Intercorporeal Biofeedback for Movement Learning

2.1 Movement Learning: A Situated Perspective

Movement is the basis and premise of our being in the world [105] and plays a primary role in human experience and communication [67]. Learning how to move and becoming aware of and controlling movement are lifelong challenges [15, 58, 93, 107]: from our earliest days of crawling and walking to sophisticated body enculturation [58], such as participating in sports and fitness.

Researchers have long investigated and debated [99] how people learn to use their bodies, and the pedagogical literature includes many theoretical perspectives [5, 99, 119, 120] (examples include behaviorism, cognitivism, constructivism, and sociocultural theories of learning; see [5, 64] for a detailed overview). These theoretical perspectives differ in the learning metaphors they adopt; their assumptions about knowledge and learning; and the roles of teacher, student and environment [5, 64]. Historically, the perspectives have shifted from foregrounding external factors (i.e., positive reinforcement in behaviorism; information processing in traditional cognitivist perspectives [64]) to progressively considering the student as an embodied subject [5, 64, 99].

Movement learning is a central part in becoming proficient in most sports and some fitness practices. This happens when the practice includes physical exercises that must be performed in particular ways [28]. Such exercises typically present movement norms—movements involving certain qualities, patterns, postures, and spatial and visual orientations [9, 20]—used to qualify what constitutes a desired or undesired performance. In such practices, learning movement is generally geared towards learning to move in the desired ways, as well as avoiding and recovering from undesired ones. Such movement learning is highly situated and social [86], often featuring an expert practitioner in a teaching role (coaches, personal trainers) who facilitates training and guides students in developing capabilities.

Since this article is concerned with such social and situated movement learning practices, it builds on recent sociocultural situated perspectives on movement learning. Drawing from anthropology and phenomenology, the situated perspective focuses on how people participate in culturally situated movement practices, viewing movement knowledge as dependent on those practices and contexts [5, 99, 102]. This perspective is permeated by an embodied understanding of decision-making, cognition, perception and action, which are seen as inseparable from each other [21, 66] and the individual's social and cultural learning environment [102].

From situated perspectives, learning manifests in observable changes in performance and in developing the ability to move in specific ways [32]. Learning is closely tied to training and practice [31]: sensorimotor capabilities and competencies are developed through and evolve from participating in established communities of practice [102]. Each movement practice is characterized by particular goals, normative movement stances, values and aesthetics [5], and movement learning in them can be viewed as a process of an individual's enculturation toward becoming a competent member of the practice [99, 102]. At the center of this process are the relations and interactions between individuals and their physical and social environment [102].

2.2 Intercorporeality and Movement Learning

The situated perspective discussed above provides insight into movement teaching and learning practices at a high level, but does not support analyzing the in-situ situation process of teaching and learning in depth. We draw from phenomenological accounts of social cognition [85, 86, 117] to look at intercorporeality as a fundamental condition through which social movement teaching and learning occur.

Intercorporeality concerns the corporeal relations that emerge in interaction, through which people build meaning and action [86], uncovering how humans are able to incorporate the perceptions and actions of other people in their experience [86]. The actions of others prompt possibilities for sensemaking and action in ourselves and vice-versa [117], based on our innate sensitivity to others’ movements [86]. Intercorporeality underpins our capacity to learn movement through observation and imitation of others [117]. The concept is closely related to the concept of second-person perspectives on body and movement [34], which describes how we, through empathic observation [74], can attune our subjective perspective to others’ bodies and movement and be sensitized to them [49]. The second-person perspective captures our ability to bridge our first-person perspective and an observational, third-person perspective on other's movements [116], and all three perspectives are held simultaneously [49].

Intercorporeality enables social sensemaking in which action and meaning are generated and sustained in a social and participatory way [53, 57]. This social sensemaking is characterized by patterns of coordination [40, 86] and a reciprocal perceptibility between bodies: in other words, people need to be able to appreciate their own sensorimotor actions, as well as those of the other [86] for mutual understanding and communication.

Intercorporeal engagements in sports can take many forms, such as agonistic as in martial arts, or symmetrical, as in synchronized swimming. The most relevant for this article is co-operative [40]: teachers and students take turns articulating, communicating, performing, assessing, and correcting sensorimotor actions. Notably, these co-operative engagements are multimodal [40]: people (both teachers and students) draw from myriad interactional resources to build meanings and action. These interactional resources include speech, demonstrations, gestures and pointing, touch and mobilization (moving somebody else), spatial elements (e.g., references in the environment), and tools and equipment [7, 21, 125]. In the empirical investigations below, we will show how intercorporeal biofeedback becomes a material, used seamlessly together with the pool of interactional resources that people already employ during teaching and learning.

2.3 Biofeedback Technologies for Body Awareness and Physical Literacy

A final strand of work that is relevant for articulating intercorporeal biofeedback is other body-centric technologies that promote body awareness and physical literacy. From this perspective, our concept builds on previous design work on biofeedback technologies. This is a broadly scoped concept for a range of technologies that can sense different aspects of our bodies (such as movement qualities or physiological processes) and provide feedback to us. Past research has shown the great potential of biofeedback technologies to enhance our body awareness and affect our actions [9, 38, 51, 61, 72, 89, 93]. Biofeedback technologies have the capacity to reveal otherwise “hidden” or “inaccessible” [93] aspects of bodily experience, such as proprioceptive sensations. Biofeedback technologies can augment bodily perception through other senses (vision, hearing, and touch), making sensation more tangible and obvious, so that we can understand it and act on it [48, 72, 89, 93].

In HCI design research, prior work has shown that biofeedback technologies can guide our attention focus between body areas or physiological functions [51, 94] and support prolonged attention to a specific part [48, 92]. They can also support us in dynamically shifting attention between the technology and our body [94].

Biofeedback technologies may foster different bodily-perceptual relations, depending on the design goals. Biofeedback works within soma design [48] often foreground a deep, prolonged felt appreciation of our bodies [51]. It can be in intimate correspondence [48] with the person, when the synchronization of biofeedback and body or movement is immediate enough to allow us to perceive the former as an extension of the body, as a mirror of the self [48]. This allows technology to be incorporated into the perceptual-bodily self-experience [55, 132], becoming an extension of our body through which we perceive and act [48].

Other design inquiries have emphasized the dynamic interplay between focus on the body and the tool. For example, the concept of present-at-body [94] centers the reflective use of wearable biofeedback tools for developing bodily self-awareness and dynamically guiding our attention. Biofeedback can be also used as a defamiliarization tool, to help people obtain a different understanding of their body and movement. For example, prior work on metaphorical sonifications [71] shows how body perception and action can be altered by augmenting movement through sounds that a body would not naturally produce, such as the sound of the wind or a mechanical gear turning. Biofeedback can also be separated from the immediate movement and bodily experience, so that the technology becomes the focus of our attention. Examples of these can be found in biofeedback technologies that provide symbolic representations to facilitate comprehensive “readings” of the referent, such as heart-rate numbers in self-tracking applications [139].

These discussions of how biofeedback technologies can mediate our perception and action will underpin the analysis of the interactive qualities required for interactional biofeedback.

3.1 Predominant Technologies for Movement Learning

Most research and design in movement learning has focused on individual experiences [10, 16, 84, 128]. We see two main types of contributions arising from such work. First, a range of technology-driven projects aim to advance technical capabilities and develop new or better ways to sense, capture, model, predict and evaluate movement (e.g., [3, 42, 68, 82, 130, 131]). These works often address a particular exercise (or small set of exercises) and perform experimental studies to prove the efficiency of their technology. Much less often is the technology trialed in authentic contexts or even with authentic users, leaving their real-world potential unclear. The second strand of projects are those supporting individual learning experiences in particular practices. These works often identify a need and contribute a technology solution that addresses it through, for instance, automated assessments on performance and feedback [11, 13, 17, 18, 27, 129, 134], and/or providing instructions and cues [4, 65, 110, 118, 140]. Such work often shows relevance in the respective practices but with limited scope.

Underpinning most of these works is a positivistic understanding on movement [79, 127, 138], an understanding under which movement can be abstracted from particular bodies and situations that give rise to it, and modelled under generic benchmarks. This understanding often disregards the sociophysical context of movement learning practices [79, 121, 125] and individual capabilities and needs [111, 127]. Technologies built on positivistic understandings of movement often fail to provide meaningful qualitative feedback and address individual needs [125, 138]. They may also restrict the quantity and quality of exercises that they can meaningfully cater to [8, 79, 125], limiting their usefulness in movement practices with diverse exercises [127].

Studies of these predominant design approaches to technology for movement learning show that they work well to address goals such as engagement [81], autonomy [129], and accessibility. However, they offer little insight or guidance into how we, as designers, can design for complex, social and situated contexts of movement learning.

3.2 Alternative Technologies for Movement Learning: Open-ended Actuations

Other technologies for movement learning align more with the design goals for intercorporeal biofeedback, in that they address situated or social contexts through open-ended feedback. Open-ended technologies provide qualitative representations without a fixed meaning and prescribed actions to be performed [12, 47], leaving feedback open to different interpretations and courses of action [12]. Hence, end users become active co-constructors of the meaning and functionality [12] of the feedback, which can be appropriated and changed over time [12, 47].

Most open-ended technologies for movement learning enhance movement awareness, with the aim of helping people understand and adapt their performance. As they are open for interpretation, they can address different performances and individual students. Some examples include open visualizations of weight distribution and center of pressure via color-coded pressure maps in weightlifting [30] and acrobatics [122]. Other visualizations augment the whole body, for example interactive mirrors for learning martial arts [43] and video-visualizations for training football [46]. Other works employ sonifications, such as augmenting the sound of a golf club's swing [95], changes in the center of gravity for skiing [44], and changes in pressure distribution in speed skating [112].

Other open-ended technologies support different forms of remote teaching. For example, a rock climbing vibrotactile wearable [83] is worn by beginner climbers and is controlled by an instructor on the ground who provides real-time feedback and guidance through a tablet. Prior to climbing, climbers and instructors jointly decide on the meaning of different vibrations.

Other designs support teaching and learning in situated contexts. An example is TacTowers [36, 76], a playified approach to training fast movement anticipation and reaction in handball. TacTowers consists of several interconnected poles rigged with touch sensors and light actuators. Lights can be tapped, and this action triggers another light somewhere else. TacTowers enables coaches to instruct specific movement patterns in the form of games (e.g., someone taps the light, the other tries to defend it) and students to practice offensive and defensive skills.

These examples show the potency of open-ended technologies to be used in social and situated practices, as they can cater to different exercises, correct or incorrect performances, and people.

4.1 Strong Concepts for Design

In this article, we articulate the specific design strategy of using biofeedback to enhance teaching and learning as a strong concept [52]. Strong concepts are a type of intermediate-level knowledge [75], resulting from Research through Design [37] processes, that aim to help other designers generate related designs. Höök and Löwgren [52] characterize strong concepts as: (1) characterizing interactive behavior as it unfolds over time; (2) possessing interactive qualities that reside between technology and people, speaking about use practices and behaviors; (3) presenting a core design idea that can cut across individual-use situations and application domains; and (4) being more abstract than particular instances, which allows the strong concept to be concretized in different design particulars with different technical implementations.

Strong concepts are primarily identified inductively, from empirical material, which distinguishes them from bridging concepts [24] which primarily draw upon theory. Their validation comes from vertical grounding: that is, relating the strong concept to theories that help articulate why and how the strong concept works [52] as well as to the empirical work from which they are derived. The validation of strong concepts is further strengthened from horizontal grounding: that is, relating the strong concept to other existing empirical examples outside from the ones it stems from as well as other intermediate-level knowledge [52]. The horizontal grounding substantiates the strong concept's potential for inspiring subsequent design as it shows the applicability of the concept outside the context in which it was developed. Together, the vertical and horizontal groundings of a strong concept support evaluating the concept according to the evaluative criteria suggested by Höök and Löwgren: to be contestable (i.e., novel), defensible (i.e., horizontally and vertically grounded) and substantive (i.e., able to inspire new designs) [52].

The concept itself emerged from our previous empirical work on movement learning [77, 121, 127], in which we designed and used a series of Training Technology Probes (or TTPs), to support movement learning practices. The TTPs are minimalist, open-ended wearables that teachers and students use during teaching and learning. We have published work on the TTPs’ design, integration in practice, and overarching effects, but in this prior work we did not develop a thorough understanding of why and how the TTPs were useful tools to support movement teaching and learning. In this article, our aim is to provide a characterization of shared design patterns between our design exemplars, which can generalize beyond both the technology and the movement learning practices of our work.

4.2 Intercorporeal Biofeedback for Movement Learning: Empirical Grounding

Although several of our design inquiries (e.g., [77, 121, 127]) have contributed to the formulation of intercorporeal biofeedback, in this article we focus on empirical material from two specific design projects. The first project is Super Trouper, a technology-supported circus training course with circus instructors for children with mild motor issues. The second is BL Strength, a series of strength-training workouts using a personal trainer with practitioners at different skill levels. In both projects, we developed and tested different TTPs to support movement teaching and learning. These two projects were selected as their studies were extensively documented through video, providing rich observational material for in-depth analysis. For this article, we revisit the empirical material from the two projects and re-analyze it through an in-depth interaction analysis [59] on the video data, focusing on the interactional qualities that made TTPs a useful tool to teach and learn movement.

The two projects have been extensively described in previous publications [77, 127], and are here summarized to provide a context for the analysis presented in this article. Both projects followed a Research through Design [37] approach, following principles of practice-based design [114] targeting human activities [80, 133] and with awareness of the fact that the studied practices already feature a rich fabric of interactions, materials, and movements. We approached design as the crafting of an experience with several important design resources: people and their bodies, practices, spatial resources, sociotechnical materials and other materials and tools [48, 80, 133]. In this understanding, designing encompasses orchestrating experiences and behaviors that result from combining this diverse set of design elements in a process that originates as much with the participants as with the designers. The TTPs as well as their use continued to be (re)designed throughout the process—by teachers and students—in an ongoing process of meaning-making in interaction.

The design process in both projects [77, 127] included an initial phase in which the designers prototyped technology together with the instructors; explored critical design decisions such as what movement qualities the TTPs could augment and the exercises in which they could be used. These explorations included details such as investigating where to position the wearable TTPs on the bodies, and their potential uses in the exercises. In a second phase, instructors deployed the TTPs in training sessions, and both instructors and students performed exercises with the TTPs. We have previously shown that open-ended feedback allowed participants to alter and adapt the original meanings and uses [77, 127], but did not previously develop an understanding of why and how the technology supported this process.

4.2.1 Super Trouper: Circus Training with Four TTPs.

Our first dataset comes from Super Trouper, a project exploring wearable technology and circus training to support children with mild motor challenges and sensory processing disorders [77, 124]. Super Trouper ran from 2017 to 2021. We partnered with Cirkus Cirkör, Sweden's largest circus company, to develop technology-supported circus training courses for children aged 7–12 with challenges in movement sensing (registering, orienting, interpreting, and organizing proprioceptive information) [106] and actuating (difficulties with controlling and eliciting motor responses) [87]. Throughout the project, we explored how to create engaging, intrinsically motivating movement learning activities that also enabled the children to work on and with movement awareness and control.

Super Trouper's different phases ranged from ideation workshops to technology development, course planning and recruiting, and data gathering and evaluation in courses with children and circus instructors. The data used in this article corresponds to design activities and courses from 2017 and 2018, illustrated in Figure 1.

Fig. 1.

View Figure

Super Trouper started in Spring 2017 with observational studies of the circus training courses that Cirkus Cirkör was offering. From those studies, we were able to understand the structure of a circus training course and also start identifying challenges in teaching and learning [77]. Based on these studies, and together with the pedagogic coordinator of Cirkus Cirkör, in Fall 2017 we planned a new course together with the circus instructors and recruited participants. In parallel, we researchers also ran a series of ideation workshops (W1, W2, W3) that resulted in a plethora of ideas for technology to augment movement awareness and control in movement learning settings [126]. Both circus and yoga instructors participated in these workshops.

In Spring 2018 and Fall 2018, we ran two courses with children. Three experienced circus artists and instructors (IN1, a woman; IN2 and IN3, men) participated. For each course, Cirkus Cirkör recruited children exhibiting mild motor challenges (5 in the first course, and 7 in the second). Both courses included training in several circus disciplines: acrobatics (e.g., headstands), aerials (e.g., silks), juggling (e.g., balls, clubs) and balance (e.g., tightwire). Both courses involved two hours of training sessions a week for six weeks, featuring 1-2 circus disciplines each. Both courses also featured the design of TTPs and their integration into the circus disciplines.

We used the Blower, FrontBalance, Laser, and Movement TTPs (Table 2) in Super Trouper. The first three TTPs were developed over the first few sessions of Course 1, implementing the ideas most relevant to our target group from the Fall 2017 ideation workshops. They were tried in the last two sessions of Course 1. The Movement TTP resulted from an ideation workshop (W4) with our instructors [104] that had as its goal suggest new technology for our target group, and we researchers implemented it in parallel to Course 2.

Table 2.

View Table

Table 2. Overview of the TTPs in Super Trouper

All the TTPs were used in Course 2, but not all at once. We explored two in detail in each training session. The use of the TTPs in each session had been loosely envisioned during a workshop with the circus instructors (W4) and further detailed in discussions between training sessions. Instructors had plenty of room to tweak the envisioned uses on the spot during sessions, as well as to explore new uses [124]. Throughout Course 2, we also tested and slightly iterated the TTPs (e.g., to improve wearability or tweak feedback precision) to better suit our target group [77, 124]. We captured action through three cameras, one on a tripod and two chest-mounted cameras worn by researchers. We also recorded semi-structured interviews with instructors after every class and a concluding interview with children after the last training session.

As discussed in prior published work [77, 124], children and instructors reported that the TTPs were useful to help increase awareness and control for children in our target group. Most notably, they reported on how children were able to focus better on the exercise at hand and control their motor responses, and they often found the TTPs playful and exciting.

4.2.2 BL Strength: Strength Training with BodyLights.

The second dataset was collected in BL Strength, a 2019 project focused on strength training through interactive technology [127]. While this was a smaller and shorter project, the BL Strength's phases still included ideation workshops, technology development, and data gathering from workouts with trainees (Figure 2(a)). As we researchers were interested in exploring the potential of the existing Super Trouper TTPs in a very different movement practice, we organized an exploratory workshop (W1) in which we brought in all our TTPS in Table 1.

Fig. 2.

View Figure

Together with a certified NSCA [142] personal trainer (PT) and two experienced practitioners (+10 years of experience), we explored the potential of each TTP to support movement teaching and learning in strength training. The results were that participants preferred the Laser TTP, as it could visualize multiple movement qualities important in strength training: movement direction and pace, body posture and alignment, and muscle engagement [123]. Yet, participants also saw limitations, such as the Laser TTP's limited capacity to project at a variety of angles from the body. These findings prompted the development of BodyLights (Figure 2(b)), an iterated version of the Laser TTP that features a 3D printed casing that lets the user manually direct the laser pointer to any point in the environment.

We implemented and refined BodyLights through an iterative process to achieve ease of interaction and projection stability. This process happened in parallel to three ideation workshops (W2, W3, W4) in which the researchers, the PT and three experienced practitioners explored and co-designed the use of BodyLights in different strength training exercises. In the workshops, we explored the best placement for the technology on the body or equipment, projection angles as well as feedforward and feedback cues that could be articulated and communicated.

These workshops and iterative implementation process resulted in the final design of BodyLights (Figure 2(b)) and its integration in 18 strength training exercises (full, upper, and lower body exercises) in collaboration with our PT. In this project, we foregrounded values such as safety, performance precision, and correct technique. Including BodyLights aimed to support the teacher and students to achieve and communicate such precision and correctness, while being mindful and respectful of safety issues.

As the basis for this collection of exercises, the researchers and the PT planned a series of workouts and recruited 15 strength trainees at different skill levels. We collected the dataset during the evaluation of the workouts with BodyLights. Each trainee received three individual sessions of workouts with the PT, featuring 4-6 exercises with BodyLights. The training sessions were captured through one fixed camera on a tripod. We also recorded semi-structured interviews with students after every session, and a final one with the PT.

As discussed in detail in prior work [127], both PT and students found that BodyLights could meaningfully augment several movement qualities (i.e., movement trajectory and pace, posture, and muscle engagement). The PT found it useful to provide instructions and monitor the students’ performance. Differently skilled students perceived that BodyLights clarified the PT's instructions, offered them a guide on how to perform the instructed movements, and helped them identify errors on their own [127].

4.3 Methods for Data Analysis

Our prior published work on Super Trouper and BL Strength [77, 124, 127] focused on the overall effects of training with TTPs on the students’ experience. For this article, we were interested instead in surfacing and understanding the interactive qualities that made the TTPs a working tool for people during teaching and learning in the different movement practices. To do so, we re-analyzed all video data captured in Super Trouper and BL Strength through structured video analysis [45] and posterior Interaction Analysis [59]. Although we analyzed all the TTPs, for space reasons our examples in the following sections are drawn only from the Blower TTP and BodyLights. We revisit the other TTPs (presented in previous work; see [77, 124]) in the horizontal grounding of our concept in Section 7.

Since we were interested in the communication and meaning-making processes happening between teachers and student, our video and interaction analysis focused on observable interactions between these. The initial video analysis was performed by the first author with the help of a research assistant, and it focused on identifying instances in our data where the TTPs were visibly being used or referred to. For each instance, we coded the timestamp, the people involved and the TTP involved. Further, informed by knowledge from movement learning literature (e.g., [32, 39, 58, 86, 125]), we also coded each of the identified instances as belonging to one of three overarching interactional processes central to movement teaching and learning [32]: (1) instructing a particular set of movements through demonstration and explanation; (2) the students subsequently performing it, through an imitation processes; and (3) assessment and feedback, in which the teacher may further guide or correct the student's performance.

The video analysis resulted in a collection of instances of the TTPs being used and referred to in the processes of movement teaching and learning. The first author of this article further performed a detailed interaction analysis on these instances. For each instance, the author noted the interactional sequence of action between teachers and students using the technology, for example: who initiated which action and what the action itself entailed (e.g., verbally referring to the TTP, visually attending to TTP, physically manipulating the TTP). The author also noted the aim (e.g., bringing the student's attention to an error by referring to the TTP feedback) and the sequential steps of that interaction (e.g., the teacher first pointing to the TTP, the students turning their head to the TTP and stopping their performance). After coding, the first author began identifying patterns of interaction across both data sets, in the process discussing and polishing the analysis and patterns with the other authors. Through such analysis, we identified recurring and prevalent strategies of technology use shared by the two projects. These strategies captured how the TTPs were used and acted upon by teachers and students to support their teaching and learning.

These resulting use strategies, which we present in the next section, gave us an empirical grounding upon which we articulated the interactive qualities of the TTPs that made them a useful tool for teaching and learning movement (which we present in Section 6). We articulated such qualities through analytical and theoretical discussions among our research group, in a back and forth process between our empirics, our theoretical framework, and related work. This process helped us trace the connection between our results and the theories that helped explain them, as well as identify similarities and differences between our work and that of others.

5.1 Instructing Movements with the TTPs

Demonstrations with explanations are one of the central methods to instruct performances [32, 58, 86]: teachers use their own body to show the students how an exercise should be performed, typically accompanied by verbal explanations. These instructions can display a whole movement or parts of it, and are also often used to bring the student's attention to particular aspects or qualities of the movement [58].

5.1.1 Establish a Benchmark on How to Perform the Exercise with the Technology.

During demonstration, teachers use and refer to the TTPs to establish a relationship between movements and the desired technology's feedback. Teachers use a host of referential cues (e.g., verbal cues, explanations, gestures) to bring attention to the movement quality that is in focus, and relate it to the desired actuation. Teachers often intertwine internal points of reference, referring to the body or movement itself, with external, referring to the effects such movement will have on the technology. This was observed in both practices.

The demonstration and explanation links feedback responses to desired performance, but teachers also map undesired performances to specific biofeedback actuations, thus bringing attention to common errors. Through demonstrating with the technology while verbally referring to it, teachers endow the TTPs’ open-ended feedback with meaning that is specific to each exercise, and communicate that meaning to the students.

Figure 3, from strength training, illustrates this recurring TTP use. The PT is demonstrating Mountain Climbers— a dynamic variation of ‘the plank’ in which students bring their legs up to their chest—to a participant (P2). A key aspect of this exercise is to maintain trunk stability, and hence, the PT starts by establishing the correct projection position. She wears the BodyLights on her chest, with its cross-shaped projection between her palms. Pointing at the projection, she instructs: keep the light here (Figure 3(a)). The PT then maps multiple erroneous performances to the TTP's feedback through demonstrations, while P2 is looking at both her body and the TTP's feedback. She demonstrates an improper back posture, saying: remember not to curve… (Figure 3(b)); She curves her back, which moves the projection slightly away from between the hands, closer to the feet. Next, she stretches the back, continuing her explanation: …or stretch the back. This moves the projection slightly ahead, beyond the hands (Figure 3(c)). The PT also sets a baseline for movement stability by referring to the projection, pointing at the light with her hand and instructing: keep the light as stable as you can (Figure 3(d)). Finally, she also demonstrates wrong hip positioning and the effect it has on the projection: raising her hips moves the projection close to the feet. She says: remember not to keep your hips up… (Figure 3(e)); she continues by bringing the hips down and saying: …or down (Figure 3(f)); moving the projection beyond the hands.

Fig. 3.

View Figure

5.1.2 Replacing Cues about the Body with Cues about Acting with the Technology.

The TTPs also provide an opportunity for instructing exercises without needing to verbally establish the relationship between the performance and the biofeedback actuation. This strategy instead favors cues with an external focus that relate to how to act with the technology. The teachers use the TTP as a medium to influence the students’ performance without making explicit, or verbally referring to, the relationship between the body and the biofeedback.

This strategy occurs abundantly in Super Trouper. An example is shown in Figure 4, where Instructor 2 (IN2) is demonstrating how to perform a crunch with the Blower TTP on his head. With children looking at him, IN2 raises his head saying: we first raise the head. He then points at his core and says: we make it strong. Maintaining the crunch, he says: blow. As he blows, the LED lights turn on one at a time. When he has lighted up four, he stops blowing and the LED lights turn off. In interview data, IN2 reflected that that the intended effect of making children blow is to emphasize the exhale, which contracts the core and helps them stabilize and maintain the posture. Yet, in the video data, the cue that he gives relates to how to interact with the technology (blow), and the relationship between core engagement and blowing is not communicated to the children.

Fig. 4.

View Figure

In BL Strength, this occurred primarily with beginner students and nuanced movement qualities, e.g.,: in Figure 3(d), the PT gives P2 an actionable cue to keep the projection stable, which will subsequently affect the body's movement pace. The relationship between the stable projection and pace is never verbalized.

5.2 Performing an Exercise with the TTPs

Students must develop the ability to perceive, identify, and recognize relevant movement qualities and desired performances to be able to act on the teachers’ instructions [7, 32, 58]. Students typically observe the teacher's demonstrations and try to imitate their performance afterwards [32].

5.2.1 Imitation: Transferring the Established Relationship Between Technology and the Body to the Student's Own Body.

In both practices, students visually attend to the teachers’ body and the TTP's biofeedback during demonstrations. The TTPs’ actuation gives students a benchmark of the feedback response they are expected to achieve if they execute the movements as demonstrated. When they perform it themselves, the biofeedback becomes a referent, and students can compare the feedback they obtain during their performance to that of the teacher's.

Figure 5 from the circus training data shows how children start performing crunches with the Blower TTP: after looking at IN2’s demonstration (Figure 4), they directly proceed to perform the same. First, P5 with the TTP on his head, does a crunch and blows into it (Figure 5(a)). His LEDs light up, as the beeping continues, and IN2 looks at him. When P3 stops blowing into the TTP and goes back to lying on the mat, the LEDs turn off and beeping stops. The other children follow P3, performing the exercise and blowing into the TTP (Figure 5(b)).

Fig. 5.

View Figure

Prior to imitation, sometimes teachers need to assist and help adjust the TTPs so students can obtain feedback as benchmarked in the demonstrations. This is particularly recurrent with BodyLights. In Figure 6, P2 is about to perform the Mountain Climbers himself. PT manually adjusts P2’s BodyLight so it projects in between his hands (Figure 6(a)), points at the projection remarking the light stays here (Figure 6(b)). P2 then proceeds to do the exercise himself (Figure 6(c)).

Fig. 6.

View Figure

5.3 Monitoring, Assessing and Correcting with the TTPs

The teachers also guide and correct the students during their performance. Monitoring and assessing performances is interconnected with providing guidance on how to correct them [32] and requires constant reflection in action [103]. Teachers assess the students’ performances, and the assessment's outcome shapes next actions (e.g., further instructions, corrective cues).

Once teachers have established the relation between a particular biofeedback actuation and a desired performance, they can choose to draw attention to it, for instance by verbally offering actionable cues referring to the biofeedback. The teacher will sometimes tighten, relax, or repeat the benchmark during guidance and correction. Figure 7 provides an example of IN2 in circus, guiding a child (P4) in maintaining balance in the rolla-bolla, which requires students to position themselves on the balancing board and maintain a constant movement to not fall. IN2, with his hand close to P4 for safety, provides an instruction on acting upon the technology (saying: Can you blow? Blow now) to make P4 to engage the core, which is key for balancing (Figure 7(a)). P4, looking at the Blower TTP feedback, starts blowing and balancing. When P4 stops exhaling into it, and the Blower TTP's LED lights start to turn off and the sound decreases, IN2 reminds him of the instruction, saying: blow (Figure 7(b)). When P4 loses balance, IN2 looking at the BlowerTTP's feedback, repeats the instruction as a correction, saying: Blow again, keep blowing (Figure 7(c)).

Fig. 7.

View Figure

Teachers sometimes also adapt or tweak the established connection between an exercise or particular movements and the TTP's biofeedback, to better suit the situation at hand. This may entail using the TTPs to bring attention to specific movement qualities or points of the exercise, concretize the appropriate biofeedback, or provide more actionable cues. Figure 8 from strength training presents an example of each. P2 is performing the Mountain Climbers (from Figure 3) with a rushed movement and poor trunk stability. The projection is not stable and moves significantly away from the benchmarked position. The PT, looking at it, offers a corrective verbal command (slow), and brings P2’s attention to a specific quality of the embodied conduct (pace), through pointing at the projection on the mat, and providing an actionable cue regarding the feedback control (Figure 8(a)). P2 continues to do the exercise but his posture does not improve enough. The PT stabilizes P2’s trunk, positioning her hands on P2’s back and chest (Figure 8(b)).

Fig. 8.

View Figure

Mobilization (the teacher physically correcting the student's body or movement) is not a sustainable method of correction and is usually resorted to when other corrective strategies fail [125]. Before the second round of repetitions, while P2 rests, the PT fetches a sticky paper note and briefly positions it exactly on the middle of the projection's cross. She tells P2 we are going to allow the cross to stay just there, try not to go across the post-it (Figure 8(c)). This establishes a concrete visual boundary on the acceptable range of movement for the projection. With the post-it in place, the P2’s movement is slower and more controlled, and the projection moves significantly less so that it barely leaves the note (Figure 8(d). During the whole exercise, PT and P2 visually orient to the projection mostly. This is an example of how a benchmark was changed (in this case, concretized).

These two examples also show that the teachers’ visual attention and observation are critical elements of their teaching work, shaping how the training session will unfold [32, 40]. Visually attending to the students’ performance is central to obtaining information on their bodily competencies, allowing teachers to identify and determining what aspects of their performance require further instruction [21, 32]. The TTPs’ biofeedback becomes one of the elements teachers use to monitor student performance, including identifying errors, checking whether they are acting on cues, and determining success.

6.1 Shared Frame of Reference

The first characteristic of the strong concept concerns the capacity of intercorporeal biofeedback to create a shared frame of reference among teachers and students. This frame of reference is made possible through the biofeedback being perceptually accessible for both parties. Intercorporeal biofeedback augments otherwise elusive aspects of the body [61], making them accessible and public. As our empirics show, the TTPs’ representations were publicly available to anyone close enough to perceive them.

Mostly, we have used visual or audial cues to this purpose, as they offer modalities that different people can access simultaneously and through the same sense (vision or hearing). For example, BodyLights employed visual augmentations to the environment; the Blower TTP included synchronized sets of LED lights—one facing the wearer and one outwards —as well as sonification that both can hear. As seen in the empirics, this shared actuation allowed teachers and students to orient simultaneously to the feedback. For example, in Super Trouper in Figure 7, the circus instructor and the children trying to balance on the rolla-bolla both look at the Blower TTP's visual feedback and attend to its sonification. In BL Strength, the PT and the student both look at BodyLight's projection while the student is performing the mountain climbers (Figure 6).

The shared frame of reference gives people an additional perspective on their own movement. It complements their felt first-person experience [105] with third-person augmentation that shows the impact of their movement as it unfolds in space and time [105]. Through the technological augmentation, people can appreciate aspects of their own body and performance through a new, different sensory modality, which can in turn lead to a new understanding of such aspects. In our empirics, BL Strength students said that BodyLights gave them a new perspective: “usually you don't see yourself from above, so you can't really realize fully your movement […] with the projection you have a good understanding on how your body moves” (P7). The shared frame of reference thus can provide people with an observational, third-person perspective on their own performance, complementing their felt sensations [34, 116].

Through the shared frame of reference, people's perception of themselves and the others is enhanced. This reciprocal perceptibility is crucial in constituting and sustaining intercorporeal engagements [85]. Teachers in both practices and students in BL Strength reflected that the TTPs enhanced their mutual understanding. The technology spotlighted selected movement qualities in a perceptually shared way that gave teachers and students a shared medium upon which they could base their communication and interaction. People could attune and appreciate both their own and the others’ movements, showing how the shared frame also supports second-person perspectives [34, 86]. For example, students in BL Strength reflected that the TTPs clarified the teachers’ explanations, as captured by P4: “we [PT and I] were using the light as the language tool to understand what I should be doing” (P4).

By enhancing people's perceptibility of themselves and the others, the shared frame of reference partially bridges the fact that teachers and students are not equally capable of perceiving and understanding movement qualities [64, 120]. For example, the Mountain Climber example from BL Strength (Figures 3 and 6) illustrates how through BodyLights the PT could articulate very nuanced body positioning issues to a student whose movement literacy was not fully developed. The PT could draw attention to many desired and undesired positionings (Figure 3) and help the student realize his performance errors through referring to the BodyLights’ feedback (Figure 6).

Further, teachers in both practices and students in BL Strength mentioned that the shared reference frame enabled what we can call a lingering perceptual imprint from the instructions. For example, as P2 performed Mountain Climbers (Figure 3), the projection gave him constant feedback on his posture, which also acted as a benchmark. Interactional biofeedback needs to offer feedback that is synchronized with the movement, so that the frame of reference dynamically evolves in space and time, as with other biofeedback technologies [115]. This ensures immediacy and synchronization [51] between what is being represented and the actuation, at the same time as feedback being available all throughout the experience, dynamically reflecting changes in performance. Participants in the empirics reflected on this quality of the shared frame of reference. As one student put it: “when you receive an instruction and you don't have the light, it's difficult to keep the instruction present […] the light keeps it present all the time, like a constant reminder [on the correct performance], the light is all the time reminding you” (P15). This points to the shared frame of reference bridging the experience of the self to that of others [74], as it not only fosters empathic observation (ibid.), but enables people to bridge the perceptual imprint of the teacher when demonstrating to their own when performing.

6.2 Fluid Meaning Allocation

The second characteristic of the strong concept concerns fluid meaning allocation, by which we mean the way teachers and students make the shared frame of reference meaningful and actionable in context through ascribing meaning to particular courses of action, and adapting these meanings as needed.

In this process, the role of the teacher is central. Teachers endow the shared frame of reference with contextual meaning, helping students making sense of it in a way that aligns with both the desired performance and students’ individual capabilities. Teachers also make the shared frame of reference actionable for students. This meaning-allocation process has been observed in previous work on movement teaching and learning [7, 21].

The empirical examples in this paper show how teachers allocated meaning and courses of action to the TTPs’ feedback, through demonstrating and explaining the connection between the performance and the biofeedback actuation. For example, the PT in BL Strength demonstrating the Mountain Climbers in Figure 3(d), saying remember to not curve or stretch the back while demonstrating an incorrect positioning of the back and the BodyLights’ projection. We also sometimes saw teachers do this without referring to the body movement, instead providing cues on what to do with the technology directly. Examples include the instructor in Super Trouper explaining how to do crunches (Figure 5) by saying blow, or the instructor for Mountain Climbers in BL Strength (Figure 3(d)) saying keep the light as stable as you can.

Through such explanations and demonstrations, teachers associate the shared frame of reference with a range of desirable and undesirable movements. These meanings are often prescriptive, related to the movement norms of a practice [22]. Interview data revealed that teachers from both practices perceived that using TTPs economized instruction, since establishing the relationship between a correct performance and TTP feedback made their instructions linger during the student's performance. BL Strength's PT captured this quality well: “[with BodyLights] I can see that students acquire [aspects of] the basic technique fast […]. I don't need to repeat the information, like the movement trajectory or posture, every time for each repetition, set, or even session […] I just [trust] the light to take care of it. I give students [more nuanced cues], but I don't repeat the basics”.

Fluid meaning allocation also concerns how the allocated meanings are adapted to best suit particular contexts. This is crucial as movement learning goals and needs change depending on particular practice's goals and values [5, 99, 102]. For example, in Super Trouper's rolla-bolla example (Figure 7), the instruction to blow did not really specify a particular pace or length of exhalation, as the main aim was to make the child to engage the core muscles and, through that, manage balancing. In comparison, BodyLights's Mountain Climbers projection (Figure 3) was linked to both posture and stability, revealing concrete correct and incorrect postures with great precision. Performance correctness was more important in BL Strength.

The meanings ascribed to the biofeedback were also adapted to the individual students and their bodies and capabilities, allowing instruction to address individual needs and goals [14, 21]. Teachers would concretize or change the meanings and actions they ascribed to the technology, by e.g., emphasizing previously instructed cues, as in the Super Trouper example with the rolla-bolla (Figure 7) when the instructor repeats the instruction blow. Teachers also linked the biofeedback's actuation to more specific aspects of the performance, as in the Mountain Climber example of BL Strength (Figure 8) where the PT brought attention to pace. Finally, teachers were able to adapt the initially allocated meanings: in the same Mountain Climber example (Figure 8(c)), the PT used a physical prop, a sticky note, to give P2 an even more actionable cue to improve his posture and stability.

Teachers can do these fluid meaning allocations due to their expertise in assessing the students’ needs in real time [21], and their perpetual reflection in and on action [39, 103]. Teachers and students negotiate meanings and courses of action to the shared frame of reference, and reach a mutual understanding of the self, the other, and the practice's norms—which is what intercorporeal engagements aim at doing [40, 85, 86] in movement teaching and learning.

6.3 Guided Attention and Action

The third characteristic of intercorporeal biofeedback concerns the multiple and changing perceptual relations [94] between people and technology, and how these were used by teachers to guide students’ attention and action. As with other biofeedback technologies, the attention of a person interacting with biofeedback dynamically fluctuates [93, 94]. A quote from one BL Strength student captures this quality: “[my attention to BodyLights’ projection] is like appearing and disappearing, it goes [in] and out” (P15).

Through being fluent and immediate, intercorporeal biofeedback has the capacity to be incorporated into the bodily experience [48], becoming an extension through which people perceive and act. Our empirical studies illustrate how both teachers and students went about their tasks (e.g., instructing, performing, assessing) while using the TTPs (Figures 3 to 8). Some participants were not only seen to be able to act and perceive through the TTPs, but they consciously experienced such incorporation. This is best captured by a BL Strength participant's quote of training with BodyLights: “I felt I used [the projection] but I was not only focusing on it […], at the same time I was focusing more on my body” (P14).

The shared frame of reference can be geared towards guiding people's attention, and/or to their inner, felt sensations, helping them hone a first-person perspective of their body movement. Intercorporeal biofeedback can be experienced, to some extent, as in an intimate correspondence [51] with the person whose movement qualities are being augmented, being merged into their perceptual-bodily selves [55, 94, 132]. BL Strength participants experienced this correspondence particularly often, as one reflected: “it's like a mirror […] it's a way in which I can see my body: it's reflecting my movements, how tired I am, how straight I am, how fast I move.” We see this as an important quality, in that it can allow people to act with technology without negatively altering their movements or practice.

Yet, intercorporeal biofeedback can also disrupt our felt sensations through acts of defamiliarization [48]. Through decoupling acts of first- and third-person perspectives on the body [49], a person checks what they sense proprioceptively against what the augmentations reveal of their performance. In our empirics, participants’ felt experience and the information the TTPs provided did not always align. A participant in BL Strength reflected: “without [BodyLights], I thought I had a good alignment in my arms and back, but with it, I saw a small twist, and [realized] I was putting too much effort on one arm in comparison to the other” (P12). The capability of creating defamiliarization is also an important quality, as this can bring attention to performance aspects and errors that are otherwise elusive.

Intercorporeal biofeedback can be also experienced as a quasi-other [132], as an external object in the world [116] that draws our explicit attention to it and makes it the focal object of awareness and action [55, 132]. In the empirics, an explicit attentional focus on the shared frame of reference was sometimes desired and fostered. The teachers replaced verbal cues with actionable cues that brought the attention and action solely to the TTPs, as in the Super Trouper examples with the Blower TTP (Figures 4 and 7) and the verbal cue to blow; or in BL Strength's Mountain Climber example (Figure 6) and the PT's instruction to control. An explicit focus on technology feedback was particularly useful for students who found it more difficult to act upon bodily cues, as with the children in the Super Trouper project. It was also more commonly used with beginner and intermediate students in BL Strength.

Explicit focus can also be used for distraction. While distraction risks disconnecting us from our body and our world [55, 132], it is sometimes sought in movement learning, for example, to distract people from pain or extend performance time. In our empirics, distraction helped children in Super Trouper to concentrate on the task at hand and control their movement and ease unpleasant sensations, and made children engage with the exercise longer (e.g., crunches with Blower TTP Figure 6, where having to blow was made a goal in itself).

Intercorporeal biofeedback can also be used by teachers to guide the students’ attention to specific body areas or movement qualities. In BL Strength, when instructing Mountain Climbers (Figure 3), the PT used verbal cues, gestures, demonstrations, and the biofeedback's behavior to bring to the foci of attention particular postural aspects (e.g., arched or curved back) and movement qualities (e.g., stability), and exemplifying desired and undesired postures. This aligns with the notion of change from motion training [49], of subdividing bodily experiences into specific areas as a way to deepen people's appreciations of their sensorimotor processes and of the practice's norms.

The use of intercorporeal biofeedback to provide very actionable instructions, as in the rolla-bolla example, helps to address challenges in articulating instructions that stem from asymmetries in movement literacy and perceptual capabilities [58, 101]. In our empirics, teachers from both practices perceived that demonstrating exercises and particular movements with the TTPs allowed them to simplify and clarify their communication. They mentioned that their explanations of the relation between a desired/undesired performance and its impact on the shared frame of reference was able to cover what otherwise would have needed several verbal cues. As the strength training PT reflected: “I connect the projection to what is a correct technique for each student, and explain it to them, and [without BodyLights] I would need to give more cues, like “try to adjust your pelvis”.

Hence, it can be argued that intercorporeal biofeedback helps to create and sustain co-operative forms of intercorporeality, namely the turn-taking processes through which teachers and students build relevant and situated meanings and action [40]. Yet, it is important to notice that, as with any technology, intercorporeal biofeedback artefacts are not neutral [132]. That is, biofeedback will encourage certain sensorimotor appreciations and responses, but might risk obfuscating others [48]. In our empirics, the teachers were key to addressing performance aspects that were not captured by the biofeedback. For example, in Super Trouper, the instructor explained how to perform a crunch with the Blower TTP (Figure 4), which included raising their heads. This was not something that the Blower TTP could augment, so the instructor brought it to their attention by using other interactional resources: he bodily demonstrated it, touched his own head with his hand, and said that we first raise the head. The fluid meaning allocation that characterizes intercorporeal biofeedback allows teachers and students to decide and negotiate in practice when to attune to and act with the biofeedback. Yet, it also allows them to decide when its use should recede to the background [77].

6.4 Interwoven Interactional Resource

Finally, intercorporeal biofeedback is an interwoven interactional resource. It is integrated with the pool of other interactional resources that teachers and students already use, and through which intercorporeality is constituted [86]. These include verbal explanations and commands, body demonstrations, gestures, mobilizations, other material elements, and so forth [58, 86].

This quality is illustrated in all the examples of our empirics (Figures 3 to 7): teachers and students use verbal cues, act with the technology, and point to it. They build meaning and action by the juxtaposition and simultaneous use of all these multimodal resources. For example, in the Mountain Climber example from BL Strength (Figure 3), the PT instructs the exercise by employing bodily demonstrations, gestures, and verbal explanations, all while using BodyLights. Hence, intercorporeal biofeedback extends the ecology of interactional resources already present in the practice with a distinct material contribution: a perceptually shared augmentation of relevant movement qualities that people render meaningful and that help guide their attention and movement.

As our examples also show, in the context of an unfolding movement learning experience, the technology can be used, ignored when not relevant [40], and then used again. For example, the instructor in Super Trouper explaining how to perform a crunch with the Blower TTP (Figure 4) first cued to raise the head by providing only bodily and verbal cues; and later cued core engagement by using the TTP.

Like other multimodal resources, intercorporeal biofeedback supports intercorporeal engagements between teachers and students. Yet, as with other resources, it alone does not sustain intercorporeality: it requires other resources to be meaningful and actionable. In our empirics, successful integration of technology with other multimodal resources required iterative exploration during their design process [77, 124, 127]. This exploration impacted not just the design of the TTPs as such but would for example serve to identify best body locations and uses of the TTPs for particular exercises, just as to identify (and in some cases modification) the exercises that benefitted most from the introduction of technology. In addition, technology needed tweaking when in use. This included hardware adjustments to mitigate bodily differences (e.g., the BL Strength PT in Figure 7, mechanically positioning P2’s BodyLights); and in the case of some of the circus TTPs, through software adjustments such as changes in sensing sensitivity [77].

7.1 Technology Characteristics of Intercorporeal Biofeedback

Several technological characteristics are shared among our work and others featuring intercorporeal biofeedback artefacts (i.e., Enlightened Yoga [121], ExoPranayama [89] Motion Echo [98], and Go-with-the-Flow [109]). We do not claim that these are the only characteristics that render an artefact a useful intercorporeal biofeedback technology. However, these characteristics do depict a particular type of technology that can support the sought interactive qualities: a shared frame of reference, fluid meaning allocation, guided attention, and action, and using technology as an interwoven interactional resource.

First, intercorporeal biofeedback artefacts benefit from being designed to augment such movement qualities or physiological processes that are relevant to the movement practice. In some cases, these are aspects that are prone to errors as in BL Strength; or physiological processes that are central to a practice, such as breathing in ExoPranayama. When given meanings, these augmentations guide attention and help establish new bodily-perceptual relationships with and through the technology. In design, it is hence important to explore the most meaningful couplings of sensing and actuation. In our work, the practice culture and the participant's values were important factors to decide what couplings to design and pursue.

Second, intercorporeal biofeedback artefacts present perceptually shared feedback to enable a shared frame of reference. All the examples revisited provide either visual or auditory feedback, which people can perceive simultaneously. Vibration or force-feedback are more difficult to work with, as these actuation forms are not immediately shareable.

Third, intercorporeal biofeedback artefacts benefit from immediacy: our example technologies synchronize movement and actuation and are perceptually present as long as a person interacts with them. This enables teachers and students to orient to it and use it as an interactional resource when they deem it relevant.

Finally, an open-ended [12, 47] actuation form is an important characteristic of intercorporeal biofeedback technologies. When technology does not prescribe the normative meaning of the feedback, it becomes possible to assign meaning to it contextually, as part of the teaching and learning process. Open-ended feedback is open to appropriation and changes.

7.2 Contribution

7.3 Limitations and Future Work

Limitations of this work revolve around the nature of our investigations. We have not performed a controlled comparative study featuring the same TTPs in different movement learning practices. The practices we report on target very different populations and training goals, and feature very different teaching and learning styles and dynamics, which makes it difficult to do a comparison. That can be seen already from the number of instances of instruction, performing, and correction that we identified in each practice, and that form our empirical base (see Table 3). In BL Strength, each student trained three sessions with BodyLights. In Super Trouper, a particular TTP was used in two or at most three sessions. There were also teaching differences between the projects, for example: the circus teachers offered instructions to the whole group, and then followed up with individual guidance and correction. Children in Super Trouper trained different exercises with different TTPs, but not every child tried each exercise or got individual guidance. In contrast, the BL Strength workout sessions were individual and the PT always offered feedback and corrections. Within BL Strength, differences in the counts stem from the students’ different skill levels, e.g., for beginners, the PT repeated demonstrations with BodyLights in every session, for more advanced students, only during the first one. Finally, the cameras in Super Trouper could not capture all the action, as the circus training hall was large. In contrast, the BL Strength camera captured everything, as the PT and the students were fairly stationary in a more confined space. All these differences can help explain unbalances in the counts of Table 3.

Further, not all the TTPs were tried in both practices. These differences between these projects anchor the generative capacity of the strong concept across different practices, participants, and technologies. However, future work is needed to determine whether each individual TTP can be included in other practices. Prior work brings indications towards this direction [123].

The work in this article has not elucidated issues of how intercorporeal biofeedback artefacts would affect learning over longer timeframes. This is a challenge shared with the area of technology-supported movement learning [84, 128]: most design work centers on the in-the-moment contexts of learning, and the role and impact of the technology over longer periods remains unanswered. Future work is needed to address how the technologies that the strong concept generates can support the development of sensorimotor competencies in such timeframes.

The empirical base from which we have articulated the strong concept stemmed from projects focused on collocated practices; yet, some of the reviewed work (i.e., Go-with-the-Flow [109]) have pointed towards the potential of using intercorporeal biofeedback artefacts to train movement also in other types of settings. In [109], the authors explored a semi-distributed setting, in which patients practiced on their own with recurring meetings with a physiotherapist. The wearable was experienced in the at-home studies as having a co-supervisory role, akin as to how our participants experienced our TTPs in the collocated setting. This example shows that while the technology was rendered meaningful and actionable during the collocated, intercorporeal engagements between the patients and the physiotherapists in clinic, it had potential to support individual practice at home once the biofeedback has been allocated with meanings and courses of action.

It would be interesting to explore the potential of intercorporeal biofeedback for fully distributed settings in which teachers and students are never physically together. Aggarwal et al. [1] elicited a series of insights and challenges on video-mediated face-to-face physiotherapy sessions, which related to how people build meaning and action on instructional cues, performance, and feedback in a setting where both teachers and students depend strongly on visuals. Intercorporeal biofeedback artefacts could potentially help teachers and students in video-mediated intercorporeal engagements, by virtue of the same characteristics they have in collocated settings, but future work is needed to explore such potential.

Finally, a limitation of our work is that we have approached learning in terms of the development of particular sensorimotor capabilities and competencies. However, there are other dimensions that are important and play a part in sensorimotor development and that our work has not touched upon, e.g., affect [97], psychological barriers to movement [71], and politics [90, 120]. It would be interesting to explore how intercorporeal biofeedback artefacts can address other dimensions of movement learning.