a full-body suit. These are compared to a Tai Chi master and to have body suit to induce students to move. Detection of movements with tactile feedback in a full-body suit to replicate touch correction to the body position.
- E2. 3D avatar like the teacher that students can pinch to zoom in to see the subtle and detailed movements.
- E3. VR to enable immersion and make it easier to use imagination and attain mindfulness, it also needs a multifaceted approach.
- E4. Create an avatar to spar with real impact on the body.
Beginners
- B1. VR would be a good solution for a virtual experience in a group. AR could also be a solution. Both VR and AR could allow you to train in a group remotely (like in in-person practice). A virtual mirror would be beneficial to reflect on their own movement or have a shadow overlay to tell your body movement. The shadow overlay would be the teacher’s movement that you have to follow. Because in videos like YouTube it is a mirror image, it would be easier to not think on which side of the limb you should be moving. The shadow is in one colour from transparent to bold opacity. Use red if the movement is wrong. It should also be fun like a game. A similar game would be like Just dance, try to follow the movement with the rhythm. As a game, it needs to have point system and rank, this would emulate the community part of Tai Chi.
- B2. The participant would like to have a projection to the wall see the 3D movements of the teacher. Participant B2 also would like to have the projection moves as they move so it would be a 3D avatar that moves with them. A controller and posture on a video with camera to record your movement. Instruction at the corner of the interface with song and rank. The avatar will be the teacher, the movement is can be interactive with direct feedback.
- B3. Practice in real environment, using hologram or VR but with the ability to connect with kinetic movement.
- B4. It should be a phone app with pre-recorded videos but it also has an option for live interaction with chat/call/hybrid and online class. A gamification would be good to have and progress tracker. It will be helpful to know your progress and how much practice you have done. It would be a multi-device system. The smartwatch to detect movement in conjunction with the system. It will have live chat to talk to your teacher. Music will be good to tie with movement, because it is easier to move with rhythm. It is also useful to track if I move too fast or too slow. Other idea would be a avatar projection or a game like just dance that makes you move like the model in the video. The tile underneath you track your movement like on a mat.
- B5. A concept for a mobile app is proposed that features a menu with different difficulty levels (for example: beginner, intermediate, or expert), a calendar, and goal setting functionalities. Users start with instructional videos, which followed by a practice mode where the app tracks movement using smartphone or smartwatch. Animations guide users by showing correct postures or "power" moves. Feedback is given based on real-time motion, such as speed or accuracy, using visual cues like a frame that changes color (e.g., green/red) when alignment is correct. The app is intended to be accessible on phones for ease of setup, while tablets and projectors offer larger displays for group or classroom use.
- B6. An ideal setup includes a smartwatch paired with an online app for training. The smartwatch tracks heart rate and movement, while projection and even drone use enhance immersion. The app includes features such as, calendars, challenges, progress badges, and instructional content. Online classes are streamed, with teachers guiding and correcting users live, sometimes supported by example students. Additional features include short 1-on-1 meetings, available through tiered pricing. The program emphasises on instructor presence, user motivation, and maintaining privacy during home use.
- B7. A VR-based system is envisioned that utilise a motion-detection platform that uses lines and character guides on screen to help users match correct postures. Deep learning is used to detect subtle movements, differences in acceleration, and quality of user performance. The system can evaluate users based on timing and precision, similar to rhythm or dance games, providing structured feedback and progression pathways for beginners to advanced users.
← PhD overview · Next: Study 3, development and evaluation →