Adopting a metaverse-based workspace to support research team collaboration: a pilot study from an academic health informatics laboratory

Abstract

Objective

This study aimed to understand how a metaverse-based (virtual) workspace can be used to support the communication and collaboration in an academic health informatics lab.

Materials and Methods

A survey of lab members (n = 14) was analyzed according to a concurrent triangulation mixed methods design. The qualitative survey data were organized according to the Capability, Opportunity, Motivation, Behavior (COM-B) model and combined to generate personas that represent the overall types of lab members. Additionally, scheduled work hours were analyzed quantitatively to complement the findings of the survey feedback.

Results

Four personas, representative of different types of virtual workers, were developed using the survey responses. These personas reflected the wide variety of opinions about virtual work among the participants and helped to categorize the most common feedback. The Work Hours Schedule Sheet analysis showed the low number of possible collaboration opportunities that were utilized compared to the number available.

Discussion

We found that informal communication and co-location were not supported by the virtual workplace as we had originally planned. To solve this issue, we offer 3 design recommendations for those looking to implement their own virtual informatics lab. First, labs should establish common goals and norms for virtual workplace interactions. Second, labs should carefully plan the virtual space layout to maximize communication opportunities. Finally, labs should work with their platform of choice to address technical limitations for their lab members to improve user experience. Future work includes a formal, theory-guided experiment with consideration on ethical and behavioral impact.

BACKGROUND AND SIGNIFICANCE

The Coronavirus Disease 2019 (COVID-19) pandemic immediately changed the way that we interact and collaborate in the workplace. In the beginning, most in-person meetings and conversation were no longer an option for workers looking to carry out their daily tasks. Thanks to today’s advanced communication technologies, remote collaboration helped mitigate the impact of social distancing. Many workers, however, feel that online collaboration is very different than in-person work. Anecdotally, most complaints about synchronous online collaboration focus on lackluster social connections and the loss of casual conversation opportunities. Currently, applications use remote collaboration techniques like audio teleconferencing, video, and web conferencing, and “telepresence” solutions like virtual reality and augmented reality.

To collaborate, workers must participate in group activities, interact with other people and synthesize new knowledge.¹ Audio and video conferencing tools are capable of supporting comprehensive collaboration on sophisticated tasks with web-based applications like Miro.² Even with progress in recent years on these technologies, informal conversation is difficult in a virtual workplace.^3,4 Another bottleneck for collaboration in virtual teams is transferring “tacit knowledge,” knowledge that is difficult to express through written or spoken word.^4–6 In an in-person workplace, teams participate in groups, interact, and synthesize information more easily¹ because all necessary team members are located in the same physical space. This concept, called “co-location,” increases collaboration opportunities, promotes socialization, and eventually facilitates tacit knowledge exchange,⁷ which plays a critical role in group innovation.⁸ When compared to remote collaboration, in-person work avoids the social isolation experienced while working from home and prevents telecommuters from “miss[ing] idle conversations in the hallway and other informal conversations.”⁹ In virtual teams, communication without seeing others’ facial expressions is often more explicit and formal than face-to-face communication.⁶ As informal communication becomes less frequent, the exchange of tacit knowledge also decreases.¹⁰ This can result in less innovative collaboration in virtual teams than in-person. Therefore, finding a way for virtual teams to recreate a sense of “co-location” may help team members communicate informally and collaborate innovatively.

Spaces for “co-location” can be virtual workplaces (eg, Meta Horizon Workrooms¹¹) or in real offices (eg, Microsoft Holoportation¹² and Telepresence Robot¹³). These so-called metaverse technologies, in addition to other kinds of telepresence technologies, help people collaborate online. Currently, metaverse technologies are defined as immersive 3-dimensional virtual worlds in which multiple users interact with each other.¹⁴ The definition of metaverse can be expanded or more accurately coined. Almoqbel et al¹⁵ conducted a systematic review to map metaverse definitions in 4 aspects, including Users and Roles, Activity, Content Creation, and Tech Specs. The idea of a metaverse is not new. As early as 1978, simple virtual worlds like Richard Bartle’s Multi-user Dungeon or Multi-user Dimension (MUD) were beginning to shape how designers approached the creation of a virtual environment. These early metaverses were usually text-based or storyboarded.¹⁶ Although more vivid 3-dimensional virtual worlds have been created on different platforms in recent years, there are still 2-dimensional virtual worlds used for low-cost metaverse experiences.¹⁷ These offer telecommuters the ability to co-locate and interact in the virtual world in a similar fashion to a face-to-face office, which theoretically makes virtual teams capable of informal communication.

Informal communication is critical in distributed scientific work and considered as a mediating factor to productivity; and affected by factors including a concentration of suitable partners, co-presence, and establishing and sharing common ground.¹⁸ Metaverse technologies provide a virtual space where people with common interest can collaborate. However, creating informal communication in a virtual space is not as simple as it is in a face-to-face workplace. Frequent communication in virtual settings can cause cognitive overload and have a negative impact on team processes and outcomes.¹⁹ Telecommuters often do not communicate with anyone outside of their team, thereby reducing their informal communication network.²⁰ This was different from what was expected of metaverse-based online collaboration.

OBJECTIVES

This pilot study addressed this gap by asking the following research questions: (1) Is co-location for virtual teams supported by a virtual workplace? (2) How is informal communication carried out in a virtual workplace? It focused on a health informatics lab at a research university and looked to generate design recommendations for a virtual workspace. While computer-supported collaborative work (CSCW) theories, frameworks, and methods have been adopted in the healthcare domain, many of them are focused on clinical work environment.^21–23 To the best of our knowledge, this is the first study focused on the health informatics research collaboration using metaverse.

MATERIALS AND METHODS

Study setting

The corresponding author’s laboratory (the lab hereafter) is an interdisciplinary data science lab at the University of Cincinnati College of Medicine. The lab consists of around 20 students at both the undergraduate and graduate level in a wide range of disciplines including health informatics, medical sciences, design, and computer science. The lab uses a technique called “matrix management”²⁴ in order to coordinate efforts between members in different teams so that the lab can effectively develop data-intensive and user-friendly systems to address clinical and health problems. Lab members are assigned into 3 skill-based teams and 8 project-based teams (Table 1). The 3 skill-based teams are the research team, the design team, and the technical team. The 8 project-based teams are labeled A–H and correspond to a team working on a single project. Project-based team members come from all 3 of the skill-based teams. Lab members have variable working hours based on their commitment.

Table 1.

Teams in the lab

Team	Number of divisions	Division of teams
Skill-based teams	3	Research, Technical, and Design Teams
Project-based teams	8	Teams A, B, C, D, E, F, G, and H

The lab members can perform tasks individually and/or collaboratively, depending on the nature of the tasks (eg, literature search, affinity diagramming, group discussion, manuscript writing). Microsoft Teams was used as the primary online platform due to the institutional support in the beginning of the pandemic (March 2020). In the middle of the pandemic (June 2021), another online platform called Gather (https://www.gather.town) was employed. Gather is a 2-dimensional pixel-style metaverse-based video chat platform for collaboration. This platform was chosen because of its convenient access, low (zero) cost for small group collaboration (25 people and below), and high customizability in the virtual space layout. Figure 1 shows an example layout of the virtual lab space.

Figure 1.

Virtual lab of the corresponding author’s lab in Gather. This figure shows the workplace layout. Area 1 shows the personal offices. Area 2 shows the divided meeting rooms for different teams. Area 3 shows the large meeting room for lab meetings.

Study design

This study used a concurrent triangulation design²⁵ to infer factors that would help facilitate collaboration in virtual academic teams. Because of the dynamic nature of academic work and online collaboration, multiple dimensions of data were required to understand and interpret observed work patterns. In parallel, we collected 2 major types of data: (1) qualitative and quantitative data from an online survey questionnaire about work experience and (2) quantitative data from the lab’s Work Hours Schedule Sheet.

The data collection was guided by the workspace awareness framework, especially the knowledge in “where” and “when”²⁶ since Gather affords co-presence and co-location in a visual space compared to traditional online meeting tools (eg, Microsoft Teams, WebEx, and, Zoom). Of note, the study did not focus on the exploration and the action aspects of the workspace awareness framework because they are based on the knowledge gained. In other words, no exploration or actions would occur until team members are co-located in a virtual space and read for task collaboration.

The qualitative and quantitative data were analyzed concurrently and then compared to identify similarities or differences in the findings. The full process is outlined in Figure 2 below and detailed methods of collection and analysis are described in the following sections. The study was reviewed by the institutional review board (IRB) and determined as not human subject research (IRB# 2021-0881).

Figure 2.

Process of this study.

Online Collaboration Experience Survey

The Online Collaboration Experience Survey collected qualitative and quantitative data related to lab members’ experiences of collaborating online. The survey questionnaire was created using Microsoft Forms (https://forms.office.com) and distributed to lab members via email. The qualitative data came from free-text long response style questions. The quantitative data came from a series of multiple-choice questions.

The focus of the qualitative section of the survey was lab members’ experience with remote work. To design the survey questions and analyze the responses, the Capability, Opportunity, Motivation, Behavior (COM-B) model²⁷ was used. The COM-B model relates a person’s capability, opportunity, and motivation to complete any given behavior to the behavior itself. It also examines how the behavior, in turn, influences those factors. The questions for this survey, then, were designed to understand the capacity, the opportunity, and the motivation that each worker had to join the virtual workspace.

The survey’s quantitative questions compared collaborative activities between the online and offline environments and identified issues pertaining to using the virtual lab for collaboration. Table 2 shows questions and choices for both sections of the survey.

Table 2.

Online Collaboration Experience Survey questions and choices

	Questions	Answers
Quantitative section
Q1	Which of the following activities have you ever taken part in online with others?	Group meeting	Individual meeting	Working together	Gathering	Games
Q2	Which of the following activities have you ever taken part in offline that can be related to collaboration?	Group meeting	Individual meeting	Working together	Gathering	Games
Qualitative section
Q3	Please describe your typical day working remotely for the lab.	Free text
Q4	Please describe the pros and cons of using the virtual lab.	Free text
Q5	What’s your opinion about online collaboration? And how does it relate to the virtual lab?	Free text
Q6	How does the virtual lab annoy when you work remotely online?	Free text
Q7	How does the virtual lab satisfy you when you work remotely online?	Free text
Q8	Please describe your goal working in the lab.	Free text
Q9	What do you think the virtual lab is most suitable to do?	Multi-choice and free text
Q10	Which of the following issues do you meet when using the virtual lab?	Multi-choice and free text

Quantitative survey data analysis

There were 2 quantitative questions asked as part of the survey: (1) online activities and (2) offline activities. Both questions had the same set of choices. The first 2 choices were “Group meeting” and “Individual meeting,” which are both related to schedule-based, formal meetings. The other 3 choices were “Working together,” “Parties/Gatherings,” and “Games,” which are related to collaboration opportunities that are more informal. Lab members’ choices in these 2 questions were summarized and compared to distinguish the differences between lab members’ experiences.

Qualitative survey data analysis

The free-text responses to the survey questions were analyzed to discover themes across participants based on the COM-B model. First, all the data collected from the survey were coded by their relevance to collaboration in a process called “open coding.” Then, these codes were grouped into categories according to the COM-B model²⁷ in a process called “axial coding.” Category 1 consists of codes that describe lab members’ behaviors. Category 2 is made up of codes that indicate lab members’ capabilities, opportunities, and motivations. Category 3 contains codes of information unrelated to the COM-B model, like career goals and other demographic information.

To provide a full picture of typical users, we constructed personas (fictional characters that represent how users might interact with a system²⁸). To form these personas, the most frequently mentioned behaviors in Category 1 were recombined with corresponding motivations in Category 2. Then, groups of behaviors that occur commonly together were created so that the personas represent how typical lab members behave in the virtual workspace. Codes in category 3 were used to provide context about what type of lab member (which skill-based team or project-based team, for example) falls into each persona.

Work Hours Schedule Sheet

The Work Hours Schedule Sheet is a shared worksheet containing the work hours commitment and semester schedule of all lab members that helps arrange potential collaboration opportunities by making all schedules available to view by all members. Since the schedule sheet records the overall work hours of lab members and overlapping hours with others, this sheet is a useful tool to facilitate possible spontaneous collaboration opportunities among the lab members.

The version used in this study was a Microsoft Excel document that contained 30-min time slots ranging from 8:30 am to 10:30 pm Monday to Friday. For each day, there were 2 columns indicating whether the work will be done in the virtual lab or in the physical lab. Using this table, the lab members put their initials on the blocks to indicate when and where they will be working. The lab members were asked to log onto the virtual lab as much as possible even when they were working in the physical lab space.

The schedule sheet focused on the collaboration opportunities each lab member had with their partners in project-based teams and skill-based teams. A lab member’s collaboration opportunities in a shared time slot were defined as a combination of the total number of members assigned for a given time slot because the minimum number of members needed for collaboration is 2. We defined the calculation of collaboration opportunities as ${\mathit{COOP}}_i$, and the number of lab members signed up for the same shared time slot as $n_{\mathit{swt}}$. The calculation of collaboration opportunities, then, is expressed as:

$${\mathit{COOP}}_i=C_2^{n_{\mathit{swt}}}.$$

This formula indicates the intensity of collaboration in ideal situation, as lab members’ willingness to collaborate is difficult to measure accurately and was not considered for this analysis. Of note, collaboration opportunities can lead to formal and/or informal collaboration.

Because the number of members in a team varies, we introduced a “collaboration opportunities rate” to perform cross-team comparisons (labeled in this model as ${\mathit{COOP}}_i\%$). This percentage indicates the proportion of collaboration opportunities in one group in a week compared to the maximum possible collaboration opportunities. The maximum possible collaboration (${\mathit{max}\_\mathit{COOP}}_i$) was the number of collaboration opportunities a team of lab members has if they are all signed for a same time slot. The number of lab members in a team was denoted as $n_{\mathit{total}}$. So, the calculation of collaboration opportunities rate is:

$${\mathit{COOP}}_i\mathrm{\%}=\frac{{\mathit{COOP}}_i}{{\mathit{max}\_\mathit{COOP}}_i}=\frac{C_2^{n_{\mathit{swt}}}}{C_2^{n_{\mathit{total}}}}\times 100\mathrm{\%}.$$

With this calculation method, the schedule sheet enabled us to compare collaboration opportunities between skill-based teams as well as project-based teams.

Data triangulation

Once the qualitative and quantitative data from the survey and the collaboration opportunity rate from the schedule sheet were analyzed separately, the 2 sources of data were compared to determine issues that prevent lab members from collaboration in the virtual lab. Each set of data revealed a different aspect of online collaboration. The qualitative data from the survey investigated how a typical worker feels about collaboration in the virtual lab. The quantitative data of the survey showed the difference between online collaboration activities and offline collaboration activities. The schedule sheet provided a numerical description of collaboration opportunities under current lab settings. Combined, these datasets provide a full and accurate picture of collaboration in the virtual lab space.

RESULTS

Online Collaboration Experience Survey

Fourteen out of 15 eligible lab members in October 2021 responded to the survey (93% response rate). Out of the 20 total lab members, 3 members of the lab did not have access to the virtual lab on Gather, and 2 members are researchers of this study and were therefore not eligible to participate. All skill-based teams were included in this survey. Of the respondents, 6 were from the research team, 5 were from the technical team, and 3 were from the design team. The respondents’ ages ranged from 18 to 26 years old, and all of respondents worked at the lab as a part-time job or as a volunteer.

Meetings were the major collaborative activity regardless of online or offline status. All respondents (n = 14, 100%) indicated that meetings were a part of their online collaboration experience. Slightly fewer respondents claimed that they collaborated with others using in-person meetings. Twelve respondents (85.71%) reported they had group meetings (meetings with more than 2 people), and 11 respondents (78.57%) had meetings with others individually.

More lab members engaged in non-meeting-related collaboration activities more often when in-person compared to when online. For example, working/learning together is a predominant activity of offline collaboration. Eleven respondents (78.57%) reported that they work and learn together with others offline. In contrast, all activities other than meeting to work or learn were reported by <3 respondents. A full summary of these results can be found in Figure 3 below.

Figure 3.

Lab members’ online collaborative activities and offline collaborative activities.

Persona generation

Eight behavior-related codes were identified in Category 1. Twenty-nine source-related codes were indentified in Category 2, which represent the capabilities, opportunities, and motivations that lead to those behaviors identified in Category 1. The 9 codes that are not related to the COM-B model²⁷ were included in Category 3. A full list of codes as they relate to the formation of personas can be made available upon request. After adding details from Category 3, 4 personas were created.

Persona I is “Collaborative” Carmen. She is a member of Design Skill-based Team. She usually works on her own. She likes to use platforms like Miro and WebEx to communicate with her partners. But as a designer, she wants to have a more active role in this lab, have more meetings on deliberating design, and dedicatedly serve as a design professional.

Persona II is “Indifferent” Isaac. He is from the Technical Skill-based Team. Collaboration is one required section of his work, but most of his work requires additional software and documents that are not on Gather. He believes using Gather would make his online workflow more complicated, so to stay efficient, he uses alternative meeting platforms that use less computing resources.

Persona III is “Online” Olivia. She is a researcher at the lab. She stays in the virtual lab during her work hours and is open to talk to people. However, she often misses notifications from Gather when she switched away from the Gather tab in her browser. It is also frustrating when she cannot get others’ responses due to the same issue of missing messages while having Gather open on another tab but not directly on the platform.

Persona IV is named “Focused” Fred. Fred is a member of the Technical Skill-based Team. For most of his work hours, he works on his own without needing to collaborate with others. He keeps Gather open occasionally. And if he gets a message, it startles him, interrupting his flow of work, as does not think that others need to reach out to him normally.

Work Hour Schedule Sheet

The results of the analysis of the Work Hour Schedule Sheet show that collaboration opportunities are generally uneven between project-based teams and skill-based teams. The maximum usage rate is from Team A, utilizing only 7.13% of their possible collaboration opportunities, followed by Team H (6.44%). The remaining project-based teams utilize <2% of their maximum possible collaboration opportunities. Project-based teams D and E (shown in Table 1) had no collaboration opportunities in a week. On the other hand, skill-based teams have less opportunities for collaboration than project-based teams. All skill-based teams have a week-wide collaboration rate around 2%. Full results are shared below in Table 3.

Table 3.

Collaboration opportunities of project-based teams and skill-based teams in the lab

	Project-based teams								Skill-based teams
	A	B	C	D	E	F	G	H	Research	Design	Technical
Number of lab members ($n_{\mathit{total}}$)	4	3	3	5	2	3	3	3	9	4	6
Total possible collaboration per week (${\mathit{COOP}}_i$)	62	9	5	0	0	7	8	28	90	21	37
Maximum possible collaboration per week (${\mathit{max}\_\mathit{COOP}}_i$)	870	435	435	1450	145	435	435	435	5220	870	2175
Collaboration opportunities rate (${\mathit{COOP}}_{ni}\mathrm{\%}$)	7.13%	2.07%	1.15%	0.00%	0.00%	1.60%	1.84%	6.44%	1.72%	2.41%	1.70%

Data triangulation

The qualitative results from the survey show that most lab members want to collaborate with others. Despite this desire, many workers feel that the lab lacks informal social opportunities that would make them more comfortable communicating with their team members. One possible cause for this discomfort could be from confusion between lab members about the expectations for communication in the virtual lab. Many of these concerns are represented by Persona II, “Indifferent Isaac,” who suggests that lab members were indifferent to collaborating in the virtual lab. A worker that fits this persona would not be interested in utilizing a virtual workspace where others could drop in unannounced and does not prefer to keep another platform open on their computer. Other personas showed that informal communication was difficult to achieve even though members were co-located in the virtual lab.

The quantitative data from the survey also supports the qualitative findings. The type of collaborative activities that lab members participated in were different between online and offline settings. Only a few members participated in online activities that support informal communication, such as working together and gaming. This contrasts with offline collaborative activities, which were experienced by most members of the lab.

The results of the schedule sheet analysis further confirmed that the overall opportunities to collaborate or have informal communication are low for all lab members. Project-based teams D and E, for example, did not have any shared time slots in which teammates could collaborate.

Comparing the 3 sets of data, the lack of collaboration in the virtual lab is not due to a lack of motivation from lab members to collaborate. This is due to (1) low scheduled time in the virtual lab means that lab members do not have the opportunity (other than in scheduled meetings) to connect with one another informally, and (2) lab members approach work in the virtual environment with varying expectations; some lab members expect their coworkers to be constantly available if their avatar is present in the virtual workspace, whereas others do not want to be bothered even if they are present in the virtual workspace.

DISCUSSION

This study used a concurrent triangulation mixed methods design to assess how co-location and informal communication are supported in a virtual workspace. The results showed that adopting a metaverse-based virtual workplace did not improve informal communication for workers collaborating remotely, even though workers were “co-located” in the same virtual space. This is supported by current literature acknowledging that the relationship between informal communication and co-location is not the same as in a face-to-face workplace^7,10 due to factors like increased cognitive load in virtual settings.

We offer 2 additional reasons why co-location alone does not create informal communication opportunities in the same way that face-to-face workplaces can. First, although many opportunities for co-location exist in our lab, lab members rarely take advantage of them because their schedules do not align. Flexibility in scheduling, however, is necessary and common in an academic health informatics lab because the majority of the workers are students and need to work around classes and other school responsibilities and commitments. Secondly, even though team members may be present in the virtual workspace, they are not always ready to collaborate. This is modeled by interactions between lab members who match the personas of “Collaborator Carmen” and “Indifferent Isaac,” for example. To mitigate these gaps, we offer 3 recommendations below for teams looking to design and implement a metaverse-based workplace.

Design recommendations for metaverse workplaces

Establish common goals and norms for the virtual workplace

In our study, some virtual workers, represented by personas “Collaborator Carmen” and “Online Olivia,” consider metaverse-based work platforms to be analogous to the face-to-face workplace. They consider their presence in the virtual workplace like punching in at a timeclock at work. Other virtual workers, represented by personas like “Focused Fred” and “Indifferent Isaac,” did not feel this way, regarding the virtual workspace as yet another window to have open on their screen; another application to minimize on their computer without paying attention to possible opportunities for collaboration. As a result, potential collaboration opportunities can be halted, and the point of the virtual workspace could be missed.

To solve this issue, organizations and virtual workplace developers should consider establishing common goals and standards for telecommuters using a virtual workplace before rolling out this new technology. Addressing expectations for collaboration may help soften areas of contention between workers. For example, after finishing this analysis, we clarified with team members that “logging-in” to the virtual workplace should be like a sitting with your “door open” in a face-to-face workplace because it provides an opportunity for meeting and exchanging knowledge rather than simply showing that someone is working on tasks. We also clarified that, if a worker did not want to be disturbed, they should use the built in “Do Not Disturb” (also called “Quiet Time”) feature to indicate that they are working on a task and wished to not be interrupted. This feature also allows for workers to write a quick message which could include the best way to contact them.

Increase communication opportunities by space planning

According to the results of the survey, scheduled meetings are the center of collaborative activity in our office. This was confirmed by the Work Hours Schedule Sheet which showed that our lab members schedule their work hours largely around showing up for meetings without considering their team members’ schedules. This leads to insufficient informal collaboration opportunities. Additionally, even when multiple workers from the same teams are present together in the virtual office, they did not often utilize their potential collaboration time. This is confirmed by the low overall “Collaborations Opportunity Rate” calculated above, with the highest rate being around 7% of opportunities to collaborate used.

One possible reason that workers do not collaborate, even when they are both present in the virtual office, is that their avatars are too far apart to interact. To increase communication opportunities in a virtual workplace, we recommend that office designers lay out the virtual space so that workers are encouraged to collaborate with one another. In other words, instead of creating divided personal office spaces (Figure 1), designers should aim to create an “Activity-Based Workplace.” Activity-Based workplaces provide different settings for each work activity.²⁹ Rooms can be created for specific teams of workers (in our case project-based teams and skill-based teams) so that collaboration on shared projects is encouraged by workers being near their peers.

Improve user experience by addressing technical limitations

There are many technical limitations for the adoption of the metaverse technologies for remote work. Staying online while working on technical tasks may require telecommuters to have high-performance computers, which would lead to an increase in cost for organizations. Other issues like audio and video quality also influence people’s experience of using metaverse-based platforms. Gather, the online meeting platform that we used for this study, is still a growing application and does not contain some of the more developed features (ie, virtual backgrounds, polls, advanced settings for meetings) of an older platform such as Skype, Microsoft Teams, WebEx, or Zoom. Additionally, some workers preferred to join the virtual space using a mobile device. At the time of conducting this study, Gather does not fully support this method, which added an additional barrier to entry for some workers. As an emerging technology, many areas for further study exist for the usability of metaverse applications for online collaboration.

Limitations and future research

This study has a few limitations. First, it is a pilot study in a single academic dry lab. The data collected here were specific to this lab and may not be representative of every experience. Next, the first layout of the virtual lab space was implemented by the director of the lab, not by a designer, nor by all lab team members (co-design). To mitigate the impact that this might have had, however, the layout of the virtual space was refined multiple times based on feedback from the rest of the team prior to the conduction of the present study. Finally, other metaverse platforms were not explored to select the best fit for our lab. We chose Gather out of convenient access and ease of use. It is possible that another platform would have solved many of the issues that we faced. It is also possible that other organizations utilizing Gather would not experience the same issues as our team. A comparison between virtual platforms would be helpful to discover which one best supports collaboration in our team. Our future work includes a redesigned lab layout to encourage informal communication and collaboration and an incentive program to promote participation in the virtual lab. The design, implementation, and evaluation of the new virtual lab space will be guided by CSCW literature. The ethical and behavioral impact of such changes on study participants (workers) will be approved by the IRB to ensure that the benefits (eg, work efficiency) outweigh the potential harms in a formal experiment.

CONCLUSION

In this study, we found that virtual collaboration tools can allow for co-location and informal communication among workers but often fall short due to differing expectations and attitudes from workers utilizing the system. Because of this, we recommend that teams looking to implement a similar system for their own workplaces establish common goals and norms for time spent in virtual workplaces, plan virtual space wisely to maximize potential collaboration opportunities, and address technical limitations whenever possible so that users can work seamlessly, without interruption.

FUNDING

This research was supported by the corresponding author’s (Wu) start-up fund and received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

AUTHOR CONTRIBUTIONS

DTYW designed the study and mentored SZ to conduct the data collection. SZ created the figures and drafted the original manuscript. SV and CX revised the original manuscript and SV drafted the final copy and significantly improved the content. CX proofread the revision of the manuscript. All authors discussed the results and provided critical feedback to the content of the manuscript.

SUPPLEMENTARY MATERIAL

Supplementary material is available at JAMIA Open online.

CONFLICT OF INTEREST STATEMENT

None declared.

DATA AVAILABILITY

The data underlying this article, including the survey questions, the survey responses, the work hour schedule sheet, are available in the article and in its online supplementary material.