Social presence

See also: Presence

Social presence is the sense of being with another person in a mediated environment. In a 2018 systematic review, Catherine Oh, Jeremy Bailenson and Gregory Welch, drawing on a 1997 definition by Frank Biocca, describe it as "the subjective experience of being present with a 'real' person and having access to his or her thoughts and emotions" and call it "a crucial component of interactions that take place in virtual reality". In 2003, Biocca, Chad Harms and Judee Burgoon described it more briefly as the "sense of being with another", and several researchers use the term co-presence for the same idea.

While Presence is the feeling of being physically and spatially in an environment, social presence is believing that you're in the presence of other people. When you feel social presence you should acquire a visceral sense of the emotional state of the person/people you're talking to or interacting with.

Social presence is extremely important for social VR apps. vTime, an early social VR app for Gear VR released in December 2015, placed avatars in predetermined seats within each environment and animated their mouths, eyes, bodies and hands from each user's audio input, because the Gear VR offered only head tracking. The app, by then named vTime XR, was taken offline on 10 October 2025, shortly before its tenth anniversary, because of a Unity vulnerability that its developer said it could not patch.

Reviewed 27 September 2026. Checked every new claim against Oh, Bailenson and Welch (2018), Biocca et al. (2003), the Garau (2003), Smith and Neff (2018), Kim (2018) and Son (2025) papers, and the Road to VR, vTime, Meta, Microsoft Research, TechCrunch, Apple Support and UploadVR sources cited. About review dates.

Definition and related concepts

The concept was first set out by Short, Williams and Christie in their 1976 book The Social Psychology of Telecommunications, which defined social presence as the salience of the interactants and their interpersonal relationship during a mediated conversation. Short and colleagues named intimacy (the feeling of connectedness between communicators) and immediacy (the psychological distance between them) as its two core components. Both depend on verbal and nonverbal cues such as facial expressions, vocal cues, gestures and physical appearance, and the authors treated social presence as a "quality of the medium itself", meaning that some media carry these cues better than others.

Early research on computer-mediated communication shared this technology-centred view: media richness theory held that media differ in their ability to carry "rich" social information, which makes some more appropriate than others for certain tasks. Social information processing theory rejected it, holding that people adapt to media with fewer cues, so that text-based communication can reach the same intimacy as face-to-face contact, though it may take longer. Walther's hyperpersonal model went further and proposed that text users can sometimes reach higher social presence than people meeting face to face by choosing which parts of themselves to reveal. In their 2003 review in the journal Presence: Teleoperators and Virtual Environments, Biocca, Harms and Burgoon argued that a detailed theory of social presence could help researchers "predict and measure differences among media interfaces" and "guide the design of new social environments and interfaces", and they proposed criteria and scope conditions for future theories and measures.

Oh, Bailenson and Welch separate social presence from two other dimensions of presence:

Dimension	Description in the 2018 review
Telepresence (spatial presence)	The extent to which a person feels present in the mediated environment rather than in the immediate physical environment
Self-presence	How connected a person feels to their virtual body, emotions or identity
Social presence (co-presence)	The "sense of being with another", which requires a co-present entity that appears to be sentient

The same review notes an overlap with the plausibility illusion described by Slater, which captures the extent to which users feel they are interacting with an actual social being. Plausibility illusion is broader, because it also covers the credibility of other parts of a virtual scenario, not only the virtual human. Social presence can be felt toward an avatar controlled by another person or toward an agent controlled by a computer.

What affects social presence

Oh, Bailenson and Welch reviewed 233 separate findings from 152 studies that tested predictors of social presence, grouping them into immersive qualities of the system, contextual factors and individual psychological traits. Their main findings were:

Factor	Finding in the 2018 review
Level of immersion	Of 10 studies comparing an immersive platform with a non-immersive one (such as a head-mounted display versus a desktop), only two found a significant difference in social presence. The authors suggest that "once a threshold is met", more immersion does not automatically raise social presence.
Visual representation	Most studies found higher social presence when a visual representation of the partner was present, for example when participants could see a partner's avatar rather than speaking with an invisible partner.
Behavioral realism	Described as "a powerful predictor of perceived social presence", with the most consistent effects when the avatar or agent shows awareness of its partner, for example through mutual gaze, nodding at appropriate times or blushing. Mutual eye contact raised social presence, but gaze held for an unrealistically long time drew negative responses.
Photographic and anthropomorphic realism	Mixed results: some studies found an increase with more realistic or humanlike appearance, others found no difference, and one found a reduction.
Haptic feedback	All but one of 10 studies found a positive relationship between haptic feedback and social presence.
Depth cues	Studies of stereoscopy and motion parallax suggest that adding depth cues increases social presence.
Audio	One study found higher social presence with binaural audio than with mono or stereo sound; another found surround sound better than two-channel sound.
Agency (avatar versus agent)	Less conclusive. Four of five studies published before 2010 found higher social presence with human-controlled avatars than with computer agents, but only two of seven studies published after 2010 did.
Physical proximity	Five studies consistently found higher social presence when participants were in the same room, although co-location was often confounded with being able to see one another.
Individual traits	Neither age nor gender had a clear effect. Traits such as immersive tendency, transportability, extraversion and need to belong tended to predict higher social presence.

The review concluded that behavioral realism may explain why earlier studies of immersive avatar-mediated VR systems failed to find higher social presence: older platforms offered a "fairly limited level of behavioral realism".

Avatar appearance and gaze

A 2003 experiment at University College London by Maia Garau, Mel Slater and colleagues paired participants in a shared immersive environment, using a CAVE-like system or a head-mounted display. Each person was represented either by a basic, genderless "match-stick" avatar or by a more photorealistic, gender-specific one, and the avatar's eye gaze was either random or inferred from voice. In an analysis of 48 post-experiment questionnaires, the higher-realism avatar received higher ratings, including for copresence, when used with inferred gaze, while inferred gaze had a consistently negative effect on every response for the lower-realism avatar, a significant interaction between appearance and behavior. The authors concluded that "the more realistic the avatar appearance, the better the gaze model that should be used". Participants in that study stood facing their partner's avatar throughout the conversation, kept a suitable interpersonal distance and, according to the authors, "felt compelled to display polite attention".

Embodied avatars

Harrison Jesse Smith and Michael Neff compared three conditions in a 2018 study of 30 pairs who negotiated an apartment layout and placed model furniture on a floor plan: face-to-face, embodied VR with visible full-body avatars, and VR with a shared virtual space but no visible avatars. They reported that embodied VR "provides a high level of social presence with conversation patterns that are very similar to face-to-face interaction", while the shared environment without avatars "was generally found to be lonely" and appeared to degrade communication.

A 2025 study by Gayoung Son, Aline Tiemann and Marius Rubo placed 128 participants in pairs in a custom social VR environment on Meta Quest Pro headsets. Spatial presence and the perceived warmth of the partner were robustly associated with social presence, while body ownership and perceived competence were not significantly related to it in the multivariate analyses. These patterns were the same whether the pairs used anonymous avatars or non-anonymous ones.

Agents in augmented reality

Social presence also applies to computer-controlled characters in augmented reality. In a 2018 study presented at the IEEE International Symposium on Mixed and Augmented Reality, Kangsoo Kim and colleagues compared three versions of an intelligent virtual assistant with 15 participants wearing a Meta 2 optical see-through headset: a disembodied voice, an embodied AR agent that stood in one place and used upper-body gestures, and an embodied agent that gestured and walked about the room. The social presence subscale of the Temple Presence Inventory differed significantly between all three conditions, and the authors reported support for their hypothesis that an agent showing awareness of, and influence over, the real surroundings improves perceived social presence, confidence and trust.

Measurement

Social presence is usually measured with self-report questionnaires. The studies in the 2018 review used several instruments, among them the Networked Minds questionnaire developed by Biocca and colleagues, the scale from Short, Williams and Christie, the Temple Presence Inventory by Lombard and colleagues, the Social Presence in Gaming Questionnaire by IJsselsteijn and colleagues, and many custom scales. The reviewers limited their review to studies with at least one self-report measure and did not include studies that used only behavioral or physiological measures. Because questionnaire wording can influence answers, they noted that the variety of measures may explain why the same feature predicted social presence in some studies but not in others, and they chose not to conduct a quantitative meta-analysis for that reason.

Applications in VR and AR

Researchers and companies treat social presence as a design goal for avatars, telepresence and collaboration tools. Oh, Bailenson and Welch report that social presence is associated with communication outcomes such as persuasion and attraction; one study they cite found it positively associated with trust, enjoyment and perceived usefulness of an online shopping website, leading to greater purchase intentions. They also argue that more social presence is not always desirable, and that research should show when it is (and is not) worth adopting the features that increase it.

Face and eye tracking

Meta introduced the Meta Quest Pro at Meta Connect in October 2022 as its first headset with inward-facing sensors for natural facial expressions and eye tracking. According to Meta, when the user raises an eyebrow, smiles or makes eye contact, the avatar does the same, which "helps improve social presence". Meta's privacy post for the headset called social presence, "feeling like you're together, despite being separated by physical distance", a "crucial stepping stone" toward its metaverse plans. Eye tracking and Natural Facial Expressions are off by default; if they are turned on, images of the user's eyes and face stay on the headset, are deleted after processing, and are not shared with Meta or third-party apps.

Photorealistic avatars and volumetric capture

Facebook Reality Labs described its Codec Avatars research in 2019 with the line: "If telepresence lets you feel like you're somewhere else, then social presence lets you share that sensation with other people." Yaser Sheikh, director of research at Facebook Reality Labs in Pittsburgh, said the avatars had to pass what the team called the "ego test" and the "mother test": "You have to love your avatar and your mother has to love your avatar before the two of you feel comfortable interacting like you would in real life."

Microsoft Research's Holoportation takes a different approach: it captures people in 3D with Azure Kinect depth and color cameras, then reconstructs, compresses and transmits the models in real time so that users of mixed reality displays such as HoloLens can "see, hear, and interact with remote participants in 3D as if they are actually present in the same physical space". Microsoft states that the technology has been used in telemedicine work with surgeons in Glasgow and incorporated into Microsoft Mesh.

Apple introduced spatial Personas for Apple Vision Pro headsets running visionOS 1.1 or later on 2 April 2024. In a FaceTime call with other Vision Pro users, spatial Personas appear in the user's space with Spatial Audio, "so it feels like you're together in the same room", and participants can move around, make eye contact and use SharePlay; the feature supports up to five participants.

At Meta Connect in September 2026, Meta announced Hologram, a realistic avatar system for calls. UploadVR reported that a shoulders-up version would reach Meta Ray-Ban Display first, for WhatsApp video calls in Early Access in the US in fall 2026, and a full-body version would ship with Meta VR Glasses in spring 2027. In his hands-on, UploadVR's David Heaney wrote that "it's compelling to see the person you're talking to at (almost) true scale in 3D".

Features that enhance social presence

• Gaze, eye contact and facial expressions of avatars
• VR audio/3D audio
• Voice chat from avatar
• Visible, tracked avatar bodies
• Touch through haptic feedback
• Stereoscopic depth and motion parallax

See also

• Presence
• Telepresence
• Social VR
• Avatar
• Codec Avatars
• Face tracking
• Uncanny valley