Doppelganger research paper

Beta Pictoris b's birth seems to have been run-of-the-mill for a gas giant. This planet formed near a star 10 times brighter than our sun at about the distance of Saturn from our sun, and it formed by grabbing gas from its star's disk and clumping that material around a rocky heart.

But 2MASS c's birth played out very differently. It orbits two small, faint brown dwarfs at a distance about 2, times that from the Earth to the sun. Those brown dwarfs, unlike beta Pictoris, weren't surrounded by much gas or dust, so their new planet couldn't have formed by vacuuming up the stellar disk. Instead, it "looks like an underweight brown dwarf that formed from the collapse of a gas cloud," University of Arizona astronomer and study co-author Kaitlin Kratter said in the statement.

The research is described in a new paper accepted for publication in The Astronomical Journal and posted to the preprint website arXiv. The team said it hopes that 2MASS c's huge distance from its faint host stars could make the planet easier to study, because it won't be blotted out by the enormous amount of light coming from a nearby star. Email Meghan Bartels at mbartels space. Moreover, the recent work of Waltemate et al. This study relies on the virtual mirror metaphor and first-person perspective, the latter being known to induce a high sense of embodiment Slater et al.

In this context, our study aims to pursue the investigations concerning the impact of truthfulness on the sense of embodiment and users' behavior in a more complex scenario relying on third-person perspective Debarba et al. This effect is based on two main theoretical foundations:. The implications of the Proteus effect in classic real-time 3D environments or immersive environments are multiple. It was initially demonstrated that avatar size and attractiveness can influence users' behavior and social interactions in immersive applications Yee and Bailenson, Many studies on virtual body ownership highlighted the impact of virtual characters' properties such as skin color Kilteni et al.

Recent work aslo highlights the influence of avatars' appearance on the emergence of common fears in immersive environments such as acrophobia Lugrin et al. Here, anthropomorphism influences the feeling of security when users are exposed to high elevation. While virtual character appearance can impact users' behavior, it has also been shown that the embodiment of characters with strong cultural connotations can have a cognitive impact.

Thus, the work of Osimo et al. In the same context, the research of Banakou et al. The literature review proposed by Hudson and Hurter also suggests that, in a learning context, avatar appearance could impact learners' attitudes and motivation. In these different use cases, deindividuation can be used as a trigger increasing the effects of self-perception theory and its influence on users' behavior. However, in the context of our experiment, using doppelgangers will not lead to deindividuation due to the lack of anonymity.

Our literature review illustrates the plurality of factors that can influence the sense of embodiment as well as the ability of avatars to affect users' behavior. We propose to pursue investigations in this direction in order to identify the impact of avatar visual fidelity, especially truthfulness, on the sense of embodiment and its effect on users' behavior in immersive virtual environment. In order to investigate the impact of avatar visual fidelity, the participants are immersed and control three virtual characters Figure 2 via a motion capture system.

Each avatar presents a different truthfulness level similarity between the avatars and the participants, Figure 1 increasing from a robot to virtual doppelgangers modeled via a 3D reconstruction process. However, each character presents a constant anthropomorphism level humanoid body schema and a realism level consistent with the virtual environment Figure 2.

Figure 2. Screenshots of the three avatars presented to the participants in the virtual environment: A robot, B suit, C doppelganger. Written informed consent was obtained from the depicted individual for the publication of this image. The scenario of the application developed for the experiment consists of an appropriation phase and five tasks Figure 5 described in the next sections. Each task is designed to potentially affect the avatars' integrity, making it possible to verify the occurrence of the Proteus effect through the analysis of the participants' behavior and reactions.

While many studies on embodiment in virtual environments are based on first-person perspective Figure 3A and real-time mirrors, the third-person perspective Figure 3B used in this experiment allows to perceive the whole embodied avatar while opening new use cases for virtual reality.

Furthermore, previous work demonstrated that the sense of embodiment could be induced in third person perspective under synchronous visuomotor conditions Debarba et al. Figure 3. The sense of embodiment is subjectively assessed through a questionnaire and post-experiment interviews. The impact of avatar appearance on users' behavior is evaluated using both objective and subjective approach based on the collected data, the post-experiment questionnaire and the interviews.

Paper statistics

The content of the application was developed using the real-time 3D engine Unity. The choice of the graphic universe and the scenario legitimates and makes plausible the presence of the three avatars and their respective level of truthfulness Figure 2 :. The artistic direction ensures the plausibility of the robot as well as the suit, because they appear coherent in this science fiction universe, which would not be the case in a contemporary environment.

Concerning the doppelgangers, we consider that the current popular culture video games, films, novels, etc. This environment consists of a training room dedicated to the appropriation phase and a main arena constituted of modular hexagonal platforms enabling the place to be dynamically configured for each task. The lighting system is designed to maximize the affordance and allows the user to quickly distinguish neutral elements from hostile ones. To prevent revulsion responses induced by this kind of environment, participants were asked to rate their level of affinity with science fiction thanks to a scale ranging from one to five in the pre-experiment questionnaire.

If it could be argued that such a rich environment could affect our results, it was nevertheless designed to maximize users' involvement in the different tasks of our experiment, which seems necessary to study user experience and behavior. We implemented the third-person perspective based on previous work Gorisse et al.

Digital Doppelgängers

The virtual camera is placed three meters behind the avatar Figure 3B. Elements intersecting the field of view become transparent to avoid occlusion problems. The rotation of the camera is centered around the character and proportional to the rotation the participants' head. The full body tracking system enables the participants to navigate freely and naturally inside the virtual environment thanks to the real time synchronization of their whole body movements.

While it requires a period of time to get accustomed to this new way of perceiving the controlled body, every participant mastered this interaction process in few minutes during the dedicated phase of the experiment, as described in the subsequent presentation of the protocol section 3. We also use two Vive controllers and three Vive Trackers to track the body movements using inverse kinematics algorithms to animate the avatars in real-time Figure 4.

Figure 4. We measured the overall latency of the system using the manual frame counting method described in He et al. Thanks to a high-speed camera Hz , we recorded simultaneously five hand-clapping motions by removing the lenses of the HTC Vive to be able to compare the delay between the real movement and its virtual counterpart displayed on the headset's screen. Analyzing the motions frame by frame, we measured an average shift of 3 frames Thus, our system benefits from these low latency devices providing an overall delay under the threshold of 20 ms recommended for virtual reality Raaen and Kjellmo, As we experienced some issues concerning 3D reconstruction of long hairs, we are currently working on a complementary study to investigate the gender impact on the sense embodiment in virtual environments.

Each subject has a correct or corrected vision thanks to glasses. All the participants have used an immersive virtual reality system at least once and are experienced with video games involving the control of avatars role playing games, first and third-person shooter, platformer, etc.

doppelganger research paper

Two reasons led to the recruitment of experienced participants. First, it limits the duration of the appropriation process, this way the participants stay focused on the content of the experiment and not on the novelty of the devices. Second, virtual reality is currently reaching the mass market and we suppose that, in few years, most of the population will have access to this technology making our results more generalizable. At first, the subjects participated in a 3D scan session. Each 3D model was then optimized to be integrated in the application dedicated to the experiment Figure 1.

They were invited, a few days later, to participate in the experiment. After completing a consent form, they filled the pre-experiment questionnaire to collect their demographic information as well as their experience of video-game and virtual reality.

Then, they were geared up with the equipment required for the experiment and received the starting instructions. According to our within-subject design, the participants perform the scenario three times to control each avatar in a counterbalanced order.

The horror of a doppelganger in documentary film

Between each session of the experiment, the participants complete a questionnaire. After passing the three conditions, a semi-directive interview is conduced to discuss about their subjective feeling and their experience with the different avatars. The goal of the five tasks consist in activating a terminal always appearing on the opposite side of the virtual arena. Once triggered, the arena starts its configuration process to dynamically adapt the environment for the next task. Each task is designed to potentially affect the avatars' integrity.

This punitive model encourages participants to perform the best of their possibility and allows to observe different behaviors and strategies. The experiment begins with the appropriation phase enabling the participants to calibrate their avatar. First, they can observe and turn around the avatar freely before they enter inside the virtual body to enable the calibration process in order to set the avatar's size and to activate the full body tracking.

When calibrated, a drone enters the room and asks the user to reach its position five times. It should be noted that most of the instructions are explicated by the drone within the virtual environment to provide the same information for all the participants. During this appropriation phase, the participants are allowed to navigate within the first room until they feel at ease with the third-person perspective and their virtual body. Thus, when asked for cyber-sickness during the post experiment interview, every subject reported being comfortable thanks to this habituation process.

After this first phase, they reach the central elevator to enter the arena to pursue the five tasks of the experiment Figure 5. Figure 5. Virtual arena during the tasks of the experiment: A appropriation phase, B task 1: bridge, C task 2: walls and ledges, D task 3: traps, E task 4: spheres, F task 5: laser. Task 1 - bridge: The first task simply consists in crossing a bridge to activate a terminal enabling the configuration of the platforms for the second task. Task 2 - walls and ledges: The second task requires the users to walk on ledges between walls and the void while being careful not to fall down.

Task 3 - traps: The third task is composed of a path with trapped platforms. Each platform can trigger the next section of the path or suddenly remove one of the next elements. The participants have to walk slowly to avoid being trapped by the moving platforms. Task 4 - spheres: The fourth task requires the subject to cross the arena while dodging or blocking spheres shot by two turrets disposed on their left and right sides.

The projectiles can only be blocked with the avatars' arms.

Short story on a confrontation with my own doppelganger

If the participant is collided by two spheres, the avatar is switched in ragdoll mode and respawns after a fade out on a previous safe platform. Task 5 - laser: During the last phase, the users face a rotating laser. At the beginning of this phase, they are protected by a pillar. They have to move or crouch to dodge the laser in order to reach the next pillars and the opposite side of the arena.

If the participant is hit by the laser, the avatar is switched in ragdoll mode and begins the respawn process. We based our measures on subjective qualitative and quantitative and objective data to collect and cross information concerning the sense of embodiment and users' behavior. After each of the three sessions of the experiment one by avatar condition , a questionnaire is filled by the participants according to the procedure presented in the previous subsection.

It consists of five-point semantic scale items divided into two dimensions Table 1 :. These items are based on previous work Debarba et al. Table 1. This file contains the following information:. It corresponds to the duration between the activation of the traveled section and the activation of the next one. The experiment process is concluded with a semi-structured interview allowing the participants to express themselves on their experience.

The questions address the sense of embodiment, as well as attitudes and strategies of the participants toward the situations that could potentially threaten their avatar. In other words, it helps to assess the occurrence of behavioral reactions induced by the avatars' appearance.

The transcript of these interviews makes it possible to carry out a semantic analysis to identify recurring remarks and to gather some relevant statements in the frame of our research. H1 : Avatar truthfulness impacts positively the sense of embodiment, especially ownership, in immersive virtual environments.

Original Research vs Research Articles

H2 : Avatar truthfulness affects users' behavior and reactions toward situations that could potentially threaten their avatar. The Shapiro-Wilk Test was carried out to check the normality of the distributions of the answers to the post experiment questionnaire. Bonferroni's correction was used to adjust alpha values for post-hoc pairwise comparisons. The results of the Friedman Test indicated that there was no significant difference in terms of self-location nor in terms of agency between the three avatar conditions.

Nevertheless, agency mean scores are very high Table 2 for each condition, confirming that visuomotor synchrony is a critical factor contributing to the sense of embodiment in immersive virtual environments. Table 2. Inspection of the median values showed an increase in ownership from the robot condition to the suit one and a further increase for the doppelganger for the four ownership items. Post-hoc Wilcoxon Signed Rank Tests did not reveal significant differences in the ownership scores between the robot and the suit conditions due to Bonferroni adjusted alpha value used to control type 1 error.

Taken together our results concerning embodiment in immersive virtual environments highlight an impact of avatar visual fidelity, especially truthfulness, on the sense of ownership Figure 6 , providing evidences of the validity of our first hypothesis. Indeed, our results revealed significant differences especially between the two extreme conditions robot vs.

Figure 6. Boxplot of the averages of the SL, A, and O factors of the embodiment dimension. However, post-hoc tests with adjusted alpha value did not indicate significant differences in pairwise comparisons. It should be noted that the dispersion of the scores concerning the Proteus Effect is very important Figure 7.

It suggests that each participant could experience the situations differently when their avatar can be threaten. We investigated further these issues during the interviews where participants provided more insights. Of those participants, 12 felt more secure using the robot due to its mechanical appearance increasing disconnection from reality.

Furthermore, 12 of those 17 subjects also underline the fact that controlling their doppelganger induced more realistic reactions in order to preserve their integrity. These results seem to demonstrate that the subjective experience lived by the participants was affected by avatar visual fidelity, but looking at objective results we cannot confirm that it affected users' performance. We discuss further the reported subjective information and their implications concerning participants' behavior in the next section.

Previous work demonstrated that multi-sensory integration like congruent visuomotor synchrony and visuotactile stimulations operate as critical bottom-up factors allowing users to embody different avatars in immersive virtual environments Kokkinara and Slater, Our study involving a full-body tracking system acknowledges that some factors of the embodiment process like agency are very high under these conditions.

On the other hand, the main focus of our experiment concerns the impact of the top-down factor that is visual fidelity. We used three avatars presenting different levels of truthfulness to evaluate the impact of similarity between users and avatars on the sense of embodiment. Our results confirm our first hypothesis and demonstrate that truthfulness positively impacts the sense of embodiment in virtual environments and especially the ownership factor. Based on complex stimuli in an immersive gaming scenario, our study corroborates and contributes to extend previous findings observed with first-person perspective Waltemate et al.

Participants of the experiment reported being more connected with their own representation and appreciated having a personalized character:.

  • importance of literature review in research report.
  • Paper statistics;
  • 50 essays a portable anthology questions answers.

You really have the impression to control your character, but it's you. It's the personalized experience […]. I was more in the skin of my character, the impact was stronger. However, as expected, some subjects noticed minor issues on their 3D model due to current limitations of scanning technology.

According to the Uncanny Valley theory, these inconsistencies could induce revulsion responses of the subjects toward their self-representation. Indeed, due to their high level of realism, participants easily noticed the mismatches between their real face and their virtual one:. While this side effect has to be considered before developing virtual reality applications involving avatars based on users' appearance, our results contribute to extend the knowledge concerning the impact of avatar truthfulness.

Moreover, they corroborate previous findings demonstrating that avatar visual fidelity can improve the sense of embodiment in immersive virtual environments Waltemate et al. Furthermore, Argelaguet et al. Considering these results together, it is now possible to confirm that the three criteria of avatar visual fidelity Garau, anthropomorphism, realism and truthfulness affect the sense of embodiment in virtual reality applications based on both first- and third-person perspectives.

Our results corroborate partially these findings.

Check out how works

Indeed, subjective data collected through the post-experiment questionnaire and the interviews demonstrate that user experience is affected by avatars' appearance as in previous study Lucas et al. Nonetheless, half of the participants consider that their avatar influenced their reactions and behavior during the different tasks of the experiment. Many participants reported feeling more secure using the robot thanks to its mechanical constitution:.

I found I walked faster on the platforms, I died fewer times. Yeah, I was more confident. I was more in a situation where I had to risk my life. With the robot I rather tried to block the laser, push the balls away. I wouldn't necessarily have tried it with my own character, […], you only have one life. That's what I like when I do this kind of experience and it's like there's interference from the physical world in the virtual world and it bothers me a bit. In the light of these insights, using abstract representations can lead to increase disconnection from reality and allows the participants to act with few restraints in the virtual world.

On the other hand, the doppelganger condition seemed to induce more realistic behaviors and a kind of preservation instinct. As expected, the participants were less deindividuated, which is one of the requirements to impact users behavior as demonstrated in previous experiments assessing the Proteus Effect Kilteni et al.

They reported being more careful in order to ensure the integrity of their avatar:. You're really more serious, you have to be careful. With the robot it might have bothered me a little less. There was less the feeling of life. The participants' comments relate the impact avatars' appearance on their behavior. Although these subjective feedback are only partially in line with our objective data, we tried to assess the impact of avatar visual fidelity on user experience and not for performance purpose. That's why in our context we consider that users' feeling takes precedence over objective observations.

Taken together, our results concerning the impact of truthfulness on users' behavior revealed that each type of avatar could be used to produce specific and expected reactions. Doppelgangers seem to induce more serious attitudes and encourage the participants to preserve the integrity of their avatar, self-representations remaining an anchor in the real world.

On the other side, more abstract representations allow users to be completely disconnected from reality and to live new experiences. One of the main limitations of our experiment concerns the 3D reconstruction of participants with long hair and relatively thin strands, thus limiting the gender mix of our panel. It should be noted that recent work aims to reproduce complex and optimized hairs for real-time 3D rendering Hu et al.

In addition, we are currently working on research investigating the impact of gender on the sense of embodiment in immersive virtual environments. Another limitation of our study concerns the virtual environment we developed for this experiment. Indeed, while the science fiction universe ensures the coherence between the avatars and the environment, it can affect users' experience and therefore the outcomes of the study.

To limit this phenomenon, we asked the participants to rate their level of affinity with this kind of universe.

source link The high overall evaluations limit the revulsion responses toward our experiment, but it can also potentially strengthen the measured factors. The study reported here investigates the impact of avatar visual fidelity on the sense of embodiment and users' behavior in virtual reality applications relying on a third-person perspective. According to our literature review, avatar visual fidelity could be divided into three factors: anthropomorphism, realism and truthfulness. Our experiment especially focuses on truthfulness, that is to say the similarity between users and avatars.

Our results demonstrated that the sense of embodiment is higher for the participants controlling their self-representation than with more abstract representations like the robot or the suit. As expected, visual fidelity does not seem to impact both self-location and agency. However, our statistical analysis revealed significant differences in terms of ownership between the robot and the doppelganger in favor of the latter. Furthermore, strong trends between the suit and the other conditions were observed and could probably be confirmed with an extended panel.

It should be noted that if doppelganger could be used as an embodiment vector, some participants noticed reconstruction flaws on their self-representation reducing their affinity toward their avatar. While it concerns a minority of subjects, it remains necessary to consider current hardware limitations before planning to develop virtual reality applications based on 3D scanning technologies.

Concerning the impact of avatar appearance on users' behavior, a phenomenon known as the Proteus Effect, we can distinguish the subjective results based on our questionnaire, the post-experiment interviews and the objective data collected by our application. Indeed, objective measures revealed few significant differences concerning users' behavior between the three avatars' truthfulness conditions. However, when interviewing the participants after the experiment, half of the panel reported having different reactions and strategies depending on the controlled character.

Abstract representations, especially the robot, allow the users to completely disconnect themselves from reality and induce the adoption of more risky behaviors. On the other hand, self-representations maintain a connection with the physical world and encourage users to ensure the integrity of their own avatar.

To conclude, avatar visual fidelity and its truthfulness factor can be a lever to improve the sense of embodiment in immersive virtual environments. However, as it seems to affect users' subjective experience, it must be used wisely to design virtual reality applications inducing expected and specific behaviors. Based on this study and in the light of our results, we planned to pursue the investigation concerning the sense of embodiment and the Proteus Effect in immersive virtual environments. Our experiment revealed specific behavior between our extreme conditions, the robot and the doppelganger.

We would like to cross these conditions to analyze users' reactions when controlling a character presenting a mechanical body, like our robot, but with their own head. Will their behavior be more affected by their mechanical body or will they focus on the preservation of their self-representation?

The panel recruited for the experiment consists of volunteer students in the frame of their educational program. The non-invasive devices of our virtual reality setup are regularly used by the subjects that we have solicited and are accessible to the general public. The protocol excludes the collection of physiological data and we ensured the anonymity of the participants. GG developed the VR application. GG and OC ran the experiment and analyzed the objective and subjective data. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

We thank all the members of our lab for their help in conducting our experiments. Many thanks also to all the people who provided us with insightful comments and advice. Argelaguet, F. Google Scholar. Aymerich-Franch, L. The relationship between virtual self similarity and social anxiety. Bailenson, J. Self-representations in immersive virtual environments. Bainbridge London: Springer , — Banakou, D.

Illusory ownership of a virtual child body causes overestimation of object sizes and implicit attitude changes. Virtually being einstein results in an improvement in cognitive task performance and a decrease in age bias. Benford, S. Embodiments, avatars, clones and agents for multi-user, multi-sensory virtual worlds. Multimedia Syst. Biocca, F. The Cyborg's Dilemma: progressive embodiment in virtual environments. Blanke, O. Full-body illusions and minimal phenomenal selfhood. Trends Cogn. Botvinick, M. Rubber hands' feel'touch that eyes see. Nature Caspar, E.

The relationship between human agency and embodiment. Consciousness Cogn. Cummings, J.