A VRE is composed of a set of qualities that make the experience more or less real. Although the minimal requirements to experience a VRE are not fully defined yet, based on the literature I identify at least four groups of interrelated contributors: (i) immersion, (ii) interaction, (iii) sensorimotor contingencies and, as consequence of these, (iv) illusions.
Immersion
Immersion is most likely the first associated term to VREs. In virtual reality, nowadays the main contribution to technological immersion comes from how visual information is provided. Immersion, however, is not just a question of display type, size and field of view, and it can be enhanced by adding other sensory cues (e.g. auditory, tactile). Indeed, immersion is not specific to electronic displays. For instance, a book can provide a certain level of engagement by which readers mentally visualizing the described world or identifying with the protagonists’ fears or pain can experience certain level of immersion [
13]. However, to elevate the reading experience to the VRE category, other elements like sensorimotor contingencies and illusions are required.
Interaction
The initial feelings generated after exposure to a new immersive environment (virtual or other) may equally vanish if this immersion is not fed by sensorimotor contingencies provided by natural and meaningful interaction with the environment. Natural interaction in VREs starts with visual exploration of the space. For instance, when using a head-mounted display (HMD) with integrated head tracker and turning the head around, participants can explore the virtual world as they would do in the real world. The level of interaction increases every time the system responds to the participants’ action, e.g. a virtual human looks back to the participant when she/he turns the gaze to the avatar. If a virtual representation of their body is included in the experience, when reaching to a virtual object, participants will expect the virtual arm to touch it and they will expect to feel its texture, shape and weight (e.g., with aid of a haptic data glove). When this tactile dimension is provided, the visuomotor interaction is enhanced with a new piece of congruent sensory information.
More elaborate ways of interaction may be provided by means of brain or other bodily inputs, mainly via brain-computer interaction and biofeedback, which enable participants to navigate through a virtual environment [
14] or, for instance, to see their respiration reproduced by the chest movements of a virtual body in a synchronous fashion to enhance self-identification with the virtual body [
15]. In another example of the use of visualization of interoceptive signals, Aspell and colleagues proposed synchronous cardio-visual stimulation, with a virtual body flashing in synchrony with participant’s heartbeats, to increase self-identification with the virtual body [
16].
Sensorimotor contingencies
Sensorimotor contingencies refer to the actions that we carry out to perceive the world [
17]. In VREs, they are often provided by our interaction with the environment, e.g. when leaning forward trying to see what there is hidden behind an object. Sensorimotor contingencies are directly involved in the generation of the sense of agency, i.e. the experience of being the author and initiator of our own actions [
18]. This is particularly important for our body interaction with the world in VREs. For instance, our perception of the virtual world depends on the congruence between our actions and the sensory feedback resulting from them, e.g. we feel the stiffness of the virtual object that our virtual body representation is touching.
The spatio-temporal correlation of the concurrent multisensory stimuli represents a critical factor for the nervous system to interpret the environment [
19]. Arbitrary spatio-temporal discrepancies created artificially in a VRE can also be used to facilitate activation of targeted brain networks to study specific pathological conditions or to potentially speed up the recovery process in rehabilitation settings (see [
20] for a review).
Illusions
Illusion is defined as “an instance of a wrong or misinterpreted perception of a sensory experience” [
21]. Within this context, VREs encompass a series of perceptual illusions related to the space, the environment and the self. The more senses congruently involved (i.e. stimulated) during the interaction, the higher the strength of the illusions generated.
Presence, also called Place Illusion, is the psychological product of technological immersion [
11]. Presence is commonly defined as “the feeling of being there” [
8,
17]. An ideal VRE would provide the ultimate level of immersion, creating an illusion of full physical presence in real or imagined worlds. More recently, Slater has further analyzed the original concept of presence, distinguishing between place illusion and plausibility illusion, i.e. the illusion that what is apparently happening is really happening to you [
17]. This includes the self as an intrinsic element of the experience, as participants stop being mere spectators or controllers (if control over the virtual world and its objects is enabled) to become actors of the new reality. Indeed, immersion provides the boundaries within which place illusion or presence can occur and can be characterized by the sensorimotor contingencies that they support [
17].
VREs can modify, add and substitute actual sensory information from reality for the experience to “feel real” [
22]. VREs can even propose to replace the own body with a virtual representation. The illusion of having (i.e. feeling as if real) a body in a VRE has been tagged as virtual embodiment. The investigation of the mechanisms and the consequences (physical, cognitive, and even legal) of manipulating body perception through bodily illusions is the ultimate goal of a dedicated research field in cognitive neuroscience that started with the “rubber hand illusion” [
23] and that rapidly expanded to virtual body illusions [
24,
25]. For instance, manipulations of the bodily self can be used to examine the brain mechanisms that support body perception and integration in healthy subjects [
26] and in the context of diverse pathologies, ranging from eating disorders [
27] to neuropathic pain in paraplegia [
28].
This sense of embodiment has been suggested to consist of three illusory components: the sense of self-location, the sense of agency, and the sense of body ownership [
29]. Several conceptual models have been proposed to address how our nervous system decides to accept a fake (e.g., rubber or virtual) body as being its own. A combination of Bayesian models characterizing bottom-up and top-down neural processes seems to explain bodily illusions obtained via VREs [
26,
30,
31]. While the illusion of presence can be easily generated by simply exposing the participant to an immersive virtual environment, the illusions related to the own body (our physical link to the outside world) require higher levels of sensorimotor contingencies.
Plausibility, place and virtual embodiment illusions together are probably the key differentiation elements of VREs versus other existing media and experiences. Although virtual embodiment (i.e., having a virtual representation of the own body) is not a sine qua non condition to elicit strong VREs, when a participant looks down when wearing an HMD, they expect to see their (virtual) body. Therefore, the occurrence of embodiment presupposes a certain level of presence, sensorimotor contingencies and plausibility illusion. Indeed, while high levels of presence can be achieved with none or low embodiment level, e.g. using third-person perspective, virtual embodiment provided through first-person perspective enables more accurate interactions in the virtual world [
32].
VRE metrics
Strategies to measure the quality and strength of VREs and their components are currently shifting from subjective scales (usually questionnaires) to more objective behavioral and neurophysiological measures. For instance, event-related brain potentials have been suggested as a non-subjective measure of virtual embodiment [
33]. In that study, brain activity in the motor cortex and readiness potential negativity were observed when an embodied virtual hand was threatened (but not the real hand), which was associated with the intention of the participant to move away the threatened (virtual) hand to avoid harm. Other examples of bodily responses used to measure the illusions generated in VREs include changes in electro-dermal activity (galvanic skin response), pupil dilation or skin temperature (see summary in [
34]). However, all these measures mostly represent indirect measures of the assessed element, as illusions like presence or body illusions are subjective experiences (qualia) arising from multisensory stimulation, which cannot be directly measured [
17].