Conference PaperPDF Available

"I need to be professional until my new team uses emoji, GIFs, or memes first": New Collaborators’ Perspectives on Using Non-Textual Communication in Virtual Workspaces

Authors:

Abstract and Figures

Virtual workspaces rapidly increased during the COVID-19 pandemic, and for many new collaborators, working remotely was their first introduction to their colleagues. Building rapport is essential for a healthy work environment, and while this can be achieved through non-textual responses within chat-based systems (e.g., emoji, GIF, stickers, memes), those non-textual responses are typically associated with personal relationships and informal settings. We studied the experiences of new collaborators (questionnaire N=49; interview N=14) in using non-textual responses to communicate with unacquainted teams and the effect of non-textual responses on new collaborators’ interpersonal bonds. We found new collaborators selectively and progressively use non-textual responses to establish interpersonal bonds. Moreover, the use of non-textual responses has exposed several limitations when used on various platforms. We conclude with design recommendations such as expanding the scope of interpretable non-textual responses and reducing selection time.
Content may be subject to copyright.
“I need to be professional until my new team uses emoji, GIFs, or
memes first”: New Collaborators’ Perspectives on Using
Non-Textual Communication in Virtual Workspaces
Esha Shandilya
School of Information
Rochester Institute of Technology
Rochester, NY, USA
es4524@rit.edu
Mingming Fan
The Hong Kong University of Science
and Technology (Guangzhou)
Guangzhou, China
The Hong Kong University of Science
and Technology
Hong Kong SAR, China
mingmingfan@ust.hk
Garreth W. Tigwell
School of Information
Rochester Institute of Technology
Rochester, NY, USA
garreth.w.tigwell@rit.edu
ABSTRACT
Virtual workspaces rapidly increased during the COVID-19 pan-
demic, and for many new collaborators, working remotely was their
rst introduction to their colleagues. Building rapport is essential
for a healthy work environment, and while this can be achieved
through non-textual responses within chat-based systems (e.g.,
emoji, GIF, stickers, memes), those non-textual responses are typi-
cally associated with personal relationships and informal settings.
We studied the experiences of new collaborators (questionnaire
N=49; interview N=14) in using non-textual responses to com-
municate with unacquainted teams and the eect of non-textual
responses on new collaborators’ interpersonal bonds. We found
new collaborators selectively and progressively use non-textual
responses to establish interpersonal bonds. Moreover, the use of
non-textual responses has exposed several limitations when used
on various platforms. We conclude with design recommendations
such as expanding the scope of interpretable non-textual responses
and reducing selection time.
CCS CONCEPTS
Human-centered computing Collaborative and social
computing.
KEYWORDS
Virtual workspaces; Computer-mediated Communication; Non-
textual Communication
ACM Reference Format:
Esha Shandilya, Mingming Fan, and Garreth W. Tigwell. 2022. “I need
to be professional until my new team uses emoji, GIFs, or memes rst”:
New Collaborators’ Perspectives on Using Non-Textual Communication
in Virtual Workspaces. In CHI Conference on Human Factors in Computing
Systems (CHI ’22), April 29-May 5, 2022, New Orleans, LA, USA. ACM, New
York, NY, USA, 13 pages. https://doi.org/10.1145/3491102.3517514
1 INTRODUCTION
Any successful team collaborating in a workplace requires individ-
uals to have strong interpersonal bonds with each other [
15
,
45
,
58
].
Phutela et al. [
67
] and Bayes et al. [
9
] state that the use of nonverbal
modes of communication like facial expressions, hand-movements,
voice-tones, and smiles assist in conveying interpersonal warmth
that contributes to developing strong interpersonal bonds. Yet many
studies on workplace bonding have primarily focused on the phys-
ical workplaces [
15
,
45
,
58
,
67
]. On the other hand, Driskell et
al. [
26
] found weaker interpersonal bonds when looking at virtual
workspaces due to the absence of nonverbal cues.
Driven by the COVID-19 pandemic, companies have not only
transitioned to completely virtual workspaces, but also had to hire
and on-board employees virtually [
17
,
24
,
27
,
52
,
53
,
55
]. This tran-
sition may be dicult for people who have no prior experience
with online collaborations [
18
]. Moreover, working in a virtual
environment can result in platform fatigue and isolation in em-
ployees [
16
,
30
]. Prior studies have also highlighted the problem
of misinterpreting a message, or an emotion associated with the
message in a virtual setting that limits nonverbal modes of commu-
nication because people do not always have access to other people’s
nonverbal cues unlike when working in physical settings [59, 75].
These drawbacks aect virtual team communication by creating
confusion, which leads to more conicts, and adversely impacts
shared interpersonal relationships in a virtual workspace [
26
]. Mes-
sages can be supplemented with non-textual responses such as
emoji, GIF, stickers, and memes to overcome misinterpretations,
communicate eectively, and increase social connections in virtual
settings [
59
,
63
,
75
,
79
], but the non-textual responses are typically
viewed as less appropriate to use in formal settings [
20
,
78
,
79
]. We
suspect that new collaborators—students or professionals who join
a new team virtually, with no familiarity to the team—likely face
challenges and tensions with using non-textual communication to
connect with colleagues, and there may be design opportunities on
various platforms to support an ecient way for new collaborators
to use non-textual responses in virtual workspaces.
We rst surveyed 49 new collaborators to understand new col-
laborators’ experiences in using non-textual responses in virtual
Permission to make digital or hard copies of all or part of this work for personal or
classroom use is granted without fee provided that copies are not made or distributed
for prot or commercial advantage and that copies bear this notice and the full citation
on the rst page. Copyrights for components of this work owned by others than the
author(s) must be honored. Abstracting with credit is permitted. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior specic permission
and/or a fee. Request permissions from permissions@acm.org.
CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA
© 2022 Copyright held by the owner/author(s). Publication rights licensed to ACM.
ACM ISBN 978-1-4503-9157-3/22/04.. . $15.00
https://doi.org/10.1145/3491102.3517514
CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA Shandilya et al.
workspaces. We dene new collaborator and virtual workspaces as:
Denition 1
(
New Collaborator).
Students or professionals who
joined a new virtual team to collaborate on a project during the
COVID-19 pandemic, without having any familiarity to the team
(e.g., interns, full-time recruits, students working on projects).
Denition 2
(
Virtual Workspaces).
Collaborative technologies
like Slack and Discord that connect and enable remote workers
of an organization to communicate, work, and achieve common
goals [87] in a virtual environment.
Our survey ndings indicated that new collaborators were using
non-textual responses, but often faced many challenges in using
those. They found their preferred method of communication was
not always supported by work platforms, the meaning behind non-
textual responses was not always clear, and unfamiliarity with the
team or organization’s work culture resulted in hesitation to use
non-textual responses.
We then ran interviews with 14 new collaborators to further
understand personal barriers, communication strategies, and tenets
of eective non-textual communication in the context of virtual
workspaces and new colleagues. Similar to previous work, we found
communication is essential for the team to thrive. More specically,
we uncovered the resourcefulness of new collaborators in using non-
textual responses eectively such as taking the lead from others
and mimicking their behaviors.
We make several design recommendations to improve how new
collaborators integrate within virtual workspaces. It is necessary
to expand the scope of non-textual responses, improve the inter-
pretability of non-textual responses, and support faster discovery
of non-textual responses.
Our contributions are as follows:
i)
We present a qualitative
research study that identies pain-points, coping strategies, and
needs of new collaborators in using non-textual responses in virtual
workspaces.
ii)
Elicit knowledge about the use of various modes
of non-textual response by new collaborators to establish interper-
sonal bonds in virtual teams.
iii)
Make design recommendation to
enhance the experience of using non-textual responses for promot-
ing interpersonal bonds in virtual workspaces.
2 RELATED WORK
Workplace communication can take many form, such as through
audio calls [
36
], blogs [
43
], emails [
62
], social media [
25
,
64
],
text messages [
54
], and video calls [
11
] (including with parallel
chat [
70
]). Collectively, those earlier studies [
21
,
25
,
43
,
54
,
64
]
demonstrate the value dierent communication mechanisms have
on enhancing social connectedness among team members and
positively inuence team bonding at the workplace; however, the
research was conducted pre-pandemic, and where possible/feasible
(i.e. fully co-located teams) interaction among team members would
be in face-to-face settings, with blogs, social media, or text messages
functioning as an additional means of connecting with colleagues.
In contrast, virtually connecting and working has become an
increasing priority, and sometimes the only feasible option in a
pandemic era. Prior studies [
12
,
21
,
25
,
43
,
45
,
54
,
64
,
67
] examined
the impact of physical and virtual workplace communication av-
enues on teams’ bonding without focusing on new collaborators
(see Denition 1). Moreover, past works [
21
,
25
,
43
,
54
,
64
] did not
study new collaborators’ use of non-textual responses to connect
with the new team (without any prior familiarity with them) in a
virtual workspace (see Denition 2).
Therefore, the focus of our paper is to study new collabora-
tors’ perceptions of using non-textual responses in a virtual-only
workspace to form interpersonal bonds. We specically studied non-
textual responses in a virtual work setting because past work has
shown that non-textual responses in text-based chat systems substi-
tute face-to-face gestures [
59
] (that lack in a virtual setting) and pos-
itively inuence relationships among people in an informal virtual
setting [
33
,
44
]. Furthermore, text-based chat systems (e.g., Slack)
provide an alternative to avoid video-calling fatigue [
7
] and support
asynchronous collaboration [42] in a virtual-only workplace.
Our related work provides background context on physical
vs. virtual workspace, the importance of interpersonal bonds for
healthy work environments, and pre-pandemic views of non-textual
communication in the workplace.
2.1 Value of Interpersonal Bonds at
Workspaces
Within the context of work, positive interpersonal bonds between
colleagues foster a safe and conducive learning environment, which
contribute to a team’s eciency [
15
]. Whereas, negative interper-
sonal bonds shared by workers due to team conicts could induce
stress, thereby making individuals more reluctant to work [58].
Interpersonal bonds can be dened by perceptions of strong
personal and social relationships and feelings of belongingness, and
they are essential for helping individuals assimilate to a specic
group and their norms [73].
People can use nonverbal communication to assist in express-
ing emotions unambiguously [
9
], which is crucial for developing
strong interpersonal bonds in a workspace [
12
,
45
,
67
]. However, in
the current COVID-19 pandemic [
17
], there has been a shift from
physical to virtual workspaces [
27
,
52
,
55
,
70
], which signicantly
changed the work dynamic and removes traditional nonverbal com-
munication (e.g., facial expressions and tone of voice). Much of the
prior work has only focused on the value of interpersonal bonds in
a physical workspace [
15
,
45
,
58
,
67
,
73
], whereas new collaborators
are now being recruited to work in virtual workspaces [24].
2.2 Non-textual Forms of Communication
Promote Interpersonal Bonds Virtually
Virtual workspaces utilizing chat and audio-only systems remove
the typical means of communicating nonverbally [
26
], which results
in challenges when interpreting the emotion associated with the
text/audio and lack of information-rich feedback [
59
,
75
]. Driskell et
al. [
26
] state that teams formed on virtual workspaces share weaker
interpersonal bonds due to the challenges of computer-mediated
communications as compared to the teams that have interacted and
formed connections at physical workspaces.
To address these challenges, non-textual forms of communica-
tion like emoji, GIFs, emoticons, memes, audios, videos, and images
may be used to enhance communication and convey emotions
New Collaborators’ Perspectives on Using Non-Textual Communication in Virtual Workspaces CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA
clearly [
59
,
75
,
79
]. The increased use of non-textual forms of com-
munication can evoke intimacy and a feeling of social connect-
edness between the users [
33
,
44
,
48
,
84
]. Thus, establishing that
frequent use of non-textual communications can promote interper-
sonal bonds in virtual settings.
2.3 New Collaborators in a Virtual Workspace
An under-researched perspective—and one that has now become
highly relevant—is to understand new collaborator’s insight on the
role of non-textual communication in forming interpersonal bonds
within virtual workspace.
Prior work [
33
,
44
,
48
,
84
] did not adequately explain how
unacquainted individuals would use non-textual modes of
communication to form bonds in virtual work settings; although
Gesselman et al. [
33
] did explore the eect of emoticon usage
on the relationship between potential romantic partners, those
scenarios were personal. Furthermore, if we focus on emoji, which
are an extremely popular mode of non-textual communication and
the successor to emoticons, people often report that they try to
avoid using emoji in work environments and with people who do
not share close bonds. As people have varied interpretations of
emoji and they consider it a non-serious and unprofessional way
of communication [5, 41, 63, 78, 79].
However, those views toward unprofessionalism may no longer
be as relevant with the transition to virtual work environments and
the need for new collaborators to connect with colleagues. There
are also many dierent types of non-textual modes of communica-
tion (e.g., GIFs, stickers, videos, memes), which may help to serve
dierent purposes.
2.4 Research Questions
We aim to understand the experiences of new collaborators in using
non-textual communication to form interpersonal bonds within
virtual teams by answering the following research questions:
RQ1:
How do new collaborators use non-textual modes of
communication to form interpersonal relationships in a vir
tual workspace?
RQ2:
Do new collaborators face any challenges in using non-
textual modes of communication in a virtual workspace? If
yes, what are those challenges?
RQ3:
What coping strategies do new collaborators employ to
overcome the challenges, if any, in using non-textual modes
of communication in a virtual workspace?
RQ4:
What design recommendations can be incorporated
in existing virtual platforms based on new collaborators’
communication strategies to promote interpersonal bonds?
-
3 NEW COLLABORATORS SURVEY METHOD
We rst distributed an online survey to understand new collabo-
rators’ experiences in using non-textual modes of communication
in virtual workspaces. We studied the experiences of new collabo-
rators in using non-textual responses as they pleased but did not
specically focus on non-textual responses with or without text.
3.1 Materials and Procedure
We advertised our questionnaire on social platforms (e.g., LinkedIn,
Facebook, Reddit) and university Slack channels during a four-week
period to collect responses. The advertisements directly dened
our target participant group—new collaborators (see Denition 1).
Participants who indicated being new collaborators were the only
ones who were prompted with the survey questions. Participants
could complete the questionnaire in their own time. There was no
reimbursement oered for completing the questionnaire.
We had 23 questions in total (19 close-ended and 4 open-ended)
Our questions were centered around collecting: i) demographic
information (age, gender, prociency with computer-mediated com-
munication in virtual work, occupation, familiarity level with the new
virtual team), ii) virtual platforms used by their team to collaborate,
iii) familiarity with the virtual platform, iv) if they employ non-
textual communication in a professional setting (and why), v) their
most preferred non-textual response types, vi) if they were hesi-
tant in using non-textual modes of communication (and why), and
vii) any challenges experienced while using preferred non-textual
communication types on virtual platforms. To ensure the validity
and clarity of survey questions, we did a read-through session with
a few researchers (not part of our research team). Furthermore, we
did not receive any responses from participants that indicated con-
fusion about what we were asking. Our questionnaire participants
could also sign up for a follow-up interview (see sections 5 and 5.3).
3.2 Participants
We had 58 participants in total, but removed 9 because they did
not acknowledge being a new collaborator. Our remaining 49 new
collaborators (Female = 26, Male = 22, preferred not to say = 1)
were aged between 18-64 year old. Out of 49 new collaborators 42
rated their familiarity to the team between 1 to 3 on a scale of 1 to
5, where 1 = ‘I did not know anyone in the new team. and 5 = ‘I
knew everyone in the team.
Our participants mostly represent younger age groups (18-34).
Specically, 18 new collaborators were 18-24 years old, 28 were
25-34, two were 35-44, and one was 55-64. We believe our group is
skewed toward a younger population because people in the 18-34
age bracket are more likely to start new jobs compared to older
people. The most frequent new collaborator category reported was
full-time employees (31 out of 49) followed by interns (7 out of 49).
We assigned the survey participants an ID letter S followed by a
number (e.g., S1, S2, S3).
3.3 Analysis
We report descriptive statistics for the close-ended responses and,
for the open-ended responses, we performed open coding [
80
].
The open coding steps were to rst familiarize ourselves with the
data by reading through all the open-ended responses. We then
assigned initial codes to the responses, then iteratively grouped
similar data points to generate high-level categories that could
provide a summarized understanding of the data.
4 NEW COLLABORATORS SURVEY FINDINGS
We present our survey ndings, using closed and open-response
data and participant quotes, under two themes: i) New Collaborators
CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA Shandilya et al.
Using Non-textual Responses and ii) Challenges New Collaborators
Face in Using Non-textual Responses.
4.1 New Collaborators Using Non-textual
Responses
We found that chat-based systems are the most used mode
of communication to collaborate with team members (40/49
respondents), as shown in Table 1. Further, the responses revealed
that our participants used a variety of platforms to ensure they
communicate eectively: Slack was most common (25/49), followed
by Microsoft Teams (17/49), and many other platforms.
All the respondents mentioned that they are comfortable using
computer-mediated communication (e.g., Slack, Google Hangouts,
audio calls, etc.) to communicate and collaborate on professional
projects. Further, we found that 38/49 respondents had high fa-
miliarity with the platform (rating themselves a 4 or 5, where 5
is I was extremely familiar with the communication platform when
I joined’) and felt condent (responding agree/strongly agree) in
using a virtual platform to communicate and collaborate.
We found that 34/49 participants used non-textual responses in
virtual workspaces. However, 34/49 were also hesitant in using non-
textual modes of communication with their virtual team, which
validates our hypothesis that new collaborators face challenges
when using non-textual response in their virtual workspace. Ad-
ditionally, the most favored and most used mode of non-textual
communication reported was emoji (36/49), which might be be-
cause of the diversity of options and it is easy to use with text. GIFS
(21/49) and memes (16/49) while being more visually engaging were
preferred after emoji.
We found many reasons for using a non-textual response in a
virtual workplace. Participants (11/49) mentioned that they prefer
using it to add humor to their conversations, for example:
S2: To make the conversation a little lighter and to indicate if some-
thing is funny or hilarious.
Similar to prior work, participants (18/49) also recognized that
non-textual responses helped to convey a wide range of emotions,
moods, and intent with a message. Some of them also reported that
non-textual responses can soften the tone (6/49) and support quick
acknowledgments (8/49). For example:
S19: I feel that they help convey emotion and provide context in
text conversations, where feelings are not as easily expressed as in
video calls. After a while on the team, it was also a means of having
a laugh or adding avor to casual conversations and getting to know
one another. For messages with a wider audience, they served as a
quick way to express agreement, looking into something or give a
quick response to indicate that attention has been given to a message
someone has put out.
Two of our participants (S3 and S25) also felt using non-textual
responses helped them feel more comfortable and connected with
their colleagues while communicating virtually.
S25: It helps to communicate emotion and create more meaningful
connections with your team. It gives the conversation a more human
and connected feel.
In addition, S48, used non-textual responses to conform to the
team’s practice of using it, for example:
S48: Other teammates and upper management uses them...
Table 1: The frequent modes of communication used to col-
laborate with team members virtually.
Modes of Communication No. of Participants
Chat Systems 40
Video Calls 36
Audio Calls 24
Emails 2
Text Messages 1
Mails 1
Webex Meetings 1
4.2 Challenges New Collaborators Face in
Using Non-textual Responses
Our online questionnaire also asked several open-ended questions
to understand the hindrances and hesitations that new collaborators
face in using non-textual responses. We summarize those insights
that new collaborators shared in using non-textual responses in
virtual workspaces.
4.2.1
Unfamiliarity:
Many participants (16/49) cited that the fear
of not knowing the team when they join acts as a hurdle in using a
non-textual response in a virtual workspace. Consequently, people
feel restricted in using a non-textual response because they are
unsure about the team/organization’s culture, resulting in appre-
hension about how they would be perceived, for example:
S21: You don’t really know these people, and so you want to give
o the right meaning of the message. Since we work remotely, what
makes sense to you may not for them... especially since I was new,
they did not know my personality.
The quote mentioned above portrays the participant’s skeptical
attitude towards using a non-textual response in a new team
virtually because the participant is new to the team and worried
if the use of a non-textual response is misconstrued.
Furthermore, S53 also shared that they are unaware of the pur-
pose of using non-textual responses at the workplace makes them
hesitant to use non-textual responses, for example:
S53: I’m very less familiar with these emojis infact till today also I
don’t know the actual need of some emojis.
4.2.2
Personal Biases and Prior Experiences:
The partici-
pants responded that they hesitate to use a non-textual response
because of their personal barriers, such as their bad past experi-
ences when using a non-textual response with a team member. A
few participants (3/49) were even more uncertain about using a
non-textual response when they were supposed to communicate
with team members in senior roles. For example, S22 answered to
the question, why were you hesitant to use non-textual modes of
communication like emojis, gifs, stickers, and memes with your
team virtually?
S22: Whenever communicating with the manager or any other
person at the higher levels.
New Collaborators’ Perspectives on Using Non-Textual Communication in Virtual Workspaces CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA
One of the most recurring reasons to not use a non-textual
response is that many participants (6/49) perceive using it as
unprofessional at work, which resonates with the ndings of
previous work [
5
,
78
]. As a result, people desire to project
themselves as professionals and convey their seriousness about
work by not using a non-textual response, for example:
S10: It’s not professional, so when joining a new team, it seems
inappropriate to use gifs/stickers when others do not. If I work with
this team longer, I’m sure I will start using them once I know everyone
well.
In the above quote, fear of being unprofessional stems from
unfamiliarity with the new team. S10 further mentions the goal is to
comply with their team’s usage of non-textual responses. However,
it is interesting that S10 would use them after working for more
time and eventually knowing the team.
4.2.3
Technological Barriers:
A few participants (S2 and S45)
mentioned that a few platforms are not compatible to support non-
textual responses of their choice when asked about the challenges
faced by them in using non-textual responses, for example:
S2: I use Microsoft teams which doesn’t support most of gifs.
Moreover, a few participants (3/49) also stated that nding a
non-textual response that could express their true emotions on the
platforms is hard:
S18: Sometimes, it is hard to nd the one emoji or GIF that expresses
what I try to communicate.
Additionally, S45 shared that some platforms automatically sug-
gest a non-textual response when they are not looking for one
while communicating, for example:
S45: [...] also if we type any text, it’s translated to the emoticon when
it is not required.
4.3 Summary of Survey Findings
Our survey revealed that most of our survey participants prefer
chat systems and use non-textual responses with texts to express
emotions, acknowledge, and connect with others. However, 34/49
new collaborators struggled using non-textual responses, especially
when new to the team, which also answered RQ2 partially that new
collaborators face hindrances in using non-textual responses to
connect with the team.
From our survey, we still did not have a deep understanding
of: i) how new collaborators use non-textual responses to bond
with the new team, ii) other personal and platform barriers in
using non-textual responses, iii) coping strategies to overcome
those, and iv) what is the best way to support eective non-textual
communication that fosters interpersonal bonds. Therefore, to
answer those questions and understand the experiences of new
collaborators in detail, we conducted online interviews.
5 NEW COLLABORATORS ONLINE
INTERVIEW METHOD
We conducted interviews to further understand the use of non-
textual responses to form interpersonal relationships with the team,
reasons for new collaborators’ personal barriers, coping strategies,
and design opportunities for an eective non-textual communica-
tion for new collaborators. The research questions mentioned in 2.4
informed our interview study.
5.1 Materials and Procedure
We conducted semi-structured interviews on Zoom and recorded
sessions for data analysis with participant consent. Out of fteen
interviewees (1 pilot, 14 participants), two did not share their videos
during the interview.
Our semi-structured interviews were designed so that we could
ask: i) clarication questions to either the interview pre-screener
for new participants or the original questionnaire for returning
participants (see sections 3 and 4); ii) their motivations for using
non-textual modes of communication and context of use; iii) if
there are challenges related to using non-textual responses and
what coping strategies are employed by the participants; iv) if
non-textual communication has helped the participants to create
interpersonal bonds with their colleagues in virtual workplaces;
and v) what principles do the new collaborators consider to be
crucial for eective non-textual communication.
We ran one pilot interview to ensure that our interview guide
was clear, that questions made sense, and to check our planned
interview would not run over the scheduled time. We also made
sure that the main interviews focused more on topics not covered
in the online questionnaire or if we still did not have a clear picture
for some of our previous inquiries. The mean recording time for
our interviews was 47.6 minutes (min = 27, max = 72).
5.2 Participants
We interviewed 14 participants (Female = 8, Male = 6) aged between
18-44 years old. Table 2 presents the demographic details of all the
interview participants. Our interviewees were participants who
scored a low familiarity rating between 1 to 3 on a scale of 1 to 5
(where 1 = I did not know anyone in the new team. and 5 = I knew
everyone in the team.’) and who started working with a new group
of people virtually during COVID-19 (i.e., only knows colleagues
virtually). We selected candidates with low familiarity ratings to
further understand our survey results indicating the eects of low
familiarity on using non-textual responses. Seven interviewees
were not from our original questionnaire, but they completed a pre-
screener questionnaire before we selected them for an interview.
5.3 Analysis
We followed the Braun and Clarke’s thematic analysis ap-
proach [
13
] to identify the themes and relationships among the
qualitative data. After the data-gathering phase, the rst author
listened back over the interviews and produced transcripts to
get familiar with the data. The next step was an iterative process
that involved coding the data for key insights from interviewees’
responses. Next, the rst author investigated underlying patterns
among the initially formed categories to classify the data into
high-level categories, while consulting with the research team.
Consequently, we looked at all the categories formed in the last
step to arrive at a nal set of themes and sub-themes. Braun and
Clarke’s highly-cited method does not call for multiple coders
or inter-rater reliability for good thematic analysis, which is also
CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA Shandilya et al.
Table 2: This table represents the demographic details like participant ID, age, gender, job role, team size, and joining period
of all the interviewees.
ID Gender Age Range Role Team Size Date of Joining
Pilot Male 25-34 Full-time Network Consultant 32 May 2020 - August 2020
P1 Male 35-44 Full-time Software Developer 25 to 30 May 2020 - August 2020
P2 Female 18-24 Ph.D. Student 5 September 2020 - December 2020
P3 Male 18-24 Full-time Data Science Associate cohort structure September 2020 - December 2020
P4 Male 25-34 Ph.D. Student 4 to 7 September 2020 - December 2020
P5 Male 25-34 Ph.D. Student 15 to 20 September 2020 - December 2020
P6 Female 25-34 Full-time Network Engineer variable team size March 2020
P7 Female 25-34 Software Developer Intern 6 to 7 May 2020 - August 2020
P8 Male 25-34 Full-time UX Designer 6 September 2020 - December 2020
P9 Female 25-34 Full-time Financial Analyst 12 to 15 May 2020 - August 2020
P10 Female 25-34 Full-time Software Developer 6 May 2020 - August 2020
P11 Male 25-34 Full-time Network Engineer 12 September 2020 - December 2020
P12 Female 18-24 Student/Research Assistant 3 to 4 September 2020 - December 2020
P13 Female 25-34 Part-time UX Designer 6 September 2020 - December 2020
P14 Female 18-24 Student/Research Assistant 6 September 2020 - December 2020
emphasized by other qualitative researchers [
66
]. The interview
participants are assigned an ID letter P, e.g., P1, P2, and so on.
6 NEW COLLABORATORS ONLINE
INTERVIEW FINDINGS
Through our thematic analysis, we answer our three research ques-
tions (RQ1, RQ2, and RQ3) under three themes: i) Using Non-textual
Modes of Communication to Form Interpersonal Bonds , ii) Chal-
lenges in Using Non-textual Modes of Communication to Form
Interpersonal bonds, and iii) Coping Strategies to Overcome the
Challenges in Using Non-textual Modes of Communication to Form
Interpersonal Bonds. We now discuss the narrative of each theme
in detail using quotes from participants.
6.1 Using Non-textual Modes of
Communication (RQ1)
This section explores unique methods adopted by participants to
use non-textual responses to form interpersonal bonds in a vir-
tual workspace. Similar to the insights gained from survey partici-
pants 4.1, all interview participants also used non-textual responses
to convey emotions accurately, mimic other people’s energy, and
conform to their team communication practices.
One of our participants, P1, mentioned that using non-textual re-
sponses with text in virtual workspaces was an outcome of the new
normal; people did not have access to any predened rules to com-
municate in virtual workspaces. Therefore, they had invented ways
to communicate eectively in the virtual workplace, for example:
P1: The emoji and then the text communication is a subset of this
new expected cultural norm of the accepted behavior for a company,
that’s now online [...] You have an expectation that’s preset on, and
we’ve been trained; you watch a movie you’re learning like oh that’s
how people behave in an oce, you know, and now it’s not like there’s
a movie that shows people behaving on slack, so like the behavior,
dierent people really varies [...]
Consequently, in the following sections, we explored unique
ways like, Gamifying Explicitly, Being Funny to Break the Ice, and
Enriching Textual Messages adopted by new collaborators to use
non-textual responses to form interpersonal bonds within their
virtual teams.
6.1.1
Gamifying Explicitly:
P2, shared how their team mem-
bers used personalized emoji to acknowledge the messages in the
communication channel, for example:
P2: When I joined, someone told me we all had chosen a specic
emoji, and whenever we have a recommended message, oh, I’ve read it
instead of like All responding separately. We respond with our personal
emoji, so that’s a super formal way, my favorite emojis were already
taken when I joined the group, so I chose, in the end, a fun video emoji
with like a star...
P1 and P9 also used emoji in their bio to imply a certain interest
that they wanted to pursue, to make quick decisions in a workplace
contest (P7), or for naming documents and products instead of using
their textual names (P13). P1 suggested using emoji for employee
ratings.
A few participants (P1, P7, P9, and P10) also mentioned they used
a GIF or sticker to represent themselves and their personalities at
the workplace. One of the participants, P1, mentioned that if an
employee used a certain emoji, then other individuals did not use
it because they felt it might be copying him.
P1: There is an emoji that’s kind of like that this guy used to send to
everybody, and it became kind of like his thing...And then now do you
know we don’t see that emoji thrown around anymore, and feel like
people don’t want to use it, because then you know; it’s copycatting
that guy
New Collaborators’ Perspectives on Using Non-Textual Communication in Virtual Workspaces CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA
In contrast to P1’s statement another participant, P7, said they
would want to use that specic emoji to connect with the person
quickly.
6.1.2
Being Funny to Break the Ice:
Many times participants
(P1, P7, P8, and P13) reported their teammates would post random
jokes or were funny to let them feel comfortable. To convey the
intent of being fun or humorous, team members used non-textual
responses like emoji and GIFs. For instance, the team lead posted
random jokes rst thing in the morning, and other team members
were supposed to react with an emoji from a set of emoji provided
by him.
P8: My team lead posts some kind of random joke every day on like
one the platform, so what he did was like he posts a joke there, and
he would add random emoji at the bottom with the post, and would
want us to vote using those emoji.
The above instance was one of the many examples where teams
tried adding humor through non-textual responses to their mes-
sages to let the new collaborators open up.
6.1.3
Enriching Textual Messages:
Participants (P1, P4, P5, P6,
P8, P9, P12, and P13) claimed that being able to use non-textual
responses in communication gave them a sense of sharing a com-
radery with their teammates. P4 also shared that it was only through
using non-textual responses in chats they were able to transition
from chats to video calls with a new team member, for example:
P4: Then slowly, then, after some time we transcended from emails
to phone calls and then to video calls like four ... And that happened
because I would sometimes use SMILEY or some emoji and they would
also reciprocate [...]
Moreover, participants (P1, P4, P5, P6, P8, P9, P12, and P13)
felt using non-textual responses promoted interpersonal bonds
that also helped their mental well-being by gaining a feeling of
companionship, for example:
P9: Yes, I would say that because previously in my last company I
have made some great friends and, and I did not met with them like
face to face ever, but I, but I still talk to them and I haven’t had any
calls with them only messages and right now, at this company, I never
had, I never had any physical Like physical connection with like a
with like anyone [...] so I would say yes, non-textual responses have
helped me in in building the relationships and I have made some great
friends in my previous [and] current company.
6.2 Challenges (RQ2)
Our survey study conrmed that new collaborators face challenges
in using non-textual responses 4.2 to form interpersonal bonds in
a virtual workspace. We also slightly discussed a few reasons for
those such as, Unfamiliarity, Personal Biases and Prior Experiences,
and Technological Barriers in section 4.2. This section extensively
examines those challenges on the basis of gathered interview data.
6.2.1 Unfamiliarity.
Similar to survey participants, even interview participants (P1,
P2, P4 - P14) also felt indecisive and hesitant to use non-textual
responses because they were new to the team. We found they were
hesitant in using non-textual responses because they did not share
a personal bond, for example:
P8: I can say, when I was not very familiar with the team, I was
hesitant to use dierent smileys. Like you know, I’m kind of like shy
in nature when it comes to like sharing a heart SMILEY with others. It
is sometimes dicult for me, so yeah, I was rst like very hesitant to
put a heart. But later as I get to know people, yeah they are like kind
of cool... I don’t mind using it now.
Moreover, almost all the participants were unsure how others
perceived their use of a non-textual response in the team; a lot of
this fear to use a non-textual response in a new team also related
to the non-textual response’s openness to interpretation based on
individuals’ personal biases and prior experiences (P4 and P13).
Consequently, such dierences in understanding a non-textual re-
sponse created misunderstandings and hurt the sentiments of team
members, as shared by P2 in the example below:
P2: There is this thing within a certain group they have like a norm
that they don’t use the praying emoji for thank you because they do
not want to use any religious emoji...so they like own a thank you
emoji and...when I entered, I didn’t know this, so I used the praying
emoji for Thank you emoji, and then some people sent me a message
saying I should use the other non-religious thank you emoji.
6.2.2 Personal Biases and Prior Experiences.
Personal Idea of Professionalism:
The perceived understand-
ing of the professional way of conducting at the workplace was
scattered. It revolved around the fact that using a non-textual re-
sponse at a virtual workplace would make the participants (P3, P9,
P10, and P11) look unprofessional. P3, also evaluated that using
fewer words and more non-textual responses in a chat-based com-
munication system showed one’s incompetency to convey ideas,
for example:
P3: [...] when you try to communicate a lot using a limited medium
like a job is essentially an instant messenger that comes from the idea
that you can’t really write letters and expect people to receive it and
all of that right [...]
The reason for contradicting ideas about professionalism was of-
ten rooted in diverse cultural backgrounds, as shown in the example
below:
P9: I would say, for using stickers and memes, they are fun, but
I don’t use those with like everyone basically I use these GIFs and
stickers approximately with only like four to ve people...I keep my
professional distance...if you would say because nobody uses those I
have back in India, so I am on the fence with that [...]
Therefore, the concept of professionalism was diverse and
depended on cultural background and exposure to dierent work
cultures.
Dierences in Seniority Age and Experience:
Younger par-
ticipants (P3, P9, P10, and P13) aged between 18 to 35 avoided using
non-textual responses with older adults at the workplace. They did
not want to take the liberty of jeopardizing their relationships with
older individuals as there was a possibility that the older adults
might not be aware of trending ways to communicate non-textually,
for example:
P9: Most people, people in my team, are above 50 or 60, so it’s a
little dicult to send like meme, stickers it can mean like something
dierent if you are not sending to the people of your own age...Because
CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA Shandilya et al.
I did not like I uh I did not get any response back, so I thought, okay,
so maybe I should not have sent that; otherwise, she would have been
responded to me...it’s a generational gap.
Moreover, a few participants (P1, P4, P5, P6, and, P12) were
more cautious in phrasing their messages and using a non-textual
response when communicating with a team member at senior roles
as they were apprehensive that it could aect their relationship
negatively, for example:
P6: The very rst time, I couldn’t use all this because I was not aware
Like if it was actually a good thing to do so, even with the manager, I
still can’t do it... though things are smooth and all but I still can’t do
it or use emoji while chatting with my manager or some other person
in the senior management, it is just that we use simple SMILEY.
The above examples demonstrated the role of age and seniority
level in determining if the non-textual response should be used.
Conformity Bias Idea of Fiing in:
P10 and P12 tended to
conform to their team’s communication practices, for example:
P10: I haven’t seen a lot of people using them already so you have to
conform with or whatever is going on in the company right, I cannot
be like Oh, you know I’m so friendly and everything and start using
it.
From the above example, we see that even when the participants
(P10 and P12) wanted to use a non-textual response, they complied
with the usage pattern of their teams or organizations.
Personalized Meanings:
The reason for varied understanding
of a non-textual response was based on participants’ association, ex-
perience, and perception of a specic non-textual response. Conse-
quently, participants dened their own denitions of a non-textual
response based on age (P9, P12, and P14), personal associations due
to prior experiences (P2, P13, and P14), and cultural backgrounds
(P4 and P9), for example:
P14: I actually mostly have that problem with the thumbs up analogy
because, like, in my family like if I use that with my mom she would
think I was really passive aggressive and like vice versa, my mom
used that with me and actually the rst time, my supervisor used it, I
thought she was mad at me, and then I realized that was her version.
Many times non-textual responses without explicit words on
them were hard to understand for participants (P4 and P13) in cul-
turally diverse teams. It is dicult to comprehend the meaning
of non-textual responses because people from dierent cultures
express emotions using non-textual responses dierently. For exam-
ple, one of the participants, P4, shared that his co-worker preferred
to thank him through a crying sticker, which puzzled the partici-
pant; the following quote captures the instance:
P4: So he would say thank you over the video, and after a few minutes
he would send a giphy...he would send that, and that sticker would be
crying. And it would seem that that person is sad, and I wouldn’t say
anything like for a few days, but then I noticed this kept happening
again. I was totally clueless...and then it turns out that he was sending
like over happiness, which I thought, like, ‘Why is he crying and why
is the person sad?’
The above anecdotal stories of participants P14 and P4 suggested
that a person’s perception of a non-textual response also depended
on their personalized experiences and the impact it had on them,
apart from their cultural background knowledge and experiences.
6.2.3 Technological Barriers.
Time-consuming Search and Select:
Another challenge faced
by the participants (P3, P5, P10, and P14) was the lack of optimal
features to nd the meaning of non-textual responses and using
them quickly, for example:
P5: Yeah, I mean one thing is that there are too many emoji and I
have to if I had to use a new one, I had to go and search what it means,
and then use it. So most of the times, I’m aware that there is a hover
over feature, it tells the meaning, but it’s just that there are too many.
So, I think it’s just a preference that I tend to not look over it since,
as I mean since I’m a Ph.D. student, I tend to have a short amount of
time to get things done in a short amount of time.
Moreover, for interactive collaborations, participants (P7, P12,
and P14) found the existing features either extremely limiting or
challenging for users to discover and use; for instance, P14 men-
tioned that even when people were aware that there was a raise
hand feature on video-calling platforms like Zoom, no one noticed
it when one used it:
P14: I think when people use the raise hand feature, nobody ever
looks at it.
Ineective Recommendation:
P13, shared that keyword
searches to locate an appropriate non-textual response matching
a user’s intent were often poorly met by the platforms. In other
words, platforms failed to suggest/recommend usable non-textual
responses in many instances. This was further aggravated by the
limited non-textual responses and the platform-specic keywords
to retrieve a specic non-textual response.
P13: When I want to search some like keywords, but I cannot get
the emoji, I don’t know if they have, like several keywords for emoji
or there’s just like a naming for the emoji say if you give one emoji
several keywords I can search it very quickly... Sometimes when I type
something, especially for the emoji. I cannot get the result. I can just
see no results.
6.3 Coping Strategies (RQ3)
In this thematic section, we present coping strategies adopted by
participants to overcome the challenges (Unfamiliarity, Personal
Biases and Prior Experiences, and Technological Barrier: Ineective
Search and Select) in using non-textual responses to form interper-
sonal bonds in a virtual workspace.
6.3.1 Unfamiliarity.
To cope with the problem of unfamiliarity within the team, par-
ticipants preferred to Observe and Adapt, Soften the Tension, and
Use Selectively and Progressively a non-textual response when new
in the team.
Observe and Adapt:
Participants (P1, P2, P4, P5, P6, P8, P12,
and P13) tried to ensure replying using appropriate non-textual
responses that would be acceptable and palatable to their collabo-
rator. They keenly observed the team’s communication practices
and adapted accordingly. This allowed them to slowly warm up to
using non-textual modes of communication within their team.
New Collaborators’ Perspectives on Using Non-Textual Communication in Virtual Workspaces CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA
Usage of non-textual response by people in senior roles further
alleviated their (P7, P8, and P13) apprehensions in using non-textual
responses, for example:
P13: [...] for emoji, I think she’s the one who made our team very
emotionally attached. She started using emoji, and everyone was free
to use emoji because I hesitated about using emoji. After all, it expresses
our emotions our attitudes, and someone could misunderstand it. I
was very cautious about using emoji. So, when your senior, as in your
project managers, started to use or gave you some sense of comfort
that Okay, maybe others could use, then, then, you start using that.
Soen the Tension:
P11, mentioned using non-textual re-
sponses like smileys in emails when reaching out to a dierent
team member helped soften the tension of unfamiliarity between
them. The quotation from the P11 is as follows:
P11: I ensured to use two smileys at least when emailing my team
mates. It softens the tension that we have because we don’t know each
other.
Use Selectively and Progressively:
We found that participants
(P1, P4, P7, and P9) used various types of non-textual responses
such as emoji and GIF dierently. When new in the team, partici-
pants started with emoji, and gradually progressed to using more
expressive non-textual responses such as GIFs when they devel-
oped closer relationships through spending time on projects and
learning common interests. For example:
P7: I think, in general, even with my team I just I started out with
emojis and get pretty crazy and then I gured out if the person and
I had the kind of relationship that was casual enough for sharing
because there’s denitely GIFs that are like very funny funny like
those are the kind that you will interact with a friend using.
P9: I would say, for using, for using like a stickers and memes, they
are, they are fun, but I don’t use use those with like, with like everyone,
basically I use these GIFs and stickers approximately with only like
four to ve people...but the people who I am like comfortable with,
whom I chat like after oce, if I share my personal things with them
to, with those people, I’m pretty comfortable and I share the stickers
and memes with them.
Moreover, participants (P2, P4, and P7) felt GIFs help a sender
share common knowledge about a popular trend or TV show
whereas emoji just lets a user express a xed emotion, for example:
P7: There is a lot more information that a GIF carries compared
to an emoji...it is just a xed kind of expression or emotion devoid
of any context of why, like a person is feeling that way, except for
like what it is that you are talking about in the conversation.
Therefore, when collaborators were new, they made selective
and careful choices to use a particular type of non-textual response,
mostly emoji, and as they spent time and learned about their teams,
they progressively used another form of non-textual response, GIFs.
6.3.2 Personal Biases and Prior Experiences.
To cope with the problem of varied interpretations of non-textual
responses participants formed due to personal biases and their prior
experiences, new collaborators Reach Out for Clarication and Use
Text-based Non-textual Responses.
Reach Out for Clarication:
A general approach followed by
participants (P2 and P13) was reaching out to the sender when they
sought clarication. Additionally, they reected on their actions
if they triggered the sender, as shown in the example below:
P13: Young people in China use slightly smiling emoji when they
do not care or are speechless, and you do not wish to engage...once a
Prof. from China used and I had no idea what he means...I asked for
clarication, suggestions to improve if something was not right.
Use Text-based Non-textual Responses:
Moreover, a few par-
ticipants (P4, P9, P13, and P14) opted for explicit text-based stickers
to avoid using culturally diverse non-textual responses that could
be interpreted dierently in a culturally diverse team.
6.3.3
Technological Barrier: Ineective Search and Select
.
Use a Closest Matching Substitute:
Many participants (P3, P5,
P10, P13, and P14) struggled to locate an appropriate non-textual
response due to ineective search and select process on communi-
cation platforms. Therefore, a few participants (P8 and P10) settled
with non-textual responses that were the closest match to their
intent. They used the rst non-textual response that appeared on
the search results to save time, for example:
P8: When I do not get a particular emoji I nd a closest match to an
emoji available on Mattermost that I can use to substitute.
6.4 Summary of Interview Findings
Our ndings revealed non-textual responses were used in creative
and resourceful ways by new collaborators to build interpersonal re-
lationships within their virtual teams. However, new collaborators
found using non-textual responses at virtual workspaces challeng-
ing due to personal hesitations and technological hindrances. We
further examined the coping strategies employed by new collabora-
tors to tackle various challenges faced by them. As a consequence,
in the next section, we discuss dierent design recommendations
for virtual workspaces that could promote interpersonal bonds.
7 DESIGN RECOMMENDATIONS (RQ4)
7.1 Expand the Scope of Non-textual Responses
Our participants indicated that the use of animated non-textual
responses makes senders and receivers happy. Animated GIFs have
been shown to be engaging and fun in the context of microblogging
platforms [
8
], though their study in a virtual workplace has
largely not been explored in prior work. Our research suggests
that animated non-textual responses should be included in the
current suite of non-textual responses as they promote feeling
good factor—including accessible non-textual responses in
communication channels. Non-textual response accessibility was
studied in various social media platforms, online blogging and
microblogging platforms [
35
,
68
,
79
]. We can build on that prior
work to enhance the accessibility of non-textual responses in the
collaborative communication channels used in virtual workspaces.
Further, the non-textual responses should be inclusive for team
members from varied cultural backgrounds. While non-textual
responses like stickers and GIFS provide a receiver with added
information about the context, situation, and relationship through
messages [
57
], though, at the same time. a non-textual response
CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA Shandilya et al.
can be misunderstood because culture can aect it’s interpretation,
therefore including text-based stickers and GIFs can avoid those
confusions. This nding is corroborated by prior work [
38
,
51
],
where authors studied the cultural gap in emoji interpretations in
personal communication and then discussed strategies to make
emoji culturally accessible.
7.2 Interpretable Non-textual Responses
Previous work has studied the interpretability issue with non-
textual responses in the context of social media and personal
communication [
49
,
51
,
78
]. Our research identied the need for
interpretable non-textual responses in virtual workspaces, and
suggests that virtual workspaces platforms should strive to design
non-textual responses that are not open to interpretations.
First, there should be a mechanism for the platform to suggest the
sender alternate forms of non-textual responses in case it understands
that the originally planned non-textual response is culturally sensitive
for the receiver. Kim et al. [
50
] attempt to bring context in emoji
recommenders that could be leveraged here as well. This would
allow people to eliminate unnecessary awkwardness.
Second, the communication platforms should suggest alternative
forms of non-textual responses that are age-neutral. Participants
were concerned that older adults may not be updated with latest
trends, therefore an alternative suggestion for trending non-textual
response would be useful in connecting with people across age
groups. Eorts have been made to understand emoji usage in older
adults [
32
,
40
,
82
]. The ndings from such studies could enable
building an age-inclusive non-textual response recommender.
Third, it should be made easier to ask/nd the meaning of a non-
textual response of the message—both for sender and receiver—and,
next, we make recommendations on how to achieve this.
Fourth, to overcome the challenge of interpretability due to un-
familiarity to the new team’s non-textual communication practice,
a user should have an option to view a trend/history of most used
non-textual responses in dierent situations such as appreciation,
gratitude, celebration, and so on in a virtual workspace.
7.3 Fast Discovery of Non-textual Response
Platforms should reduce the time to nd and select a non-textual
response. There should be an easy way to use non-textual response
recommendations—when to use or not to use—in a given context or
for a given search keyword. The current study suggests providing
keywords based on the user’s geographical location, language, cultural
sensitivity, mood, and context of the conversation. Similarly, the non-
textual response recommendations should also consider the cultural
acceptance according to a person’s location while recommending
non-textual responses; moreover, before a sender sends a culturally
sensitive non-textual response it should even ag them about their
inappropriate use of a non-textual response, which might hurt the
sentiments of the receiver.
Previous work by Cunha et al. [
23
] used a blend of emoji to
represent a concept like cold, world peace, and so on. Other stud-
ies [
37
,
85
] predict certain emoji by determining the sentiment
associated with a message by analyzing the content of the message;
another study [
85
] recommends emoji based on understanding the
entire dialogue; the user will be suggested an emoji depending on
the content of the last message. Such recommendations would allow
users to select an appropriate form of non-textual response without
investing a lot of time. However, the past works [
23
,
37
,
85
] do not
take into account cultural acceptance of a non-textual response in
a particular region while recommending it.
More recently, Feng et al. [
29
] assessed the request of new emoji
on Twitter data to gure out what people look for, when they look
for it, and what does not exist. This study is an initiation to expand
the keywords or provide more emoji. However, only one of the
studies [
72
] discussed the user’s awareness to retrieve a non-textual
response using keywords in their native language.
During our interviews, P12 mentioned the issue of not being
able to nd an appropriate emoji that she was looking for. This
could be due to the user’s language, dierent ways to perceive the
non-textual response, or cultural background knowledge, which
might not return the expected result for her search as shown in the
following quote:
P12: Maybe I have dierent feelings about the emojis because the
emoji I see after the search is something I don’t know, like the meaning
behind it or the text to describe an emoji that I was looking for. We
have dierent feelings about this, so when I searched like type the
text to search for emoji, I probably won’t get the correct result or any
results.
Therefore, to address this issue, enhancing alternate keywords
corresponding to a particular non-textual response based on lan-
guage or geographical location might help. For instance, if a native
Hindi speaker searches the ‘smile’ emoji in their local language,
the platform does not provide any non-textual response sugges-
tions. Currently, on platforms like Slack [
1
] and WhatsApp [
3
],
if a user searches a non-textual response using keyword in their
native language other than English, then the platform recommends
appropriate GIFs, but no results for emoji. Additionally, for Apple
products, there is a predictive emoji feature [
2
], which suggests
emoji only for a phrase or word in English to its users.
7.4 Standardize the Issue Reporting and
Automate the Linguistic Corrections
Our participants shared that a lot of time is lost in explaining issues
over textual modes in these cases. Moreover, at times the actual
issue is neglected while watching out for correct intonation and
grammar. Therefore, the platforms should enable template creation
for communicating issues and roadblocks eectively. These plat-
forms should also include automatic spelling, grammar, intonation
checks, and a relevant non-textual response. Currently, various inde-
pendent services provide virtual team template services for example
monday.com [
28
]. Besides that, Microsoft Teams also provides inte-
grated templates, SalesTim [
61
], on their platform to make remote
collaboration easier and smoother. However, these services do not
provide a template that could help users write their impediments
quickly without worrying about spelling, grammar, and intonation
checks. On the other hand, certain video conferencing tools like
Zoom provide whiteboards [
81
] where the presenter can draw to
explain their ideas. However, these features are fairly limited in
their capabilities. Participant P4 suggested that there should be i)
some predened shapes and controls for use, ii) touch-based controls
New Collaborators’ Perspectives on Using Non-Textual Communication in Virtual Workspaces CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA
to allow for eective and ecient freehand drawing on the boards,
and iii) automatic creation of shareable documents from these boards.
8 DISCUSSION
Our research examined new collaborators’ perspectives on using
non-textual responses in virtual workspaces. Contradictory to some
prior work (e.g., [
5
,
20
,
39
,
41
,
78
,
79
]), our survey reveals a shift
in attitude where the majority of new collaborators (34/49) prefer
using non-textual responses in virtual workspaces as virtual settings
have become the default workspace. Our ndings suggest that new
collaborators use non-textual responses to connect better and to
be able to share a bond with a teammate, which is in line with the
past studies [
33
,
44
]. However, prior work has not studied the use
of non-textual responses in building interpersonal relationships
between the team and new collaborators [
33
,
44
,
78
,
79
], though
a few studies focused on the implications of computer-mediated
communication without including non-textual responses [
14
,
19
,
21, 25, 43, 46, 54, 64, 65].
New collaborators’ unfamiliarity to their teams made them appre-
hensive about using non-textual responses because they are unsure
how others would perceive them in the team. This uncertainty
in using non-textual responses stems from non-textual responses’
openness to various interpretations that team members have due
to personal biases, experiences and cultural knowledge, which is
also in line with Bai et al.’s nding [
6
]. Herring et al. [
41
] and
Tigwell et al. [
78
] briey discussed that people only preferred to
use non-textual responses with whom they shared a close relation-
ship. Moreover, we discovered technological barriers like limited
keywords to search and select a non-textual response hampers a
user’s willingness to use it, which has not been studied by past
work [
22
,
37
,
83
,
85
]. However, the challenges such as nding an ap-
propriate non-textual response to match sender’s intent [
74
,
79
,
86
],
and platform compatibility to support the visuals of a non-textual
response [
47
,
60
,
63
,
69
,
76
,
78
] are known issues. Moreover, the
recent works by Tigwell et al. [
79
] and Zhang et al. [
86
] actively
pursued this problem for people with vision impairments.
Our work further highlights that new collaborators put eorts to
cope with their personal hesitations and technological barriers. One
of the most prominent strategies to tackle personal hesitations was
observing and adapting non-textual communication practices in a
team, reaching out to people to seek clarication, and using a par-
ticular mode of non-textual response selectively and progressively
depending on the bond they share within their team.
As a new collaborator who has to juggle constantly to form inter-
personal relationships within a virtual workspace using non-textual
responses could aect the mental well-being of a user, echoing the
thought of Gilson et al. [
34
] that computer-mediated communi-
cation impacts the mental health of a virtual team. Our study is
a starting point in that direction, exploring the communication
strategies used by the new collaborators to address their hesitance
in using a non-textual response. Moreover, those strategies would
help to inspire and inform the designers’ design decisions in a way
that does not put collaborators in situations that might take a toll
on their mental health.
9 LIMITATION AND FUTURE WORK
We acknowledge that our participant sample was skewed towards
a younger age group from 18 to 44 years of age. Therefore, all the
ndings were grounded on the experiences of younger adults in
using non-textual responses on virtual workspaces. However, a
majority of new collaborators come from the age-group considered
in this study because those people are more prone to internships and
switching jobs [
10
]. Many of these young collaborators come in the
job market for the rst time. It would have been interesting to learn
about older age groups and their insights using non-textual virtual
platforms. Additionally, all the survey and interview participants
were located either in India or in the US, keeping the scope of the
study to two countries. However, we note that typical workforces in
the US can still be very diverse and comprise of people from dierent
nationalities [
4
]. Therefore, it exposed us to gather perspectives
from dierent work cultures even when participants came from
only two countries.
Our future work will focus more on understanding the perspec-
tives of older age groups, people from dierent countries, and per-
spective of teams (where the new collaborator joins) in contrast
to current study which is one-sided focusing only on the percep-
tions of new collaborators. The insights from these user groups will
allow us to understand better the eectiveness of our design rec-
ommendations for non-textual responses in building interpersonal
relationships. A longitudinal diary study [
77
] would be a useful
method to further examine the behavior of new collaborators with
respect to their time spent in the team [
56
,
71
] and transition from
virtual to a face-to-face work setting on the usage of non-textual re-
sponses. We could more closely understand the nuance surrounding
familiarity gained over time vs. the support of non-textual response,
although we are condent that non-textual responses are facilitat-
ing connections because new collaborators are adopting a common
language with their colleagues, which increases relatability through
rapport [
31
]. In contrast, we designed our study to understand the
perception of new collaborators in using a non-textual response
with a new team in a virtual-only workspace.
10 CONCLUSION
We present the experiences of new collaborators in using non-
textual responses in a virtual work environment to build interper-
sonal relationships with their team members. Even though people
still perceive using a non-textual response as unprofessional in the
workplace, we found new collaborators heavily use non-textual
responses in virtual work environment during the COVID-19 pan-
demic. Our work identied the hindrances of new collaborators in
using non-textual responses, coping strategies employed by them
to address the challenges, and how virtual platforms could provide
an elevated experience to new collaborators in using a non-textual
response eciently to form interpersonal relationships with their
colleagues quickly.
ACKNOWLEDGMENTS
We thank all our participants for taking the time to share their
experiences.
CHI ’22, April 29-May 5, 2022, New Orleans, LA, USA Shandilya et al.
REFERENCES
[1] 2022. Slack is your digital HQ. https://www.slack.com/.
[2]
2022. Use emoji on your iPhone, iPad, and iPod touch. https://support.apple.
com/en-us/HT202332.
[3] 2022. WhatsApp. https://www.whatsapp.com/.
[4]
AAUW. 2020. The Future Workforce: More Diverse than Ever. (2020). https:
//www.aauw.org/resources/article/future-workforce- diverse/
[5]
Sarah Alismail and Hengwei Zhang. 2018. The use of emoji in electronic user
experience questionnaire: An exploratory case study. In Proceedings of the 51st
Hawaii International Conference on System Sciences. https://doi.org/10.24251/
HICSS.2018.427
[6]
Qiyu Bai, Qi Dan, Zhe Mu, and Maokun Yang. 2019. A systematic review of emoji:
Current research and future perspectives. Frontiers in psychology 10 (2019), 2221.
https://doi.org/10.3389/fpsyg.2019.02221
[7]
Jeremy N Bailenson. 2021. Nonverbal overload: A theoretical argument for the
causes of Zoom fatigue. Technology, Mind, and Behavior 2, 1 (2021). https:
//doi.org/10.1037/TMB0000030
[8]
Saeideh Bakhshi, David A Shamma, Lyndon Kennedy, Yale Song, Paloma De Juan,
and Joseph’Josh’ Kaye. 2016. Fast, cheap, and good: Why animated GIFs engage
us. In Proceedings of the 2016 chi conference on human factors in computing systems.
575–586. https://doi.org/10.1145/2858036.2858532
[9]
Marjorie A Bayes. 1972. Behavioral cues of interpersonal warmth. Journal of
Consulting and clinical Psychology 39, 2 (1972), 333. https://doi.org/10.1037/
h0033367
[10]
Guy Berger. 2016. Will This Year’s College Grads Job-Hop More Than Previous
Grads? https://blog.linkedin.com/2016/04/12/will-this- year_s-college- grads-
job-hop- more-than- previous-grads. Accessed: 2021-8-28.
[11]
Anna Bleakley, Daniel Rough, Justin Edwards, Philip Doyle, Odile Dumbleton,
Leigh Clark, Sean Rintel, Vincent Wade, and Benjamin R Cowan. 2021. Bridg-
ing social distance during social distancing: exploring social talk and remote
collegiality in video conferencing. Human–Computer Interaction (2021), 1–29.
[12]
Silvia Bonaccio, Jane O’Reilly, Sharon L O’Sullivan, and François Chiocchio. 2016.
Nonverbal behavior and communication in the workplace: A review and an
agenda for research. Journal of Management 42, 5 (2016), 1044–1074. https:
//doi.org/10.1177/0149206315621146
[13]
Virginia Braun and Victoria Clarke. 2006. Using thematic analysis in psychology.
Qualitative research in psychology 3, 2 (2006), 77–101. https://doi.org/10.1191/
1478088706qp063oa
[14]
Amparo Caballer, Francisco Gracia, and José-María Peiró. 2005. Aective re-
sponses to work process and outcomes in virtual teams: Eects of commu-
nication media and time pressure. Journal of Managerial Psychology (2005).
https://doi.org/10.1108/02683940510589037
[15]
Abraham Carmeli, Daphna Brueller, and Jane E Dutton. 2009. Learning be-
haviours in the workplace: The role of high-quality interpersonal relationships
and psychological safety. Systems Research and Behavioral Science: The Ocial
Journal of the International Federation for Systems Research 26, 1 (2009), 81–98.
https://doi.org/10.1002/sres.932
[16]
Wayne F Cascio. 2000. Managing a virtual workplace. Academy of Management
Perspectives 14, 3 (2000), 81–90. https://doi.org/10.5465/AME.2000.4468068
[17] CDC. 2020. CDC’s Response to COVID-19. (2020).
[18]
Shayna Chapel. 2020. New Study: Nearly one-third of workers expect to work
remotely full-time after the pandemic. (2020).
[19]
Arik Cheshin, Anat Rafaeli, and Nathan Bos. 2011. Anger and happiness in virtual
teams: Emotional inuences of text and behavior on others’ aect in the absence
of non-verbal cues. Organizational behavior and human decision processes 116, 1
(2011), 2–16. https://doi.org/10.1016/j.obhdp.2011.06.002
[20]
Henriette Cramer, Paloma de Juan, and Joel Tetreault. 2016. Sender-intended
functions of emojis in US messaging. In Proceedings of the 18th International
Conference on Human-Computer Interaction with Mobile Devices and Services.
504–509. https://doi.org/10.1145/2935334.2935370
[21]
Ioana C. Cristea, Paul M. Leonardi, and Emmanuelle Vaast. 2019. Social Media
and Teamwork: Formation, Process, and Outcomes. Cambridge University Press,
480–508. https://doi.org/10.1017/9781108649636.019
[22]
João M Cunha, Nuno Lourenço, Pedro Martins, and Penousal Machado. 2020.
Visual Blending for Concept Representation: A Case Study on Emoji Generation.
New Generation Computing 38, 4 (2020), 739–771. https://doi.org/10.1007/s00354-
020-00107- x
[23]
João Miguel Cunha, Pedro Martins, and Penousal Machado. 2018. How Shell and
Horn make a Unicorn: Experimenting with Visual Blending in Emoji.. In ICCC.
145–152.
[24]
Dill. 2020. It’s Not Just Working Remotely; Hiring and Onboarding Go Virtual,
Too. Wall Street Journal (2020).
[25]
Joan DiMicco, David R Millen, Werner Geyer, Casey Dugan, Beth Brownholtz, and
Michael Muller. 2008. Motivations for social networking at work. In Proceedings
of the 2008 ACM conference on Computer supported cooperative work. 711–720.
https://doi.org/10.1145/1460563.1460674
[26]
James E Driskell, Paul H Radtke, and Eduardo Salas. 2003. Virtual teams: Eects of
technological mediation on team performance. Group Dynamics: Theory, Research,
and Practice 7, 4 (2003), 297. https://doi.org/10.1037/1089-2699.7.4.297
[27]
Dain Evans. 2020. How Zoom became so popular during social distanc-
ing. https://www.cnbc.com/2020/04/03/how-zoom- rose-to- the-top- during-
the-coronavirus- pandemic.html. Accessed: 2021-8-28.
[28]
Oren Ezra. [n.d.]. How fast-paced teams thrive while WFH (+tem-
plates). https://monday.com/blog/remote-work/10-templates- to-manage- covid-
19-and- get-your- team- wfh- ready/. Accessed: 2021-8-28.
[29]
Yunhe Feng, Zheng Lu, Wenjun Zhou, Zhibo Wang, and Qing Cao. 2020. New
emoji requests from Twitter users: when, where, why, and what we can do
about them. ACM Transactions on Social Computing 3, 2 (2020), 1–25. https:
//doi.org/10.1080/19386389.2020.1907893
[30]
Liz Fosslien and Mollie West Duy. 2020. How to combat zoom fatigue. Harvard
Business Review 29 (2020).
[31]
Vittorio Gallese. 2003. The manifold nature of interpersonal relations: the quest
for a common mechanism. Philosophical Transactions of the Royal Society of
London. Series B: Biological Sciences 358, 1431 (2003), 517–528.
[32]
Jose A Gallud, Habib M Fardoun, Fernando Andres, and Nehme Safa. 2018. A study
on how older people use emojis. In Proceedings of the XIX International Conference
on Human Computer Interaction. 1–4. https://doi.org/10.1145/3233824.3233861
[33]
Amanda N Gesselman, Vivian P Ta, and Justin R Garcia. 2019. Worth a thousand
interpersonal words: Emoji as aective signals for relationship-oriented digital
communication. PloS one 14, 8 (2019), e0221297. https://doi.org/10.1371/journal.
pone.0221297
[34]
Lucy L Gilson, M Travis Maynard, Nicole C Jones Young, Matti Vartiainen, and
Marko Hakonen. 2015. Virtual teams research: 10 years, 10 themes, and 10
opportunities. Journal of management 41, 5 (2015), 1313–1337. https://doi.org/
10.1177/0149206314559946
[35]
Cole Gleason, Amy Pavel, Xingyu Liu, Patrick Carrington, Lydia B Chilton, and
Jerey P Bigham. 2019. Making memes accessible. In The 21st International
ACM SIGACCESS Conference on Computers and Accessibility. 367–376. https:
//doi.org/10.1145/3308561.3353792
[36]
Steve Greenspan, David Goldberg, David Weimer, and Andrea Basso. 2000. Inter-
personal trust and common ground in electronically mediated communication.
In Proceedings of the 2000 ACM conference on Computer supported cooperative
work. 251–260.
[37]
Gaël Guibon, Magalie Ochs, and Patrice Bellot. 2018. Emoji recommendation in
private instant messages. In Proceedings of the 33rd annual acm symposium on
applied computing. 1821–1823. https://doi.org/10.1145/3167132.3167430
[38]
Sharath Chandra Guntuku, Mingyang Li, Louis Tay, and Lyle H Ungar. 2019.
Studying cultural dierences in emoji usage across the east and the west. In
Proceedings of the International AAAI Conference on Web and Social