Virtual Reality (VR) is currently on everyone’s lips for various purposes. Although the hype is currently nearing its productive plateau, there are still uncertainties about the utilization and potential of the technology in research and educational contexts. Together with the two following articles on the subject of VR research, this article provides an overview of the inherent potential as well as remaining challenges in the use of VR as an instrument for research and education. These will be clarified and critically examined on the basis of specific academic projects. Based on this, the article concludes with a comprehensive look on the need for further discussion of ethical issues as the key for VR success in research and education.
The use of digital media is nowadays common in knowledge-based professions. Particularly, this development was driven not only by adventurous scientists, but also by major public funding schemes, for instance regarding IT-based teaching and learning. In the early days of personal computing the term multimedia was dominant, as in so-called computer-based trainings. With the advent of networked systems, web-based trainings and, later on, mobile learning applications became popular. Furthermore, the big gain in performance and drop in prices of hard- and software led to an increased usage of complex infrastructures. Typical examples are large-scale educational platforms (from web-based learning management systems to massive open online courses) as well as sophisticated local installations (like smart rooms). Against this background, this article is dedicated to the field of virtual reality (VR) in higher education institutions.
VR is not a new technology. Early prototypes originate from the late 60s , and several applications in entertainment and arts  or in tele-robotics  appeared in the following decades. Though the hardware installations during these days were comparatively large and heavy and the software was rather slow – and thus the overall experience for the users was rather far from reality – the fundamental concepts of these ancestors are still the same in current VR systems. The idea is to replace the natural environment of the user by an artificial one, resulting in the strong impression of being present somewhere else (presence ). This is mainly achieved via the visual system, i. e. the user is wearing a head-mounted display (HMD) in order to block visual cues from reality and to bring in a three-dimensional representation instead, which is virtually created . As familiar from real-life, rotating the head results in a different perspective within the scene, and moving the body leads to a different viewpoint. This sensory-motor contingency emphasizes the need to track the position of relevant points of the user’s body, which can be realized either by the HMD itself or by a surrounding infrastructure (e. g. body tracking by a Kinect sensor , target tracking in a Cave Automatic Virtual Environment ).
It should be noted that this VR approach is different from so-called augmented reality , where the current environment is still part of the scene and can be perceived by the user, but is enriched by artificial objects. Here, the challenge is to identify and locate environmental objects or structures in order to determine the position, size and orientation of related virtual objects as a visual overlay.
Beyond the visual system, the user’s perception in VR can be deceived by adding artificial cues for the other senses. This may include sound, but also tactile impressions. As an example, handheld controllers or data gloves can help to provide tactile feedback while the user is approaching a virtual object . Even larger movements in the scene can be supported by using an omnidirectional treadmill . This opens up several possibilities for the user to truly interact with the scene (in terms of selection, navigation, and modification) instead of just watching it. In fact, the illusion of plausibility of a VR scenario is closely coupled with the plausibility of interactivity , requiring an virtual environment which immediately and meaningful responds to an action of the user.
Today, the VR market is still driven by the entertainment sector. Besides commercial uses, there are several examples of VR case studies from the academic sector that try to explore the potential of this technology in education and research. However, this exploration requires interdisciplinary efforts. From a pedagogical perspective, the critical, responsible and productive handling of media is crucial, while certain characteristics of digital media (being processible, algorithmic and formalized) as key factors of context and culture are less reflected . On the other hand, technological developments and especially interactional designs are hard to link with pedagogical outcomes, since the latter are hard to measure and thus hard to translate into design guidelines . This is especially true for emerging technologies like VR. Thus, the aim of this article is to identify the potential of VR for education and research by a structured analysis of case studies. This brings together not only aspects from pedagogy and technology, but also reflects on the impact of VR on humans from a sociological perspective. Our mission is to contribute to a critical and systematic reflection on the benefits and drawbacks of VR. On this basis, we intend to derive guidelines for future development and use of such technologies.
The following section provides six VR-based case studies originate from the groups of the authors in cooperation with partners from different fields of application. Afterwards, the presumed potentials and challenges of VR as known from current state of research are assembled and compared against these case studies, leading to a structured view on reasonable expectations and existing limitations of current VR scenarios. This comparison is followed by a critical discussion of the effects of VR on human perception, self-conception and interrelation within different communities. From this we motivate the need for further discussion of ethical issues, which we argue to be the key for successful VR usage in education and research.
2 Case Studies
2.1 VR Teacher Training
Successful classroom management results in lectures where all pupils cooperate well, and the disturbance rate is low. In order to develop the necessary interdisciplinary teachers’ competence, different instruments for university courses have been developed, such as texts, lesson videos, role plays and more. Currently, there is little empirical evidence on how effective student teachers are in their classroom management after being trained with these instruments. VR classroom trainings for aspiring teachers such as the VR Teacher Training  provide an intra-university test environment to perform a preventative, action-oriented training on how to deal with disturbances under highly authentic stress.
The project works on the simulation of virtual microteaching event with several scalable educational disruptions. It focuses on the training of cognitive competences (classroom management and diagnostics) and situational competences (preventative monitoring and adaptive, targeted interventions). Disturbances are most likely to be observed in the example of method and/or situation changes during teaching (e. g. start of lessons, new thematic sequences, group work etc.). There is always an interaction between a lecturer (coach) and the teaching student mediated by the VR technology. The student enters the VR environment and gives a short lesson on his or her subject (e. g. starting a lesson in history). During this time, the coach can modify the behavior of the virtual pupils from the outside and may purposefully monitor and promote the students’ competencies. In interesting interaction situations the program can be paused, so that the coach (or possibly the auditorium following the interaction on a monitor) can provide feedback to the student.
Figure 1 shows the two interfaces of the training tool. Using the student view, the teaching student is immersed in the VR classroom. He acts in the simulation by real movements of the hand controllers or position changes by moving around or teleporting. His actions (e. g. dialogs with the virtual pupils) have no automatic effects. All pupil behaviour is simulated by the coach who monitors the student’s activities and his reactions to disturbances. The coach has an overview of the classroom in the web-based coach view. He is able to select specific virtual pupils by clicking on them and so to and trigger a disturbance via a dropdown menu. This creates an indirect interaction between the coach and the student via the virtual students.
Through the detailed character models, the realistic soundscape and the own visible, first-person avatar, the student experiences a noticeable level of presence . Thus, a better preparation for real-life classroom experiences can be expected, leading to smoother and more successful practical phases in school.
Currently, a systematic evaluation of the environment under realistic conditions with 25 teacher students is carried out as part of an educational science course in the winter semester 2018/19 at the University of Potsdam.
This project considerably benefits from VR technology. First of all, the freedom of design allows the modelling of nearly real-world classrooms, heterogeneous groups of pupils, a variety of disturbances (from eating apples to whipping fellow pupils) and other components of authentic classroom situations. The presence generated by VR allows students to experience authentic teaching situations, in contrast to a video recording with a low level of interactivity and personal engagement. Moreover, privacy aspects will not be a problem in VR, since no real students are recorded. Last but not least, a simulation is associated with significantly lower costs and risks than a placement of teaching students in a real school. Of course, such a tool should not replace real teaching experiences but can only be an aid to the theoretical part of teacher training.
2.2 Hard Disk Change in a VR Server Room
VR learning environments can provide learners with the opportunity to safely apply content from theory in a wide range of simulated practice situations. As an example, the prototype developed in the HardDrive Exchange project  simulates the complex task of hard disk changes in a storage area network (SAN) and is intended to enable students and trainees from computer science disciplines to practically experience theoretical learning content on network-based storage systems.
For the implementation of this training scenario, a pedagogical concept was developed according to the constructivist learning approach. Because of the heterogeneous target group, the learners begin with an introduction into the problem situation. First of all, the basic theoretical knowledge of SAN functionality and redundant arrays of independent disks (RAID) are explained. Afterwards, the learners have to actively identify and exchange a defective hard disk step by step under the guidance of the learning application. This clarifies the central work steps, processes and operating concepts. Afterwards, the application goes into the exploratory phase. During this time, learners can generate arbitrary error constellations and observe the visual and logical responses of the SAN by monitoring its behavior. They can replace other defective disks to better internalize the work steps. In addition, they can experiment in different constellations about the consequences of removing hard disks that are still being accessed, or what happens if a hard disk of the wrong type would be used.
An efficient operating concept for this scenario was implemented using a natural user interface (NUI), which was realized by means of the CORPUS framework  that fuses multiple gesture sensors (e. g. Kinect, LEAP Motion) into a single, consistent body model. Figure 2 shows a screenshot of the application with the learner’s virtual hand in the center. Beside the virtual server rack (shown on the left side) and a shelf with different replacement disks (shown on the right side) the simulation includes a large admin screen (not shown), that is inspired by typical administration software in a data center.
In a typical learning scenario, the learner has to recognize and understand an error displayed on the admin screen. Afterwards, the affected hard disk in the rack causing the error has to be identified. If that makes sense to solve the problem, the learner removes the defective hard disk and replaces it with a matching one from the shelf. This may either result in faultless operation of the SAN or may trigger other problems.
Based on expert interviews and application tests, the suitability of the application for the target group, the achievable added value for teaching, and the operating concept were confirmed.
The main VR benefit of this project is the reduction of risks and costs in comparison to an equivalent training with a real SAN infrastructure. SAN hardware is expensive and can usually not be provided just for training purposes. Moreover, existing SANs are crucial elements in IT infrastructures and should not be used for training sessions. In addition, this case study benefits from the arbitrary replicability of learning actions and their results. Thus, learning processes of different learners become better observable as well as comparable than in non-VR based equivalents.
2.3 Simulated Musical Instruments
The VR simulation of musical instruments for learning purposes is also motivated by a reduction of costs in comparison to physical instruments. As examples, Figure 3 displays two musical instruments in VR.
The piano simulator has been developed at the University of Potsdam as a demonstrator for the CORPUS framework  that fuses multiple gesture sensors. It makes possible full body tracking with the option of a more detailed tracking of specific body parts (e. g. the fingers). For the piano simulation, the fingers have been tracked in detail and the feet just regarding their position.
The theremin simulator has been developed in a student project at the University of Potsdam. Theremins are contact-free electronic musical instruments that produce different sounds depending on the player’s hand position (which affects the surrounding electromagnetic field). Since the purchase of the instrument is very expensive, the use of a VR simulator is of great advantage.
Especially with musical instruments, VR offers another big advantage besides the cost reduction: standardization. Simulated instruments always react in a similar way. Peculiarities of specific real instruments do not occur here. This unifies and potentially helps in music lessons, even if the artistic character of an individual instrument is unquestionably lost.
2.4 Natural Hazards Assessment Using VR
The analysis and assessment of natural hazards requires skills that can only be acquired in the field of geosciences through practical learning units in the field. Above all, this poses a major financial challenge for learners and teachers and is very time-consuming. VR has the potential to simulate similar experiences and can therefore face the obstacles in the teaching of natural hazards.
In the project Natural Hazards Assessment , a range of natural hazard assessment tasks was implemented with a VR application based on 3D models of the region near the Chaitén volcano (Chile). As an example, in one task the north direction should be marked in a free area. The learner is located on a mountain or can move around in space in virtual flight mode. He has to find the correct north direction based on the terrain forms and his knowledge of the region with an interactive direction marker. Another typical task is to estimate the volume and mass of dead wood nests in a river, which is important to assess the likelihood and amount of damages downstream. Figure 4 illustrates the simulation as an overview of one scene as well as one view on the generated 3D model.
A special feature in this project is the automated creation of the 3D models. These were generated from two general sources. Firstly, the project uses images of a small-scale drone flight in 2016, resulting in approximately 400 high-resolution aerial images. Secondly, freely accessible geodata (e. g. Landsat satellite images, OpenStreetMap data) has been utilized. All datasets have been prepared and merged with the free geographic information system QuantumGIS. A special challenge for this process was to map the different scales of these data to the resolution as used in the VR application.
Beside a reduction of risks and costs in comparison to a real-life excursion, this project benefits from the VR advantage of replicability. All learners may learn under the same conditions (e. g. tasks and VR models). Thus, their activities and learning outcomes will be much more comparable than in a real-world learning scenario (e. g. with different weather conditions).
2.5 Embodied Cognition
From the descriptions of VR applications so far it is already clear that they demand a certain adaptability on the part of their users. In particular, VR applications require additional cognitive processes, some of which are not present when interacting directly with the world. In order to understand the cognitive demands of VR it is helpful to consider a hierarchical relationship between different types of cognitive operations.
Our minds are well tuned for un-mediated interactions because they were built on a long history of successful perception and action. On an evolutionary timescale, our environment adapted us to develop cognitive mechanisms that expect certain constraints, such as the vertical pull of gravity, the presence of light sources above our heads, and the impossibility of two solids to occupy the same space at the same time. Within each individual’s lifetime, these “grounded” constraints on cognition support our sensory-motor interactions with the world, allowing for the build-up of idiosyncratic expertise and experiences. These “embodied” constraints encompass both biological and cultural factors, ranging from gender and body size to hand preferences and habitual reading direction. Finally, the current task at hand imposes “situated” mental sets that optimally support cognition, be it to apply established performance patterns or to carefully control new behaviours. An example of the former is eye-hand coordination such that the eyes lead the hand to the target and allow for error minimization during hand approach. An example of situated learning is to establish novel patterns of sensory-motor contingencies that have not previously been experienced, such as detecting a light that appears somewhere after presentation of a digit at the point of fixation.
Figure 5 illustrates the three aspects of cognition.
Virtual reality is a formidable research tool to investigate this hierarchical interplay of grounded, embodied and situated cognition. Starting at the situated level, it becomes possible to dissect established mental sets by introducing novel sensory-motor mappings. This has been done, for example, in a study of the relationship between lateralized hand actions and the cognitive representation of good and bad . In their study, the authors examined a prediction derived from previous work of Casasanto  who had earlier documented that right-handed adults associate the right side of space with good things and the left side of space with bad things. For example, when asked to place “good” or “bad” cartoon figures arbitrarily into boxes located on their left or right sides, right-handed adults prefer to place positively valenced figures into their right-side box in about 70 % of cases, while left-handers have the same preference to their left side. Casasanto  had interpreted this asymmetric mapping of valence onto space as an expression of his participants’ motor fluency, i. e. a memory of the fact that it had previously been easy to manipulate and accomplish things on their dominant or preferred side while it had been less easy on their non-preferred side. Since right-handers are more dextrous with their right hand they think of this as their “good” side.
How could this hypothesis be tested? It is difficult (although not impossible ) to manipulate hand preference because it reflects a combination of genetic bias and long-standing motor habits . Yet, VR enables researchers to easily dissociate motor activities from their spatial effects by mirroring one’s real arm and displaying it as an appendage on the opposite side of a first-person avatar. This is what Bailey et al.  did: Their 155 right-handed students each wore a HMD plus arm trackers that recorded the positions of their two arms in space while they repeatedly pointed at a target block that appeared in random positions in a virtual 3 × 3 grid in front of them. Importantly, participants performed with each arm separately in separate blocks and the visualization of that arm was either switched (active left arm rendered as a right arm) or not switched for separate groups. In a subsequent assessment of the space-valence mapping only the group with the normal mapping showed the asymmetry (good animal to the right box in 51 and to the left box in 25 cases), while the asymmetry was no longer statistically reliable in the group with the switched mapping between their own arm and its visualization (44 vs 34 cases). The group difference was not attributable to differential immersion experience, which the authors also registered per questionnaire; but it is unclear whether the difference was already present prior to the experimental manipulation because no pre-test was performed.
Nevertheless, the study illustrates an important advantage of VR-based research over other methods, namely the excellent manipulability of otherwise hard to manipulate aspects of sensor-motor experience.
2.6 Replicability of Experiments in VR
Another valuable aspect of VR-based research was illustrated by the work of Lachmair et al. , namely the ability of other researchers to produce a close replication of a study that is of particular importance or that may have yielded unexpected results. Replicability of scientific studies is becoming increasingly important in the light of high-profile reports of reproducibility failures in all scientific disciplines .
The Lachmair study was aimed at the surprising finding that comprehending the meaning of a number can lead to predictable changes in the visual sensitivity of the comprehender . Specifically, small numbers (such as 1 or 2) seem to enhance visual sensitivity in the left and lower visual field while larger numbers (such as 8 or 9) apparently enhance visual sensitivity in the right or upper visual field. In the original study, this fact was documented by presenting a number symbol at the centre of a CRT screen and measuring the time it took observers to notice a light that appeared randomly and unpredictable on the screen. Participants quickly pressed a button whenever they detected the light and were faster in this task when a small number was followed by left-side lights or a large number was followed by right-side lights, compared to the opposite combinations. Thus, we seem to have learned to associate “more” with “right”, perhaps as a result of reading habits or finger counting . The extension of this horizontal association to the vertical dimension was predicted by the hierarchical relationship between grounded and situated cognitive processes because we universally associate “more” with “up”.
Lachmair et al.  implemented a version of this task in VR. In their task, centrally presented numbers 1 or 9 were followed by centrally presented letter strings, half of which were words and the other half non-words. Participants’ task was to classify each letter string accordingly. Real words were names of objects associated with upper or lower space, and the authors found indeed faster responses for lower-space words following 1 and upper-space words following 9, compared to the opposite combinations. The authors offer their software and explanations for its use to all interested parties. By wearing HMDs and using identical programs and hardware for stimulus presentation and response recording, replications of experiments which were initially conducted in VR environments can be more diagnostic because several obvious explanations for replication failure can be excluded: differences in the test environment and the experimental set-up are largely removed, and instructions as well as stimulus materials can be held comparable across multiple sensory and motor dimensions.
3 Identified Potentials and Challenges
Let us recapitulate the potentials and challenges of VR technology from our perspective. Figure 6 summarizes the specific VR benefits, related to the described case studies. It is obvious that not every case study makes use of or discloses all the benefits associated with VR. Thus it can be reasoned that the potential lies not in the technology itself but in their targeted application when designing a certain scenario.
Clearly, the most prominent advantage of virtual over non-virtual environments is the flexibility we have in designing the scene. Virtually, “anything goes”  – if it can be visualized it can be virtualized by using VR technology. Methodologically, there are also no limits to the design of VR settings in education and research. Guidance in the implementation of VR scenarios comes from considering their benefits, but maybe societal and ethical implications must also be taken into consideration. As researcher, do we need, for instance, to reflect the impact of the VR scenarios on constructivist learning approaches? The virtual classroom example we described first is likely to become a useful training tool for future teachers: It simulates their immersion in quasi-realistic conflictive interaction situations and enables them to acquire adaptive coping strategies. Thus, VR is able to prepare trainee teachers for unusual social situations without exposing any pupils to possibly maladaptive responses. If desired, the virtual training environment can be further enhanced by adding physiological stress monitoring devices; it can also be used to formally compare problem solving strategies used by differently experienced teachers while holding the challenge constant, thus enabling quasi-experimental studies. This comes along with adaptivity of VR settings – live manipulations by external persons or the software itself. For example, due to the students activities in the VR teaching scenario, the coach is able to manipulate the scenario (pupils behaviour) at runtime.
Related to this benefit of VR from its support function for a scientific approach to training is its potential for replicability of studies across laboratories. As explained above, the replication of important empirical results is desirable but currently hampered by small differences in testing across laboratories. Adopting standardized VR equipment to control sensory inputs and sharing experiment software between labs will be a step towards removing many of these small differences – and thus providing more evidence to research.
From our second example, the practicing of maintenance in a storage server environment, it should be clear that the economic potential for VR-based trainings is huge. Once the trainees are successfully immersed in the training environment, it should be possible to simulate pretty much all workplace-related tasks and challenges without putting the trainee or the actual infrastructure into any danger. This is also a principal advantage of the hazard assessment example we described later in our survey. More generally, such VR-based approaches permit comparing the efficiency of different instructions and training methods. However, little is known so far about possible difficulties in and long-term effects on VR’s acquisition of methodical competencies. What is the relationship between VR-supported learning and practical application in non-VR-based situations? Of course these approaches must be validated by assessing learning outcomes and their transfer to the professional everyday-life equivalent of the training scenarios.
Training of musical skills, our third VR application scenario described above, is a related example and served to highlight the benefits of standardization of trainings. It also makes clear that VR-based skills trainings should, in principle, become more effective than traditional trainings because it will be possible to monitor the trainee’s behaviour at the microscopic level, tracking each finger movement with millisecond precision and millimetre accuracy. This minute recording of performance will enable adaptive trainings that focus on each individual’s strengths and weaknesses.
While these scenarios served mostly as motivators for the broader use of VR, we are of course aware of current limitations for this vision. Some points have been already touched. Aside from the broader availability of VR technology outside of research labs and large companies we acknowledge technological limitations of the current hardware with regard to updating and rendering. These limitations make the resulting scenarios occasionally unstable and might even introduce artificial behaviours on the part of their users, merely to cope with lagging display changes and other technical shortcomings. Finally, there is the issue of deceiving the user into believing to be present in a situation that is not real. Fundamental questions about how people open up the world cognitively and process it socially are thus touched upon. These questions mainly concern the multidisciplinary co-construction of VR environments for research in a methodological-ethical dimension. In this respect, attention must also be drawn to the training of scientists, especially, but not only in the field of computational and computer science. Research activities going back to genomics and nanotechnology are continuously anticipating and addressing ethical, legal, and social implications (ELSI) of their professional work. It remains to be discussed how existing reflective approaches such as the MEESTAR model  apply to development and design processes as well as practical research actions with VR settings, or whether such settings are so complex that they require new ethical orientation tools. It is undisputed that VR technologies have an enormous potential both for research and for the transformation of social order. This technology, which, as the case studies show, is initially elaborated on very limited experimental fields, has an inherent vision of designing the societal future. More research needs to be done to develop this vision and the changes it might bring.
4 Researching Digitalised Life-Worlds: A Sociological Exploration of Reality Through VR Technology
Analytically, the social construction of reality , as well as Schütz and Luckmann’s concept of the “life-world”  and their thoughts on empirical analysis and theoretical description of human-social experience, serve as a starting point to explore the potentials of VR environments for social science research. The mentioned theoretical strands make it possible to re-examine the classical question of the production of the social world as a socially shared system of taken for granted institutions, rules, and ideas from a new perspective. We subjectively produce reality by turning our consciousness towards objects. These objects of experience do all belong to the same reality. We do live in our subjective life-worlds that segments into “finite provinces of meaning“ . That means, humans perceive the world as self-evident or given. Usually, humans do not question the existence and the social nature of their reality. By choosing experience as experienced and interpreted reality as the theoretical background, it is possible to question the accent on reality and the specificity of digitized life worlds. Terms such as immersion, which is used in a variety of disciplines to specifically describe the experience of digitized reality, can thus be used in a research-opening and reflective manner. In other words, the reality of digitized worlds can be operationalized as experiential spaces in order to empirically investigate the specific styles of experience and knowledge. For this purpose, the “structures of everyday life”  can be used heuristically. Already at the end of the 1960s, Schütz and Luckmann investigated the role technology plays to transform basic elements of human social experience, namely spatial, temporal, and social structures that constitute the limits of everyday situations. However, the relationship between technology and society is fundamentally different today under the conditions of VR. VR is not only a medium. The technology itself functions as a digital basis for social experiences and such for the social construction of (virtual) reality. Technologically, the most different boundaries of everyday-life can thus be transcended. An avatar, i. e. a digitized representation, can be “embodied” as desired, e. g. as a human being, dragon, object/object or fantasy being. The same applies to the degrees of freedom and behavioral possibilities assigned to the actors in VR. It is just as possible to fly through the VR space as it is to pull a fuel rod out of the cooling water of a nuclear reactor unharmed or to meet with friends in the virtual cinema hall independent of the location of one’s own body and discuss its content for hours following the film.
All of this has far-reaching implications that can be explained using the already presented case studies focussing on research and education. When it comes to learning, the hard disk change example, as well as the teacher training scenario, are theoretically based on a constructivist learning approach. Learning is understood here as the process of self-organization of knowledge, which takes place on the basis of the construction of reality and meaning of every single learning individual and is, therefore, relatively individual and unpredictable . The diversity of what has been learned thus depends to a large extent on individual experiences gained in the VR setting. With regard to constructivist learning concepts, this means that VR technology, in particular, can be used to create multimodal and communication-oriented environments that address subjective areas of experience and at the same time contain new “puzzles” that invite people to pragmatically, interactively, and creatively orientate themselves. Computers can process data from a variety of data sources and generate highly aggregated information from it. The natural hazards example shows how different data sources can be used to modulate and therefore simulate phenomena such as natural hazards. But what does happen on the way from gathering, processing, and analysing data? Will it be necessary for the future to know the origin of data and how they are collected? How will VR environments contribute to changing the roles of scientific experts and so-called laymen?
On the side of research, replicability is a strong argument for experimental VR settings. Almost each context condition as well as the defined variable can be controlled within those experiments. Completely new data sets can be generated. This offers the possibility to increasingly capture the individual experience in terms of measurement and evaluation. This might be a threat not only for the individual but also with regard to the interactive social construction of reality. What reality do people actually share then? Do VR settings lead to new finite provinces of meaning? Gesture sensors, as shown within the simulated musical instruments example, for instance, allow to precisely measure hand gestures in order to standardise the body movement techniques. The same standard of perfection also applies to the produced sound. If the quality of the sound output device does not deteriorate, then the digital signal of the VR environment will always reliably produce the same perfect sound. This often leads to irritations for the learners, because they do not expect a constant perfect tone, but more fluctuations and multifactorial unpredictability. Usually, the complexity of societies can only be reproduced to a limited extent. The possibility of overcoming social contingency is both tempting and frightening.
IT-developers and other scientists who work with VR technology and co-construct VR settings are thus able to design countless interaction scenarios. Either they are based on the digital duplication of analog (non-digital) scenarios, e. g. by creating a teacher training scenario with VR technology, in which future teachers can make virtualized classroom experiences, or they aim at the development of completely new VR environments with their own spatial and object logic, which can elude both the everyday experience and the special knowledge of groups of experts. For both extremes, it is exciting to investigate how certain interaction situations present themselves.
In order to better understand VR phenomena and their interactions, different forms of (mediated) interaction can analytically be distinguished. In VR-based worlds such as the virtual classroom or hard disk change, humans act and interact as digital representations (avatars) either with each other or with more or less behaviorally restricted (scripted) avatars. (1) Depending on the technically modeled forms of communication (language, facial expressions, gestures, sound, other standardized sign systems), digitally represented persons have great degrees of freedom and, associated with this, almost infinite interaction options which therefore could be the bases for new finite provinces of meaning. (2) In the case of scripted avatars, the clearly defined options for action are either automated or controlled depending on the situation interpretation of an avatar controlling person, reducing complexity and drastically limiting the diversity and length of potentially possible interaction sequences. It becomes much more complex when the digital representations no longer represent only people, but machines. Avatars motivated by artificial intelligence also open up potentially infinite possibilities between humans and machines. In order to decipher the blueprints of socio-technical arrangements and their social meaning, processes of knowledge production should be investigated. The decoupling of a human and its avatar serves as a good starting point.
The ability to realize inter-actions is not tied to ego’s animate organism, but to its virtual representation. To which extent are body and avatar connected in appearance and behavior (see Figure 7)? The question of embodiment also applies for the mentioned cognitive mechanisms that make us expect certain “embodied” constraints such as our (gender) identity, ethnicity or age. Descartes’ classical philosophical mind-body problem could be reinvented by looking at VR settings.
VR as a research tool in social sciences allows transcending boundaries in many ways. New insights should be gained in the reciprocal construction of interaction scenarios. To condense these reflective, sociological thoughts, special attention has to be paid to three characteristics of VR environments:
While in everyday life symbols take us outside the province of everyday-life, in VR the use of symbols is quintessential for all experience. Moreover, VR is constituted by symbols. VR cannot be experienced immediately since the objects in VR are not part of the everyday life-world but are constituted symbolically.
Animate organisms and the outer world are decoupled. The outer world of VR cannot be experienced immediately since it is a symbolic world made out for digital data rather than for physical objects.
VR is a new, quasi-circular or paradox type of symbolism or a symbolic world that only refers to itself (and not, for example, to the zeros and ones that lie behind them).
This article has specifically dealt with the advantages of VR technologies from computer science, cognitive science and social science points of view. For creating some empirical ground, we presented six case studies which illustrate selected application possibilities in research and education. Also, remaining challenges, namely shortcomings of existing equipment as well as problems of ethical orientation for IT developers and users as co-constructors of socio-technical arrangements, have been addressed.
The first three scenarios, teacher training, hard disk change, and musical instruments are mainly in the area of simulation. This means that an attempt is made here to represent and recreate areas, principles and interaction situations of the everyday life by means of VR technology. Everyday routines in various areas of society form the model for simulation, with which situation-adequate interactions with other persons or given objects, in particular, are to be learned and trained. Natural hazards, the fourth case, is slightly different because it is based on highly aggregated 3D data that are used to simulate natural hazards and their development over time. Not social behavior in-situ is the target to be simulated and trained, but time-based developments and assessment strategies on a higher level of abstraction. The last two cases refer to cognition including high-level mental constructs such as categories and concepts. These are basic elements of perceiving the world and generating the knowledge basis of the world which becomes our social reservoir for acting successfully in everyday life even in unmediated interactions. VR enables researchers to manipulate individual preferences by dissociating motor activities from their spatial effects. The senso-motoric case expands the horizon of manipulation possibilities tremendously. The last case demonstrates the strengths of VR technology to increase the quality of methodological research designs. Replicability of scientific studies becomes progressively important. Requirements on the technological instrumentation as well as experimental conditions become more transparent and easier to replicate. At least due to the open source availability of developed VR experimental scenarios. Methods and procedures of quality assurance of research should be essential parts of higher education.
Future research in the VR domain should intensify its focus on the benefits and current limitations as well as on upcoming risks of this technology and its applications, respectively. This includes research on our self-conception as a human being (from an individual and a professional perspective) and on the associated transformation potential of social order, which is generally inherent in VR technology. This concerns the increasing everyday use of VR hardware and software for purposes of entertainment as well as their use in professional fields of education, medicine, care or exercises, to name just a few examples. This raises the question of structures, orientation problems and orientation possibilities in complex digitalised worlds, for which in the context of this article ethics is used as a disciplinary partner. The question determining the research agenda is thus: What structural conflicts and associated orientation problems are apparent between the construction of and user practices in complex digitized worlds in research and education, and what clues do they provide for the design of ethical orientation for the constructors of such worlds? We strongly encourage all fellow researchers to intensify their work and to share their insights on these issues in order to promote a more targeted and reflected development and use of VR technology – not only in research and education, but in general.
About the authors
Dr. Raphael Zender studied Information Technology/Computer Engineering and obtained his doctorate for innovative e-learning infrastructures. As a research associate at the chair for Complex Multimedia Application Architectures at the University of Potsdam, he performs various tasks in the field of teaching and research. He is currently researching the use of virtual reality in education. He is also spokesman of the working group VR/AR-Learning of the Gesellschaft für Informatik.
Alexander Henning Knoth, Sociologist. He studied Social Sciences, Law and historical Anthropology at the University of Erfurt. From 2009 to 2018 he worked as researcher at the chairs of sociology of gender and complex multimedia application infrastructures at the University of Potsdam. After being an E-Learning Coordinator for the Faculty of Economics and Social Sciences as well as Science he moved to the President’s Office as an Advisor on the Digitalisation of Teaching and International Affairs.
Prof. Dr. Martin H. Fischer is a Professor of Cognitive Sciences at the University of Potsdam. He previously worked in Dundee/Scotland and in Munich/Germany on various aspects of cognition, including attention allocation and mental arithmetic. He published around 150 peer-reviewed papers and recently co-authored the two-volume “Foundations of Embodied Cognition”.
Prof. Dr.-Ing. habil. Ulrike Lucke holds the chair for Complex Multimedia Application Architectures at the University of Potsdam since 2010. She studied computer science at the University of Rostock and also completed her PhD and habilitation in this discipline. Her areas of scientific interest include complex applications and infrastructures for media-based education and research. In the German Informatics society she was speaker of the SIG E-Learning from 2008 to 2014, and she is active as a member of the executive board since 2014 and of the SIG E-Science since 2015. Moreover, she was Chief Information Officer of the University of Potsdam from 2010 to 2018 and is vice chair-person of the German association of CIOs at universities.
 Berger, P. L. & Luckmann, T. (1966). The Social Construction of Reality. A Treatise in the Sociology of Knowledge. Garden City, NY: Anchor Books.Suche in Google Scholar
 Camerer, C. F., et al. (2018). Evaluating the replicability of social science experiments in Nature and Science between 2010 and 2015. Nature Human Behavior, 2, pp. 637–644. doi:10.1038/s41562-018-0399-z.Suche in Google Scholar PubMed
 Casasanto, D. (2009). Embodiment of abstract concepts: Good and bad in right- and left-handers. Journal of Experimental Psychology: General, 138(3), pp. 351–367. doi:10.1037/a0015854.Suche in Google Scholar PubMed
 Casasanto, D., & Chrysikou, E. G. (2011). When left is “right”: Motor fluency shapes abstract concepts. Psychological Science, 22, pp. 419–422. doi:10.1177/0956797611401755.Suche in Google Scholar PubMed
 Creagh, H. (2003). Cave Automatic Virtual Environment. In Proceedings of the Electrical Insulation Conference and Electrical Manufacturing and Coil Winding Technology Conference (pp. 499–504). Indianapolis, Indiana, USA: IEEE. doi:10.1109/EICEMC.2003.1247937.Suche in Google Scholar
 Darken, R. P., Cockayne, W. R., & Carmein, D. (1997). The omni-directional treadmill: a locomotion device for virtual worlds. In Proceedings of the 10th annual ACM symposium on User interface software and technology (pp. 213–221). New York, NY, USA: ACM.10.1145/263407.263550Suche in Google Scholar
 Feyerabend, P. K. (1970). Wider den Methodenzwang. Frankfurt: Suhrkamp.Suche in Google Scholar
 Fischer, M. H., Castel, A. D., Dodd, M. D., & Pratt, J. (2003). Perceiving numbers causes spatial shifts of attention. Nature Neuroscience, 6(6), pp. 555–556.10.1038/nn1066Suche in Google Scholar PubMed
 Fischer, M. H., & Brugger, P. (2011). When digits help digits: Spatial-numerical associations point to finger counting as prime example of embodied cognition. Frontiers in Psychology, 2, pp. 1–7. doi:10.3389/fpsyg.2011.00260.Suche in Google Scholar PubMed PubMed Central
 Fischer, M. H., & Shaki, S. (2018). Number concepts – abstract and embodied. Philosophical Transactions of the Royal Society B, 373(1752), 20170125. doi:10.1098/rstb.2017.0125.Suche in Google Scholar PubMed PubMed Central
 Fisher, S. S., McGreevy, M., Humphries, J. & Robinett, W. (1987). Virtual environment display system. In Proc Interactive 3D graphics (I3D ’86) (pp. 77–87). New York, NY, USA: ACM.10.1145/319120.319127Suche in Google Scholar
 Krueger, M. W., Gionfriddo, T. & Hinrichsen, K. (1985). VIDEOPLACE – an artificial reality. In Proc. SIGCHI Conference on Human Factors in Computing Systems (CHI ’85) (pp. 35–40). New York, NY, USA: ACM.10.1145/317456.317463Suche in Google Scholar
 Lachmair, M., Ruiz-Fernandez, S., & Gerjets, P. (2018). Does grammatical number influence the semantic priming between number cues and words related to vertical space? An investigation using virtual reality. Frontiers in Psychology, 9(573). doi:10.3389/fpsyg.2018.00573.Suche in Google Scholar PubMed PubMed Central
 Lutze, F., Zender, R. & Lucke, U. (2016). HardDrive Exchange – Eine VR-Lernanwendung zur Durchführung von Festplattenwechseln in Speichersystemen. In Lucke, U., Schwill, A. & Zender, R. (Eds.), DeLFI 2016 – Die 14. E-Learning Fachtagung Informatik (pp. 179–190). Bonn, Germany: Gesellschaft für Informatik e.V.Suche in Google Scholar
 Manzeschke, A., Weber, K., Rother, E. & Fangerau H. (2015). Ethical questions in the area of age appropriate assisting systems. Ludwigsfelde: Thiel Gruppe. https://www.nks-mtidw.de/dokumente/meestar-studie-englisch.Suche in Google Scholar
 McCarthy Ch., Sears, E. (2010). Science Education. Constructing a True View of the Real World? In Philosophy of Education Yearbook (pp. 369–377).Suche in Google Scholar
 McManus, C. (2002). Right hand, left hand: The origins of asymmetry in brains bodies, atoms and cultures. London, UK: BCA Publishers.Suche in Google Scholar
 Milgram, P., Takemura, H., Utsumi, A., & Kishino, F. (1995). Augmented reality: A class of displays on the reality-virtuality continuum. Telemanipulator and telepresence technologies, 2351, pp. 282–293.10.1117/12.197321Suche in Google Scholar
 Richter, A. & Weise, M. (2018). Bewertung von Naturgefahren mit Hilfe von Virtual Reality. In Krömker, D. & Schroeder, U. (Eds.), DeLFI 2018 – Die 16. E-Learning Fachtagung Informatik (pp. 123–128). Bonn, Germany: Gesellschaft für Informatik e.V.Suche in Google Scholar
 Salomon, G. (1994). Interaction of Media, Cognition, and Learning. New York: Routledge.Suche in Google Scholar
 Schütz, A. (1962). Collected Papers V.I: The Problem of Social Reality. The Hague: Martinus Nijhoff.Suche in Google Scholar
 Schütz, A. & Luckmann, T. (1979). The Structures of the Life-world. Evanstone: Northwestern Univ. Press.Suche in Google Scholar
 Sutherland, I. E. (1968). A head-mounted three dimensional display. In Proc. Fall Joint Computer Conference (AFIPS ‘68) (pp. 757–764). American Federation of Information Processing Societies.10.1145/1476589.1476686Suche in Google Scholar
 Slater, M. (2009). Place illusion and plausibility can lead to realistic behaviour in immersive virtual environments. Philosophical Transactions of the Royal Society of London B: Biological Sciences, 364(535), pp. 3549–3557.10.1098/rstb.2009.0138Suche in Google Scholar PubMed PubMed Central
 Weise, M., Zender, R., & Lucke, U. (2015). Systematic Fusion of Gesture Sensors for Practical Learning in Virtual Environments, In IEEE 15th International Conference on Advanced Learning Technologies (pp. 83–87). Hualien: IEEE. doi:10.1109/ICALT.2015.47.Suche in Google Scholar
 Wiepke, A. & Paulicke, P. (2018). Aufbau von Professionswissen zum Umgang mit Unterrichtsstörungen im VR-Klassenraum. Special award for a successful VR learning scenario in university teaching, Working Group VR/AR-Learning. Retrieved from http://www.uni-potsdam.de/vrarl/downloads/VRARL_Wettbewerb2018-Klassenraum-20180522-final.pdf on 2018-10-20.Suche in Google Scholar
© 2019 Walter de Gruyter GmbH, Berlin/Boston