Virtual Reality and Productivity in Knowledge Workers

Productivity has a significant impact on success and monetary wellbeing of every organisation. Over the past few years, the substantial developments of digital technologies have encouraged a shift in the way we work and produce, from an office-based environment to “virtual work”. However, very little is known as to how virtual work and productivity can be supported by virtual reality (VR). We conducted two studies to extend previous productivity research in relation to VR: Study one examined the routes that connect the organisational context with the individual productivity position through the lens of remote working and distributed collaboration; Study two explored the nature of and connections between productivity in individuals and teams working in VR. Based on the findings we explored how the future of VR could enact in knowledge workers’ daily productivity. This was done by developing a VR productivity framework that represents physical, environmental, cognitive, and behavioural needs to ensure productivity and organisational growth. Theoretical and practical implications of the findings are discussed.


INTRODUCTION
Productivity is an important factor of every organisation and has a significant impact on success and monetary wellbeing. Being "productive", traditionally, refers to the ratio between output and input (Tangen, 2002) quantified through real, tangible units, as well as by understanding the way an organisation uses resources to meet company goals (Diewert, 1992). More modern methods use proxies, by closely examining the relationship between internal (motivation) and external (office layout) factors on employees' cognitive performance, mood, and overall job satisfaction and engagement (Zhou and Shalley, 2003;Chandrasekar, 2011).
Over the past few years, substantial developments of digital technologies have encouraged a shift in the way we work and produce, from an office-based environment to "virtual work" (Wang et al., 2021). Virtual work differs from traditional work in that workers are physically dispersed, communicating, and working mostly via and assisted by digital technology. They, furthermore, exhibit a relationship with their employers away from a command-and-control process to more independent working, with greater control of the direction and process of tasks and executions (Watson-Manheim et al., 2002). This change in working has attracted research specifically trying to understand associations between virtual employment and productivity (Aimee, 2020), showing, at an individual level, a positive trend towards job satisfaction and self-empowerment, as well as reduced work-related expenses, and the ability to get more quality work done (Eddleston and Mulki, 2017). On a team-level, such as working in distributed virtual teams across geographical locations and/or time-zones, it has been demonstrated to lead to lower absenteeism, increased productivity, and quicker responsiveness to client needs (Lilian, 2014).
Alongside virtual employment, virtual reality (VR) has been argued to provide great potential for use in work-related applications due to its unique features and flexibility (Weiss and Jessel, 1998). VR environments are usually classified as immersive environments, including interactive 3D visualisation and graphical displays, provided through a head-mounted display and handhold, position-tracked devices with one or more position trackers (Cipresso et al., 2018). It permits users to experience and interact with life-like environments, in safe and convenient times, while providing a degree of control over the simulation. Suitable work-related applications have been suggested in relation to visualisation and representation; distance communication and education; hands-on training; and orientation and navigation (Weiss and Jessel, 1998;Antonov and Hristov, 2020). Overall, findings suggest that VR has been effective for these purposes in areas such as in sports, psychology, and medicine, showing significant enhancements in technical and non-technical skills (Haluck and Krummel, 2000;Ahir et al., 2019). However, very little is known as to how VR can be used to facilitate virtual working effectively, and literature on VR specifically related to productivity is sparse.
To illustrate, research on distinctive VR features has shown that the display and manipulation of VR time and equipment such as controlling the time rate of a virtual clock (Ban et al., 2015) and keyboards of different sizes (Kim et al., 2014) can enhance productivity by increasing typing speed/accuracy and click rate, while environmental features such as the colours of the walls as well as room temperature do not create such effect (Latini et al., 2021). On a systemic level, studies have highlighted the suitability of VR especially for knowledge workers (Hansen et al., 2008)-a group of employees which represents a unique set of work characteristics, including but not limited to complex problem solving and information processing opportunities (Hernaus and Mikulic, 2014). In this context, VR is believed to enable knowledge workers to "dive into their own world of concentration through an environment that creates less distraction and more focus, compared to nowadays laptop or smartphone usage." (Li et al., 2020). However, no direct assessment of VR for productivity has been carried out, such as through ethnographic observations and/or interviews of knowledge workers having performed team and task work in VR.
In the light of the above, the aim of this study is to 1): closely examine the routes that connect the organisational context with the individual productivity position through remote working and distributed collaboration (without VR) through observations and interviews 2); explore the connections between productivity in relation to VR; and 3) demonstrate how the future of VR could enact in knowledge workers' daily productivity. The utilisation of VR for team and task work was assessed through the lens of Horizon Workrooms (WR). WR is a VR platform first released by Meta in 2021, and offers a virtual office space for traditional desktop work as well as meetings via VR telepresence, with tools such as a remote desktop and a whiteboard to 'ideate' and work on ( Figure 1). Accessed through an Oculus headset, which is head-mounted device that provides virtual reality for the wearer, it enables users to imitate, visualise, and simulate the workplace design to the perspective of the user who uses it.

Study 1: Understanding and Conceptualising Productivity in the Context of Remote Work and Distributed Collaboration
We conducted 90-min observations and interviews, respectively, with 47 knowledge workers across four countries, including Norway, Singapore, the United States, and the United States, to examine productivity in the context of remote working and distributed collaboration (without VR). Areas of interest were to understand differences of synchronous and asynchronous working, including remote individual working versus distributed collaboration; tools used for effective working; and areas of improvement and potential design implications for VR (cf. interview guide for team collaboration can be found in Supplementary Appendix A). Participants worked in product gaming; online media publishing; furniture; manufacturing; IT; and logistics. Company sizes ranged from 50 to 200+. All participants were active Slack, Microsoft, and Teams users, and used both desktop and mobile platforms for work activities.

Study 2: Examining Team and Individual Productivity in the Context of Virtual Reality
Study two focused on observing and discussing productivity in VR and through the utilisation of Horizon Workrooms. Thirteen knowledge workers from a large social network company, distributed over two teams (Global Policy and EMEA Data Centre), were observed during their teamwork activities (>16 h of video and audio material), taking place over a 2-month period. In addition, interviews were conducted with 25 knowledge workers using VR for teamwork, including 12 knowledge workers who executed individual productivity activities in Workrooms for at least 30 min once a week over the past 2 months (cf. interview guide for individual productivity in VR in Supplementary Appendix B). Participants were distributed across three locations, including the United States, the United States, and Ireland, and worked across fields, including engineering, operations, legal, data science, and business.
Ethical approval was sought and obtained by Meta's Ethical Research Authority before the data collection started. Written consent was obtained from all participants.

Data Processing
The data was transcribed verbatim by an external company for an agreed fee. Both observational and interview data were analysed using a framework analysis and journey mapping was executed where possible. The framework analysis is a method/technique first used in the 1980s to analyse large-scale social policy research (Ritchie and Lewis, 2003;Gale et al., 2013). Similar to content analyses, the first steps are to transcribe data, followed by familiarisation with the interview and coding of the data. In this study, both the audio and video recordings of the observations and interviews were transcribed verbatim (word for word), and audio recordings and transcripts read multiple times, before initial contextual or reflective notes, such as analytical notes, thoughts, and impressions were formulated by the researchers. After the familiarisation, the researchers conducted a mixture of inductive and deductive "open coding" of the data line-by-line. The deductive/inductive approach helps to code data by not only having pre-defined areas of interest for this study, but also to ensure that important aspects of the data that were not considered are not missed. Open coding refers to anything that may be relevant in relation to VR and productivity. This could be related to behaviours, values, including beliefs of how VR can support/hinder productivity, emotions (e.g. frustrations by missing aspects of the VR environment to support productivity). Example quotes of extracted codes are provided throughout the manuscript, and participants were numbered to ensure anonymity. Once the coding was done, a working analytical framework was developed, by discussions of initially developed codes from each researcher being grouped into clearly defined categories. The framework was then applied by indexing subsequent transcripts using the existing codes and categories, before being charted into a framework matrix. Charting involves summarising the data by category from each transcript, which can then be used as a supporting feature in exploring interesting ideas/concepts/themes that highlight (a) characteristics of and differences between the data; (b) theoretical concepts (either prior concepts or ones emerging from the data); or (c) mapping connections between categories to explore relationships and/or causality. This allowed the authors to understand and predict how participants may respond to VR for productivity, and to identify areas that could be enhanced as part of the VR office concept.

RESULTS
By drawing on the findings, we developed a productivity framework in relation to virtual work, productivity, and VR. First, we examined the routes that connect the organisational context with the individual productivity position through the lens of remote working and distributed collaboration; second, we explored the nature of and connections between productivity in individuals and teams working in VR and demonstrated how the future of VR could enact in knowledge workers' daily productivity. Tables 1 and 2 provide an example of the codes and exemplar quotes; Figure 2 a journey mapping of remote working and distributed collaboration; Figure 3 depicts a VR productivity framework developed from the findings, and which is supported by existing productivity framework in non-VR related fields (Haynes, 2007). This was done by mapping connections between categories and to explore relationships.

Virtual Work: Conceptualisation
Findings from the observations showed that virtual work was characterized by transparency; distributed connection; and responsiveness. All three were reflected in the need of synchronous and asynchronous collaboration and visibility to team members' availability and working progress, through information sharing and exchange and simultaneous task management and execution. Synchronous collaboration was mostly conducted when high-stake decisions were made, content or projects were in the initial stage of developments, and when tasks were yet to be delegated and responsibilities distributed. In contrast, asynchronous working was observed when projects were already formulated and defined through clear ownership and associated experience (cf. Figure 2). Individual work and productivity, mainly executed during asynchronous working, were defined by participants working on single, focused tasks activities, such as programming codes or preparing and reviewing documents. Ideally, they were executed in a certain state of mind, i.e. "in the zone" (Csikszentmihalyi, 1990), in order to achieve high-quality work. Collaborative work and productivity were reflected in multi-tasking, which, depending on the nature of the role, meant executing tasks with multiple objectives at the same time; coordinating projects; and managing people. Collaborative productivity was, furthermore, shaped by: (a) external monitoring and validation by, for instance, managers; and (b) shared task responsibilities with team members working in different time zones.

Virtual Work: Productivity Tools
Remote workers usually utilised tools that were cloud-based and collaboration focused, including Office 365, WhatsApp, Messenger, or Zoom. Inherent of each tool was the desire of users to make collaboration easier and more efficient, including fewer meetings and calls, through greater asynchronous communication, and less reliance on a single ownership as well as reduced feedback processes, by enabling multi-party access and responsibility distribution for content creation and review. Tools were further highlighted to maintain productivity through productivity prioritisation and monitoring, respectively, which involved a decision-making process based on focus adjustments in relation to task urgencies and deadlines. Prioritisation and monitoring occurred at both individual and team level-internally (e.g. anxiety, motivation) and externally (e.g. project deadline)-driving the direction and process of task execution. Tasks priorities were monitored by the types of set goals; the reduction of the number of priorities for the day and week, respectively, both of which were usually tracked through tools such as OneNote or Post-It notes; and whether tasks were ahead of schedule or behind track.

Virtual Work: Availability and Notification Management
Observations have also shown that current indications of team members availability and notifications was a poor proxy for presence, and that current tools did not reflect the layers of social complexity around one's perception of how a person saw one's own availability to others. In particular, priorities around availability were dynamic and dependent on who a knowledge worker wanted to be available for, as well as what tasks they wanted to be available for. Furthermore, notifications of availability were observed to be too simplistic; while notification management was often used for focused productivity time or the ability to execute 'undisrupted' communication with team members during collaboration activities, they did not offer smart filtering, such as appropriate timing based on one's degree of workload, feedback on who the notification is from, or how important or urgent the notification was for the knowledge worker. This, alongside a constant feeling of 'having to be available', has shown that notifications could create a significant interruption to one's productivity.

Study 2: Examining Individual and Teamwork and Productivity in the Context of Virtual Reality
Study one focused on the conceptualisation of productivity and work in the context of remote working and distributed collaboration. With this in mind, study two investigated remote working and distributed collaboration in the context of VR, and how VR can enact in knowledge workers' daily productivity. The latter was done by highlighting opportunities, challenges, and implications for future VR developments (cf. Figure 3).

Individual Work and Virtual Reality
Participants' reasons for participating in VR for individual productivity was the opportunity to have a private space, a personal desk, and the physical separation from one's home environment. All three created the feeling of "being somewhere else", without distractions. They also appreciated the level of immersion VR offered them, with an 'infinite' space at their disposal, as well as the ability to shut out external distractions. Through the headset a physical separation between their external environment was created, enabling productivity through the mental state they wished to achieve in VR. Expressed constraints of VR for individual work were mentioned in relation to the physical setup and environmental needs; and the tools needed to be productive. Physical setup and environmental needs in VR were: lighter hardware weight, longer battery life, and better comfort, allowing to utilise VR for prolonged amount of time; personalised work space by having pens, papers, and a picture of the family on their virtual desk, as well as the ability to change environmental features such as a beach with ocean noises, coffee-house environments, chirping birds, or calming background music as a way to relax; the ability to move around and interact with the VR space itself, such as going to the window for a break, or by changing venues to allow for interaction with others; and safety concerns, by not knowing where the real space ends and the VR space begins.
Participants, furthermore, expressed the importance of work, health-and well-being tool integrations. Work tools included an enhanced ability to interact with and draw on the whiteboard, for individual and team activities, as well as smaller, multiple groups; multiple monitors to execute different tasks; task monitoring and meeting tracking tools such as calendars; and, a timer that enables participants to increase productivity and mindfulness of the time spent in VR. Health and well-being tools were mentioned in relation to having background music or white noise to enhance one's focus; reduced notifications to permit distractions and the risk of multi-tasking; breaks, either by creating opportunities for (non-) work-related interactions with colleagues or alone; and taking part in meditation or stretching activities, provided through tailored notifications in VR.

Teamwork and Virtual Reality
Benefits of working in VR was the ability of sharing an office space, enhanced perceptions of joyful, energised teamwork sessions, and an equal feeling of presence and voice. Findings showed that shared space mattered, because it supported teams to align on task and project content, direction, and progress, and enabled sharing content that was visible for everyone, while not losing sight of the person in the room. As such, participants were more engaged with the work as well as one another. Creating personalised avatars meant for participants to share aspects of their personality by emulating themselves through their virtual representation, which participants perceived as a significant contributor to team cohesiveness. Over the course of 2 months, participants expressed and exhibited a feeling of pro-active involvement, team inclusion, social presence (Greenwald et al., 2017), as well as the ability to mutually influence on another (Carson et al., 2007). Overall, it was expressed that VR supports team collaboration by helping team members to align on the same content, but in a better setup (cf. Table 2 for examples).
Challenges and opportunities for VR and teamwork were emphasised by (a) reduced hardware weight and better comfort; (b) enhanced ability of note taking and modification; (c) greater multi-tasking-ideally, by having multiple screens for different purposes, such as screens for sharing, searching for content, taking notes, etc.; (d) better facial expressions of avatars, and a stronger representation of workers' physical appearance, including clothing and accessories. Lastly, future usage of VR for collaboration was dependent on the ability to allow for VR customization, such as by having break-out rooms for smaller teamwork activities and discussions.

DISCUSSION
Investigating how knowledge workers conceptualise virtual work including individual and team productivity is important to understand how VR's current productivity (tool) design can be enhanced and benefit a range of different activities (Kim et al., 2019). The aim of this study was two-fold: Study one examined the routes that connect the organisational context with the individual productivity position through the lens of remote working and distributed collaboration; Study two explored the nature of and connections between productivity in individuals and teams working in VR. Based on the findings we explored how the future of VR could enact in knowledge workers' daily productivity. This was done by developing a VR productivity framework that represents physical, environmental, cognitive, and behavioural needs to ensure productivity and organisational growth (cf. Figure 3 for an overview). In addition, we provide the following theoretical and practical implications: Firstly, findings from study one showed that productivity is a multifaceted concept, which, in line with Kim et al. (2019) (Kim et al., 2019), was largely determined by perceived efficiency and quality of work, set by deadlines, and monitoring that was selfimposed; knowledge workers' mental state (e.g. attention, motivation, tasks satisfaction); type of task (single tasks versus multi-tasking); and self-regulation behaviour (e.g., task tracking versus external validation). In addition, we explored productivity within the organisational context, which shaped and impacted the need of synchronous and asynchronous collaboration and visibility to team members' availability and working progress, through effective and efficient information sharing and exchange, as well as simultaneous project management and execution.
Previous research has shown that productivity is mediated by knowledge workers' engagement and self-regulatory behaviour, both of which have significant impact towards creating organisational growth (Stander et al., 2014). Work engagement is defined as a 'positive, fulfilling, work-related state of mind characterised by vigour, dedication, and absorption' (Schaufeli and Bakker, 2004), including aspects such as positive affect and energy, psychological involvement in one's role, as well as psychological flow (van Woerkom et al., 2016). Self-regulation, which is the ability to control one's behaviour, emotions, and thoughts helps to continuously engage in the direction, intensity, and persistence of effort, with the goal to strive for internal representations of their desired end states, such as collaborative task achievements (Austin and Vancouver, 1996;Vancouver, 2008).
Navigating to the point when these end states are accomplished depend on the feedback loop they receive, enabling a comparison between personal and externally validated performance, and taking corrective actions if discrepancies arise. For this to happen, developing realistic perceptions on goal progresses and velocity, which is the rate to which these goals are achieved, matter (Johnson et al., 2013). This requires effective time management planning, such as task lists, prioritising tasks, and determining how and when to perform them, as well as contingent planning, in which employees anticipate possible interruptions in their work and plan for them, as a mechanism to stay engaged, on track, and perform well.
Our findings are in alignment with the above, showing a positive association between time management, contingency planning, and productivity; as well as conditional effect between effective time management and the number but also type of interruptions throughout the day. Sentiments towards interruptions were dependent on its nature, such as impromptu interactions for problem-solving or learning activities, or scheduled meetings to maintain organisational productivity. These were perceived less affecting compared to distractions, which were described as interfering stimuli one wished to ignore, such as the exposure to work-irrelevant conversations through notifications coming from, for instance, online chats including Workplace or Microsoft Teams.
Results from study two showed that VR positively contributed to a perceived productivity and overall quality, by providing a distraction-free and focused work environment; and the perception of a shared office space that created joyful, energised teamworking, and an equal feeling of social presence and proactive involvement.
Research has shown that equal abilities to contribute to team activities increases effective team performance (Driskell and Salas, 1992;Butchibabu et al., 2016) by providing a strong sense of camaraderie, support and participation, which, in turn, elevates individuals' work cooperation, and the development of shared responsibilities for team outcomes (Elsaied, 2018). It furthermore supports members to ask questions, seek help, report mistakes and raise concerns without having to fear negative consequences as a result of their behaviour (i.e. psychological safety; (Edmondson, 1999)). In contrast, social presence, although dependent on various factors such as the type of task, quality of avatars, level of interactivity, haptic feedback, etc., has been associated with a greater perception of trust in team members and perceived usefulness of team activities (Oh et al., 2018). Future research should therefore investigate how VR exactly shapes such group behaviour and perception, by acknowledging individual roles, and the multilevel linking mechanism between individual traits and team outcomes via psychological and social processes (Stewart et al., 2005). This will help identify predictors at individual and team level on an individual-level outcome, and the moderating effects of team level variables on relationships between individual-level variables (Oh et al., 2018), including aspects such as shared leadership (Carson et al., 2007), psychological safety (Edmondson, 1999), task, interpersonal and process conflicts (Jehn and Mannix, 2001), as well as an adaptive cooperative attitude-all assessed in the context of VR. By providing an environment that encourages individuals to both manage and develop within a team; team bonding through active involvement and emotional support; and that offers avatars for an enhanced feeling of social presence, VR has the potential to achieve enhanced productivity through greater individual expressivity and commitment, and active participation in decision making.

Implications for Future Virtual Reality Applications
Because both individual and team productivity are needed to ensure organisational success, future developments in VR office spaces are advised to find the right balance between overall (individual) productivity and the complexity, duration, timing, and frequency of collaborative activities throughout the day. This includes 1) easily accessible, usable, and safe hardware and software applications; 2); customisable room configurations (e.g. personalised versus team working space); and 3) productivity enhancing role-and team-dependent work tools, including prioritisation and monitoring applications that use smart filtering suitable for the nature and content of the task activity (e.g. synchronous and asynchronous work, brainstorming, etc.).
For the latter, VR applications, in particular, need to address challenges that are inherent in current organisational productivity paradigms, which requires ad-hoc availability for information interaction and exchange, by attending team meetings and engaging in (synchronous and asynchronous) activities such as planning, delegating, and sharing tasks and information. Such working creates the risk of content fragmentation and lack of clear ownership, which, in this study, was observed by teams operating on numerous 'living' documents of different formats, and an enhanced need for content tracking and archiving. Multiple ownerships caused conflicts and discomfort in relation to the quality of content output and project accountability, with tools not being able to account for nuances and changes in ownerships over time. Considering these needs alongside adequate time management and contingent management plans to be productive, future VR applications will need to be tested via e.g. simulated series of predetermined work activities; productivity success measures (e.g. efficiency); perceived effectiveness of the simulated workplace; and, a validation of the designed features through e.g. Multi Criteria Decision Making, in order to establish the order of preference in selecting the best option among many alternatives based on the desired productivity outcome (Muttaqin et al., 2020).

Limitations and Further Research
While this study is one of the first to assess the concept of productivity in the context of VR, the following limitations are worth mentioning: This study was conducted with a diverse sample using qualitative methods to understand productivity in VR. Future studies are encouraged to investigate productivity in VR with a larger sample of similar professional background, using a mixed methods methodology including objective performance markers as well as productivity behaviour to assess the value of VR for teamwork and task work, and by directly comparing the VR productivity to real life activities.
We explored team and task work using a specific VR application, namely Horizon Workrooms. Using this platform to explore productivity may have limited participants' views on opportunities and constraints experienced in VR. Future studies are, therefore, advised to either collect data on productivity using a variety of platforms, or by exploring productivity in VR more broadly, to allow for more generalizable interpretations of results in the context of VR.
While we observed teams in VR over a period of 2 months, future studies are encouraged to develop a longitudinal study that robustly tracks user sentiment data, where changes of perceptions, effectiveness, and usefulness of VR for work and productivity are recorded over a longer period. This will help overcome the risk of biased assessments, and to design a VR space that has been design iteratively, with user perceptions in mind.
Our framework offers first insights into how productivity can be supported through VR, through the lens of Meta's Workrooms; however, future studies are advised to carry out further validations, continuing to revise and adapt content depending on the nature of the productivity tasks and work profession, as well as by examining the causal relationship between productivity themes (e.g. transactional knowledge, concentrated study, group processes versus individual processes) and productivity evaluation, such as the relationship between the worker's state and the productivity in VR.
Lastly, we wish to acknowledge the limitations of current technologies itself. For instance, the vergence-accommodation conflict remains a cause for eye fatigue and discomfort for both virtual and augmented reality applications (Hoffman et al., 2008;Kramida, 2016). These constraints limit the ability of users to engage full-time in teamwork and task work using VR, and, as such, does not offer a replacement to other online productivity and collaboration tools. Instead, it represents an additional means to engage oneself in individual taskwork and collaboration activities, respectively, through immersive three-dimensional surroundings and content display. Future studies who apply VR for work will have to explore these health and safety implications in greater detail, and design and execute policies that support users in the appropriate use of VR for productivity.

CONCLUSION
This research has determined a range of factors that may be helpful for individual and teamwork and productivity conceptualisation within the virtual environment. Future studies are advised to further explore the need of planning and productivity tools using smart technology such as artificial intelligence to ensure optimised and individually tailored productivity within VR. Ideally, this will be achieved by a better awareness of goal-discrepancy induced task pressures and urgencies, and the environmental, cognitive, and behavioural needs of VR to help increase work engagement and overall productivity accomplishment (Parke et al., 2017).

DATA AVAILABILITY STATEMENT
The datasets presented in this article are not readily available because of data privacy polices at Meta.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Meta Ethical Research Authority. The patients/ participants provided their written informed consent to participate in this study.

FUNDING
This study received funding from Meta. The funder was not involved in the study design, collection, analysis, interpretation of data, the writing of this article or the decision to submit it for publication. All authors declare no other competing interests.