RESEARCH DOSSIER
Immersive Media, Presence, and Virtual Embodiment
Abstract. Immersive media organize perception and action through technologically generated environments in which spatial location, bodily ownership, agency, and the credibility of events can be experimentally dissociated. This monograph investigates how virtual reality and related extended-reality systems produce spatial presence, plausible events, avatar embodiment, perceptual recalibration, and behavioral change. Its method is a critical narrative synthesis of foundational theoretical works, controlled multisensory and avatar experiments, psychophysical studies, quantitative meta-analyses, reviews of measurement validity, and recent integrative research. The resulting account distinguishes physical immersion from experienced presence, place illusion from plausibility, body ownership from agency and self-location, and experimental effects from claims about durable real-world transfer. Evidence indicates that tracking and sensorimotor contingencies shape presence; temporally coordinated visual, motor, and tactile information can induce virtual body ownership; avatar characteristics can alter selected behaviors; and spatial judgments, learning, social responses, and discomfort depend on context and design. The monograph specifies methodological requirements for causal identification and examines safety, representation, consent, and research opportunities arising with embodied artificial agents in the Artificial Era.
Keywords: immersive media; virtual reality; extended reality; spatial presence; telepresence; place illusion; plausibility illusion; virtual embodiment; body ownership; sense of agency; self-location; avatar; Proteus effect; visuomotor synchrony; visuotactile synchrony; peripersonal space; egocentric distance; cybersickness; perceptual adaptation; human–artificial interaction.
The Research Object and Scientific Questions
Immersive media transform the relationship between the perceiving body and its represented environment. A participant can see a remote landscape from an egocentric viewpoint, move through synthetic space with tracked gestures, encounter a responsive person represented by an avatar, or perceive a digitally constructed body occupying the position normally reserved for the biological body. These events are scientifically tractable because the experimenter controls visual perspective, temporal contingency, movement feedback, apparent morphology, spatial scale, social response, and sometimes touch and sound. A psychologically meaningful investigation therefore examines an organized relationship among sensory information, motor prediction, bodily representation, perceived place, and situated behavior.
The object of this dossier is the human psychological organization of spatial presence and bodily experience in immersive and extended-reality media. Four linked questions govern the inquiry. Under what technological and perceptual conditions does a person experience a virtual place as the currently inhabited environment? How can a represented body become the temporary object of ownership, control, and self-location? Which perceptual and social responses transfer from ordinary environments into virtual ones, and which depend on artificial task structure? Finally, how can researchers distinguish repeatable causal effects from expectancy, demand characteristics, novelty, hardware-specific artifacts, and selection of highly tolerant participants? The review by Bailenson, J. N., DeVeaux, C., Han, E., Markowitz, D. M., Sa (2025) provides a contemporary synthesis of canonical findings that connects these questions without treating every kind of VR application as psychologically equivalent.
The dossier belongs primarily to Media and Technology Psychology. Its experimental foundations also intersect Perception and Psychophysics, Cognitive Psychology, Human Factors and Engineering Psychology, and research on bodily self-consciousness. The focus remains media-specific presence and embodiment. Clinical virtual-reality exposure is scientifically related because it recruits perceptual and affective responses, while the comparative efficacy, indications, and protocols of treatment are independent clinical subjects.
Definitions and Conceptual Architecture
Immersive media are technologies that place participants within coordinated mediated sensory and action spaces. Virtual reality (VR) typically substitutes a computer-generated world for the immediately visible physical surroundings; augmented reality (AR) overlays digital elements within a view of the physical world; mixed reality (MR) combines digital and physical referents with degrees of spatial anchoring and interaction; extended reality (XR) is the umbrella used across these arrangements. These technical descriptions require operational specification: a stereoscopic headset with six-degree-of-freedom tracking supports experiments unavailable to a stationary monitor, but binocular imagery alone does not identify the psychological process responsible for an effect.
Technological immersion describes relatively objective capabilities of a display and interaction system: field of view, stereoscopy, update latency, the extent and fidelity of tracking, sensory channels, and the correspondence between action and feedback. Spatial presence designates the participant's experienced location in the represented world. Telepresence has historically characterized being experienced at a mediated rather than immediately physical site. Steuer, J (1992) proposed vividness and interactivity as relevant dimensions, while Slater, M., & Wilbur, S (1997) separated technological immersion from the psychological experience of presence.
A usable vocabulary needs further differentiation. Spatial presence concerns perceived place; plausibility concerns whether depicted events seem to be occurring within the enacted situation; social presence concerns experiencing others as socially available; self-presence concerns the experience of one's mediated representation as personally relevant; embodiment concerns ownership, agency, and experienced bodily location. Attention, engagement, emotional intensity, task success, and enjoyment are possible correlates, outcomes, or moderators, and each requires separate measurement. Calling all of these states “immersion” hides the very differences experimental research must explain.
The distinction has a causal implication. A hardware manipulation can increase sensory immersion without increasing the credibility of a narrative, and an engrossing narrative may create involvement with limited tracking. A strongly owned virtual body may coexist with poor distance calibration; a convincing environment may produce sickness; an engaging learning scene may impose extraneous cognitive load. Research becomes cumulative when it identifies manipulable inputs, hypothesized mediating processes, measured outcomes, and the boundaries under which the relations recur.
A Historical Genealogy of Immersive Psychology
The scientific lineage combines teleoperation, visual simulation, experimental perception, human-computer interaction, communication research, and bodily self-consciousness. Early immersive apparatus made a decisive experimental move: the visual world surrounding an observer could be generated, altered, and coupled to measured head motion. That architecture connected naturalistic behavior with controlled stimulus manipulation. Loomis, J. M., Blascovich, J. J., & Beall, A. C (1999) formulated the methodological promise of immersive virtual environments for perception, spatial cognition, and social interaction, stressing both the experimental control and the new kinds of artifacts that such systems introduce.
Communication scholarship meanwhile established presence as an independent psychological problem. Steuer, J (1992) located the issue in mediated experience and interaction rather than a particular commercial technology. The Slater, M., & Wilbur, S (1997) formalized the separation between system properties and experienced presence. Witmer, B. G., & Singer, M. J (1998) developed a presence questionnaire and an immersive-tendencies questionnaire, helping make subjective reports part of experimental measurement while opening questions about construct validity and comparability.
In parallel, researchers studied bodily self-identification using spatially and temporally controlled multisensory conflicts. The Botvinick, M., & Cohen, J (1998) demonstrated that synchronous visible and felt touch could shift experienced ownership toward an artificial hand. The Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O (2007) extended the experimental problem toward self-location. These findings gave virtual embodiment research a mechanistic foundation independent of particular headset brands.
A further transition joined bodily experiments to fully simulated scenarios. Sanchez-Vives, M. V., & Slater, M (2005) argued that presence could serve as an experimental window into perception and consciousness. Slater, M., Antley, A., Davison, A., Swapp, D., Guger, C., B (2006) illustrated that virtual social situations could elicit consequential responses while maintaining experimental control. Slater, M (2009) clarified why people can act realistically toward simulated events they explicitly recognize as simulations.
The 2010s added controlled changes to a participant's represented body. Slater, M., Spanlang, B., Sanchez-Vives, M. V., & Blanke, O (2010) demonstrated body-transfer illusions with a first-person virtual body; Banakou, D., Groten, R., & Slater, M (2013) examined consequences of embodying a childlike form; Peck, T. C., Seinfeld, S., Aglioti, S. M., & Slater, M (2013) examined implicit racial attitudes following embodiment. Contemporary research increasingly integrates controlled VR experiments with meta-analyses, revised psychometrics, and scrutiny of generalization. This history is a convergence of scientific traditions, not a sequence in which each newer device automatically invalidates older theory.
Immersion as an Experimental System
An immersive apparatus is a generator of controlled relationships between action and sensation. The researcher can hold the represented room constant while varying whether head translation produces the expected perspective shift, or can retain head tracking while varying the temporal relationship between a participant's movement and an avatar's movement. Such factorial designs allow a hypothesis about tracking to be tested separately from content attractiveness, rendering quality, or task narrative. Cummings, J. J., & Bailenson, J. N (2016) synthesized 115 effect sizes from 83 studies and found a medium overall influence of technological immersion on presence, with tracking, stereoscopy, and field of view especially consequential in the literature they examined.
Experimentally useful variables include display configuration, six-degree-of-freedom tracking, update rate, tracking accuracy, delay, apparent eye height, avatar morphology, audiovisual congruence, tactile timing, physical walking relative to joystick locomotion, and the consistency of collision or contact feedback. Their effects must be interpreted relative to a clearly identified comparator. A modern headset compared with an older monitor varies several properties simultaneously; claims that one particular feature caused a difference require direct isolation, random assignment, or a sufficiently informative counterbalanced design.
A credible virtual laboratory also calibrates the participant's position and body scale. Incorrect eye height changes the apparent size of the environment; a misaligned avatar alters action feedback; a shifted floor affects perceived clearance; inconsistent audio distance can influence orientation. Experimental records need hardware models and firmware, rendering frame rates, tracking modes, locomotion schemes, content builds, and versioned software conditions. A report that says only “participants used VR” leaves essential causal variables undisclosed.
Immersive experiments offer a distinctive balance of environmental richness and repeatability. An identical event can occur for every participant, while behavior and physiological response can be captured continuously. Yet repeatability of the software is not equivalent to repeatability of the experience: adaptation, expectations, discomfort, prior gaming experience, and anthropometric fit may vary widely. Bohil, C. J., Alicea, B., & Biocca, F. A (2011) emphasized that VR combines contextual stimulation and laboratory control; the broader methodological lesson is to document both sides rather than assuming ecological validity from visual realism alone.
Place Illusion, Plausibility, and Behavioral Realism
Place illusion is the experience of occupying a represented location as if it were the current place of perceptual engagement. Plausibility illusion concerns the apparent occurrence and relevance of events taking place there. In Slater, M (2009), place illusion depends heavily on sensorimotor contingencies, while plausibility depends on events coherently responding to the participant and making sense in the depicted situation. The same person may know that a virtual precipice is computer-generated and nonetheless hesitate at its edge. An enduring factual belief about the world and an immediate situated response can operate at different levels of processing.
This distinction explains why graphical realism is an incomplete design variable. A visually detailed environment can be unconvincing when objects ignore the user's actions, and a simpler environment may support strong spatial presence if head motion, orientation, perspective, and interaction remain coherent. Conversely, an environment with excellent tracking can depict events whose behavior is inconsistent or socially implausible. Research should manipulate these components independently and measure perceived place, event credibility, and behavior rather than compress them into a single global score.
A psychological explanation of “realistic response” also requires careful outcome selection. Physiological arousal during a virtual hazard can establish sensitivity to a depicted threat without proving that the participant believed actual injury was imminent. A participant who talks to an avatar may respond to ordinary conversational cues or the experimenter's expectations. Observed behavior is the outcome of interest, but its causal interpretation depends on plausible alternative mechanisms and the comparison condition. Slater, M., Antley, A., Davison, A., Swapp, D., Guger, C., B (2006) used an ethically controlled simulation of obedience-related circumstances; the study cannot simply be read as an unrestricted replication of the original historical paradigm.
The most useful experimental test contrasts matched environments that differ in event contingency, relational responsiveness, or sensory-motor consistency. Repeated measures can identify the moments when an illusion breaks; subjective reports and behavioral outcomes can establish whether the same manipulation changes both felt presence and action. These analyses make presence an experimentally structured research construct instead of a synonym for technological sophistication.
Sensorimotor Contingencies and Perceptual Coupling
Sensorimotor contingencies are systematic relations between action and resulting sensation. When an observer moves the head to the right, nearer surfaces usually shift in a different retinal pattern from distant surfaces; when the body rotates, the surrounding visual field changes coherently; when a hand reaches toward an object, visual and proprioceptive information support a prediction about the movement's outcome. A tracked environment can reproduce many of these relations. The resulting correspondence is central to explaining why some virtual places feel spatially inhabitable.
These contingencies support active perception. The observer samples the represented world by moving, looking, reaching, and testing predicted relationships. Presence therefore has a temporal and action-dependent aspect that static image quality alone cannot explain. In research, this predicts different experiences under passive panoramic video, head-tracked six-degree-of-freedom navigation, and body-tracked manipulation, even when the imagery shares objects and narrative content. It also predicts sensitivity to latency, mismatched physical/virtual motion, and insufficient feedback about contact.
Sensorimotor coupling should be characterized by which channel is modified and by its temporal profile. Artificial hand control can be synchronous for small gestures but delayed for larger ones; body tracking can match orientation while misrepresenting stride; a virtual arm can appear anatomically plausible while its movement violates the participant's proprioceptive estimate. Such disruptions offer experimentally informative dissociations between agency and ownership. They can also provoke discomfort or attention to the display as an apparatus.
The experience of coupling emerges from the whole action-perception loop. An attractive avatar cannot compensate for persistent spatial inconsistencies in every task. Equally, a coherent loop does not guarantee comprehension of a learning lesson or a lasting social attitude change. The causal value of sensorimotor design lies in allowing precisely identified relationships to be tested, rather than in assuming a uniform psychological benefit from adding more sensors.
Spatial Presence and the Geometry of the Virtual Environment
The represented environment provides spatial reference frames: world-centered coordinates, body-centered directions, virtual eye height, perceived ground plane, motion parallax, horizon, and apparent object scale. Presence depends partly on the coordination of these cues into a workable location for the perceiver. Egocentric perspective is especially consequential because the participant encounters the world from the apparent position of the current body rather than from a detached camera. The perceived place can remain stable even as the person moves through the room physically or uses artificial locomotion.
Spatial presence has behavioral implications. A person may step around a virtual obstacle, regulate distance from another represented agent, or lean toward an object to inspect it. Such responses can provide evidence that the mediated environment participates in action planning. They do not automatically prove accurate metric perception; one can inhabit a virtual room while misjudging distances. Renner, R. S., Velichkovsky, B. M., & Helmert, J. R (2013) documented a substantial history of underestimation of egocentric distance in virtual environments, with wide methodological and contextual variation.
Experimental studies should distinguish orientation, route choice, presence ratings, spatial memory, perceived scale, and absolute distance estimation. Relative spatial order may be preserved while distance magnitude is compressed. A participant may show natural avoidance behavior without reproducing exact real-world interpersonal spacing. Differing metrics map onto distinct perceptual processes and design affordances.
The scientific status of spatial presence consequently rests on multiple measurements and explicit geometrical calibration. Depth cues, interpupillary distance settings, display optics, texture, floor representation, apparent body scale, task goal, and movement mode should be reported. Spatial judgments are especially relevant to engineering applications because navigational errors that appear small in a laboratory rating can become consequential when a person interacts with physical equipment through an augmented or mixed-reality overlay.
Multisensory Integration and Bodily Self-Representation
The experience of a body combines visual appearance, proprioception, touch, movement prediction, vestibular information, interoception, and the continuity of self-location. The brain normally integrates these cues because they provide convergent evidence about a single organism. Altering their relationship can induce experimentally informative illusions. In the classic Botvinick, M., & Cohen, J (1998), synchronous visual and tactile events were sufficient to elicit the experience that touch on a visible artificial hand belonged to the participant's own bodily representation.
Theoretical significance follows from the possibility of separating visual bodily identity from the physical limb. A person retains factual knowledge of the body while undergoing a changed immediate experience of ownership. The experiment thus gives access to inferential processes through which sensory evidence contributes to bodily self-identification. Its result also serves as a methodological warning: ownership questionnaires, pointing errors, and reactions to a threat near a dummy hand may reflect partly different processes.
Full-body manipulations extend this logic to the experienced position of the self. Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O (2007) investigated changes in bodily self-consciousness with experimentally controlled visual perspectives and tactile stimulation. The underlying question became whether self-location and body ownership necessarily shift together. Virtual reality expands the available manipulations because a participant can see a coherent whole body, move it through a tracked mapping, and experience transformations of its appearance while physical action continues.
Contemporary experiments must distinguish the induction procedure from the state claimed to result. Synchronizing a virtual hand is a manipulation; the participant's reported ownership is a measured response. A difference in skin conductance near an apparent threat may support an affective consequence but should not be treated as a unique physiological signature of ownership. Strong interpretation requires converging measurements, manipulation checks, and designs that can test competing explanations.
The Components of Virtual Embodiment
Virtual embodiment describes the experienced relation between a participant and a represented body. Kilteni, K., Groten, R., & Slater, M (2012) distinguish body ownership, agency, and self-location. Ownership concerns experiencing the body as one's own; agency concerns the felt authorship or control of its actions; self-location concerns where the participant experiences being situated. These components commonly interact, yet a system may change one without producing an equivalent change in the others.
Ownership often depends on first-person viewpoint, plausible alignment, and cross-modal congruence. Agency can arise when observed movement corresponds to an intended motor command. Self-location depends on the integration of viewpoint and bodily information, including the position from which events appear to be experienced. An avatar can look like the participant without being strongly owned; it can move under accurate control without seeming to be located where the participant's biological body is located. The formulation is useful precisely because it resists equating avatar customization with achieved embodiment.
Gonzalez-Franco, M., & Peck, T. C (2018) proposed a structured approach to avatar-embodiment questionnaire content, emphasizing multidimensional assessment. The operational consequence is to score relevant facets separately and examine their covariance, rather than treating a single total as a universal index of embodiment. Self-report items should be interpreted according to the scenario and the participant's understanding of language such as “my body,” “control,” and “being there.”
Virtual embodiment can be partial or dynamically interrupted. A user may own an avatar's torso while perceiving its extremities as misaligned, or experience convincing action control until a tracking failure exposes the surrogate character of the representation. Experiments recording temporal breaks in the illusion, as well as post-session scores, permit a more faithful representation of this process. The science of embodiment is thereby a science of transient configurations of perception and action, not a binary judgment that a virtual body “became real.”
Visuomotor and Visuotactile Embodiment Experiments
Virtual embodiment experiments create a controlled correspondence, or discrepancy, between a biological participant and the represented body. In visuomotor conditions, the participant moves and observes an avatar moving concurrently; in visuotactile conditions, stimulation on the biological body is paired with a visible event on the virtual body. The relevant experimental comparison often varies temporal synchrony, viewpoint, or anatomical alignment while keeping the virtual scene stable. The outcome is assessed through reported ownership, agency, localization, response to apparent threat, or performance in an embodied task.
The Slater, M., Spanlang, B., Sanchez-Vives, M. V., & Blanke, O (2010) demonstrated that a life-size virtual body viewed from the participant's apparent bodily position could support striking ownership experiences. Their contribution lies in showing how viewpoint and multisensory consistency provide a manipulable basis for identifying with a digital body. A virtual body can be treated experimentally as a component of the perceptual system rather than a merely decorative image of the participant.
Kokkinara, E., & Slater, M (2014) examined the evolution of body ownership under synchronous visuomotor and visuotactile stimulation. Their work matters because embodiment is often measured only after a session, despite being induced and maintained by a changing temporal relationship. The effects of moving and feeling need to be examined across exposure intervals and compared against appropriately asynchronous conditions. Tracking breaks, fatigue, and adaptation may change the experience during the same trial.
Causal interpretation requires attention to expectancy. Participants may infer the experiment's purpose when the avatar mimics their movements and then endorse ownership statements because those statements appear to be expected. Randomized condition order, concealed hypotheses when feasible, open-ended phenomenological reports, and behavioral tests help evaluate this possibility. Proprioceptive drift, a change in estimated bodily location, provides a complementary indicator but is not equivalent by definition to ownership. A strong experiment establishes the relation among manipulation, reported experience, and predicted downstream behavior without presuming that one measure validates all others.
Embodiment research gains further power through factorial designs crossing first-person and third-person perspective with synchronized or delayed movement, humanlike versus altered anatomy, and matched versus mismatched tactile feedback. Such designs can establish whether apparently similar ownership ratings arise through distinct perceptual pathways. These are the foundations for assessing applications in training, accessibility, spatial interaction, and encounters with artificial social partners.
Avatar Form, Scale, and Self-Perception
A virtual avatar can preserve the participant's form, modify its proportions, change apparent age or body size, or represent a body from another social category. Such transformations allow researchers to investigate how visible body structure enters perceptual judgments, action planning, and self-evaluation. An avatar's characteristics form part of the task environment because the participant may act through that representation, see it from an embodied viewpoint, and experience it as linked to ongoing movement.
Banakou, D., Groten, R., & Slater, M (2013) tested the consequences of embodying a childlike virtual body. The experiment connected virtual body representation to changes in apparent object size and implicit associations. Its central theoretical relevance is that self-representation can alter the scale against which environmental objects are perceived. Yet it does not establish that all avatar transformations produce equivalent changes, nor that effects persist after the body representation is removed.
An avatar's scale can affect judgments indirectly through eye height, hand size, reach, and the virtual ground plane. If an apparently small body coincides with a lower viewpoint, the experimental manipulation contains both a bodily and a geometrical component. Researchers need conditions that distinguish the effects of body appearance from those of camera height and world scale. The issue also matters for avatar-based social psychology: visual category cues may affect how participants expect others to react, while embodied motor coupling may affect how they respond to their own representation.
Self-perception in VR is a layered construct. Visual recognition of an avatar as personally similar differs from accepting its hand as part of the current bodily model; perceived authenticity differs from feeling control; social identification differs from perceptual ownership. Measuring these distinctions can reveal whether a behavioral shift follows altered body scale, internalized social meaning, immediate motor affordances, or a combination. Body transformation thereby provides a useful experimental probe into the relation between perceived body, represented identity, and environmental judgment.
The Proteus Effect and Behavioral Transformation
The Proteus effect refers to changes in behavior associated with characteristics of the avatar through which a participant interacts. Yee, N., & Bailenson, J. N (2007) introduced the term in experimental research on transformed self-representation, connecting avatar cues to self-perception and behavior. The phenomenon belongs at the intersection of mediated social cognition, situational self-representation, and embodied action. The fact that a participant knows the avatar is digitally generated does not eliminate the possibility that its apparent attributes provide salient behavioral cues.
A quantitative test of the broader claim appears in the Ratan, R., Beyea, D., Li, B. J., & Graciano, L (2020). Their synthesis covered 46 quantitative experiments using randomized avatar attributes and found small-to-medium effects, with estimates approximately 0.22 to 0.26 across analytic subsets. This supports a measurable family of avatar-characteristic effects while constraining any claim that body representation generally transforms personality or causes large, persistent changes in conduct.
Several mechanisms are plausible. Avatar appearance may prime social expectations; a participant may infer an appropriate role from the represented body; changes in perceived physical capability may alter action opportunities; social feedback from other avatars may reinforce a role; and task instructions may amplify or suppress the effect. The comparative empirical problem is to distinguish these pathways. A random avatar assignment establishes the influence of the assignment under that condition, while stronger mediational claims require independently measured and experimentally separable mechanisms.
Duration and transfer are critical. A person might negotiate more assertively during an avatar-mediated task and return to ordinary behavior afterward. Within-session actions, post-session attitudes, delayed behaviors, and independent real-world outcomes are scientifically distinct. Longitudinal tests and pre-registered outcome hierarchies are therefore needed before avatar effects can be interpreted as lasting transformations of identity, prejudice, confidence, or social behavior.
Avatar research also raises an interpretive boundary in studies of stereotyped social categories. Apparent race, age, gender, disability, or status can recruit historically learned associations; reproducing those associations experimentally can expose their causal operation without making them intrinsic properties of persons. Ethical design requires explaining the hypothesis, limiting degrading role assignments, and avoiding conclusions that treat social stereotypes as natural characteristics of the groups being represented.
Emotion, Appraisal, and Realistic Response
Virtual scenes can evoke fear, relief, interest, embarrassment, enjoyment, and other emotions when participants appraise depicted events as relevant to their current goals. Presence may contribute to the vividness and salience of those events, but the emotional outcome depends on meaning, personal history, learning, social context, and the perceived possibility of acting. The scientific object is an interaction between perceptual presentation and appraisal processes.
Diemer, J., Alpers, G. W., Peperkorn, H. M., Shiban, Y., & M (2015) reviewed the relation of perception and presence to emotional response in VR. Their synthesis highlights a field in which emotional reactions and felt presence can covary but are not interchangeable constructs. An arousing scene may heighten attention and reported presence; stronger presence may in some situations intensify appraisal; and a third variable, such as vivid threat content, may influence both. Correlation alone does not establish the direction of these influences.
Ethically appropriate experiments can vary the location, speed, or predictability of simulated hazards while monitoring avoidance behavior and physiological signals. Such tasks may reveal sensory-motor responses that participants cannot completely control through explicit knowledge that the danger is simulated. Nonetheless, an elevated heart rate or larger skin conductance response is compatible with multiple processes, including surprise, exertion, temperature, novelty, or social pressure. Strong claims require specific predictions and controls.
Emotional engagement is important in education, training, social perspective taking, and clinical research, but the functions differ. A strong affective response may support episodic encoding while narrowing attentional scope; a simulated threat may improve procedural rehearsal while discouraging exploration; a powerful social scene may prompt immediate action without durable change in attitudes. Consequently, responsible scientific evaluation reports the target behavior, the emotional state, the follow-up interval, and any adverse effects separately.
Spatial Perception, Distance, and Action Calibration
Spatial behavior in XR depends on the coordination of optical information, body scale, proprioception, locomotion, and learned action parameters. Immersive displays can supply stereoscopic disparity and motion parallax while still producing systematic differences from physical-world distance perception. Some judgments arise from visual cues, others from the movements used to approach a target. Experimental tasks must distinguish perceived distance, remembered distance, walking-to-target performance, and the selection of a navigation action.
The Renner, R. S., Velichkovsky, B. M., & Helmert, J. R (2013) documented a longstanding tendency toward underestimation of egocentric distance in many immersive virtual environments, together with sensitivity to measurement method and apparatus. The important scientific finding is the pattern of conditional distortion, not a timeless universal correction factor. Contemporary rendering, tracking, familiar-size cues, walking experience, and display calibration change the problem.
Perceptual calibration can improve with task feedback or physical locomotion, yet a participant's adjustments may be specific to the environment and response measure. A VR system that trains reaching could facilitate performance through repeated practice with its own controller, leaving physical reaching unchanged. Transfer requires a physically distinct test or another valid criterion. This methodological distinction is indispensable in safety-critical simulation, industrial training, and human-factors evaluation.
Virtual embodiment adds a body-relative frame to spatial measurement. The represented arm's length can make a target appear reachable, while altered eye height can change visual scale. A participant acting through an avatar has to resolve discrepancies between seen limb and proprioceptive estimates. Factorial manipulations of avatar dimensions and environmental geometry can reveal whether the participant adapts body schema, world scale, or both.
AR and MR introduce an additional issue: digital objects must appear spatially stable relative to the physical world. Tracking drift, occlusion errors, focal-distance conflict, and differences between virtual and physical contact can change decisions at close range. Evaluating mixed-reality guidance therefore requires measurements of physical task error and user safety in addition to ratings of visual coherence.
Learning, Training, and Cognitive Load
Immersive learning can provide situated practice, spatial demonstrations, procedural rehearsal, and opportunities to explore otherwise inaccessible environments. It can also consume attentional resources through novelty, navigation, input complexity, and the need to manage an unfamiliar body representation. The educational value of VR consequently depends on instructional design, learning objective, cognitive load, and a comparison condition appropriate to the claimed benefit.
Makransky, G., Terkildsen, T. S., & Mayer, R. E (2019) experimentally compared an immersive science-laboratory simulation with a slideshow-based alternative. Their study found greater presence alongside less learning in the immersive condition, demonstrating that a vivid instructional experience can impose costs on knowledge acquisition. It also analyzed cognitive-load indicators. The result is especially valuable because it identifies a concrete condition in which more technologically immersive instruction produced an educational disadvantage rather than merely failing to produce an advantage.
Parong, J., & Mayer, R. E (2018) investigated learning science in immersive virtual reality and offered further evidence that learning outcomes depend on the cognitive organization of immersive instruction. Procedural skills, conceptual explanation, spatial reasoning, and transfer should be treated as different endpoints. A learner may improve at a specific simulated action while gaining limited ability to explain the underlying concept.
The Bailenson, J. N., DeVeaux, C., Han, E., Markowitz, D. M., Sa (2025) reinforce the importance of distinguishing types of learning and task fit. Presence may motivate attention to an experience while also drawing attention away from verbal or abstract content. Repeated realistic practice may be more relevant to a task involving sensorimotor coordination than a lesson in symbolic relationships. Instructional interventions should therefore be designed around intended cognitive operations rather than display novelty.
Good experimental comparisons equate exposure duration, instructional content, prior knowledge, feedback, and assessment opportunities as far as feasible. They assess retention and transfer after a delay. They also measure sickness, cognitive load, and learner effort, because the same system can have benefits for one subgroup and burdens for another. The strongest educational conclusion identifies a particular task, learner population, design feature, and outcome that improves under testable conditions.
Perspective Taking, Social Attitudes, and Transfer
Embodied perspective taking can allow a participant to encounter a scenario from a visually and spatially different point of view. Social-psychological applications aim to influence how people interpret another person's circumstances, appraise social distances, or recognize consequences of unequal treatment. Virtual perspective differs in principle from passive description because the participant can act and receive feedback within the scene. Yet adopting a visual perspective is not equivalent to accurately knowing the thoughts, history, or lived conditions of the group represented.
Peck, T. C., Seinfeld, S., Aglioti, S. M., & Slater, M (2013) investigated implicit racial bias following ownership of a virtual body represented as belonging to a different racial category. The experiment provided evidence for an attitude-related consequence under a specified embodiment procedure. It does not warrant extrapolation to permanent prejudice reduction or general improvement in intergroup relations. The construct measured, the participant sample, the experimental timing, and the social meaning of the avatar limit what the data can establish.
Herrera, F., Bailenson, J., Weisz, E., Ogle, E., & Zaki, J (2018) compared immersive and nonimmersive perspective-taking interventions in studies of homelessness-related empathy and prosocial action. Their results distinguished self-reported empathy from particular behavioral outcomes; a more immersive intervention did not simply raise every empathy measure above its comparators. The scientific lesson is to measure what participants do, what they report feeling, and how long any difference persists.
Social intervention effects may be mediated by attention, involvement, narrative appraisal, personal identification, perceived norms, or the opportunity to rehearse action. Studies must report baseline attitudes and distinguish between an avatar representing the participant and a simulated person encountered as another. Cultural and ethical considerations are central because a brief simulation can produce a misleading impression of complete understanding of another's experience. A well-designed intervention communicates the limits of its representation and uses evidence of actual outcomes to evaluate effects.
Transfer beyond a laboratory session remains a stringent criterion. Durable effects on interpersonal behavior, decisions, or civic action require appropriate follow-up, realistic opportunities to act, and independent replication. Some VR interventions may be useful as elements of broader educational or social programs. Their value is established by controlled outcomes and meaningful durability, not by the intensity of the initial immersive encounter.
Experimental Paradigms and Causal Inference
Research on immersive experience employs a family of experimental paradigms with different inferential strengths. A technical-immersion paradigm varies display or tracking characteristics and measures presence. An embodiment paradigm manipulates sensory synchrony, avatar perspective, or morphology and measures bodily ownership or behavior. A spatial-calibration paradigm varies perceived geometry and measures distance estimates, navigation, or reaching. A social paradigm varies an avatar's appearance, operator, behavior, or apparent authority. A learning paradigm compares instructional conditions and measures retention, transfer, and load. These paradigms produce different kinds of evidence and should not be pooled conceptually merely because all use head-mounted displays.
Random assignment supports causal inference about a particular experimental manipulation when assignment is implemented and outcomes are interpretable. But many VR comparisons are complex interventions. A new headset can differ from the control in display field of view, comfort, novelty, image latency, visual resolution, interaction freedom, and expectations. A positive difference in presence establishes the effect of the entire configuration; identifying the responsible constituent requires additional experiments. Factorial designs, component ablations, within-participant counterbalancing, and preregistered manipulation checks can help disaggregate these influences.
Within-subject comparisons increase statistical efficiency for many perceptual endpoints, yet carryover may be pronounced. Once participants recognize an ownership illusion, experience sickness, or learn the geometry of a virtual corridor, later conditions may be contaminated. Counterbalancing addresses order on average but cannot eliminate persistent learning or adaptation. Between-subject comparisons reduce direct carryover but require adequate sample size and baseline balance. Researchers should justify the design relative to the psychological time scale of the effect.
Masking presents unusual problems. A participant generally notices a change from a desktop screen to a headset and may detect whether tracking is responsive. Researchers can nevertheless blind outcome scorers, conceal specific hypotheses, standardize experimenter interactions, and provide credible instructions across conditions. Demand effects can be studied through independent expectancy measurements and open-ended reports. A control condition should be chosen to test a specific mechanism, not simply to make the treatment appear more elaborate.
Sample-size planning should reflect the primary effect, expected variation, and multiple outcomes. Embodiment research often collects several ownership items, behavioral indices, and physiological signals. If a paper tests many correlations without defining the primary endpoint, some apparently impressive relationships can arise by chance. Preregistration, multiplicity control, effect estimates with confidence intervals, and shared analysis scripts strengthen interpretation. Generalizability remains a separate question from the internal validity of a single well-controlled laboratory comparison.
Measurement of Presence and Embodiment
A measurement instrument operationalizes a theoretical construct; its title does not guarantee that it measures a uniform psychological state. The Witmer, B. G., & Singer, M. J (1998) became influential by providing structured subjective items about virtual experience. Yet the field also employs other scales that emphasize spatial location, engagement, plausibility, or general involvement. A study must therefore specify which dimensions are being measured and why the chosen scale corresponds to its hypothesis.
The problem has recently received an unusually direct methodological examination. Kukshinov, E., Tu, J., Szita, K., Nathan, K. S., & Nacke, L (2026) systematically analyzed 397 empirical quantitative presence studies. They found widespread modification of presence questionnaires without adequate revalidation and incomplete reporting of instrument use. The study supports a concrete corrective principle: publish the actual item wording, scoring algorithm, transformations, missing-data procedure, and validity rationale. An adapted measure should be assessed psychometrically in the context in which it is used.
Embodiment scales face a related but distinct issue. Gonzalez-Franco, M., & Peck, T. C (2018) addressed avatar embodiment through questionnaire development and standardization. Items probing ownership, motor agency, location, and appearance must be related to relevant experimental manipulations. A person may feel control over a robot-like arm while rejecting an item asserting anatomical similarity. Aggregating the responses can conceal theoretically informative dissociation.
Psychometric evaluation entails more than an internal-consistency coefficient. Researchers need to examine factor structure, test–retest behavior where stable states are expected, sensitivity to known manipulations, and discriminant validity relative to enjoyment, immersion propensity, negative affect, and cybersickness. Comparisons across headset types, age groups, languages, and cultures require consideration of measurement invariance. Where a questionnaire describes changing states, repeated brief measures or continuous probes may be useful, but the probes themselves can interrupt presence.
The methodological priority is therefore construct-specific measurement. A scale intended to capture momentary place illusion should not be interpreted as a general measure of social presence, body ownership, or motivation. Statistical relationships between scales can inform a theory, but they cannot substitute for defining the constructs in advance.
Physiological and Behavioral Indicators
Immersive experiments generate continuous traces that are often unavailable in traditional questionnaire research. Head position and rotation, eye direction, hand trajectories, interpersonal distance, navigational choices, reaction times, object manipulation, and speech behavior can be recorded with millisecond or frame-level timestamps. Physiological indicators may include electrodermal activity, cardiac measures, and other signals suitable to the experiment. These measurements offer a way to investigate how action unfolds before a participant gives a retrospective summary.
Each behavioral or physiological indicator remains indirect. Recoiling from a virtual threat may indicate avoidance under the depicted conditions; it does not uniquely measure presence. Increased electrodermal activity can signal arousal without identifying fear or ownership. Head movement can reflect exploration, discomfort, task demands, or efforts to compensate for display limitations. Sensor logs should be interpreted against explicit mechanistic predictions and comparison conditions.
A useful multimethod design aligns distinct signals with distinct parts of a theory. If synchronous touch is predicted to increase virtual hand ownership, the design can collect ownership reports, localization judgments, and a prespecified protective response to a virtual threat. Agreement across these outcomes would support a coherent pattern; disagreement may reveal that the manipulation altered one component of embodiment more than another. The discordance is theoretically informative and should be reported.
Continuous data also create analytic risks. Dozens of motion variables, physiological windows, and machine-learning features can produce seemingly predictive models through overfitting. Signal preprocessing, artifact correction, temporal aggregation, and classifier validation must be documented before inspecting confirmatory outcomes. Subject-level train–test separation matters when repeated windows are recorded from the same participant. External validation in a different laboratory or headset can establish whether a model learned a general pattern rather than idiosyncratic instrumentation.
In mixed-reality and social-agent research, sensor traces can contain identifiable movement patterns, voice, gaze, and inferred emotional states. Their collection therefore requires specific consent, retention limits, access controls, and consideration of how apparently anonymous behavioral data could be reidentified. Measurement design and ethical governance are part of the same scientific protocol.
Evidence Synthesis and the Limits of Generalization
The evidence base contains strong local demonstrations, quantitative syntheses, and several unsettled construct-level questions. Cummings, J. J., & Bailenson, J. N (2016) offers evidence that technological immersion can increase reported presence across the earlier literature. Ratan, R., Beyea, D., Li, B. J., & Graciano, L (2020) offers quantitative support for avatar-characteristic effects with relatively modest aggregate effect sizes. Saredakis, D., Szpak, A., Birckhead, B., Keage, H. A. D., Ri (2020) shows that discomfort is a consequential and heterogeneous part of the experience. Their populations, manipulations, and outcomes differ, so their numerical summaries cannot be combined into a single verdict about VR's psychological efficacy.
Primary studies establish several useful dissociations. Presence can increase while learning decreases, as illustrated by Makransky, G., Terkildsen, T. S., & Mayer, R. E (2019). Body form can influence spatial scale judgments in Banakou, D., Groten, R., & Slater, M (2013) paradigm. Virtual perspective taking can produce outcome-specific social effects, as in Herrera, F., Bailenson, J., Weisz, E., Ogle, E., & Zaki, J (2018). These findings jointly suggest that the psychologically relevant variable is the particular affordance or mechanism activated by an immersive configuration, together with the outcome being measured.
Generalization faces four especially important constraints. First, many studies recruit students, technologically interested volunteers, or participants comfortable with headsets. Second, head-mounted displays evolve rapidly, changing optical, tracking, ergonomic, and latency characteristics. Third, laboratory experiences are often brief and highly structured, while ordinary media use is prolonged, social, and shaped by personal goals. Fourth, the outcome called “presence” or “embodiment” may have been assessed with incompatible instruments. A robust theory must state which elements of a finding are expected to survive these changes.
Independent replication requires recovering the intended manipulation, not merely reimplementing the narrative of a study. Researchers should deposit stimulus specifications, device settings, avatar dimensions, analysis code, and appropriately deidentified outcome definitions. When software cannot be shared, an auditable description and reproducible behavioral task may still permit close replication. A field-wide evidence synthesis should record risk of bias, sampling, comparator quality, effect size, uncertainty, and publication year before claiming a general pattern.
The present monograph is a critical narrative synthesis. It selects and compares relevant original studies and structured reviews to develop a coherent explanatory account; it does not provide a preregistered exhaustive search, formal risk-of-bias coding of every eligible study, or a new pooled effect estimate. Its scientific conclusions therefore derive from the cited primary experiments and existing systematic syntheses rather than a claim to a newly completed systematic review.
Cybersickness and Sensorimotor Conflict
Cybersickness is a constellation of symptoms, often including nausea, dizziness, disorientation, and visual discomfort, associated with particular immersive configurations. It can affect who completes an experiment, how long they participate, and how they behave. It is also an outcome in its own right because a participant's comfort contributes directly to the usability and accessibility of the technology.
Saredakis, D., Szpak, A., Birckhead, B., Keage, H. A. D., Ri (2020) systematically reviewed factors associated with VR sickness in head-mounted displays and performed meta-analytic assessments. Their evidence supports consideration of exposure duration, control mode, hardware, and participant characteristics. Studies using the Simulator Sickness Questionnaire or related instruments should distinguish nausea, oculomotor symptoms, and disorientation where possible rather than assuming one global symptom mechanism.
Sensory conflict is an influential explanatory family. Visual flow can suggest acceleration or translation that vestibular information does not support. Mismatched head tracking and display latency can disrupt expected correlations between action and sensation. But multiple influences are possible, including postural demands, individual susceptibility, accommodation-related issues, motion characteristics, and attention. A single explanation should not be presumed to capture every sickness condition.
Weech, S., Kenny, S., & Barnett-Cowan, M (2019) reviewed evidence that presence and cybersickness are often negatively related. This relation must be interpreted carefully: discomfort may draw attention away from the environment, while particular technical conditions can independently affect both presence and symptoms. A correlation between the experiences does not identify the causal direction.
Research ethics and statistical integrity converge here. Participants need withdrawal opportunities and clear stopping rules, while investigators need to document adverse effects, early termination, and missingness by condition. Excluding those who become sick can make a system appear more effective than it is for the target population. Sensitivity analyses can address attrition and the possible selection of unusually tolerant users. Technological success should be judged in relation to meaningful outcomes and tolerable exposure.
Individual Differences, Development, and Accessibility
Virtual experiences vary with sensory capacities, prior technology use, motion sensitivity, cultural familiarity, body dimensions, cognitive resources, and personal goals. Some participants enter VR with extensive experience navigating digital environments; others must first learn how to use tracked controllers or how to resolve conflicts between physical and virtual movement. The same instructional or experimental apparatus can therefore provide different psychological tasks to different users.
Developmental research must attend to age-appropriate measures and consent. Children and adolescents may differ in physical fit, task comprehension, susceptibility to social influence, and the way they describe body ownership or presence. Older adults may differ in sensory functioning, mobility, and visual comfort, while also benefiting from tailored simulation or accessible interaction. Strong developmental inference requires age-comparable tasks, validated measures, and appropriate controls for familiarity and motor performance. Demonstrations from adult student samples cannot simply be transferred to children or clinical populations.
Accessibility is integral to experimental validity. A system requiring unsupported standing, two-handed gestures, or specific visual acuity excludes users for reasons unrelated to the construct under study. Seated locomotion, alternative controls, captioning, adjustable motion, high-contrast cues, and scalable avatars can broaden participation. Yet accommodations sometimes change sensorimotor contingencies and therefore require transparent reporting rather than being treated as invisible technical substitutions.
Individual differences in the tendency to become engaged with media can influence questionnaire reports, but an individual's trait score should not be mistaken for the momentary causal effect of a display. Witmer, B. G., & Singer, M. J (1998) distinguished an immersive-tendencies instrument from reported presence. Modern research benefits from separating baseline familiarity or disposition, experienced state, objective task performance, and adverse effects.
Cultural analysis also matters in socially significant embodiments. An avatar's appearance and the social meaning of its features depend on the local context. Role expectations and interpretations of personal space differ across settings. The correct research response is to specify the population and social meanings involved, examine moderation and measurement invariance, and replicate in independent contexts before generalizing behavioral outcomes.
Ethics, Data Governance, and Experimental Responsibility
Immersive media can place participants in compelling situations involving fear, authority, social humiliation, stigma, exclusion, or apparent physical risk. The intensity and controllability of the scene create distinctive responsibilities for informed consent, monitoring, debriefing, and withdrawal. A well-designed protocol describes the nature of virtual events and possible discomfort without destroying every experimental manipulation; independent ethics review and scientifically appropriate deception procedures may be necessary for particular studies.
Embodiment experiments can carry moral and social meanings beyond a neutral change in avatar geometry. Participants may be placed in bodies that represent historically stigmatized groups, childhood, disability, or other socially consequential identities. An ethical protocol evaluates potential harm, avoids reinforcing stereotypes, and preserves the participant's dignity during debriefing. It also distinguishes observed changes in a laboratory attitude measure from claims about understanding another community's lived reality.
The data infrastructure introduces additional risks. Immersive systems can collect fine-grained head and hand movements, gaze, voice, interactions, and room geometry. Such records may be individually distinctive and can reveal behavior beyond the stated primary outcome. Data minimization is a scientific design decision: a study testing body ownership may not need indefinite retention of raw audio or room scans. Collection should be tied to hypotheses and justified by the analysis plan.
When commercial headsets, cloud rendering, or external agent services are used, investigators must document which data leave the laboratory, which providers process them, and how participant rights are protected. Secure storage, access limitation, deletion schedules, and consent for secondary research require explicit governance. Open-science objectives should be met with deidentified derived datasets or controlled-access procedures when raw sensor trajectories present reidentification risks.
Finally, researchers should report both benefits and harms. A VR training system that improves one performance score but produces significant discomfort, inaccessible interactions, or adverse aftereffects requires evaluation of the full outcome profile. Experimental responsibility includes realistic risk assessment, transparent reporting, and attention to who is excluded from the evidence base.
Immersion in the Artificial Era
In the Artificial Era, immersive interfaces increasingly connect human perception and action to environments populated by adaptive artificial systems. An embodied agent may respond to eye direction, gesture, speech, spatial proximity, and continuing interaction history. The relevant scientific relationship extends beyond rendering a synthetic scene: the environment selects information, infers intentions, changes its behavior, and provides perceptual consequences to the human participant.
This development creates a distinction between experienced agency and actual system agency. A person can feel that a virtual body acts under personal control while some movements are automatically assisted or constrained. An artificial social agent may appear autonomous while its behavior is scripted; another may be genuinely adaptive in limited operational respects while appearing visually simple. Human reports of presence, trust, attachment, or cooperation are therefore measures of the participant's organization of experience. Claims about an artificial system's capacities require independent technical and behavioral evaluation.
Embodied human–artificial interaction offers experimental opportunities. Researchers can vary whether an assistant has a humanlike body, a geometric marker, or no visible body; whether it anticipates actions or waits for requests; whether its gestures are aligned with task information; and whether the user can inspect its reasoning or correct its actions. The outcomes include task success, perceived control, cognitive workload, reliance, error recovery, interpersonal distance, and learning. The design supports hypotheses about how embodied cues alter cooperation without presuming that social experience alone determines the underlying ontology of artificial systems.
The philosophical horizon From Homo to Artificial provides a context for examining these transitions in psychological organization. In scientific terms, the immediate research task is operational: specify which capacities are attributed to a human participant, which operations are carried out by an artificial agent, and how the interaction changes the participant's cognition, affect, and behavior. A philosophy of Artificial as an independent historical order can generate research questions, while empirical claims about particular systems remain accountable to observable evidence.
Immersion also facilitates distributed action. A person may perceive a distant environment through sensors, manipulate a virtual replica, and rely on artificial assistance to interpret incomplete information. Spatial presence, bodily ownership, and functional dependence can then diverge. A participant can feel situated in a remote scene while the system provides inaccurate guidance, or can perform accurately without experiencing strong embodiment. This divergence makes trust calibration and inspectable system behavior crucial outcomes for future immersive psychology.
Open Scientific Problems and Research Priorities
The field's first unresolved task is conceptual integration. Presence, plausibility, body ownership, agency, self-location, and social experience share relationships, yet their boundaries are often obscured by broad subjective scales. A stronger theoretical program specifies the minimum distinguishable constructs and tests whether changes in each have separable behavioral consequences. Designs that independently manipulate viewpoint, sensory synchrony, social contingency, and source expectations can establish the conditions for meaningful distinctions.
The second task concerns longitudinal durability. Most controlled embodiment paradigms examine immediate experience or behavior during a short session. Researchers need delayed outcomes, repeated exposure designs, and opportunities for real-world behavior. The questions include whether avatar-induced effects habituate, whether perceptual recalibration persists, whether social attitudes change beyond an instrument, and whether immersive learning transfers to unaided physical performance. Each requires a distinct endpoint and plausible causal comparison.
A third priority is cross-technology replication. Claims should be tested across headsets, mixed-reality displays, motion-tracking systems, software engines, and control methods. A consistent effect across implementations would strengthen the case for a general perceptual mechanism; an effect limited to one device or interaction protocol would refine the boundary of a theory. Technical metadata and reproducible stimulus descriptions should become standard parts of scientific reporting.
A fourth priority is measurement reconstruction. The Kukshinov, E., Tu, J., Szita, K., Nathan, K. S., & Nacke, L (2026) identifies concrete weaknesses in how questionnaires have been modified and reported. Future work should validate theory-specific scales across independent samples, use multimethod convergence, and test whether responses can be meaningfully compared between VR, AR, and conventional media. The aim is stronger construct validity, not merely shorter questionnaires or more convenient indices.
Fifth, inclusivity and safety require representative designs. Systematically measuring susceptibility, physical accommodations, developmental differences, and dropout will help determine for whom an immersive intervention is usable. The same attention is needed for adverse emotional effects and data governance. Accessibility and privacy should enter the explanatory model as variables influencing participation, exposure, and interpretation.
Finally, adaptive artificial agents make experimental environments dynamic in new ways. Future work must preserve causal interpretability when the environment changes in response to the participant. Versioned policies, logged agent decisions, controlled counterfactual conditions, and independent evaluations of system behavior will allow the psychological science of immersive media to develop alongside the technologies it studies. The central research question remains precise: how do controlled arrangements of perception, action, representation, and feedback produce particular experiences and behaviors, under which conditions, and with what enduring consequences?
Conclusion
Immersive media create an experimental arena in which the psychological relation between body, space, and action can be systematically reconfigured. The central finding of this monograph is that virtual presence and embodiment emerge from identifiable combinations of sensorimotor consistency, spatial perspective, bodily representation, event plausibility, and task engagement. Immersion is an enabling technical architecture; the psychological processes it recruits are distinct and demand independent definitions, manipulations, and measures.
The historical research tradition has established powerful experimental demonstrations. Multisensory synchrony can alter ownership of an artificial or virtual body. First-person tracking can support an experienced location in a represented environment. Avatar appearance can influence selected behaviors and judgments, with meta-analytic evidence indicating effects that are meaningful but generally modest. Virtual scenes can elicit emotional and social responses, yet the strength of those responses varies with appraisal, perceived contingency, and participant characteristics. Spatial judgments can be biased even while an environment is experienced as present, and greater subjective presence can coexist with poorer conceptual learning.
The synthesis leads to a structured explanatory position. Presence concerns the organization of experienced place; embodiment concerns ownership, agency, and location of the represented body; social response concerns interpreting and acting toward other agents; learning and behavior change concern outcomes that must be evaluated separately. Their relations are empirical questions. The strongest research designs isolate causal components, triangulate measurement, assess adverse effects, and establish the durability and external validity of the intended result. Slater, M (2018) revisited the conceptual relationship between immersion and presence, while Wilson, C. J., & Soranzo, A (2015) illustrated how VR can serve precisely specified perceptual research. These contributions reinforce the principle that the explanation must be built around the phenomenon being measured.
The Artificial Era adds an expanding class of adaptive environments and embodied artificial interlocutors. The resulting scientific questions concern how people experience, trust, control, and cooperate with systems that mediate perception and action. Studying those questions requires separate evaluation of subjective experience and artificial operational capacities, joined by experimentally testable relations. The lasting contribution of immersive psychological research is its ability to show how bodily and spatial experiences depend on structured relations among information, movement, expectation, and response. A mature science of virtual embodiment will determine when such arrangements produce reliable immediate experience, useful behavior, safe interaction, and durable transformation.
References
- Bailenson, J. N., DeVeaux, C., Han, E., Markowitz, D. M., Santoso, M., & Wang, P. (2025). Five canonical findings from 30 years of psychological experimentation in virtual reality. Nature Human Behaviour, 9, 1328–1338. https://doi.org/10.1038/s41562-025-02216-3
- Banakou, D., Groten, R., & Slater, M. (2013). Illusory ownership of a virtual child body causes overestimation of object sizes and implicit attitude changes. Proceedings of the National Academy of Sciences, 110(31), 12846–12851. https://doi.org/10.1073/pnas.1306779110
- Bohil, C. J., Alicea, B., & Biocca, F. A. (2011). Virtual reality in neuroscience research and therapy. Nature Reviews Neuroscience, 12, 752–762. https://doi.org/10.1038/nrn3122
- Botvinick, M., & Cohen, J. (1998). Rubber hands ‘feel’ touch that eyes see. Nature, 391, 756. https://doi.org/10.1038/35784
- Cummings, J. J., & Bailenson, J. N. (2016). How immersive is enough? A meta-analysis of the effect of immersive technology on user presence. Media Psychology, 19(2), 272–309. https://doi.org/10.1080/15213269.2015.1015740
- Diemer, J., Alpers, G. W., Peperkorn, H. M., Shiban, Y., & Mühlberger, A. (2015). The impact of perception and presence on emotional reactions: A review of research in virtual reality. Frontiers in Psychology, 6, Article 26. https://doi.org/10.3389/fpsyg.2015.00026
- Gonzalez-Franco, M., & Peck, T. C. (2018). Avatar embodiment. Towards a standardized questionnaire. Frontiers in Robotics and AI, 5, Article 74. https://doi.org/10.3389/frobt.2018.00074
- Herrera, F., Bailenson, J., Weisz, E., Ogle, E., & Zaki, J. (2018). Building long-term empathy: A large-scale comparison of traditional and virtual reality perspective-taking. PLOS ONE, 13(10), e0204494. https://doi.org/10.1371/journal.pone.0204494
- Kilteni, K., Groten, R., & Slater, M. (2012). The sense of embodiment in virtual reality. Presence: Teleoperators and Virtual Environments, 21(4), 373–387. https://doi.org/10.1162/pres_a_00124
- Kokkinara, E., & Slater, M. (2014). Measuring the effects through time of the influence of visuomotor and visuotactile synchronous stimulation on a virtual body ownership illusion. Perception, 43(1), 43–58. https://doi.org/10.1068/p7545
- Kukshinov, E., Tu, J., Szita, K., Nathan, K. S., & Nacke, L. E. (2026). Widespread yet unreliable: A systematic analysis of the use of presence questionnaires. Interacting with Computers, 38(2), 163–178. Advance online publication February 1, 2025. https://doi.org/10.1093/iwc/iwae064
- Lenggenhager, B., Tadi, T., Metzinger, T., & Blanke, O. (2007). Video ergo sum: Manipulating bodily self-consciousness. Science, 317(5841), 1096–1099. https://doi.org/10.1126/science.1143439
- Loomis, J. M., Blascovich, J. J., & Beall, A. C. (1999). Immersive virtual environment technology as a basic research tool in psychology. Behavior Research Methods, Instruments, & Computers, 31(4), 557–564. https://doi.org/10.3758/BF03200735
- Makransky, G., Terkildsen, T. S., & Mayer, R. E. (2019). Adding immersive virtual reality to a science lab simulation causes more presence but less learning. Learning and Instruction, 60, 225–236. https://doi.org/10.1016/j.learninstruc.2017.12.007
- Parong, J., & Mayer, R. E. (2018). Learning science in immersive virtual reality. Journal of Educational Psychology, 110(6), 785–797. https://doi.org/10.1037/edu0000241
- Peck, T. C., Seinfeld, S., Aglioti, S. M., & Slater, M. (2013). Putting yourself in the skin of a black avatar reduces implicit racial bias. Consciousness and Cognition, 22(3), 779–787. https://doi.org/10.1016/j.concog.2013.04.016
- Ratan, R., Beyea, D., Li, B. J., & Graciano, L. (2020). Avatar characteristics induce users’ behavioral conformity with small-to-medium effect sizes: A meta-analysis of the Proteus effect. Media Psychology, 23(5), 651–675. https://doi.org/10.1080/15213269.2019.1623698
- Renner, R. S., Velichkovsky, B. M., & Helmert, J. R. (2013). The perception of egocentric distances in virtual environments—A review. ACM Computing Surveys, 46(2), Article 23. https://doi.org/10.1145/2543581.2543590
- Sanchez-Vives, M. V., & Slater, M. (2005). From presence to consciousness through virtual reality. Nature Reviews Neuroscience, 6, 332–339. https://doi.org/10.1038/nrn1651
- Saredakis, D., Szpak, A., Birckhead, B., Keage, H. A. D., Rizzo, A., & Loetscher, T. (2020). Factors associated with virtual reality sickness in head-mounted displays: A systematic review and meta-analysis. Frontiers in Human Neuroscience, 14, Article 96. https://doi.org/10.3389/fnhum.2020.00096
- Slater, M. (2009). Place illusion and plausibility can lead to realistic behaviour in immersive virtual environments. Philosophical Transactions of the Royal Society B: Biological Sciences, 364(1535), 3549–3557. https://doi.org/10.1098/rstb.2009.0138
- Slater, M. (2018). Immersion and the illusion of presence in virtual reality. British Journal of Psychology, 109(3), 431–433. https://doi.org/10.1111/bjop.12305
- Slater, M., Antley, A., Davison, A., Swapp, D., Guger, C., Barker, C., Pistrang, N., & Sanchez-Vives, M. V. (2006). A virtual reprise of the Stanley Milgram obedience experiments. PLOS ONE, 1(1), e39. https://doi.org/10.1371/journal.pone.0000039
- Slater, M., Spanlang, B., Sanchez-Vives, M. V., & Blanke, O. (2010). First person experience of body transfer in virtual reality. PLOS ONE, 5(5), e10564. https://doi.org/10.1371/journal.pone.0010564
- Slater, M., & Wilbur, S. (1997). A framework for immersive virtual environments (FIVE): Speculations on the role of presence in virtual environments. Presence: Teleoperators and Virtual Environments, 6(6), 603–616. https://doi.org/10.1162/pres.1997.6.6.603
- Steuer, J. (1992). Defining virtual reality: Dimensions determining telepresence. Journal of Communication, 42(4), 73–93. https://doi.org/10.1111/j.1460-2466.1992.tb00812.x
- Weech, S., Kenny, S., & Barnett-Cowan, M. (2019). Presence and cybersickness in virtual reality are negatively related: A review. Frontiers in Psychology, 10, Article 158. https://doi.org/10.3389/fpsyg.2019.00158
- Wilson, C. J., & Soranzo, A. (2015). The use of virtual reality in psychology: A case study in visual perception. Computational and Mathematical Methods in Medicine, 2015, Article 151702. https://doi.org/10.1155/2015/151702
- Witmer, B. G., & Singer, M. J. (1998). Measuring presence in virtual environments: A presence questionnaire. Presence: Teleoperators and Virtual Environments, 7(3), 225–240. https://doi.org/10.1162/105474698565686
- Yee, N., & Bailenson, J. N. (2007). The Proteus effect: The effect of transformed self-representation on behavior. Human Communication Research, 33(3), 271–290. https://doi.org/10.1111/j.1468-2958.2007.00299.x
Social Presence and Responses to Virtual Others
Immersive environments often place the participant among other represented agents. Those agents may be controlled by other people, scripted software, or adaptive artificial systems. Social presence describes the degree to which an interaction partner is experienced as socially available within the mediated setting. It depends on responsive turn-taking, spatial orientation, timing, gaze behavior, interpersonal distance, role expectations, and the perceived contingency of the other's actions. A lifelike visual model without reciprocal responsiveness may be less socially convincing than a simpler representation with coherent interaction.
The distinction between an avatar and an agent matters. An avatar communicates another person's actions through a mediated representation; an agent may generate actions autonomously. The same visual form can be presented in either role, and participants' knowledge about agency can itself influence trust, disclosure, evaluation, and social attribution. Experimental designs should manipulate actual control separately from descriptions of control, because anthropomorphic appearance and beliefs about the operator can have different behavioral consequences.
Social VR supports controlled examinations of spacing, approach, imitation, conformity, and responses to perceived evaluation. The Slater, M., Antley, A., Davison, A., Swapp, D., Guger, C., B (2006) illustrated the possibility of observing responses to socially consequential simulated events. The study's ethical and technical configuration is historically significant, but the broader causal hypothesis requires more than observing distress: experimenter authority, plausibility of the scripted situation, and participant awareness must all be considered.
Social presence also varies with conversational latency and synchrony. Delays can disrupt mutual coordination even when visual rendering remains high quality. Eye-contact cues, body orientation, facial animation, and apparent shared attention may support different levels of interpretation depending on cultural expectations and the social relationship. The relevant outcomes include interaction quality, task cooperation, trust calibration, and comfort rather than a single impression of realism.
The research implication for an Artificial Era is direct. Embodied artificial agents can enter human perceptual and social environments in forms that promote immediate responsiveness. Scientific inquiry must therefore examine the participant's social experience, the artificial system's actual capacities, and the reliability of its behavior as separate variables. A compelling interface can be experienced as socially present while remaining unreliable in practical judgment, and reliable task assistance may be delivered by a visually minimal agent.