Content uploaded by Yung-Ju Chang
Author content
All content in this area was uploaded by Yung-Ju Chang on May 03, 2023
Content may be subject to copyright.
Get Distracted or Missed the Stop? Investigating Public Transit
Passengers’ Travel-Based Multitasking Behaviors, Motives, and
Challenges
Hsin-Ju, Lee Fang-Hsin, Hsu∗ Wei-Ko, Li∗
hsinju.lee@iss.nthu.edu.tw vivi160638.cs09@nycu.edu.tw weiaquarius.cs10@nycu.edu.tw
National Tsing Hua University National Yang Ming Chiao Tung National Yang Ming Chiao Tung
Hsinchu, Taiwan, Taiwan University University
Hsinchu, Taiwan, Taiwan Hsinchu, Taiwan, Taiwan
Jie, Tsai∗ Ying-Yu, Chen Yung-Ju, Chang†
jessic462071@gmail.com yingyuchen@nycu.edu.tw armuro@nycu.edu.tw
National Yang Ming Chiao Tung National Yang Ming Chiao Tung National Yang Ming Chiao Tung
University University University
Hsinchu, Taiwan, Taiwan Hsinchu, Taiwan, Taiwan Hsinchu, Taiwan, Taiwan
ABSTRACT ACM Reference Format:
Mobile users commonly multitask during travel, but doing so on
Hsin-Ju, Lee, Fang-Hsin, Hsu, Wei-Ko, Li, Jie, Tsai, Ying-Yu, Chen, and Yung-
Ju, Chang. 2023. Get Distracted or Missed the Stop? Investigating Public
public transit can be challenging due to the dynamic nature of the
Transit Passengers’ Travel-Based Multitasking Behaviors,
environment as well as long-standing lack of infrastructural sup-
Motives, and
Challenges. In Proceedings of the 2023 CHI Conference on Human Factors in
port. Nevertheless, HCI scholars and practitioners have devoted
Computing Systems (CHI ’23), April 23–28, 2023, Hamburg, Germany. ACM,
relatively little attention to developing technology for enhancing
New York, NY, USA, 14 pages. https://doi.org/10.1145/3544548.3581391
travel multitasking. To facilitate such development, we sought to
understand travel multitaskers’ practices and challenges while on
public transit, and to that end, conducted a multi-methods study
1 INTRODUCTION
that involved shadowing and interviewing 30 of them. We iden- Amid technological advancements and a growing emphasis on
tied four travel-multitasking patterns, characterized by distinct eciency, people today are able to multitask almost anywhere
motives that aected these travelers’ multitasking practices, recep-
and at any time. Multitasking on public transit is a key exam-
tivity to environmental stimuli, and task persistence. The two main
ple of this [
25
,
30
,
32
]. Travel-based multitasking behavior is de-
challenges they encountered during travel multitasking resulted
ned as when “individuals endeavor to do multiple things concur-
from mutual interference from their tasks and from the dynamic rently while en route to destinations” [
8
,
26
,
41
]. Previous stud-
nature of transit environments. Based on these ndings, design ies have shown that, in addition to accomplishing their primary
recommendations for public-transit agencies and mobile services
task, i.e., reaching the destination [
29
], travel is utilized by many
are also provided.
travel multitaskers for various work-related and leisure purposes
[
28
,
30
,
41
,
42
,
66
]. While some travelers use a range of technologies
CCS CONCEPTS
to engage in productive tasks that transform their “dead time” into
• Human-centered computing
→
Empirical studies in ubiq-
meaningful time [
24
,
66
] , others use tech for enjoyable activities
uitous and mobile computing.
such as listening to music, playing digital games, and viewing media
on mobile devices [23, 30].
KEYWORDS
However, travel multitaskers’ fragmented attention is likely to
Travel-Based Multitasking; HCI; Public Transit
negatively impact their travel task: e.g., by missing their stop [
15
],
losing belongings due to packing them up in a rush, or even hurting
themselves due to disembarking too hurriedly. In addition, it may
∗
cause problems to their tasks-at-hand, i.e., the work the multitaskers
Contributed equally to this research.
†Corresponding author.
are doing and paying attention to, since having to frequently divert
one’s attention to the travel task can impede these other tasks’
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
progress and reduce their quality [
57
]. Moreover, active smartphone
for prot or commercial advantage and that copies bear this notice and the full citation
users who feel that public-transit environments are interruptive
on the rst page. Copyrights for components of this work owned by others than the
may develop negative perceptions of the environments [28].
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 specic permission
While public transit has become a space in which people use a
and/or a fee. Request permissions from permissions@acm.org.
wide variety of information technologies, infrastructure and ser-
CHI ’23, April 23–28, 2023, Hamburg, Germany
vices that support this ubiquitous behavior have lagged far behind.
© 2023 Copyright held by the owner/author(s). Publication rights licensed to ACM.
ACM ISBN 978-1-4503-9421-5/23/04.. . $15.00
To improve travel multitasking experiences via technological sup-
https://doi.org/10.1145/3544548.3581391
port for this behavior, more research attention and eorts from the
CHI ’23, April 23–28, 2023, Hamburg, Germany Lee, Hsu, Li, et al.
HCI community is needed. In particular, detailed understanding of
travel multitaskers’ practices on public transit and the challenges
they encounter in this dynamic environment is largely absent from
the literature, yet crucial to informing the kinds of infrastructural
and technological support researchers, practitioners, and public
servants need to develop. The present paper therefore explores a
range of travel multitasking behavior patterns, identies their un-
derlying causes, and delineates the challenges travel multitaskers
encounter on public transit, guided by the following two research
questions:
•
RQ1: What are travel multitaskers’ behavioral patterns on
public transit and the causes underlying them?
•
RQ2: What challenges to travel multitasking do travel multi-
taskers frequently encounter?
To answer these questions, we conducted a qualitative study
consisting of shadowing and semi-structured interviews with 30
travel-multitaskers in 6 cities in Taiwan. The former approach
was adopted to observe travel multitaskers’ in situ multitasking
practices and reactions to environmental stimuli during their public-
transit journeys, while the latter was intended to shed light on why
they made certain decisions and how they perceived and felt about
such decisions and travel experiences.
This paper makes the following three important contributions
to the HCI and travel-multitasking literature.
(1)
It identies four main patterns of travel-multitasking behav-
ior, each of which is characterized by a distinct motivation
that is manifested in their task choices, their expectations
of those tasks’ progress and quality, and their receptivity to
travel-task-related signals on public transit.
(2)
It identies three main types of challenges to travel multi-
taskers’ travel tasks and tasks-at-hand that can result from
mutual interference between these two classes of tasks.
(3)
It identies an additional three main types of concerns aris-
ing from the public transit’s ever-changing environmental
and interpersonal surrounding.
Based on our results, we also oer four high-level design rec-
ommendations to future transportation and technology providers,
which we hope will improve travel multitaskers’ experiences on
public transit.
2 RELATED WORK
2.1 Attention and Multitasking
Attention determines what is perceived and what is not [
14
]. It
is linked closely to “the voluntary and eortful control of action”
[
51
], which means that it is associated with states of consciousness.
Petersen and Posner, however, proposed that there is an attention
system composed of an alerting network, an orienting network, and
an executive-function network [
47
]. The rst increases humans’
sensitivity to external stimuli; the second manages the process
of selecting the desired information from among various stimuli;
and the third excludes irrelevant information and manages goal-
oriented behaviors [
6
]. Clearly, multitasking – dened by Delbridge
as engaging in “multiple task goals in the same general time period”
with “frequent switches between individual tasks” [
20
] – has a
close relationship with attention, insofar as people who multitask
have multiple targets to process. Benbunan-Fich et al. [
5
] dened
multitasking based on the principles of task independence and
concurrency. Task independence refers to ongoing tasks being self-
contained, while concurrency refers to multiple tasks being carried
out with some temporal overlap. Depending on the amount of
overlap, multiple tasks can be executed sequentially, starting one
task after the completion of the previous one, where only one
task is attended to at a time [
5
,
9
]; in contrast, when multiple
tasks are executed at the same time, they are performed in parallel.
Salvucci et al. [
53
,
54
] later proposed a "multitasking continuum"
that characterizes multitasking practices based on the frequency
of task switching. At one end of the continuum are multitasking
practices that require nearly simultaneous processing, while at
the other end are those with longer spans between switches. This
continuum therefore accommodates both concurrent and sequential
multitasking. In examining the interrelationship of the multiple
tasks involved in multitasking, it can be useful to divide tasks
into primary and secondary tasks: for example, according to the
order in which they are mentioned, their perceived importance, the
length of time people engaged in them, and/or the degree of their
respective attentional-resource demands [
29
]. Given that people
often have diculty distinguishing between their own primary and
secondary tasks unless given specic guidance, Kenyon [
29
] also
recommended that "activities that a person will do anyway" can
be regarded as primary tasks, while other overlaid or interleaved
activities should be seen as secondary [29].
In order to achieve good multitasking performance, previous
studies have pointed out that the competition among multiple tasks
on cognitive resources [
53
] and the problem of residual attention
during task switching [
36
] should be avoided. However, despite a
common notion that multitasking refers to switching between tasks
in the same time [
20
] , it has been argued that there is no universal
agreement on the denition of multitasking. For this, Circella et
al. [
16
] classied the relationship between the primary and the
secondary tasks into three categories: tasks switching, which refers
to alternation of tasks, but with only one task being performed
at once; tasks interleaving, which refers to one task consumes the
majority of the resources, while another remains in the background,
and tasks overlaying, which refers to both tasks are performed at the
same. The topic "travel-based multitasking" usually be categorized
in the third category, where activities are overlaid on travels [7].
Apart from the perspective in neurophysiology, the notion of
attention, which is seen as a limited ability and resource [
59
] from
the eld of economy is also core to multitasking. That is, according
to Simon [
56
], people’s attention and actions are competed by vari-
ous received stimuli and inputs, and that people are only aware of
the stimuli that appeal to them [
27
]. Nowadays, the pervasiveness
of the Internet, the plethora of information and media, as well as
the increasing availability of ubiquitous and mobile technology
have allowed people to multitask nearly anywhere and anytime
[
1
,
16
,
21
,
22
,
46
,
55
,
58
]. For example, multitasking during work has
drawn considerable research attention including HCI and CSCW,
as there is a general concern that multitasking in this context may
harm the worker’s productivity and performance. Investigating
the impact of multitasking in workspace, Mark et al. found that
interruptions, such as unavoidable task switching, reduce produc-
tivity at work [
43
]. Likewise, Leroy et al. showed that workplace
Exploring Travel-Based Multitasking Behaviors in Public Transportation Context CHI ’23, April 23–28, 2023, Hamburg, Germany
multitasking was linked to more emotional exhaustion [
37
]. Czer-
winski et al. [
17
], who used a diary study to investigate multitasking
behavior at work, suggested that the phenomenon of information
workers sometimes altering their tasks-at-hand was a result of
constant shift of the context at work, which inescapably caused
interruptions to their tasks-at-hand. On the other hand, beyond
workspace, researchers have also investigated multitasking behav-
iors during remote meeting. For example, Cao et al. studied workers’
multitasking behaviors in this context and found that it is impor-
tant to consider when and how much people are distracted when
scheduling remote meetings [11].
In this study, we focused on public transit passengers’ multitask-
ing behaviors, commonly noted as travel-based multitasking in the
literature, for the purpose of highlighting its distinct characteristics
of multitasking in a rapidly changing and unstable physical and
social environment compared to workspace or home. We provide a
brief literature review in this line of research in the next section.
2.2 Travel-Based Multitasking
Information technology and digital services have gradually changed
the way and places where people perform their activities [
16
].
Smartphones, laptops, tablets and other internet-enabled portable
devices oer new possibilities for work and entertainment during
travel, as well as increasing the multitasking ability of passen-
gers [
25
,
30
,
32
,
42
]. Travel-based multitasking behavior has been
referred to as "individuals’ endeavor to do multiple things concur-
rently while enroute to destinations" [
16
,
34
,
60
]. Dierent trans-
portation modes and environments will lead to dierent travel-
based multitasking practices and activities prevalence. For example,
while car users must actively participate in driving and navigation
[
25
,
32
,
42
,
44
], public transit oers its passengers an opportunity to
undertake productivity tasks along the travel [
25
,
32
]. In addition,
research has also shown that train travel results in more productive
tasks than bus travel does[
25
,
61
,
67
]; specically, train riders are
found to be more inclined to read, use a computer, sleep, write,
or work, whereas bus riders are more likely to enjoy the scenery
[
25
,
61
]. Lyons and Urry [
42
] pointed out that the limits of public
transit, such as the level of crowdedness, seat availability, and facil-
ity characteristics are the core causes behind dierent levels and
kinds of multitasking behaviors on public transit.
In addition to the inuence of the physical environments, re-
searchers have also found other concerns that had impacted pas-
sengers’ practices of multitasking on public transit. For example,
Axtell et al. [
3
] suggested that limited internet access and privacy
concerns during train travel made it dicult for train passengers to
make phone calls. Also due to privacy concerns, Tillema et al. [
62
]
indicated that people preferred to have condential conversations
via quiet communication channels such as SMS, email on the train
over phone calls.
Safety concern has also been found inuential in public tran-
sit environment. Newton [
45
] showed that the ease of passenger
distraction can be a predictor variable for theft at transit stations,
suggesting that people may encounter property safety concerns
when attention is frequently switched. Keseru et al. [
31
] conducted a
eld study to investigate the impact of public public’s environment
characteristics to the type of activities carried out by passengers
during travels; they suggested that safety concerns inuence the
use of digital services, and thus it should be taken into account in
public transit multitasking. Since the 2020 COVID-19 outbreak, De
Vos [
19
] showed that passengers’ concern with the risk of infection
has aected their public transit behavior, such as choose to keep a
social distance from other passengers [
19
]. Lastly, social acceptance
also plays a role. As Campbell [
10
] suggested, passengers deemed
making calls to be less acceptable on crowded Japanese trains and
buses, and thus preferred to use their mobile phones quietly.
However, despite the number of studies investigating travel-
based multitasking behavior in the transportation domain, most
of these studies, if not all, focused on the passengers’ choices of
activities, and how dierent external factors and their concerns
inuenced their choices. Thus far, there is limited understanding
of what makes passengers want to multitask in such a rapidly
changing, unstable, and sometimes even risky and dangerous, envi-
ronment, and how the reasons and causes behind their decisions
to multitask on public transit in turn shape their multitasking be-
havior pattern on public transit and aect challenges they would
encounter during multitasking, respectively. The current paper
provide these insights.
3 METHODOLOGY
3.1 Recruitment
Participant recruitment for this study was divided into two broad
stages, based on our aims for and approaches to data collection
and analysis. The rst stage, which involved shadowing, debrieng
interviews, and semi-structured interviews, helped us to capture the
broadest possible range of multitasking behaviors and challenges
on public transit. As a result, the selection process focused on
reaching participants with diverse characteristics and experiences
with multitasking on public transit.
The second stage of data collection and analysis commenced
after we had established a set of theoretical categories and codes
based on the data generated by the 22 individuals who made up the
rst stage’s nal participant pool. In it, our focus has been on testing
the saturation of our theoretical categories: a process referred to as
theoretical sampling [
12
]. In this case, this process included looking
for additional examples of multitasking scenarios that had appeared
infrequently in our data up to that point. Because we found that
our understanding of variations in both multitasking practices and
challenges was mainly derived from semi-structured interviews,
participants recruited in the second stage have only participated in
semi-structured interviews, i.e., are not being shadowed.
For each wave of recruitment, we distributed recruiting ads via
four main channels: Facebook groups with themes related to public
transit, Facebook pages intended for recruiting research partici-
pants in our country, the researchers’ personal networks, and word
of mouth. Each ad contained a link to a screening questionnaire
whose questions covered the respondents’ demographic informa-
tion; their travel behavior in the past six months (i.e., types of public
transit they frequently took, and their journey purposes, frequen-
cies, and timings); their frequent choices of tasks on public transit
(adopted from [
32
]); and their tendency to multitask, as measured
via the Polychronic-Monochronic Tendency Scale (PMTS) [
38
]. We
then primarily selected participants who were diverse in terms
CHI ’23, April 23–28, 2023, Hamburg, Germany Lee, Hsu, Li, et al.
of their public-transit use and demographic characteristics, but
who mostly had a high polychronic propensity, as individuals with
such a propensity were more likely than others to have numerous
and complex multitasking experiences. However, we also recruited
some participants with a monochronic propensity to enrich our
data.
In the rst stage of the recruitment, we initially recruited 22
participants, 9 males and 13 females, ranging in age from 21 to 50
(M=26.3). In the second, we recruited 8 participants, 6 males, and 2
females, ranging in age from 21 to 40 (M=29.9). The participants’
details are shown in Table 1.The public transit mode seen in the
proles were selected by participants based on the three modes of
public transit they were most familiar with.
3.2 Shadowing Study
The main objective of the shadowing activity in phase 1 of our
study was to observe people’s multitasking behaviors, including
their attention-switching, task choices, technology choices, and re-
sponses to the dynamic environments within public-transit vehicles.
These observations allowed us to investigate their in situ behav-
iors that would have been dicult to obtain through retrospection,
such as autonomous responses to environmental stimuli [
33
] (e.g.,
attention-switching triggered by vehicle-generated alerts ), and be-
haviors linked to procedure memory [
2
] (e.g., habitual phone use).
Our sites of observation consisted of one public-transit journey per
participant, chosen by them. Among the 22 participants who were
shadowed, 16 chose their routine commutes, and the remainder,
occasional travel such as business trips or going to meet someone
for social purposes.
Each shadowing session began as the shadowed participant
waited to enter a public-transit vehicle, and ended when they ex-
ited it. Throughout this process, they were shadowed and observed
by one member of the research team, who positioned themselves
unobtrusively at least 5 but not more than 10 feet away to mini-
mize the participant’s awareness of their presence. Each observer
observed and recorded eld notes of the participant’ multitask-
ing behaviors and their context, including activities, devices used,
attention-switching, progression of the journey, incidents within
the participant’s immediate surroundings, and other aspects of the
vehicle environment such as its crowdedness and the nature, fre-
quency and clarity of stop/station-arrival alerts. Immediately after
the observation (i.e., when the participant successfully arrived at
their intended destination ), the observer walked with the partici-
pant to a quiet place and gave them a debrieng interview. Since all
these debriengs took place in the middle of the participants’ wider
journeys, they were limited to a duration of 30 minutes to mini-
mize interference with the participant’s schedule for the rest of the
day; and all participants were informed in advance about this time
commitment. Specically, the observer’s questioning focused on
quick clarications of the participant’s in situ perceptions, feelings,
experiences, and rationales behind their multitasking decisions that
might have been dicult to recall in a subsequent semi-structured
interview, the questions for which were based on the data gathered
during the observation and debrieng interview. Basic descriptions
of the participants’ shadowing sessions can be found in Table 1,
above.
3.3 Stage 2: Semi-Structured Interview
Regardless of their participation in shadowing/debrieng, we in-
vited each participants to a semi-structured interview lasting be-
tween 90 minutes and two hours, aimed at capturing their multitask-
ing experiences on public transit over the preceding six months.
Those who had participated in the shadowing study were also
asked about specic behaviors and incidents the research team had
recorded in eld notes during their journeys. More specically, the
semi-structured interview questioning asked participants to walk
through their multitasking experiences when taking public transit,
including their procedures; their choices of tasks and technology;
the rationales behind those choices; their feelings/attitudes toward
incidents; their multitasking processes and outcomes; and the chal-
lenges and barriers they had encountered along dierent stages of
their journeys, including waiting, riding, and disembarking.
3.4 Study Procedure
The study was conducted between late October 2021 and mid-June
2022. The research team contacted people who had completed the
screening questionnaire and selected them based on the aforemen-
tioned selection criteria. The research team then walked the selected
individuals through the study’s objectives and procedures. They
were informed that the study included being shadowed on one
public-transit journey of their choice, and that this would be im-
mediately followed by a 30 minute debrieng interview. After they
provided their informed consent to participate, the research team
agreed times and locations to meet each of them, such that the
researchers could observe their entire journeys, from waiting for
the vehicle to arriving at the destination. Before their respective
meetings with the observer, each participants was sent an email
reminder about the debrieng interview so that they could plan
their schedules accordingly.
The participants were given the option of either in-person or
online semi-structured interviews; ultimately, they all chose to par-
ticipate online. All these interviews were therefore conducted via
Google Meet, and all gave their consent to being recorded. They
were compensated for their time based on the length of their par-
ticipation in the eld observation and semi-structured interviews,
with compensation ranging between NT$400 and NT$480 (US$13-
16) based on whether or not to participate in the shadowing study.
Participants who participated in the shadowing study could receive
an additional $80.
3.5 Qualitative Analysis
Our shadowing data covered a total of 18.2 hours on ve types of
public transit in six cities. Data collected during shadowing and the
transcriptions of both types of interviews were entered and coded
in ATLAS.ti, an online application for qualitative-data analysis (AT-
LAS.ti Scientic Software Development GmbH, Berlin). Our data-
analysis process was guided by Charmaz’s grounded approach [
12
].
In it, open coding started as soon as the research team had begun col-
lecting data. The rst set of codes was developed from our rst three
interviewees’ data, and the creation of their high-level categories
was guided by Keseru et al.’s [
32
] work, from which we adopted
codes including Duration of Travel (Long/Short), Trip Purpose (Com-
muting/Business/Leisure), and Activity Type (Work/Leisure), among
Exploring Travel-Based Multitasking Behaviors in Public Transportation Context CHI ’23, April 23–28, 2023, Hamburg, Germany
Table 1: Participant Proles
ID Age / Gender/ Ocuupation/ Polychronicity Prefered Tasks-By-Hand Shadowing Shadowing Semi-Structured Interview
Location Transit Duration Interview Duration
Mode Transit Mode
P01 23/ F/ Student/ Hsinchu Polychron Relaxing, Media Use, Mobile Phone Use,
Working/Studying, ICT Use, Talking Bus 30 min Intercity Bus, Metro, Train 94 mins
P02
P03 24/ M/ Student/ Taipei
22/ F/ Student/ Hsinchu Polychron
Polychron Relaxing, Media Use, Working/Studying, Other
Relaxing, Media Use, Playing, Mobile Phone Use, Intercity
Bus Bus 90 min
20 min Bus, Intercity Bus, Metro
Bus, Intercity Bus, Metro 91 mins
91 mins
P04 26/ M/ Worker/ Taipei Polychron ICT Use, Talking
Relaxing, Media Use, Mobile Phone Use, Metro 20 min Bus, Metro, HSR 72 mins
Working/Studying, ICT Use, Eating and Drinking,
Talking
P05 24/ M/ Student/ Taipei Polychron Reading, Relaxing, Media Use, Playing,
Mobile Phone Use, Working/Studying, ICT Use, Metro 20 min Bus, Metro, Train 102 mins
P06 24/ F/ Worker/ Hsinchu Polychron Eating and Drinking, Talking
Reading, Mobile Phone Use, ICT Use, Talking Bus 45 min Bus, Intercity Bus, Metro 83 mins
P07 25/ M/ Student/ Taipei Monochron Reading, Relaxing, Media Use, Mobile Phone Use,
ICT Use, Talking Metro 15 min Bus, Intercity Bus, Metro 73 mins
P08 25/ F/ Worker/ Taipei Polychron Reading, Relaxing, Media Use, Mobile Phone Use,
Working/Studying, ICT Use, Talking Metro 25 min Bus, Metro, Train 117 mins
P09 23/ F/ Worker/ Taichung Polychron Reading, Relaxing, Media Use, Mobile Phone Use,
Working/Studying, ICT Use, Eating and Drinking, HSR 54 min Bus, Intercity Bus, Metro 102 mins
P10 32/ F/ Self-Employed/ Polychron Talking
Reading, Relaxing, Media Use, Playing, Bus 50 min Bus, Metro 91 mins
New Taipei Mobile Phone Use, Working/Studying, ICT Use,
Eating and Drinking, Talking
P11 24/ F/ Student/ Taipei Polychron Relaxing, Playing, Mobile Phone Use,
Working/Studying, ICT Use, Talking Bus 20 min Bus, Intercity Bus, Metro 82 mins
P12 22/ F/ Student/ Taipei Polychron Relaxing, Media Use, Working/Studying,
Eating and Drinking Intercity Bus 50 min Bus, Intercity Bus, Metro 81 mins
P13 21/ F/ Student/ New Taipei Polychron Relaxing, Media Use, Mobile Phone Use,
Working/Studying, Eating and Drinking, Talking Train 38 min Bus, Metro, Train 155 mins
P14
P15 24/ F/ Worker/ Taipei
24/ F/ Worker/ Taipei Polychron
Polychron Reading, Mobile Phone Use
Reading, Relaxing, Media Use, Mobile Phone Use,
Working/Studying
HSR
HSR 100 min
110 min Bus, Metro, HSR
Bus, Metro, HSR 105 mins
91 mins
P16 22/ F/ Student/ Hsinchu Monochron Relaxing, Media Use, Mobile Phone Use,
Eating and Drinking, Talking Intercity Bus 90 min Bus, Intercity Bus, Metro 84 mins
P17 24/ F/ Worker/ New Taipei Monochron Relaxing, Mobile Phone Use, ICT Use,
Eating and Drinking, Talking Train 145 min Intercity Bus, Metro, Train 84 mins
P18 22/ M/ Student/ Taoyuan Polychron Reading, Media Use, Mobile Phone Use,
Working/Studying, ICT Use, Talking Train 30 min Train, Metro, HSR 93 mins
P19
P20 48/ M/ Worker/ Kaohsiung
27/ M/ Worker/ Kaohsiung Polychron
Polychron Reading, Media Use
Relaxing, Media Use, Mobile Phone Use, Metro
Metro 30 min
30 min Intercity Bus, Metro, HSR
Metro, Train 89 mins
90 mins
Working/Studying, ICT Use, Eating and Drinking,
Talking
P21
P22 22/ M/ Student/ Taipei
50/ M/ Worker/ Kaohsiung Polychron
Monochron Media Use, Mobile Phone Use, Working/Studying
Reading, Relaxing HSR
Metro 70 min
30 min Bus, Metro, HSR
Metro 70 mins
42 mins
P23 24/ F/ Worker/ Taoyuan Monochron Reading, Media Use, Mobile Phone Use,
Working/Studying, Talking N/A N/A Intercity Bus, Metro, Train 81 mins
P24 25/ M/ Worker/ Taipei Polychron Relaxing, Media Use, Playing, Mobile Phone Use,
Working/Studying N/A N/A Intercity Bus, Metro, Train 54 mins
P25 21/ F/ Student/ New Taipei Polychron Reading, Relaxing, Media Use, Playing,
Mobile Phone Use, Working/Studying, N/A N/A Bus, Metro, Train 70 mins
P26 30/ M/ Worker/ New Taipei Polychron ICT Use, Talking
Reading, Media Use, Working/Studying N/A N/A Bus, Intercity Bus, Metro 67 mins
P27 40/ M/ Worker/ New Taipei Polychron Reading, Relaxing, Media Use, Playing,
Mobile Phone Use, Working/Studying, ICT Use, N/A N/A Bus, Metro, Train 55 mins
P28 26/ M/ Student/ Nantou Polychron Eating and Drinking, Talking
Reading, Relaxing, Media Use, Playing, N/A N/A Bus, Intercity Bus, Metro 54 mins
P29 40/ M/ Worker/ Taipei Polychron Mobile Phone Use, Working/Studying, ICT Use
Reading, Relaxing, Media Use, ICT Use, N/A N/A Metro, Bus, Train 60 mins
P30 33/ M/ Worker/ Taipei Monochron Eating and Drinking
Reading, Relaxing, Media Use, Playing, N/A N/A Metro, Train, Bus 57 mins
Mobile Phone Use, Working/Studying, ICT Use,
Eating and Drinking, Talking
others. The rst three transcriptions were coded by the rst author
and three co-authors. Throughout the coding process, new code
category such as Challenge were added. Another two co-authors
would join each time a transcription was compiled to discuss the
current state of our codes and synchronize the team’sunderstanding
of coding semantics. The codebook was then iteratively revised by
the research team until all team members agreed with all of its con-
tent. The remaining transcription data were then coded by all six
researchers separately using the codebook, and they met regularly
to discuss any questions so as to guarantee consensus on the codes.
CHI ’23, April 23–28, 2023, Hamburg, Germany Lee, Hsu, Li, et al.
When new codes or revisions to codes were proposed, they were
also examined against existing data. When a circumstance that was
challenging to code arose, the researchers replayed the interview
recording le and attempted to comprehend the context of what
the participant had said. At this stage, improving the code was our
main priority. For instance, the previously mentioned code Activity
Type was eventually renamed Task Feature, and extra features were
added to it, e.g., Concentration/ Duration/ Expectation/ Importance/
Operation/ Urgency/ Complication. During this process, we wrote
memos and drew diagrams to depict the relationships among the
codes and code categories. This revealed that both the dynamics of
the environment and the reasons behind the participants’ multitask-
ing on public transit had profound inuences on their multitasking
practices, choices, adaptation to the environment, and challenges.
This led us to focused coding [
12
], a process whereby we started
establishing our theoretical codes and categories by focusing on
participants’ motivations for multitasking, multitasking behaviors,
task performance, challenges, and coping strategies. Based on those
theoretical codes and categories, we then recruited the second wave
of participants to assess saturation, as explained above.
4 FINDINGS
In this section, we present our key ndings regarding the partici-
pants’ multitasking practices and challenges on public transit. For
simplicity’s sake, we use the term travel task to refer to actions
whose ultimate purpose is safe, timely arrival at one’s destination.
Typically, a travel task’s subtasks include monitoring the progres-
sion of the journey, preparing to disembark from the vehicle, actu-
ally doing so, and so forth. The term task-at-hand, in contrast, is
used to refer to any non-travel-related task that a participant en-
gaged in during his/her journey. Travel-based multitasking [
32
] is
thus the process whereby participants performed a travel task and
a task-at-hand simultaneously; and we regarded their travel-based
multitasking experiences as successful only if both these classes of
tasks were satisfactorily completed.We did not adopt Kenyon’s [
29
]
primary vs. secondary task classication in our study because, in
the context of travel-based multitasking, the travel task can always
be considered primary in theoretical terms, whereas some travelers
might view it as a secondary task, e.g., because it is part of a familiar
routine. Thus, looking at travel tasks vs. tasks-at-hand both aligns
better with travelers’ subjective perceptions and enables us to dis-
cuss task types with greater precision than if Kenyon’s typology
were used.
The rst of the following subsections describes the four travel-
multitasking behavioral patterns we identied, which are distin-
guished from one another by dierences in motivation. Subsections
4.2 and 4.3 then explain the challenges the participants encoun-
tered, including those that arose due to interference between their
travel tasks and their tasks-at-hand, and those resulting from the
dynamics of the public-transit environment.
4.1 Four Multitasking Patterns on Public
Transit
As noted above, classifying multitasking patterns according to the
participants task motivations and the meanings they assigned to
tasks resulted in a four-part typology. However, they also diered
in their choices of tasks-at-hand, levels of concentration on them,
insistence about engaging in particular ones, persistence in complet-
ing them, and expectations regarding the completed ones’ quality.
These dierences, in turn, resulted in the participants exhibiting
variation both in their levels of receptivity to environmental stim-
uli and in the task challenges they experienced. Each of the four
patterns is described in turn below.
4.1.1 Habitual Behavior, Part of the Daily Routine.
This travel-based multitasking was characterized by the partici-
pants’ lack of intention to accomplish a particular task-at-hand;
rather, they just habitually performed such tasks while traveling,
without much regard for whether they would be completed. Proba-
bly because of this, when we asked them in interviews about the
multitasking behaviors we had observed while shadowing them,
the participants who followed this pattern usually told us that
their task-at-hand choices were habitual, spontaneous, and without
particular intention; and some did not recall what they had done
during their journeys. For example, during the shadowing study,
P05 was observed browsing a stock-trading app, but he seemed
to be unaware of this when he asked about it in the debrieng
interview. "I probably have checked some stock stu at that time. [...].
It just occurred to me that I haven’t done this today and so I did it”.
Then, when asked about browsing social-media apps during this
time, he commented, “It’s probably just an unconscious thing, for no
particular reason".
However, some participants who followed this multitasking pat-
tern expressed much greater awareness that their tasks-at-hand
were parts of their daily routine, e.g., catching up with news or in-
coming messages during the morning commute, applying makeup
before arriving at work, accumulating reward points, and so on.
Importantly, those participants whose tasks-at-hand were habitual
reported that they were not committed to them, and could start,
stop, and resume them at any time.
4.1.2 Making the Most of Travel Time.
This pattern of multitasking was motivated by utilizing one’s time
fully, regardless of task-at-hand type. Interestingly, the participants
who exhibited this multitasking pattern expressed their motivations
both negatively (e.g., avoidance of idleness) and positively (e.g.,
seeking a feeling of being productive). Unsurprisingly, given that
their main aim was simply to ll dead time, these participants were
quite open both about task choices and task outcomes. That is, they
were neither insistent nor persistent about undertaking a particular
task-at-hand, but rather, were open to switching to other tasks, as
long as doing them helped them ll the time. As P10 told us, "I
am a person who can’t tolerate being idle [...] I feel that I just had to
nd something to do, even including playing games". Similarly, P05
commented, "It is extra time that is freed up where I had nothing
to work on but do want to do something to leverage it." In light of
such motivations, members of this class of participants did not have
high expectations about the quality or progression of their tasks-
at-hand. As P23 put it, "I usually post what I ate today and comment
on it on Instagram [during the after-work commute...]. Some typos in
there are okay, not work you need to be serious about." This attitude
freed these participants to intermittently shift their attention to
the environment and to monitoring the progression of their travel
tasks.
Exploring Travel-Based Multitasking Behaviors in Public Transportation Context CHI ’23, April 23–28, 2023, Hamburg, Germany
Some participants in this category mentioned that they usually
thought about or even planned what they would do while waiting to
arrive at their travel destinations. This helped them feel "productive"
(P09). "I’d think about what I was going to do before I took public
transit. [...] It was sort of dependent on where I was going and how
many things I could do during this time. I’d quickly walk through
them before I took it" (P09). According to several participants, this
type of planning sometimes occurred many days before their travel,
or even before they had determined what mode of transportation
they would be using. For instance, P18 had a tendency to seek
optimization, and therefore, when using an expensive mode of
transportation such as high-speed rail, he planned carefully how he
was going to leverage the travel time to make the expense worth
it. As he put it, "Mostly I see taking high-speed rail as a luxury, so I
would hope to be productive when I do it, like making PowerPoints
or something. [...] It’s like the extra amount of money could let you
stay in a coee shop. So I will try my best to make this worth it."
He even stated that he was willing to buy additional hardware to
optimize his tasks-at-hand in light of high-speed trains’ table-size
constraints: "I try not to use a mouse on the high-speed train, but
sometimes I still have to use it, which is troublesome. I may buy a
small mouse in the future."
4.1.3 Completing Last minute Work or Clearing Work Backlogs.
This pattern of travel-multitasking was motivated by the partic-
ipants having heavy workloads and wanting to complete them
during the travel. Often, people in this category needed to deliver
completed tasks very soon after their travel, or even immediately
upon arrival at their destinations. This pattern diered sharply
from the two described above, insofar as the importance assigned to
completing tasks-at-hand was very strong, even when performing
them caused discomfort or was otherwise ill-suited to the public-
transit environment. For example, P02 was a student who needed
to prepare for an upcoming meeting on the day we shadowed him.
Despite the intense vibration of the bus, he worked on his laptop
during the trip, writing reports and responding to emails, and at one
point, zooming in to read small text. Later, he told us that, despite
being aware that the movement of the bus might negatively aect
his vision, "making progress on this [task-at-hand] was way more
important than my health. I felt that if I didn’t do this well, I’d have a
mental breakdown. It’s like I wanted to borrow a little health from the
forty-year-old me for this moment". Similarly, P15 reported squatting
down among a crowd of fellow passengers in a moving train to use
her computer: "There happened to be some urgent business to deal
with at that time. I had to turn on the computer to send a document
to a client because there was no way to use my phone to do this. But
I was standing in the Metro. All the seats were occupied, so I had to
squat down and take out my computer, nd a space, and then put the
computer on my knee to do my work. It was not an ideal situation.
Very crowded, but I had to do it." On other occasions, when the train
was not crowded but she also felt it was urgent to work, she moved
to areas where she thought fewer people would go: "The area where
cars are connected [...] is very unstable, and there are usually not
many people standing there, so the space is relatively large". She also
told us that she wished the Metro service would provide a work-
friendly area so that people with similar needs could have a more
comfortable and safe environment in which to perform their tasks.
P02 told us that he had worked on a last-minute report using his
computer while standing up: "There was still a little bit of the task
left. It was extremely dicult. I thought my computer would bend
and break. But that was the only way to do this. [...] It was hard to
use the mouse and it was very crowded". The same participant also
told us that the urgency he felt about nishing his task-at-hand on
time made him be willing to run the risk of his computer falling.
4.1.4 Performing Tasks Suited to Public Transit’s Rhythm or Envi-
ronment.
This pattern of travel-multitasking involved choosing tasks the
participants perceived as especially well-suited to public transit’s
temporal rhythm or environment. Several participants mentioned
that they intentionally deferred certain tasks until they were on
public transit because they preferred to perform them there. Partic-
ipants who mentioned this motivation were highly familiar with
the temporal rhythms of the public transit they took: i.e., not only
its duration, but also the time intervals between each pair of stops.
This familiarity enabled them to allocate tasks that t into the time
available. P15, for example, reported a preference for listening to
podcasts during her commute on the Metro. "I usually commute on
the MRT from Monday to Friday. I use this time to listen to podcasts
related to my work. [...] Most podcast shows are thirty to forty minutes
long, and the commute time is the same, so I have this clear schedule
in which I need to do this. But if I’m not commuting, like taking a
walk, my behavior is less predictable and so I wouldn’t want to listen
to podcasts during that time." In addition, P15 commented that the
unpredictability of the Metro environment enhanced her alertness,
enabling her to pay more attention to podcast content: "I think when
I’m in the oce listening to a podcast for thirty or forty minutes, it’s
hard to focus and ensure that I’m paying attention the whole time.
[... But on the MRT] you have to keep your eye on the situation. So
many things are happening around you. [...] Your attention is always
drawn back to your current environment, and this makes you also
quite alert and attentive to what is on the podcast."
P11, on the other hand, was observed using the short intervals
in her commute to quickly respond to messages, and commented:
"If I spend all my [non-travel] time replying to messages, I feel like it
is a waste of time [... So] I usually do not reply until I’m waiting for
the bus or commuting."
Interestingly, the desire to perform specic activities while trav-
eling inuenced some of these participants’ transportation choices.
P09, for example, anticipating that she would rather take a rest
during her journey than arrive at the destination earlier, chose to
take the normal train instead of high-speed rail. As she explained:
"Although the train takes a long time, I feel that if I’m not in a hurry,
taking the train is pretty nice, because I can rest and eat in an easy
way, and feel that ’Oh! I’m able to do so many things on the train’."
To sum up, the four travel-multitasking patterns we discerned
were linked to dierent motivations as well as variation in their task
choices and expectations of their completed tasks’ quality. Such
variation, in turn, aected their levels of concentration on tasks-at-
hand and their ability to monitor the progress of their travel tasks,
as explained below.
CHI ’23, April 23–28, 2023, Hamburg, Germany Lee, Hsu, Li, et al.
Figure 1: Images of travel multitaskers engaged in travel multitasking on dierent modes of public transit
4.2 Challenges to Carrying out the Travel Task
and the Task-at-Hand Simultaneously
When multitasking, the participants divided some of their attention,
or switched their attention quickly, between their task-at-hand and
their travel task: in particular, monitoring cues indicating how much
time was left before they needed to disembark. Such monitoring
can be challenging at times irrespective of one’s other activities or
the lack thereof. As we will see, however, it can be especially di-
cult when multitasking, because the sharing of cognitive resources
for processing information between two tasks can result in low
receptivity to travel-progression cues, especially when people are
highly engaged in their tasks-at-hand. This interference becomes
more severe when the cues are hard to perceive, which can be for
a variety of reasons including their unavailability, low clarity, low
salience, and/or low reliability, as well as the diculty of mapping
them onto key journey timings. Below, we explore each of these
challenges in turn.
4.2.1 Unavailability of Clear Cues. Sometimes, the cues partici-
pants used to rely on for journey-monitoring, such as those deliv-
ered by public-transit stop-announcement systems, were simply
not present. Even when they were, however, a variety of environ-
mental constraints made them unclear and/or not salient enough
for participants to notice or understand clearly. For example, when
relying on visual information such as LED displays of stop names,
one commonly mentioned obstacle was the distance between the
passenger and the signal source. On buses, P04 told us, “the place
where it shows the information is too far away. There is only one
place that shows the information, and that is above the driver”. Other
participants mentioned that crowding on public transit not only
made the displays dicult to see, but also hard to move closer to.
“Sometimes I don’t really want to look at that map on the Metro, be-
cause there may be too many people standing in front of it. Or every
time I look at it, I have to tell other passengers to let me go through”
(P13).
When relying on auditory information from stop-announcement
systems, noises in the environment commonly interfered. This was
especially challenging when the participants’ tasks-at-hand were
also auditory in nature, e.g., listening to a podcast, as P18 noted:
“You need to pay attention to the podcast when it’s noisy, and you don’t
know where you are, especially when you need to wear headphones [. . .
as doing so] makes you unable to hear the Metro announcements. Then
you miss the stop.” To overcome this challenge, some participants
developed strategies such as trying to sense changes in the ow of
the crowd. Others reported that they simply had to abort their tasks-
at-hand: “The bus is quite noisy, and then the volume of the station
announcement is very low, so if it’s rush hour, I can’t hear clearly, so
I will watch the electronic scroll” (P14). Such obstacles caused some
participants to be uncertain about when to disembark, causing them
to become anxious and, in some cases, to choose to complete their
tasks-at-hand immediately and exit the vehicle sooner than was
strictly necessary. And most of the time, the participants had to
expend extra eort to seek other, clearer cues; and this additional
eort negatively aected the quality of their tasks-at-hand. As P02
told us, “I had to look out the windows. Some buses don’t broadcast
[stop announcements], don’t know why, maybe it’s broken, so I can’t
focus on listening to the radio.” Others felt they had to check the scroll
frequently, which was burdensome and did not allow continuous
task focus, making trips especially stressful for participants with
urgent tasks.
4.2.2 Receptivity. Even when the cues they relied on were avail-
able and clear, our participants were not always receptive to them,
normally because they were concentrating on their tasks-at-hand.
Many reported that this was especially common when the task-at-
hand was urgent, leading them to be so focused on it that they did
not notice whether the vehicle was emitting journey-monitoring
cues or not. P02, for example, told us: “When I arrived at the station,
I was like, ‘Oh my God! I got here!’, and I grabbed all my belongings
and threw them into the bag. But it turned out that I had missed my
stop.” Some said they concentrated so hard on their tasks-at-hand
that they often missed opportunities to get o the vehicle in time,
even when they had noticed their stop being announced. “I heard
that we were about two stops away from where I should get o, and I
thought I should wait and listen to the announcement for two more
stops before preparing. It turned out that I exceeded three stops. I’m
like, ‘Oh no! I didn’t hear that’” (P06).
Travel monitoring became particularly dicult when the modal-
ity of travel-porgression cues overlapped with that of the partici-
pants’ tasks-at-hand. Some even adjusted their task choices to keep
their cue receptivity high. As P10 explained, “I choose not to play
games that require my concentration.” P04, on the other hand, said
that he purposefully made himself uncomfortable to avoid failing
to notice announcements. “I prefer to stand now. Standing itself is
not that comfortable and costs you more eort. This makes you more
alert and so less likely to miss stops.” Some participants, however,
were able to use pre-planning to balance tasks-at-hand and travel
tasks without anything being sacriced. For example, anticipating
that she would have to concentrate on her work while taking the
Metro, P08 set a phone alarm to remind her it was time to leave.
Exploring Travel-Based Multitasking Behaviors in Public Transportation Context CHI ’23, April 23–28, 2023, Hamburg, Germany
“When I’m so focused on work, I don’t pay attention to the time on
my phone, I don’t want to pay attention to whether there is anyone
around me who wants me to give up my seat, or how crowded it is, or
what everyone else is doing. I don’t care about things like that. That’s
when I [...] estimate the time and set the alarm to remind myself to
get o.”
4.2.3 Cues Not Reflecting the Actual Situation. In some cases, par-
ticipants were not condent that a travel-progression cue reected
the actual progress of their trip. This was mainly for two reasons.
First, some participants did not nd it easy to map the cues they re-
ceived onto their actual travel task. For instance, the scenery outside
the window was sometimes used as a reference for when to get o
a bus, but some participants noted that this was only useful when
they were familiar with the route. As P12 said, “I’m not familiar
with the route. During the bus ride, it’s hard to tell when I will arrive. I
can’t identify where I am now by the change of scene.” This prompted
them to actively seek other cues, especially when they suddenly
realized that they might have already missed their stop. Second,
some participants mentioned that certain cues, such as information
on both in-vehicle displays and bus-tracking apps, tended to be un-
reliable or inaccurate, or that the information provided by dierent
sources was inconsistent. This not only worsened their task-at-
hand performance at the moment of its occurrence, but also had a
long-term inuence on their activity choices and transportation-
mode preferences. The unreliability of one cue source tended to
prompt the participants to expend extra eort on cross-validating
its information using multiple sources. As P10 explained, "There are
two ways to display arrival information on the bus in Taipei. One is
an electronic scroll placed in the front of the bus, showing only the
next stop. The other is an electronic board that tells you the estimated
time of arrival at each place. Normally, the scroll is not accurate, so
I usually don’t rely on it. But if there is a board that tells you the
arrival time of the coming bus, I will accept it.” Even so, however,
P10 said she was prone to using GPS information from Google
Maps to identify when to get o. She also noted that having to
cross-validate the progress of her journeys via multiple platforms
and information sources fragmented her attention, and thereby
rendered her task-at-hand performance lower than expected.
4.3 Concerns and Challenges Arising from
Public Transit’s Dynamic Environment
Last but not least, challenges arising from the constantly changing
environment of public transit itself also impacted our participants’
travel multitasking practices and performance. Details of these
challenges are presented below.
4.3.1 Unstable and Vibrating Vehicles. The most frequently men-
tioned transit-environment challenge was executing tasks-at-hand
in moving vehicles beset by heavy vibration, which at best had a
slightly negative impact on task quality, and at worst, placed the
participants in danger. Specically, although not all participants’
tasks on their phones were inuenced by vehicular vibration, some
mentioned that they often needed to perform tasks that required
precision and care. As P06 noted, “Making slides requires more skills.
You must precisely move the cursor to a certain location.” In addition,
several participants mentioned that, because of vibration, they had
to hold onto handgrips/handrails to keep their balance, with the
result that they could only perform a limited range of tasks. P11,
for instance, commented, “I could only do things that were conve-
nient to do while holding a mobile phone with one hand, like replying
to messages.” However, using one hand was still perceived as less
ecient, as P08 noted: “I didn’t have any door nearby, and worse,
nothing to lean on. All I could do was grab the handrail, and you just
swing with the carriage. I could thus only use my left hand to scroll
the phone, but it was slower.”
Additionally, participants mentioned that vehicular vibration
made it dicult to keep track of their progress when reading. To
help themselves track it, they had to highlight text constantly, as
P02 explained, "When the intercity bus wobbled, it would make me
suddenly unable to nd which line I was reading [...]. So, I had to
highlight the line I’m reading, so I can know if I lose my place later, I’ll
know where to continue reading from." Others mentioned that when
vehicles shook severely, they had to cease their tasks, as persisting
would have made them feel uncomfortable, or even nauseated: "I
don’t stare at the screen while the vehicle is moving or vibrating.
[...] It keeps wobbling and makes me dizzy" (P12). To overcome this
challenge, P19 said, he "bought a small tool with clips at both ends,
one for the phone and the other for the handle of the chair. [...] When
the bus vibrates or is unstable, the phone and clips move with the
bus and me simultaneously so you won’t feel nausea." However, in
vehicles that vibrated so severely that the participants perceived
a risk of falling, they had to pause their task-at-hand until the
vibration stopped (e.g.,P18).
4.3.2 Concerns about Constantly Changing Passengers and Close
Contact with Them. The high turnover rate of passengers in the
small and constrained public-transit environment led many par-
ticipants to express concerns about multi-tasking in such environ-
ments. These concerns involved their property, health, and safety,
and arose due to their awareness that other passengers were con-
stantly changing and that there were many strangers with whom
they might come into physical contact. These concerns not only
inuenced participants’ seat choices but also distracted them from
their tasks-at-hand, as they felt they had to increase their alert-
ness to their surroundings to facilitate quick reactions if needed.
For example, P18 mentioned that when he was reading, being sur-
rounded by a crowd of people made him worry about his personal
belongings being stolen. This worry made it dicult for him to
concentrate on the task-at-hand because he had to repeatedly check
his belongings: "When you’re focused, you would fear things being
taken away. [...] Like your phone or keys, you put them in a place
where your backpack can be locked, but when there are many people,
you’re still afraid of things being stolen, so you will still check again.
[...] I had to put them where I could easily touch them.”
Several participants mentioned that due to the COVID-19 pan-
demic, they had become more protective of their personal health.
For example, P13 mentioned that she would change her seat or
stand away from a crowd of people, which inuenced whether
she had a seat in which she could undertake her task-at-hand. "I
want to avoid people who are sneezing, and if they don’t cover their
mouths, I’m more likely to be infected. So, I used to stand at the end
of the car because I didn’t want to be exposed to this kind of cough
or sneeze." P09 also considered health issues when choosing a seat
CHI ’23, April 23–28, 2023, Hamburg, Germany Lee, Hsu, Li, et al.
on high-speed rail: “If alone, I prefer to pick a window seat, since it
makes me feel less in contact with other passengers."
Several others mentioned the Taipei Metro attack in 2014
1
, and
particularly, how it had led to them developing the habit of intermit-
tently checking their surroundings and reminding themselves not
to concentrate too much on their tasks-at-hand on public transit.
"Since that incident happened, I’ll be more aware of the environment
while watching a video [on my phone]. I’ll watch the video a little
bit and then look around at what’s happening nearby, and I’ll be
less focused on what I’m doing. I tend to focus more on observing
people around me, like fty-fty, to see if there are any weird people"
(P06). Similarly, when shadowing P04 on the MRT, we noted that he
regularly looked up and watched people getting on and o at every
stop. In his debrieng interview, he told us that he raised his head
frequently because he wanted to be alert to any potential dangers
or unexpected situations. "I like to see what the other passengers look
like, to avoid danger [...]. I assume what they might want to do or if
they have an emergency. Just in case, I’ll take a look. [...] I wanted to
wait until the door was closed". This tendency had accustomed him
to switching his attention whenever the vehicle arrived at a stop,
and as a result, we observed that his attention to his task-at-hand
was never longer than two minutes.
4.3.3 Concerns about Privacy and Personal Image. Finally, some
participants also expressed concerns about their privacy and per-
sonal image when they were aware of the presence of many other
people on public transit, and told us that such concerns aected
their task choices. For example, P26 said, “If I need to pay with a
credit card for online shopping, I will try to use it at home, because
I have no way to fully ensure my privacy on public transportation”
(P26). He later noted that, “as a journalist, I would try not to let other
people see the content of my interviews and the issues I am preparing
before publishing.” Indeed, above a certain level of crowdedness,
some participants chose not to perform any tasks-at-hand at all, not
because they did not want to, but because they perceived that their
action might negatively aect other passengers. As P13 explained,
“if it is crowded, taking out my phone might lead to me accidentally
hitting people, so sometimes I avoid this kind of action, and then I
think it’s better to do nothing.” P02, on the other hand, preferred to
only undertake simpler tasks-at-hand to avoid spending too much
time packing when someone needed his seat. “If I want to give my
seat to someone else but have to pack up a lot of stu, by the time I’m
done, he may have passed out or he’d be really pissed o.”
Interestingly, however, some participants regarded the presence
of many other people on public transit as positive pressure to per-
form tasks-at-hand, and said they would deliberately engage in
certain activities on a vehicle to present a positive image to its
other passengers. P14, for example, mentioned one occasion on
which she wanted to project a diligent attitude toward learning: “I
have some strange insistence, that is to be perfect in front of strangers,
so that people next to me will think, ‘Oh, this young lady is so hard-
working, she is not looking at the people around her but at her books.’
Anyway, at that time, I forgot to get out of the [train and...] was late
for my event because of this.” P18 also noted that he “rarely read
books in daily life, but do so on public transit. [...] Reading books looks
cool, I think [laughter].”
1https://en.wikipedia.org/wiki/2014_Taipei_Metro_attack
5 DISCUSSION
5.1 Motivation is Key in Travel-Based
Multitasking
As noted in our literature review, prior research on travel multitask-
ing has largely focused on the activities people engage in during
their journeys to leverage the travel time [
52
], and the eects on
those activities of factors predetermined by the researchers [
32
].
However, none seems to have explored the inuence of motiva-
tions behind travel multitasking behaviors. Our results suggest
that such motivations are deserving of considerable attention –
indeed, more attention than multitasking activities themselves –
because the former more clearly reect travel multitaskers’ support
requirements.
Specically, we have shown that these motivations played the
key role in determining how insistent the participants were about
performing a specic task-at-hand (vs. changing to a dierent one);
how strongly they concentrated while performing it (vs. being easily
distracted); and how much task progress they expected to make,
and how high they expected the quality of their work to be (vs.
no expectations). All of these factors inuenced their receptivity
to cues/signals relevant to their travel tasks, their task-at-hand
performance, and their ability to adapt their tasks-at-hand to their
public-transit environments. Thus, while Axtell et al. concluded
that people would adapt both their tasks and contexts to overcome
obstacles to working while traveling [
3
], our results instead suggest
that whether public-transit passengers change their task choices
may depend chiey on their motives for multitasking.
Consequently, to the best of our knowledge, the current study
is the rst to have analyzed and recognized the underlying moti-
vations of travel multitasking; and, based on its results, we argue
that recognition of such motivations has important implications
for design. We discuss those and other implications in the next
subsection.
5.2 The Mixture of Inuences on and
Challenges to Multitasking on Public
Transit
Our observational and interview data both indicate the variety of
inuences on participants’ multitasking practices and performance
while on public transit, and that these inuences arose both from
environmental dynamics, and personal motivations. However, what
makes the challenges of travel multitasking distinct from those of
multitasking in other contexts such as the workplace [
11
,
43
] is
that travelers all share the same primary task – arriving at their
destinations – which not only requires them to take actions at
very specic timings that are beyond their control, but also can
be disrupted by varying, and sometimes unpredictable, changes in
the public-transit environment. Thus, when the participants found
themselves losing track of the progression of their journeys, and
also perceived that missing their stop would be costly (whether in
terms of delaying their subsequently scheduled events, or losing
face in front of other passengers), nearly all of them assigned a
higher priority to their travel tasks than to their tasks-at-hand,
sacricing the quality and/or progress of the latter.
Exploring Travel-Based Multitasking Behaviors in Public Transportation Context CHI ’23, April 23–28, 2023, Hamburg, Germany
However, these conicts were not rare. Many participants in
our study had experienced rushing to disembark or even missing
their station/stop because cues about when to exit the vehicle were
unavailable, not salient, non-straightforward, or unreliable, and
thus failed to help them establish awareness of their journey’s pro-
gression. These cues were especially crucial to participants who
had become immersed in their tasks-at-hand. Additionally, some
such cues were delivered in modalities that overlapped with those
of the participants’ tasks-at-hand, while others had lost the par-
ticipants’ trust due to their past unreliability, and cross-validating
their information was a further distraction from tasks-at-hand. For
many participants, the trade-os between the travel task and the
task-at-hand that resulted from the aforementioned challenges and
issues could have been avoided, if the public-transit systems they
were using had provided cues that were more readily perceivable,
more reliable, and delivered at more opportune moments.
Finally, the fact that the characteristics of Taiwanese public-
transit contexts varied rapidly and profoundly further complicated
and hindered multitasking. Environmental and social challenges –
including vehicle vibration and lurching, crowdedness, fellow pas-
sengers’ unpredictable behavior, and the participants’ own desire
to sustain their personal images – could arise at any time during
journeys on any of our ve categories of public transit. While some
of these challenges did not have direct negative impacts on tasks-at-
hand, they heightened participants’ concerns about their personal
safety and well-being, and this in turn distracted their attention
from such tasks.
It is noteworthy that, although the literature on travel multitask-
ing has mentioned some of these or similar challenges, including
the need for privacy [
3
,
62
], social perceptions [
10
], and personal
security [
45
], none appear to have been investigated in depth; and
therefore, questions about their eects on travelers’ multitasking
practices have long remained unanswered. Through the identi-
cation of challenges to travel multitasking in this study, we hope
to draw transit-service providers’ and other relevant practitioners’
attention to ways in which the infrastructure and environment of
public transit, as well as novel mobile services, could better support
this increasingly prevalent behavior.
5.3 Which Task Should I Pay Attention to? The
Potential Inuences of Familiarity and
Transportation
Not all participants appeared to have encountered the same level
of conict between travel tasks and tasks-at-hand, and thus, the
amount of attention-switching they reported also varied sharply.
While various factors play a role in attention-switching, we found
its major drivers in the context of travel multitasking were 1) peo-
ple’s familiarity with the travel task and/or the task-at-hand, and 2)
how concerned they were about changes during the journey that
were specic to the mode of transportation they had chosen.
Specically, according to our observations and our participants’
self-reports, some participants seemed to manage disembarkation
well without having to frequently check the progress of their trips,
despite performing tasks-at-hand, whereas others frequently self-
interrupted to check such progress, or even paid attention to the
travel task most of the time. These data resonate well with the
concept of a multitasking continuum [
54
], with one end being con-
current multitasking (in which there is no need to frequently switch
attention) and the other being sequential multitasking (in which
frequent attention-switching is absolutely necessary). Specically,
we found that in cases where the participants reported being highly
familiar with the critical timings of their travel tasks, including
when to pack up and/or when to disembark, such tasks did not de-
mand much in the way of attention and other cognitive resources;
and thus, they were able to undertake the two tasks concurrently
and successfully. Likewise, when participants’ tasks-at-hand were
simply habitual, they were also able to perform both tasks concur-
rently and eectively. In contrast, if the participants perceived that
both tasks demanded their attention, time, or cognitive eort, they
tended to adopt sequential multitasking behavior: i.e., switched
their attention between the two tasks. Because many of the par-
ticipants’ motivations for travel multitasking were not to kill time
but to make progress on or complete specic tasks that demanded
their cognitive resources, their familiarity with the journey became
crucial to whether and to what extent they could concurrently mul-
titask; and therefore, whether they had to frequently switch their
attention between their two tasks. As noted by Salvucci et al. [
54
],
when performing sequential multitasking, retrieving a representa-
tion from memory takes time, and attempts to do so are not always
successful. In this vein, some of our participants who engaged in
sequential multitasking told us that their task-at-hand performance
was not as expected because they had needed to pay too much at-
tention to changes in their external environments. Yet, while such
diculty might make some people change their task choices [
18
],
it is important to recognize that, as our results have shown, people
are often strongly motivated to undertake specic tasks-at-hand
that lead (or indeed force) them to sequentially perform those tasks
and their travel tasks. As a result, we deem travel multitaskers who
are unfamiliar with their routes as most in need of technological
assistance to make their travel multitasking more eective.
Another notable factor was the participants’ concern about jour-
ney unpredictability. For example, they seemed to generally per-
ceive that buses/shuttles were more unpredictable than other forms
of public transit, not only in their uctuating schedules, which are
easily aected by trac conditions, but also in their jolting and
vibration that can be caused unexpectedly by poor road conditions.
Due to such unpredictable variations, many of our participants
felt that they needed to be more alert in general and to pay more
frequent attention to the progression of their trips when taking
buses or shuttles than they would otherwise. The Metro was re-
garded as largely free of the same unpredictable factors, but had
its own, in the form of constantly changing passengers. Though
of course bus passengers also change regularly, the participants
perceived that, on the Metro, they encountered much larger and
more diverse crowds of strangers, whom they were also more likely
to be physically close to.
It is noteworthy that, while our participants felt a need to in-
crease their alertness and had switched their attention away from
their task-at-hand on both buses and the Metro, the target to
which they switched their attention, and the reason for which
they switched it, diered in each case. This dierence might sug-
gest a dierent set of design implications for the dierent classes of
public-transit service providers involved. Unfortunately, however,
CHI ’23, April 23–28, 2023, Hamburg, Germany Lee, Hsu, Li, et al.
our qualitative data do not allow us to make quantitative compar-
isons of these aspects. Therefore, we encourage future researchers
to extend our investigation to the question of how the presented
challenges and concerns dier quantitatively across dierent modes
of transportation.
5.4 Design Implications
By their sheer quantity, our ndings about challenges imply that
there is considerable room for improvement in the technological
facilitation of multitasking on public transit. Our rst high-level
recommendation is therefore to take people’s motivations for travel
multitasking into account when designing or redesigning public-
transit infrastructure. Specically, future technology designed to
support multitasking on public transit should be context-aware
and employ techniques to either collect or autonomously learn
individuals’ primary or frequent motivations. This information
can be collected through prole-building during onboarding or
prompted questionnaires. Alternatively, the information can be
learned through training a machine-learning model based on fea-
tures extracted from people’s patterns of task choice, task length,
task switching, and sensor data from their devices, among others.
In addition, prompted questionnaires can be utilized to gather
labels for task sessions. Prior research has successfully character-
ized app sessions [
64
] and distinguished moments when people are
bored [
48
] or “killing time” [
13
]. We believe that a motivational
model of travel multitasking can be built upon these previous stud-
ies. Once individuals’ motivations are recognized, trip-planning
apps or public-transit services can provide suggestions of tasks or
locations tailored to them, according to whether the motivation is
related more to the available time or to the task itself. For example,
passengers who have an explicit need to complete a specic task,
such as catching up on urgent work, will be more likely to accept
location suggestions than task suggestions. In addition, travel multi-
taskers motivated by the urgency of work are likely to concentrate
strongly on their tasks-at-hand, to the extent that they might miss
their stops. As such, they may need more salient cues reminding
them to disembark than passengers with other motivations do.
Second, we learned from our participants that when travel mul-
titaskers were familiar with both the temporal and environmental
characteristics of their journeys, they could – and often did – plan
their tasks-at-hand in advance. Therefore, our second high-level
recommendation is to enhance travel multitaskers’ awareness of
the characteristics of their upcoming public-transit journeys, in-
cluding their overall durations, the lengths of the intervals between
stops, noise levels, the spatial distribution of seating and standing
areas, the locations of stop-announcement displays, etc. Such in-
formation can be supplied either directly by public-transit services
or via crowdsourcing, which has already been leveraged to obtain
public-transit crowdedness information [49].
Third, we have shown that travel multitaskers confront three
main types of challenges to their travel tasks, and that their worries
about losing track of their journey progression – and, in particu-
lar, about missing their stops – repeatedly and negatively impact
the progress and quality of their tasks-at-hand. Thus, our third
high-level recommendation is that travel multitaskers should be
supported in tracking their journey progression by addressing all
of those challenges.
Fourth, it should also be borne in mind that travel multitaskers
can be at various places within a vehicle and perform tasks involv-
ing a variety of information modalities, and that our participants
frequently switched their attention among many displays, including
but not limited to the devices they were using for their tasks-at-
hand and vehicle-mounted electronic scroll screens. As such, we
recommend that public-transit services recognize this increasingly
prevalent second-screen phenomenon [
39
,
40
] and diversify the
channels whereby real-time updates about a vehicle’s current lo-
cation, speed, and/or upcoming destinations can be accessed by
its passengers. This could be accomplished via browsers, ocial
transit-service apps, third-party apps, and vehicle-mounted physi-
cal displays of existing and new types. This would allow passengers
exibility in accessing the required information across channels,
and reduce the unnecessary attentional cost of switching between
devices.
Fifth, we recommend that passengers be enabled to set their own
time- and/or position-based reminders, which would alert them to
travel-progression information via the same devices on which their
tasks-at-hand are being conducted. Such alerts should allow cong-
uration of their dismissability, among other forms of customization
to be established by future user research. A countdown feature,
for example, might usefully be included to give passengers a sense
of urgency [
36
] about wrapping up their tasks and packing their
belongings.
If adopted, the above design recommendations can reasonably
be expected to promote less attention-switching between travel
tasks and tasks-at-hand. However, we would like to stress that such
recommendations should not be taken as encouraging people to
devote all their attention to their tasks-at-hand. Rather, based on
our observations that some participants concentrated so hard on
their tasks-at-hand that they seemed to become totally unaware of
their surroundings, our sixth and nal high-level recommendation
is to enhance travel multitaskers’ awareness of potential dangers
around them. Specically, real-time safety and security information
and warning alerts could be provided via transit services, according
to the location of the vehicle or the direction it is heading in, based
on governmental statistics and/or crowd-sourced information about
the relevant areas.
5.5 Research Limitations and Future Work
There are several limitations of this research that could negatively
inuence the generalizability of our paper’s ndings and/or the
strength of its claims. First, as it is a qualitative study based on
shadowing and interview data, its results do not allow us to make
quantitative claims about the impact or prevalence of specic fac-
tors we have identied as impacting travel-multitasking practices,
or about such factors’ interrelationships. Thus, although our par-
ticipants mentioned certain challenges and concerns more often
when discussing some transportation modes than others, and some
travel-multitasking situations than others, we cannot draw any
rm conclusions about frequency from those data. Additional quan-
titative research is therefore needed to address questions of how
the factors we have identied – such as transportation mode [
16
],
familiarity/novelty of the route [
18
], and companions [
63
,
65
,
67
] –
are correlated to the motives, challenges, and concerns identied
in our study.
Exploring Travel-Based Multitasking Behaviors in Public Transportation Context CHI ’23, April 23–28, 2023, Hamburg, Germany
Second, our data-collection approaches, including shadowing
and interviews, primarily focused on lone-passenger scenarios.
As a result, the extent to which our results apply to multitasking
behavior that occurs when passengers are traveling in pairs or
groups remains unclear.
Third, our results might have been aected by recall bias [
50
],
insofar as a considerable proportion of our ndings were derived
from the participants’ recall of their previous public-transit expe-
riences. Additionally, because our research was conducted in the
midst of the COVID-19 outbreak, some participants had not used
public transit for a long time prior to their interviews, and this time-
lag would have tended to magnify such an eect. Although a key
strength of interviewing as a method is that it enables researchers
to obtain people’s in-depth reections on their motivations, needs,
attitudes, and feelings, as well as details of their specic experi-
ences, multitasking behaviors are nonetheless situated behaviors,
and one’s own in situ reactions to environmental stimuli are often
infeasible to recall. To address this limitation, future researchers on
this topic should consider including relatively more introspective
methods such as experience sampling [
35
] and participant obser-
vation [
4
] to record people’s public-transit travel experiences in
situ.
Fourth, the durations of our observations of the participants’
behavior via shadowing were quite limited in duration. Specically,
each participant was only shadowed on one trip lasting between 15
and 145 minutes. This prevented us from observing their behavior
across multiple trips, trip purposes, transportation modes, times of
day, and times of year. As a result, we had to rely on their self-reports
in interviews to obtain such data. In addition, during shadowing,
the presence of researchers might have inuenced the participants’
decisions about what tasks to perform, and/or other aspects of their
behavior. Although we told the participants that they should behave
as they would normally do, and that we would not interfere with
them and try to stay out of their lines of sight, two participants told
us in their debrieng interviews that they were unable to ignore
the fact that they were being shadowed. Although the majority of
the ndings were based on participants’ reections, this limitation
could have prevented us from observing some natural multitasking
behaviors.
Finally, in terms of generalizability, the majority of our partic-
ipants were selected as having high levels of polychronicity. So,
despite the targeted population of the current study being travel
multitaskers, our ndings should not be assumed to be generalizable
to all public-transit passengers. In addition, our sample size of par-
ticipants was small and limited to people in one Asian country. The
participants’ age-range was also fairly limited, with none under age
21 or over 50, and an average age of 26.7. Although we can discern
similar patterns of travel multitasking behavior between the partici-
pants in our study and those in previous studies conducted in other
countries, it is unclear whether our ndings can be generalizable
to travel multitaskers who are older or younger, and/or who live
in countries where public-transit infrastructure and/or health-and-
safety policies dier sharply from ours. Moreover, given that the
current study was conducted against the backdrop of the COVID-19
pandemic, it is likely that the participants were engaging in social
distancing, further limiting the generalizability of our results to
periods without infectious-disease epidemics. That being said, how-
ever, we believe the results of the current study, though preliminary,
are an important step toward drawing research attention to the
complex focal phenomena.
6 CONCLUSION
This qualitative research on multitasking behavior on public transit
makes three main contributions to the literature. First, it appears
to be the rst study to identify and discuss the inuential role of
travel-multitasking motivations, for which it provides a novel four-
part typology. It shows how dierences in these motivations shape
travel multitaskers’ task choices, expectations about their tasks’
progress and outcomes, and receptivity to travel-task-related public-
transit cues. Second, it delineates the three main types of challenges
to travel tasks and tasks-at-hand caused by these two task classes’
mutual interference. And third, it identies three additional types
of challenges arising from instability and rapid variation in public-
transit systems’ physical and interpersonal surroundings. In addi-
tion to these contributions, we have oered ve practical high-level
design recommendations for public-transit services and other ser-
vice providers who hope to improve people’s travel-multitasking
experiences.
ACKNOWLEDGMENTS
We greatly thank all of our study participants. This project is sup-
ported by National Science Technology Council, Taiwan (110-2222-
E-A49-008-MY3, 111-2221-E-A49 -164), as well as the Higher Educa-
tion Sprout Project of National Yang Ming Chiao Tung University
and Ministry of Education (MOE), Taiwan.
REFERENCES
[1]
Reem Alzahabi, Mark W Becker, and David Z Hambrick. 2017. Investigating
the relationship between media multitasking and processes involved in task-
switching. Journal of experimental psychology: human perception and performance
43, 11 (2017), 1872.
[2]
Sally Andrews, David A Ellis, Heather Shaw, and Lukasz Piwek. 2015. Beyond
self-report: Tools to compare estimated and real-world smartphone use. PloS one
10, 10 (2015), e0139004.
[3]
Carolyn Axtell, Donald Hislop, and Steve Whittaker. 2008. Mobile technologies
in mobile spaces: Findings from the context of train travel. International Journal
of Human-Computer Studies 66, 12 (2008), 902–915.
[4]
Lynda Baker. 2006. Observation: A complex research method. Library trends 55,
1 (2006), 171–189.
[5]
Raquel Benbunan-Fich, Rachel F Adler, and Tamilla Mavlanova. 2011. Measur-
ing multitasking behavior with activity-based metrics. ACM Transactions on
Computer-Human Interaction (TOCHI) 18, 2 (2011), 1–22.
[6]
Andrea Berger, Laura Jones, Mary K Rothbart, and Michael I Posner. 2000. Com-
puterized games to study the development of attention in childhood. Behavior
Research Methods, Instruments, & Computers 32, 2 (2000), 297–303.
[7]
Rosaria M Berliner, Aliaksandr Malokin, Giovanni Circella, and Patricia L
Mokhtarian. 2015. Travel-based multitasking: modeling the propensity to con-
duct activities while commuting. Technical Report.
[8]
Thomas Bjørner. 2016. Time use on trains: media use/non-use and complex shifts
in activities. Mobilities 11, 5 (2016), 681–702.
[9]
Allen C Bluedorn, Carol Felker Kaufman, and Paul M Lane. 1992. How many
things do you like to do at once? An introduction to monochronic and polychronic
time. Academy of Management Perspectives 6, 4 (1992), 17–26.
[10]
Scott W Campbell. 2007. Perceptions of mobile phone use in public settings: A
cross-cultural comparison. International Journal of Communication 1, 1 (2007),
20.
[11]
Hancheng Cao, Chia-Jung Lee, Shamsi Iqbal, Mary Czerwinski, Priscilla NY
Wong, Sean Rintel, Brent Hecht, Jaime Teevan, and Longqi Yang. 2021. Large
scale analysis of multitasking behavior during remote meetings. In Proceedings
of the 2021 CHI Conference on Human Factors in Computing Systems. 1–13.
[12]
Kathy Charmaz. 2006. Constructing grounded theory: A practical guide through
qualitative analysis. sage.
CHI ’23, April 23–28, 2023, Hamburg, Germany
[13]
Yu-Chun Chen, Keui-Chun Kao, Yu-Jen Lee, Faye Shih, Wei-Chen Chiu, and Yung-
Ju Chang. 2021. Killing-Time Detection from Smartphone Screenshots. In Adjunct
Proceedings of the 2021 ACM International Joint Conference on Pervasive and
Ubiquitous Computing and Proceedings of the 2021 ACM International Symposium
on Wearable Computers. 15–16.
[14]
Marvin M Chun and Nicholas B Turk-Browne. 2007. Interactions between atten-
tion and memory. Current opinion in neurobiology 17, 2 (2007), 177–184.
[15]
Myoungbeom Chung. 2022. A Disembarking Notication Application for Public
Transportation Using Inaudible Frequency. Mobile Information Systems 2022
(2022).
[16]
Giovanni Circella, Patricia L Mokhtarian, and Laura K Po. 2012. A conceptual
typology of multitasking behavior and polychronicity preferences. Electronic
International Journal of Time Use Research 9, 1 (2012).
[17]
M Czerwinski, E Horvitz, et al
.
2004. A diary study of task switching and
interruptions.’CHI.
[18]
Gabrielle C Danna, Jason G Randall, and Bhindai K Mahabir. 2022. Commute
based learning: Integrating literature across transportation, education, and io
psychology. Organizational Psychology Review (2022), 20413866221132060.
[19]
Jonas De Vos. 2020. The eect of COVID-19 and subsequent social distancing on
travel behavior. Transportation Research Interdisciplinary Perspectives 5 (2020),
100121.
[20]
Kerry Allison Delbridge. 2000. Individual dierences in multi-tasking ability:
Exploring a nomological network. Michigan State University.
[21]
Jason R Finley, Aaron S Benjamin, and Jason S McCarley. 2014. Metacognition of
multitasking: How well do we predict the costs of divided attention? Journal of
experimental psychology: applied 20, 2 (2014), 158.
[22]
Rico Fischer and Gesine Dreisbach. 2015. Predicting high levels of multitasking
reduces between-tasks interactions. Journal of Experimental Psychology: Human
Perception and Performance 41, 6 (2015), 1482.
[23]
Charlotte Frei, Hani S Mahmassani, and Andreas Frei. 2015. Making time count:
Traveler activity engagement on urban transit. Transportation Research Part A:
Policy and Practice 76 (2015), 58–70.
[24]
Mattias Gripsrud and Randi Hjorthol. 2012. Working on the train: from ‘dead
time’to productive and vital time. Transportation 39, 5 (2012), 941–956.
[25]
Zhan Guo, Alexandra Derian, and Jinhua Zhao. 2015. Smart devices and travel
time use by bus passengers in Vancouver, Canada. International Journal of
Sustainable Transportation 9, 5 (2015), 335–347.
[26]
Juliet Jain and Glenn Lyons. 2008. The gift of travel time. Journal of transport
geography 16, 2 (2008), 81–89.
[27]
W James. 1890. The principles of psychology, Vol I. Henry Holt and Co. https.
doi. org/10.1037/10538-000 (1890).
[28]
Tom Erik Julsrud and Jon Martin Denstadli. 2017. Smartphones, travel time-use,
and attitudes to public transport services. insights from an explorative study
of urban dwellers in two norwegian cities. International Journal of Sustainable
Transportation 11, 8 (2017), 602–610.
[29]
Susan Kenyon. 2010. What do we mean by multitasking?–Exploring the need for
methodological clarication in time use research. Electronic International Journal
of Time Use Research 7, 1 (2010).
[30]
Susan Kenyon and Glenn Lyons. 2007. Introducing multitasking to the study
of travel and ICT: Examining its extent and assessing its potential importance.
Transportation Research Part A: Policy and Practice 41, 2 (2007), 161–175.
[31]
Imre Keseru, Ewoud Heyndels, Tan Dat Ton, and Cathy Macharis. 2020. Multi-
tasking on the go: An observation study on local public transport in Brussels.
Travel Behaviour and Society 18 (2020), 106–116.
[32]
Imre Keseru and Cathy Macharis. 2018. Travel-based multitasking: review of the
empirical evidence. Transport reviews 38, 2 (2018), 162–183.
[33]
Benjamin Kommey, Elvis Tamakloe, and Eric Tutu Tchao. 2021. Passenger Des-
tination System for Commercial Vehicles. Int. J. Appl. Inf. Technol. Vol 5, 02
(2021).
[34]
Cornelius J König and Mary J Waller. 2010. Time for reection: A critical exami-
nation of polychronicity. Human Performance 23, 2 (2010), 173–190.
[35]
Reed Larson and Mihaly Csikszentmihalyi. 2014. The experience sampling
method. In Flow and the foundations of positive psychology. Springer, 21–34.
[36]
Sophie Leroy. 2009. Why is it so hard to do my work? The challenge of attention
residue when switching between work tasks. Organizational Behavior and Human
Decision Processes 109, 2 (2009), 168–181.
[37]
Sophie Leroy, Aaron M Schmidt, and Nora Madjar. 2021. Working from home
during COVID-19: A study of the interruption landscape. Journal of Applied
Psychology 106, 10 (2021), 1448.
[38]
Jay D Lindquist and Carol Kaufman-Scarborough. 2007. The poly-
chronic—Monochronic tendency model: Pmts scale development and validation.
Time & Society 16, 2-3 (2007), 253–285.
[39]
Valentin Lohmüller, Philip Eiermann, Peter Zeitlhöer, and Christian Wol. 2019.
Attention Guidance in Second Screen Applications. In Proceedings of Mensch und
Computer 2019. 179–187.
[40]
Hila Lowenstein-Barkai and Azi Lev-On. 2018. Complementing or substituting?
News in an era of multiple platforms and second screens. International Journal
of Human–Computer Interaction 34, 10 (2018), 922–931.
Lee, Hsu, Li, et al.
[41]
Glenn Lyons and Kiron Chatterjee. 2008. A human perspective on the daily
commute: costs, benets and trade-os. Transport reviews 28, 2 (2008), 181–198.
[42]
Glenn Lyons and John Urry. 2005. Travel time use in the information age.
Transportation Research Part A: Policy and Practice 39, 2-3 (2005), 257–276.
[43]
Gloria Mark, Shamsi Iqbal, Mary Czerwinski, and Paul Johns. 2015. Focused,
aroused, but so distractible: Temporal perspectives on multitasking and com-
munications. In Proceedings of the 18th ACM Conference on Computer Supported
Cooperative Work & Social Computing. 903–916.
[44]
Patricia L Mokhtarian and Ilan Salomon. 2001. How derived is the demand
for travel? Some conceptual and measurement considerations. Transportation
research part A: Policy and practice 35, 8 (2001), 695–719.
[45] Andrew D Newton. 2014. Crime on public transport. Springer.
[46]
Jeremy Packer and Stephen B Crofts Wiley. 2013. Communication matters: Mate-
rialist approaches to media, mobility and networks. Routledge.
[47]
Steven E Petersen and Michael I Posner. 2012. The attention system of the human
brain: 20 years after. Annual review of neuroscience 35 (2012), 73.
[48]
Martin Pielot, Tilman Dingler, Jose San Pedro, and Nuria Oliver. 2015. When
attention is not scarce-detecting boredom from mobile phone usage. In Proceed-
ings of the 2015 ACM international joint conference on pervasive and ubiquitous
computing. 825–836.
[49]
Pruthvish Rajput and Manish Chaturvedi. 2019. Automatic detection of bus-
stops and bus-crowdedness using crowdsourced data. In 2019 IEEE Intelligent
Transportation Systems Conference (ITSC). IEEE, 740–745.
[50]
Karen Raphael. 1987. Recall bias: a proposal for assessment and control. Interna-
tional journal of epidemiology 16, 2 (1987), 167–170.
[51]
M Rosario Rueda, Joan P Pozuelos, and Lina M Cómbita. 2015. Cognitive neuro-
science of attention from brain mechanisms to individual dierences in eciency.
AIMS Neuroscience 2, 4 (2015), 183–202.
[52]
Marie Russell, Rachel Price, Louise Signal, James Stanley, Zachery Gerring, and
Jacqueline Cumming. 2011. What do passengers do during travel time? Structured
observations on buses and trains. Journal of Public Transportation 14, 3 (2011), 7.
[53]
Dario D Salvucci and Niels A Taatgen. 2010. The multitasking mind. Oxford
University Press.
[54]
Dario D Salvucci, Niels A Taatgen, and Jelmer P Borst. 2009. Toward a unied
theory of the multitasking continuum: From concurrent performance to task
switching, interruption, and resumption. In Proceedings of the SIGCHI conference
on human factors in computing systems. 1819–1828.
[55]
Carl Shapiro, Hal R Varian, and WE Becker. 1999. Information rules: a strategic
guide to the network economy. Journal of Economic Education 30 (1999), 189–190.
[56]
Herbert A Simon et al
.
1971. Designing organizations for an information-rich
world. Computers, communications, and the public interest 72 (1971), 37.
[57]
Jonathan B Spira and Joshua B Feintuch. 2005. The cost of not paying attention:
How interruptions impact knowledge worker productivity. Report from Basex
(2005).
[58]
Tilo Strobach, Tiina Salminen, Julia Karbach, and Torsten Schubert. 2014. Practice-
related optimization and transfer of executive functions: a general review and a
specic realization of their mechanisms in dual tasks. Psychological research 78,
6 (2014), 836–851.
[59]
Elizabeth A Styles. 2010. Psicología de la atención. Editorial Centro de Estudios
Ramón Areces Madrid, Spain.
[60]
Shanshan Sun and Yiik Diew Wong. 2022. Travel-based multitasking behaviour
in Singapore: Determinants and impacts on money-time-seat trade-os. Travel
Behaviour and Society 26 (2022), 84–95.
[61]
Jia Tang, Feng Zhen, Jason Cao, and Patricia L Mokhtarian. 2018. How do
passengers use travel time? A case study of Shanghai–Nanjing high speed rail.
Transportation 45, 2 (2018), 451–477.
[62]
Taede Tillema, Tim Schwanen, and Martin Dijst. 2009. Communicating something
condential while travelling by train: the use of a telephone conversation versus
silent modes. Transportation 36, 5 (2009), 541–564.
[63]
Harry Timmermans and Peter Van der Waerden. 2008. Synchronicity of activity
engagement and travel in time and space: descriptors and correlates of eld
observations. Transportation Research Record 2054, 1 (2008), 1–9.
[64] Niels Van Berkel, Chu Luo, Theodoros Anagnostopoulos, Denzil Ferreira, Jorge
Goncalves, Simo Hosio, and Vassilis Kostakos. 2016. A systematic assessment
of smartphone usage gaps. In Proceedings of the 2016 CHI conference on human
factors in computing systems. 4711–4721.
[65]
Peter Van der Waerden, Harry Timmermans, and Rob van Neerven. 2009. Extent,
nature, and covariates of multitasking of rail passengers in an urban corridor: A
Dutch case study. Transportation research record 2110, 1 (2009), 106–111.
[66]
Mark Wardman and Glenn Lyons. 2016. The digital revolution and worthwhile
use of travel time: implications for appraisal and forecasting. Transportation 43,
3 (2016), 507–530.
[67]
Junyi Zhang and Harry Timmermans. 2010. Scobit-based panel analysis of
multitasking behavior of public transport users. Transportation research record
2157, 1 (2010), 46–53.
