Extended Reality (XR)

The umbrella term used for Virtual Reality (VR), Augmented Reality (AR), and Mixed Reality (MR), as well as future realities immersive technologies might create. XR continuum covers the full spectrum of real and virtual environments. The design and implementation of such environments faces many challenges that Sharespace studies and takes up, for instance allowing for Hybrid Co-location of users from different physical spaces, optimizing the accuracy of their Movement Representation and thereby guaranteeing a certain Authenticity of XR Experience, or providing users with means of Action-Perception Coupling, Sensimotor Propagation and Hybrid Synchronization, which fosters a sense of Cooperation, Social Presence and Sociality more generally.

Illustrative reference: https://digital-strategy.ec.europa.eu/en/policies/extended-reality

Avatar (L1)

An avatar is a three-dimensional digital character that represents a physical person within social XR environments. While not always an exact replica, an avatar typically resembles the person’s appearance (morphological similarity) and movements (kinematic similarity through accurate movement representations). The Avatar serves as a digital proxy of the individual, and, through the Social Information it provides to others in shared hybrid spaces, it also constitutes a vehicle of Sociality.

Illustrative reference: https://link.springer.com/article/10.1007/s12144-022-03232-6

Semi-Autonomous Virtual Character (L2)

A semi-autonomous virtual character occupies a hybrid position between an avatar (L1) and an autonomous virtual character (L3). Its behaviour is partially controlled by a real person, and partially generated algorithmically, for instance by the cognitive architectures designed in Sharespace.

Illustrative reference: https://research.gold.ac.uk/id/eprint/2292/

Autonomous Virtual Character (L3)

An autonomous virtual character is an Embodied autonomous agent, that is, an agent that interacts with others in a social XR environment through a virtual body, and that yet retains its own way of operating and behaving in the process. In other words, it perceives, acts upon, and adapts to the environment over time following a pre-defined set of rules implemented with reinforcement learning processes.

Illustrative reference: https://doi.org/10.1016/j.cosrev.2016.09.001

Autonomization Continuum

Digital agents can be categorized according to incremental levels of autonomization (autonomy in artificial agents) inspired from autonomous vehicles classification: avatar (L1) are fully driven by humans, typically implemented through the optimization of movement representation in real time; Semi-autonomous virtual characters (L2) are partially human-controlled, and partially governed by a modulation of human users information, for instance through movement amplification/attenuation of sensorimotor primitives; Autonomous virtual character (L3) are fully independent, powered by AI cognitive architectures, and capable of real-time adjustments of their behaviour.

Illustrative reference: https://www.researchgate.net/publication/339371847_A_Standard_Driven_Software_Architecture_for_Fully_Autonomous_Vehicles

Remote Social Connectedness

It refers to the sense of belonging and closeness people experience when interacting with others in XR, . It is reinforced by strong Action-Perception Coupling, making Hybrid Synchronisation a key process. In return, it fosters Cooperation, motivation, emotional support, and adherence by strengthening bonds between individuals, or between individuals and digital agents: Avatars (L1), Semi-Autonomous Virtual Character (L2) and Autonomous Virtual Character (L3).

Illustrative reference: https://www.tandfonline.com/doi/full/10.1080/17483107.2024.2310262

Hybrid Synchronisation

A phenomenon whereby the movements of individuals and/or virtual agents (Avatars (L1), Semi-Autonomous Virtual Character (L2) and Autonomous Virtual Character (L3)) are temporally aligned, thanks to their Action-Perception Coupling and mechanisms of Sensorimotor Propagation. Hybrid Synchronisation can be defined by phase-locking (i.e., the matching of time signals), allowing for minor differences or consistent time lags. Synchronisation being considered as a marker of social connectedness in human-human interactions, Sharespace explores the potential of Hybrid Synchronisation to foster Remote Social Connectedness.

Illustrative reference: https://ieeexplore.ieee.org/document/11220424

Sensorimotor Primitives

The foundational components of intentional bodily actions. These include coordinated patterns of kinematic, dynamic, and sensory variables. Sensorimotor primitives encode movement-based Social Information such as intentionality and emotion. A technical challenge that Sharespace promises to take up is to compute and render accurate Movement Representations of these primitives.

Illustrative reference: https://www.sciencedirect.com/science/article/pii/S1571064524001477

Social Information

Information embedded in Sensorimotor Primitives —including intentions, emotions, and subjective feelings— that enable individuals to infer others’ states and roles, fostering behavioral alignment, shared understanding, and collective action, without explicit verbal communication, but rather through Sensorimotor Propagation.

Illustrative reference: https://pubmed.ncbi.nlm.nih.gov/30072239/

Sensorimotor Propagation

The transmission and entrainment of Social Information, encoded in Sensorimotor Primitives, across Avatars (L1), Semi-Autonomous Virtual Character (L2) and/or Autonomous Virtual Character (L3).. The extent of propagation is rooted into the amount of Social Information encoded and transmitted. In XR, sensorimotor propagation through action-perception coupling is also fundamental to Hybrid Synchronisation.

Amplification/Attenuation of Sensorimotor Primitives

Amplification involves enhancing sensorimotor cues to facilitate the readout of Social Information by human interactants, for instance through a dedicated Cognitive Architecture. Attenuation reduces or filters such cues, decreasing the transmission of Social Information in XR. Amplification and Attenuation are key to Sensimotor Propagation in XR.

Embodiment/Embodied

A pre-reflective subjective experience combining self-location (i.e., feeling located where the avatar is), agency (feeling in control of the avatar’s actions) and ownership (i.e., feeling that the avatar’s body is one’s own). Embodiment can nurture a sense of presence in XR environments, the subjective sense that the experiences in XR are authentic.

Illustrative reference: https://www.researchgate.net/publication/243056510_The_Sense_of_Embodiment_in_Virtual_Reality

Social Presence

The subjective sense of perceiving virtual others [Avatars (L1), Semi-Autonomous Virtual Character (L2) and/or Autonomous Virtual Character (L3)] as intentional and being connected to one’s digital self across dimensions of body, emotion, and identity. It includes both the psychological and physical sense of “being there” in a shared hybrid space. Social presence is paramount in triggering the motivation to read Social Information out and to feel Remote Social Connectedness.

Illustrative reference: https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2018.00114/full

Cybersickness

A form of motion sickness experienced in XR, caused by lack of control, postural instability, sensory conflicts, and prolonged immersion. Typical symptoms include nausea, dizziness, disorientation, eye strain, and fatigue, which can disrupt presence and social interaction. In social XR, reducing cybersickness involves optimizing frame rate and latency, providing stable reference cues, supporting active control of movement, optimizing Movement Representations and Perception-Action Coupling, and enabling Embodied Hybrid Synchronization.

Illustrative reference: https://www.tandfonline.com/doi/full/10.1080/10447318.2020.1828535

Sociality (L1)

The capacity or tendency of humans to live in social groups and engage in interactions with others, including forming relationships, communicating, and cooperating. Sociality represents the social, cultural, biological and personal bases to human coexistence as a social animal relevant to reproduction and survival. In Social XR, a digital replica — Avatar (L1) — acts as a proxy for the user and subsequently is an extended vehicle for Remote Social Connectedness.

Illustrative reference: https://www.annualreviews.org/content/journals/10.1146/annurev.anthro.32.061002.093158

Dignity (L1)

The inherent worth of individuals, protected by legal and ethical standards. Dignity is a foundation of human rights and guides moral decisions to guarantee respect and autonomy. In Social XR, digital replicas — Avatar (L1)— are extensions of the personhood(s) of the users and therefore are constrained and influenced by social norms and practices.

Illustrative reference: https://link.springer.com/chapter/10.1007/978-981-96-1154-6_9

Self-Identity (L2)

The ontological state of self-awareness which differentiates one individual from another and marks out self-ownership. In Social XR, preserving and/or protecting self-identity is critical when users engage with Semi-Autonomous Virtual Characters (L2) or when their identity is conveyed by Avatar (L1) by Embodiment/Embodied processes.

Illustrative reference: https://www.researchgate.net/publication/374562269_eXtended_Reality_of_socio-motor_interactions_Current_Trends_and_Ethical_Considerations_for_Mixed_Reality_Environments_Design

Transparency (L2)

Transparency refers to the understandability accountability of AI system and its development process. In transparency concern involves clear awareness of the Cognitive Architectures used for both human and autonomous agents —Semi-Autonomous Virtual Characters (L2) and Autonomous Virtual Character (L3)— including how interactions are moderated, and how decisions are made.

Responsability (L2/L3)

The obligation to ensure that AI technologies are developed and used in ways that respect human rights, minimize harm, and promote fairness, transparency, and accountability. Offsetting future harms by conscious and ethical deliberation and implementing cautionary approaches such as “do not do significant harm”. In Social XR in which Semi-Autonomous Virtual Character (L2) and Autonomous Virtual Character (L3) are implemented, this can mean considering potential or undesirable outcomes arising from its expansion to broader applications.

Cooperation (L2/L3)

Cooperation is a fundamental aspect of human Sociality, understood as reciprocal acts shaped over time for mutual benefit. In Social XR, it includes the key alignment principle of human and autonomous characters’ interactions[Semi-Autonomous Virtual Characters (L2) and Autonomous Virtual Character (L3)]. Cooperation can also refer to the collaborative effort among various stakeholders, such as researchers, developers, governments, companies, and the public, to ensure that AI is developed and used in safe, fair, and beneficial ways.

Illustrative reference: https://www.researchgate.net/publication/394175272_Breaking_the_mould_of_Social_Mixed_Reality_--_State-of-the-Art_and_Glossary

Attachment (L3)

Emotional bonds that develops between people, especially in close relationships. It plays a critical role in how we form connections, respond to others, and regulate emotions. In Social XMR, decreasing face-to-face interactions in favour of computer-mediated interactions means social entanglement with Autonomous Virtual Character (L3). This might hold implications for attachment characteristics such as human social skills, emotional regulation, and bonding. Similarly, interactions with Avatar (L1) have implications in Remote Social Connectedness.

Illustrative reference: https://awspntest.apa.org/record/2025-47797-001

Trust (L3)

Trust refers to the belief that an AI system will act reliably, ethically, and in the user's best interest. There is an inherent opaqueness of AI systems, leading to potential issues of deception and humans being unable to predict what Autonomous Virtual Character (L3) will do, and how reliable and secure the given system is. To promote synthetic cooperation between humans and Autonomous Virtual Characters (L3), it is vital to know the identity of agents, their capabilities, and their interaction roles.

Ego-Centric Visual-Inertial Tracking

A method for estimating the position and orientation of body segments using a combination of visual and inertial sensors. In egocentric setups, the camera is mounted on a head-mounted display (HMD), while minimal additional sensors are placed on the self-occluded segments (e.g., lower-limbs body). This method is optimal to drive the rendering of Avatar (L1) in XR and allow Sensorimotor Propagation.

3D Digitization

A process by which a real-world 3D space with the corresponding content (boundaries, objects, and persons) is recorded in XR by means of direct capture (direct point-cloud registration) or computationally reconstructed from a set of images (for example, using methods of photogrammetry).

Scene Neural Rendering

An class of image and video generation techniques that use deep learning to synthesize images from real-world data. It leverages generative machine learning, Neural Radiance Fields (NeRF), or Gaussian Splatting techniques and enables high-fidelity reconstructions (photos and videos) of complex scenes in XR.

Varifocal Display

A single focal plane display, in which the position (distance) of the focal plane can be changed. With respect to stereoscopic 3D displays, varifocal display architecture is coupled with a gaze or eye position tracking system to estimate the eye vergence angle and to match the distance of a focal plane to a viewer's vergence distance in a given instant of time - thus overcoming vergence - accommodation conflict and respective negative effects. Varifocal displays do not convey natural focus cues and for realism must rely on computationally synthesized defocus for realism. When implemented in optically see-through stereoscopic displays - varifocal architecture can address focal rivalry.

Movement Representation

Unsupervised or semi-supervised machine learning techniques that reduce the dimensionality of high-dimensional kinematic data. These representations facilitate understanding, reconstruction, and generation of movement patterns for [Avatars (L1), Semi-Autonomous Virtual Character (L2) and/or Autonomous Virtual Character (L3)] in social XR environments.

Digital Posture Modeling

The process of creating the appearance and whole morphology of a digital representation of a human [Avatars (L1), Semi-Autonomous Virtual Character (L2) and/or Autonomous Virtual Character (L3)] for use in a XR. Posture refers to the position and alignment of the body parts, such as the spine, shoulders, hips, and limbs, in relation to each other and to gravity. Part of the process includes the creation of a skeletal rig to animate the character.

Illustrative reference: https://www.researchgate.net/publication/393743022_Multi_-_Modal_Signal_Processing_for_Avatar_Motion_Adaptation

Cognitive Architecture

A set of algorithms, datasets, and feedback control strategies that a computer uses to determine the behaviour of a virtual human, say X, in the shared space. It can take as input recorded motion data from one or more humans, and past motion data generated by X and by other virtual humans, such as Avatars (L1), Semi-Autonomous Virtual Character (L2) and/or Autonomous Virtual Character (L3). It outputs (parts of) the motion data to be followed by X. When deciding this motion data, the cognitive architecture follows an objective that depends on the level of autonomy within the Autonomization Continuum.

Illustrative reference: https://www.sciencedirect.com/science/article/pii/S2405896325001533

Pain Unlearning

Pain can be a conditioned response —a learned behavior— rather than solely a physical problem. It often begins as a direct reaction to injury but can become reinforced over time, evolving into a conditioned response (i.e., Chronic Pain). Pain unlearning with XR involves reshaping these learned experiences through innovative interactive scenarios using Avatars (L1), Semi-Autonomous Virtual Character (L2) and/or Autonomous Virtual Character (L3) driven by Cognitive Architectures, and re-educating the perception of pain to enhance patient well-being.

Illustrative reference: https://www.researchgate.net/publication/377705399_Perspective_The_Promise_of_Virtual_Reality_as_an_Immersive_Therapeutic

Technostress

Technostress refers to the physical, cognitive, and emotional strain that arises from interacting with digital technologies. In low back pain rehabilitation using XR, it describes the stress patients and health professionals may experience when adapting to immersive systems — such as increased cognitive load, postural demands, or usability challenges — which can influence engagement, adherence, and therapeutic outcomes. Adopting ethical standards such as Dignity, Cooperation, are crucial to prevent technostress.

Illustrative reference: https://www.researchgate.net/publication/285608418_Technostress_The_Dark_Side_of_Technologies

Virtual Pills

Brief, targeted XR-based interventions designed to address a specific low-back pain problem with the patient. Much like taking medication, these short and focused immersive experiences, exploiting for instance Hybrid Synchronization and Sensorimotor Propagation, aim to deliver precise therapeutic effects—such as reducing pain, anxiety, or maladaptive movement patterns—within a limited timeframe.

Perceived Usability

The extent to which a patient believes that using technology will be free of effort. In XR rehabilitation, higher perceived usability increases patient engagement, adherence, and confidence in the therapeutic process. Perceived usability can be increased by adopting an ethical approach, reinforcing Self-identity, Trust and Transparency.

Illustrative reference: https://pubmed.ncbi.nlm.nih.gov/37279039/

Social Pain Exercise

The physical activity performed while interacting with other patients who share similar conditions, exploiting Hybrid Synchronization, and the transmission of Social Information. This social dimension reduces the perception of pain, increases Remote Social Connectedness, and fosters broader health benefits, enhancing both physical and emotional well-being.

Anticipatory Perception

The ability to detect early cues (e.g., posture shifts, micro-accelerations, SensoriMotor Primitives) of Social Information that signal a forthcoming action. In XR cycling, realistic modeling, Amplification/Attenuation of sensorimotor primitives is critical for training defenders to anticipate attacks of their opponents in the peloton.

Situation Awareness in XR

The capacity to perceive, comprehend, and predict evolving situations in immersive simulations. Challenges include ensuring athletes can generalize what they learn with virtual cues (e.g., the kinematic read-out of Social Information that signal an upcoming attack in a racing scenario) to real-world contexts.

Illustrative reference: https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2015.00010/full

Fidelity of Micro-Movements in Avatars

The precision with which XR systems capture and render subtle movement cues in Avatars (L1), Semi-Autonomous Virtual Character (L2) and/or Autonomous Virtual Character (L3), and therefore facilitate Sensorimotor Propagation of Social Information. In the sport domain, micro-movements such as shoulder twitches or changes in pedal rhythm, visible through the body deformation can signal an attack. High-fidelity capture and rendering through accurate Movement Representation is therefore essential for athletes to learn to detect and act on these early signs.

Action-Perception Coupling

The tight linkage between what an agent perceives (e.g., visual flow, body language of others) and how they act (e.g., adaptation, reaction, changing the visual perspective itself..). In racing sports, perception can be the Social Information embedded in others’ kinematics, and coupled actions can be shifting cadence, or blocking positions for instance. In XR, maintaining the natural timing of this coupling is key to effective Sensorimotor Propagation and transfer of training.

Illustrative reference: https://pubmed.ncbi.nlm.nih.gov/40279790/

Stress and Decision Load

The heightened cognitive and physiological demands of monitoring and reading out the Social Information of multiple partners (e.g., rival and team mate riders), anticipating their action (e.g., attacks), and responding under time pressure. XR scenarios must balance realism with manageable cognitive load to avoid overwhelming athletes during training.

Illustrative reference: https://www.frontiersin.org/journals/virtual-reality/articles/10.3389/frvir.2022.995090/full

Hybrid Co-location

Commonly, co-location means that multiple people are in the same physical space. In the context of ShareSpace, Hybrid Co-location applies to participants who are present in the same physical and virtual space (e.g., in a VR cave such as the Ars Electronica’s Deep Space 8K that was used in Sharespace artistic applications), or in the same virtual space as remote users with whom they do not share the physical space. A challenge in designing shared hybrid spaces is to implement Hybrid Co-location in a way that enables feelings of social presence and remote social connectedness.

Illustrative reference: https://ars.electronica.art/futurelab/en/projects-converge/

Collaborative Aesthetics (CA)

Building upon the term Cooperative Aesthetics, defined by media art expert Gerhard Funk, collaborative aesthetics describes an interactive yet creative process of unique shared experiences in XR. The Action–Perception Coupling between members of a group allows the product of their interaction to bring forth a new system of interconnected participants which generates a unique outcome (e.g., an audio-visual art piece, a collective narrative and many more). Within a shared hybrid space, CA empowers each and every users to actively contribute their ideas and perspectives to a collective experience of creation.

Illustrative reference: https://ars.electronica.art/hope/en/events/cooperative-aesthetics/

Authenticity of XR Experience

The challenge of preserving the emotional depth, immediacy, and aura of live performance when mediated by digital tools. It addresses concerns about whether digital art can evoke the same intensity and Social Presence as in-person encounters. One way to take up this challenge and that Sharespace employs is to design and use human-inspired Cognitive Architectures to drive the behaviours of Semi-Autonomous Virtual Character (L2) and/or Autonomous Virtual Character (L3) and, in particular, to model their ways of interacting with Avatars driven by humans (L1).

Immersive Audience Engagement

The active involvement of spectators in XR performances through sensory immersion, detection and transmission of Social Information,and participation through Hybrid Synchronization. It shifts the audience from passive viewers to co-experiencers, deepening emotional resonance but also raising challenges of agency, accessibility, and design.

Hybrid Liveness

The blending of physical and virtual presence in XR performance, where digital and Embodied elements coexist. It highlights both the creative potential and technical barriers of maintaining real-time Hybrid Synchronization across physical and virtual stages.