Bored Mondays and Focused Afternoons: The Rhythm of
Attention and Online Activity in the Workplace
Gloria Mark1, Shamsi T. Iqbal2, Mary Czerwinski2, Paul Johns2
1Department of Informatics
University of California, Irvine
Irvine, CA 92697 USA
One Microsoft Way
Redmond, WA 98052 USA
While distractions using digital media have received
attention in HCI, understanding engagement in workplace
activities has been little explored. We logged digital activity
and continually probed perspectives of 32 information
workers for five days in situ to understand how attentional
states change with context. We present a framework of how
engagement and challenge in work relate to focus, boredom,
and rote work. Overall, we find more focused attention than
boredom in the workplace. Focus peaks mid-afternoon while
boredom is highest in early afternoon. People are happiest
doing rote work and most stressed doing focused work. On
Mondays people are most bored but also most focused.
Online activities are associated with different attentional
states, showing different patterns at beginning and end of
day, and before and after a mid-day break. Our study shows
how rhythms of attentional states are associated with context
and time, even in a dynamic workplace environment.
Engagement; Attention; Multi-tasking; Focus; empirical
study; workplace; computer logging; experience sampling
ACM Classification Keywords
H.5.3 [Information Interfaces and Presentation]: Group and
Organization Interfaces; K.4.m [Computers and Society]:
In recent years, a great deal of attention in the CHI
community has been directed to understanding disruptions
in the workplace, due to interruptions and task-switching,
e.g. [4, 7, 10]. While it is important to investigate how a
digital environment can introduce distractions, little
research has been directed to the converse: understanding
the nature of engagement in activity in the workplace. This
is important because if we can gain insight into when
people are engaged and involved in their work, this can
inform the design of tools and interfaces to promote a better
The dynamic nature of the workplace can cause attentional
states of information workers to change depending on many
factors: the task-at-hand, interactions, their affective state,
interruptions, and other contextual conditions, as well as
online activities which constitute a large part of their work.
Studies in the field of organizational and management
science have investigated how people allocate their
attention in the workplace, e.g. , but have mostly
ignored online activity. However, given that information
workers mostly engage in digital activities and tend to
multitask frequently, digital work patterns can cause
fragmented attention and changes in engagement in work.
Under these premises, we feel that it is important to
understand, broadly speaking, how people's attentional
behaviors, and consequently, a notion of engagement,
changes across activities and contexts in a real-world
workplace environment. In the field of HCI, precision
tracking methods are being developed to study in situ
behavior including online activity, allowing us to gain a
fairly precise "micro-view" into how human behavior and
online activity are related (e.g. ).
In this current paper we report results from an in situ
tracking study using online activity logging and experience
sampling (i.e., probing the user throughout the day) that
enabled us to discover how online activity is related to
different attentional states. We first present a theoretical
framework of four different attentional states derived from
different combinations of engagement and challenge
experienced while performing online activities. We then
further characterize each state in terms of the online
activities that people perform as they report experiencing
that state. We utilize the attentional states to explain how
people’s behavior changes over the course of a day, and at
periods contiguous to a break in activities. We discuss how
the proposed attentional states can be used in real life
settings to better understand and improve the workplace
Permission to make digital o r h ard copies of all or part o f th is w ork for personal or
classroom use is granted without fee provided that copies are not made or distributed
for profit or commercial advantage and that copies bear this notice and the full citation
on the first page. Copyrights for components of this work owned by others than ACM
must be honored. Abstracting with credit is permitted. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior specific
permission and/or a fee. Request permissions from
CHI 2014 , April 26 - May 01 2014, Toronto, ON, Canada
Copyright 2014 AC M 978-1-4503-2473-1/14/04…$15.00.
RELATED WORK: MULTITASKING AND ATTENTION
We first review studies related to attention and work in the
HCI field, and then concepts closely related to engagement.
Multitasking and Disruption
A large body of work has focused on how multitasking
impacts attention in the workplace, primarily focusing on
the distraction caused to an ongoing task that is interrupted
by another activity. Czerwinski et al.  showed from a
diary study how information workers switch activities due
to interruptions in the workplace, focusing on the difficulty
of the continuous switching of context. Iqbal and Horvitz
 studied how external interruptions cause information
workers to enter into a ‘chain of distraction’ where stages of
preparation, diversion, resumption and recovery can
describe the time away from an ongoing task. Gonzalez and
Mark  reported on how information workers
conceptualize and organize basic units of tasks and how
switching occurs across these conceptual units. In the
mobile domain, Karlson et al.  found that tasks on
mobile phones become fragmented across devices and they
identified challenges that exist in resuming these tasks.
While most studies have looked at distraction due to
multitasking, no study to our knowledge in the HCI field
has focused on the converse--how engagement and
challenge are associated with a person's current type of
activity. Such characterizations can provide insight into
when a person is focused and consequently more productive
as well as providing an understanding of when downtime
occurs and what types of activities entail lack of focus.
Engagement in the workplace
Other theoretical constructs exist that can be related to
engagement in activity. Cognitive absorption refers to when
people experience total immersion in an activity,
characterized by deep enjoyment, a feeling of control,
curiosity, and not realizing the passing of time. Cognitive
absorption has been shown to be associated with ease of use
and perceived usefulness of IT . Cognitive engagement is
similar to absorption, involving curiosity, deep interest and
attentional focus, but without a feeling of control of the
Mindfulness refers to a psychological state focused on
phenomena (both externally and internally) with the
emphasis that attention is geared to the present moment .
Weick characterizes mindfulness in organizational work as
being aware of fine detail, affording the capacity to
discover and manage unexpected events .
Flow refers to a state of total immersion in an activity,
where according to Csikszentmihalyi : "Nothing else
seems to matter." High challenge and high use of one's
skills are preconditions for the flow state; however, their
presence do not guarantee that the flow state will occur.
Tasks that are not challenging rarely are associated with
flow, whereas tasks that present challenges, utilizing one's
skills, and that require attention, can be associated with the
flow experience . Time spent with engaging and
challenging activities is positively correlated with a high
quality of experience .
Flow, absorption, and cognitive engagement have been
found to be associated with high positive affect [1, 3, 25],
while states of boredom are associated with negative affect
. On the other hand, Schallberger  found that
challenge in work could involve both high positive and
negative affect. The underlying dimension is activation
which could relate to either type of affect. Testing this idea,
Gross et al.,  found that positive events in the workplace
result in resource replenishment, especially under
conditions of chronic stress or duress. They argue that
positive events could either replenish or deplete cognitive
The concepts of flow, engagement and absorption are
relevant to our work as they refer to active states of
attention, e.g., as Weick and Sutcliffe  describe: "the
capacity to take action." More specifically, they are
associated with times when people are highly engrossed in
their activity. However, we are also interested in states of
attention when engagement in work may not be high.
ACTIVITY ENGAGEMENT AND CHALLENGE IN WORK
We are interested in gaining a perspective on the counter
phenomenon to distractedness due to digital media (e.g. [4,
7, 10]): what is associated with people's engagement in their
digital activity at work? For example, if we consider email
usage, on the average, are people engaged and challenged
in managing their email or is this more of a mechanical
task? Are there certain times of the day when people have
more of a focused effort in their work and other times when
they tend to be less engaged? Are people happier when they
are focused in online activity or rather when the work they
are doing is more of a rote task?
We first began conceptualizing this problem by measuring
engagement in activity. However, measuring engagement
does not reveal a full picture about how one relates to an
activity. One can be engaged in work that is quite effortless,
such as copying figures, or filling out forms. On the other
hand, one might be engaged in a task that is more
consuming, presenting a challenge to their skills, i.e., that
involves a mental effort such as writing an article. It is
important to consider both engagement and challenge
together, as these have been associated with creativity in
Therefore, to measure engagement in work where people
are also expending mental effort, we also consider how
much of a challenge that activity presents to the user. We
define challenge as the amount of mental effort that one
must exert to perform an activity. We therefore measured
two dimensions that we feel are highly relevant in capturing
task involvement the workplace: the degree to which one is
engaged in the activity, and the degree to which one is
challenged in the activity. The choice of these dimensions is
informed by those used commonly in studies measuring
quality of experience, for example, in relation to work and
leisure (for a review see ).
Engagement in work has been studied extensively. We
follow the definition of Schaufelli et al. , who define
engagement as a state of mind where one feels absorbed
and dedicated in work. For a review on how engagement
has been characterized and studied with work, see .
Importantly, users’ self-ratings of engagement, found to
have situational validity, have been used in numerous
experience sampling studies--see  for a review.
Challenge has also been studied thoroughly in the
workplace and has been validated as a construct in
experience-sampling studies as part of the experience of
"flow." See  for a review of different contexts in which
challenge has been measured.
Focus, Boredom, and Rote work: A theoretical
To visually conceptualize different types of task
involvement that people might experience in the workplace
at different times in terms of engagement and challenge, we
present a theoretical framework, as shown in Figure 1. We
expect that people fluctuate across these attentional state
boundaries throughout the day, depending on the task,
interactions, and other contextual factors.
The upper right quadrant (Q1) indicates that at times people
may be highly engaged in an activity and also challenged.
This quadrant represents a temporal state when people feel
absorbed in an activity, i.e., are "active" in their focus of
attention because the activity requires some amount of
mental effort. In English slang terms, perhaps the best
characterization is that people are "into" their work. We are
interested in examining those times during the workday
when people experience activities of this nature. We apply
the label of "Focus" for this quadrant to refer to a state
where people are actively focused and feel that the activity
affords some degree of challenge to their particular skill set.
The upper left quadrant (Q2) refers to times when people
feel engaged but not challenged. This state can characterize
an activity that requires attention but requires little mental
effort to accomplish. An example might be transcribing
numbers or playing an online game such as solitaire. We
label this quadrant "Rote" to refer to a state where people
are engaged, but the work is not challenging. Rote work is
defined as "mechanical or unthinking routine or repetition"
The lower left quadrant (Q3) depicts those times during the
workday when one feels neither engaged nor challenged in
their work. These feelings could be consistent with a feeling
of boredom and we label this quadrant "Bored". The lower
right quadrant (Q4) describes a state where one feels
challenged but is not engaged in work. An example of such
activity is when a software developer feels that a bug is
very difficult to solve and has little to no interest in working
on it. We label this quadrant "Frustrated."
We emphasize that our labels are merely used as referents
and may not fully characterize the definitions precisely. It is
worth mentioning that this set of dimensions is related to
the dimensions used to measure the preconditions for the
experience of flow . People who experience flow
describe the experience as one involving high
concentration, engagement, absorption, and challenge in the
activity. It is important to note that our framework does not
specifically identify a flow experience. Rather we are
simply using the concept of flow as an example of the type
of experience that people could experience if people's self-
reports occur in the upper right quadrant. While the Focus
quadrant in our framework is most relevant to flow, it
captures a subset of characteristics specific to flow, namely
engagement and challenge. We feel that it is important to
understand when people are experiencing a "Focus" in work
activity because it has been shown that being highly
engaged and challenged in work is correlated with
motivation, activation, concentration, creativity, and
How do people's engagement and feeling of challenge
correspond with their task activity? The workplace is
dynamic and we expect that information workers can
change their psychological states of attention depending on
a host of factors: their task-at-hand, interactions, their
affective state, interruptions, and other contextual
conditions. Our goal is to understand how a feeling of being
involved in work is related to the use of digital technology.
We have broken this broad question down into several
RQ1. How is affect, in terms of valence, associated with
different attentional states? Are people happiest when their
attention is focused in the workplace? Valence  refers
Figure 1. A theoretical framework of quadrants
representing different attentional states in the
to the range of positive to negative emotions one might feel
(usually together with the dimensions of arousal) and has
been widely used to study affective state . We will
investigate the relationship of the different attentional states
in our framework using valence. In particular, we will
examine whether focused effort in work is associated with
positive affect as flow, engagement and absorption studies
suggest [1, 3, 25], or rather, because it also encompasses
challenge , is it associated with negative affect? Also,
as Tschan et al.  have shown that one's organizational
work role is related to the emotional quality of life, we will
examine the influence of work role as well.
RQ2. How do people's attentional states change over the
course of the day? Can we find temporal patterns that
correspond to focus and boredom in individuals? Some
work has suggested that people may have different
"rhythms" of work. Begole and Tang  looked at people's
email usage and showed that people tend to exhibit fairly
regular rhythms for some digital activity. With this research
question we examine whether we can detect discernible
temporal patterns of behavior of focused attention and
boredom concurrent with other media usage.
RQ3. How are different attentional states related to online
activities? Related to RQ2, we examine how different types
of digital activity relate to measures of engagement and
challenge, i.e., to the quadrants as shown in Fig. 1. It is an
open question to what extent people feel engaged and
challenged when they conduct activities such as reading
email or using Facebook in the workplace.
RQ4. How do people's attentional states change over the
course of the week? Might people's attentional states vary
depending on the day of the week? Does the so-called Blue
Monday effect, where people are in a bad mood on
Mondays , affect the ability to focus, or after a weekend
break might people be more focused? Does focused
behavior wax or wane over the week? Here we compare
attentional states to online behaviors over the week.
RQ5. Does a break in work replenish attentional
resources? This question addresses two times of the
workday: (a) Are people more (or less) focused at the
beginning or end of day? (b) Are people more (or less)
focused before or after lunch, which represents a mid-day
break? There is some reason to believe that people might be
more focused when starting work at the beginning of the
day or after a break. A study of rulings of Israeli judges
found that more favorable rulings were given at the
beginning of the day, and after a lunch break. . One
explanation for this result could be that breaks can lead to
higher positive affect (possibly accompanied by lower
mental workload) which in turn can restore people's mental
resources when depleted .
We conducted an in situ study in the fall of 2012 at a large
U.S. corporation. We used a mixed-methods approach
where we combined automatic data collection of digital
activity with experience sampling. The automated data
collection allowed us to track a wide range of digital
activities with detailed precision. Experience sampling was
used to collect user perceptions of engagement and
challenge, as well as other self-report measures at intervals
throughout the day. We also deployed surveys for other
subjective and demographic measures. Further details of
these, and other, measures not reported in this paper can be
found in  .
Participants were recruited through advertising,
convenience sampling and recommendations of
participants. Thirty-two people (17 females, 15 males)
participated. Participants included researchers (15),
managers, administrators, an engineer, a department
director, a designer, and a consultant.
Methodology. Each participant was observed for a period of
five days, Monday through Friday, for most people. When
participants traveled or missed a day, they made up the
missed day the following week (in most cases). The
computer logging software and experience sampling
software were installed on participants' computers the
Friday before the study began. Participants were assured of
anonymity in their data.
We logged online interactions with custom-built software
that captured all activity in the Windows 7.0 Operating
System. This included beginning and end times for the
lifespan of every window, and the beginning and end times
for every instance of every foreground window. Mouse and
keyboard activity were captured, as was computer sleep
mode, so that we could ignore periods of time when a
window was open but was not being used in the foreground.
Capturing what email was being read or any other
application interaction was not collected due to privacy and
We used experience sampling, in the form of a small pop-
up window that appeared on the computer screen to capture
the participants' perspective in situ, i.e., as the situated
nature of the environment changes. Experience sampling
has been shown to have internal validity as well as external
validity . Experience sampling has been used extensively
in studies to capture the flow experience . We used a
hybrid interval-contingent and event-contingent sampling
approach . The sampling was done: 1) whenever a user
left email after uninterrupted active use in that application
for at least three consecutive minutes or when in Facebook
after a full minute, and 2) whenever a user logged into
Windows or unlocked the screen saver (event-contingent).
If 15 minutes passed without a sampling, then a probe was
Participants were instructed to go about their usual workday
activities and were told to answer the experience sampling
probes when the probe windows popped up on their
computer screens. We emphasized that they should answer
the probe questions as accurately as possible but they could
cancel the probe window at any time. Subjects were given
the following verbal and written instructions:
"Sometimes( the( rating( scale( will( pop( up( and( may( annoy( you,(
especially( if( you( were( in( the( middle( of( doing( something.( If( you(
feel(annoyed,( do(not( rate(your(mood( based(on( the(annoyance(of(
the(pop:up( window.(Instead,(rate( your( experience( based( on( the(
task( or( interaction( you( were( doing( at( the( time( of( the( pop:up(
the( annoyance( of( the( pop:up( window,( then( hit( ‘cancel’( and( the(
We used rating scales used in other experience sampling
approaches  to measure the following: for Engagement,
participants were asked 'In the task/interaction you were
just doing: How Engaged Were You?' using a 6-point
Likert scale (0=Not at All; 5=Extremely). To measure
Challenge, participants were asked the same question as
above, but instead: "How Challenged Were You?' using the
same Likert scale: (0=Not at All; 5=Extremely). We also
measured Valence (positive and negative affect) and
Arousal using the question “Please rate how you feel right
now”, based on Russell’s 2-dimensional Circumplex model
. Valence was measured on a horizontal scale which
corresponded to a range of -200 (negative affect) to +200
(positive affect). Arousal was measured on a vertical axis
that crossed the Valence axis using a range of -200 (low
arousal) to +200 (high arousal). Subjects were asked to
click with their cursor on that point in the 2x2 grid that best
expressed their feeling "right now." The timestamp when
participants submitted the probe was recorded. Valence
measures have been reported to have high internal
consistency . For a review of the Circumplex measure
for Valence/Arousal, including its validity, see .
We collected data on each of the 32 participants for 5 days
each, for a total of 160 person-days, or 1,509 hours of data
collection. Our computer logging software collected 91,409
computer window switches. We collected 2,809 experience
sampling probes. Each person averaged 17.56 probe
responses per day, for an average of 87.8 probe responses
per participant over the five study days.
Experience sampling studies of flow have mostly used
normalized scores in analyses . We thus normalized all
responses. We chose to exclude the mid-range values and
just use the top and bottom thirds of the normalized
responses as our intent was to investigate those aspects of
the participant experience which we felt better
corresponded to our framework in Figure 1. Mid-scale
ratings are more ambiguous in their interpretation. We thus
combined the top third of the normalized Engagement
ratings and top third of the normalized Challenge ratings to
create the category of "Focus" (Q1, see Fig. 1). The top
third of normalized Engagement ratings and bottom third of
normalized Challenge ratings were combined for the
category of "Rote" (Q2). The bottom third of normalized
Engagement ratings and bottom third of normalized
Challenge ratings were combined for the category of
"Bored" (Q3) and the bottom third of normalized
Engagement and top third of normalized Challenge ratings
were combined to yield the category of "Frustrated" (Q4).
As only seven responses occurred in Q4 ("Frustrated"), we
disregarded this category for the rest of our reported
analyses. Of all the probe responses, 42.9% occurred in one
of the four quadrants in Fig. 1. All participants gave ratings
in two or more quadrants; only five participants did not
have a rating in all three quadrants during the study period.
Participants' end-of-day ratings on their feeling of
productivity for the day showed no significant association
with any of the four quadrants.
RQ1. Valence and activity involvement
Our first research question asked how attentional state is
associated with positive affect, represented by the Valence
measure. We compared Valence self-reports between the
three quadrants. We used a linear mixed model (LMM),
with Subjects as random effects, to handle the correlated
data. There was a significant difference of Valence levels
among quadrants: F(2, 1134)= 53.17, p<.0001. A
Bonferroni test set at .05 showed a significant difference
among all means. Contrary to our expectation, participants
had the highest positive affect when they were doing "Rote"
work: being highly engaged but not very challenged, and
were the least happy when bored (Q3). Mean Valence (non-
normalized) ratings (on a scale of -200 to +200) are:
Focus=34.49, SE=7.15, Rote=77.36, SE= 7.76
Prior work in flow suggests that being in a state of flow
causes people to be happy ; however, our results did not
find this to be the case for Focus, the state in our framework
closest to flow. To investigate this further, we reasoned that
focused activities may occasionally cause stress, which may
be responsible for why people are not happiest when
reporting they are focused (Q1). Depending on the
situation, stress can influence affect .
To further understand the relationship between affect and
the attentional states in our framework, and stress in
particular, we looked deeper into the mood ratings of
Valence and Arousal collected through the self-reports.
Stress is defined as high arousal and low valence  and
has been well validated with experience sampling . We
normalized the Valence and Arousal ratings. We divided up
our valence and arousal measures into four categories,
generically labeled: "Happy" (Valence >0 and Arousal >0);
"Stressed" (Valence < 0 and Arousal > 0); "Calm"
(Valence > 0, Arousal < 0) and "Bad Mood" (Valence < 0,
Arousal < 0). Note that self-reports of '0' are not included in
the mood ratings. Again, we use these terms simply as
referents for readability; mood states associated with the
circumplex model are more nuanced .
Table 1 shows the counts of all participants' self-reports of
Mood Types x Attentional states. To handle the correlated
data within participants, we conducted a generalized linear
mixed model analysis (GLMM) in SPSS which can be used
for categorical dependent variables. We found a significant
relationship between Mood Type and Quadrant: F(6,
1132)=45.76, p<.0001. In Q1 (Focus), most people self-
reported as happy. Yet when people do rote work (Q2), they
also mostly report being happy. When people are bored
(Q3), they also mostly report being in a bad mood. Of all
stress self-reports, most occurred in Focus (61.3%),
whereas only 16.0% occurred in Rote and 22.6% in Bored.
Therefore, because a higher percentage of stress reports
occurred in Focus, this could be an explanation for why
people did not report having the most positive affect in this
state. When people are consumed by an activity, it can be
either gratifying or stressful, depending on the context .
As degree of work involvement could be tied to gender or
one's work role , we examined these factors. In the
survey, participants identified their work roles and these
were coded into three categories: concerning
Administration and technical support (5 people), Research
(19), and Management (8). A multivariate GLM test
showed no significant difference for Gender or Work Role
with attentional state.
RQ2. Focus at 3 p.m.: Time of day and activity
In this research question we reasoned that Focus and
Boredom reports range over the time of day in relation to
other digital activity (and other contextual factors which
could be related to time). Fig. 2 shows how self-reports in
the Bored and Focus quadrants change over the course of
the day, averaged over all days and all subjects. The time
span is 7 a.m. to 9 p.m.
Overall, participants report being more focused than bored
in the workplace. People are most focused in their work
mid-afternoon, with a peak at 2-3 p.m. when the use of
productivity apps (e.g., Word, Excel, Visual studio), Email,
and viewing the Inbox/Calendar app are at their highest
usage. Focus is also high at 11 a.m., which is generally
Table 1. Counts of self-reports: Mood Type over the
different quadrants. Row percentages are above column
percentages in parentheses.
Figure 2. Focus, Rote and Boredom ratings over the course of the day, in relation to other digital activity, averaged over 32
subjects, 5 days. Error bars for Focus, Rote and Bored show SE of the mean.
Table 2. Counts of activity reports by day of week. Row
percentages are above column percentages in parentheses
before a break for lunch, when the reports then dip. After
peaking mid-afternoon, Focus reports continue to decline
until when most people typically leave work. The majority
of participants report being most Bored at the beginning of
the day (9 a.m.), and Bored reports peak at 1 p.m. Boredom
is at the lowest at 2 p.m. Remote communication (e.g.,
Skype, Instant messaging) is highest at 10 a.m. and between
2 and 3 p.m. The use of FB, non-work email (i.e., web
email), and information seeking (i.e., web search) is done
continually throughout the day in a fairly uniform manner.
RQ3. The not-so-boring work of email: Digital activity
While fig. 2 shows the data averaged over all participants,
in this research question we take individual differences into
account, investigating how different types of online activity
relate to the amount of engagement and challenge
experienced. We compared computer activity that occurred
in a window of time 10 minutes prior to each probe, for the
most frequently used applications selected from fig. 2
(duration measured in seconds): Email reading/writing
(Email), Facebook (FB), Email inbox and Calendar
viewing1 (Inbox/Cal), and counts of: switches on the
Internet (i.e., Internet surfing), and computer window
switches (Win Switches) were analyzed. Means are shown
in Figure 3.
Using a linear mixed model (LMM) with Subjects as
random effects, significant differences were found among
quadrants in Figure 1 with Email: F(2,1122)=4.59, p<.01. A
Bonferroni comparison set at .05 showed that users spent
significantly less time on Email while reporting Bored
compared to Focused (see Fig. 3). There were also
significant differences with FB (F(2, 1055)=12.08,
p<.0001), and a Bonferroni test at .05 showed that users
spent significantly less time on FB in the Focus state,
compared to both the Bored and the Rote states. Internet
surfing showed a difference among quadrants: F(2,
1 Note that Email refers to reading and writing email; Inbox/Cal
refers to only when the Email Inbox is in the active window.
1134)=6.46, p<.002, and a Bonferroni test (.05) showed that
participants spent more time Internet Surfing while in the
Bored state compared to the Focused state. Using a log
transform of the amount of Win Switches due to lack of
normality, differences were also found: F(2, 1115)=5.19,
p<.006, and a Bonferroni test (.05) showed more Win
Switches in the Bored state compared to the Focused state.
A log transform of the amount of Inbox/Calendar use
showed a trend for a difference: F(2, 1127)=2.78, p<.06. A
Bonferroni test showed a trend (p<.06) that more
Inbox/Calendar use occurred in the Bored than Focus state.
Therefore, we found that attentional states vary with types
of digital activity. With email, on the average, people report
least being in a Bored state. In contrast, switching windows,
surfing the Internet, and using the Inbox/Calendar are
associated with a Bored state. When people use FB, they
generally do not report being in a Focused state.
RQ4. Bored Mondays: Days of week and activity
Feelings of boredom and focus may vary depending on the
day of the week. Table 2 shows a breakdown of self-report
counts in each quadrant, by day of the week. A GLMM (to
handle the correlations within participants) shows a
significant relationship of Day of Week with attentional
state: F(8, 1130)=4.86, p<.0001. Participants report most in
the Focus quadrant on Mondays but also they report most
being Bored on Mondays. People do most rote work on
Thursdays. A Bonferroni test set at .05 showed reports on
Monday and Tuesday are significantly different than reports
on Wednesday, Thursday, and Friday.
To investigate further whether attentional states might be
tied to specific online activity, we compared the means of
different types of computer usage over Day of the Week:
Email, Facebook, Inbox/Cal, Internet Surfing and Win
Switches. Using a LMM, we found that only Win Switches
showed a significant difference (F(4, 157)=3.03, p<.02) and
Internet surfing showed a trend (F(4, 157)=2.21, p<.07),
over Day of Week. A Bonferroni test (.05) showed Win
Switches were significantly higher on Monday (M=661.2
switches/day, SE=69.60) than Friday (M=390.7 Win
Figure 3. Means of online activity (sec. and counts 10 min.
prior to probes) for the quadrants. Error bars show SE.
switches/day, SE=28.7), and also that Internet surfing is
higher on Monday (M=280.8 switches/day, SE=42.9) than
Friday (M=151.3 switches/day, SE=16.2). Interestingly,
Table 2 shows about double the incidence of reports in the
Bored quadrant on Monday (27.8%) compared to Friday
(13.3%), along with higher window switching and Internet
surfing. Thus, we find a relationship with online activity
and Mondays, the day when people report being the most
bored, but at the same time also the most focused.
RQ5a. Activity involvement: beginning and end of day
Are people more focused or bored at the beginning or end
of the day? We contrasted self-reports at the beginning and
end of the day along with the change in online activity. For
beginning and end times of day, we used the first and last
hour of each participant's data. To correct for a lack of
normality in the distributions, we did log transforms of
Email and Win Switches. We also analyzed Web email as
this is non-work email and could be related to boredom.
A related-samples Wilcoxon-signed rank test showed no
difference in proportion of self-reports in the Focused or
Rote quadrants in the first and last hour of the day. A slight
trend shows that participants reported being more Bored at
the beginning rather than at the end of the day, p=.10. A
paired t-test showed no significant difference in Valence or
A paired t-test of First and Last Hour (Table 3) shows that
significantly more time is invested in managing corporate
email in the first, compared to the last, hour. We also find
that more time is spent with the Inbox/Cal as an active
window in the first, compared to the last, hour. Our
participants switched windows significantly more in the
first hour, compared to the last hour. Productivity App
usage shows the contrary: significantly more time is spent
in the last, compared to the first hour of the day.
Q5b. Activity involvement: before and after a mid-day
We examined self-reports in the hour before and after a
mid-day break (i.e., lunch). Mid-day break was defined as:
1) a break of 20 minutes or longer in computer activity)
between the hours of 11 a.m. and 2 p.m. (which is when the
company cafeteria was open), and 2) if participants
responded in the first probe after the break that they had
had a scheduled face-to-face meeting, then this break was
excluded from analysis. As not all people who took a break
may have eaten lunch, we label this as a "mid-day" break.
A related-samples Wilcoxon-signed rank test showed no
difference in proportion of self-reports in any of the
quadrants before and after the mid-day break (Table 4). A
paired t-test showed no difference in Valence or Arousal.
A paired t-test revealed that only Web email and Facebook
showed significant differences before and after a mid-day
break. In both cases, the usage increased after the break. A
weak trend showed that productivity apps increased as well.
DISCUSSION AND CONCLUSIONS
While a large body of research in multitasking has focused
on distractions, our research examines the alternative view–
-various levels of activity engagement in the workplace, in
particular, an attentional state of 'Focus.' Our study provides
three main contributions: a theoretical framework to explain
different attentional states in the workplace; a novel
methodology combining computer activity logging with the
user’s perceptions; and empirical results showing how
different attentional states are associated with contextual
factors in the workplace: valence and mood, online activity,
time of day, time during the week, and the role of breaks.
In sum, our results show that overall, our participants had
more focused attention than boredom in the workplace.
Table 4. Means (SE) in seconds of online activity in the hour
before and after a mid-day break. *=p<.05.
Table 3. Means (SE) in seconds of online activity in the first
and last hour of the day. *=p<.05.
Focus peaks mid-afternoon while boredom peaks earlier in
the afternoon. Unexpectedly, we found that people are
happiest doing rote work; we explain this by showing that
focused work can involve stress. We also found that
people's attentional states shift as their online activities
change, e.g., email can be rote or focused work while
Facebook does not require focused attention. We also found
that day of the week is associated with attentional state:
Mondays appear to be people's most bored day. Our result
contributes to the debate on whether a "Blue Monday"
effect exists, cf : perhaps on Mondays people are not
"blue" but rather bored.
Previous studies of tracking workplace behavior with
ethnographic and automated methods (e.g. [7, 15]) did not
capture the user perspective. For example, it could not be
known how engaged a user was with a window in active
use. With our experience sampling method we were able to
periodically gain insight into what the user was
experiencing at the time that the computer windows were
actively being used. This enabled us to understand on the
average how people experience online activity, e.g., that
when people switch windows they are bored. We hope that
these results can lead to further research.
While being in a state of flow (high challenge and high
skill, associated with high engagement) is thought to result
in increased happiness and satisfaction , the
corresponding state of Focus in our framework does not
yield the most happiness. In fact, our participants were
happiest when doing rote work. Our result of Focus self-
reports is consistent with findings of Schallberger , who
found that high challenging activities at work are associated
with both negative and positive "activation", which refers to
both high positive energy and high stressful feelings.
Similarly, our results are loosely related to findings by
Tschal et al. , who cite evidence supporting both
replenishing and depleting effects of positive events in the
workplace. Our results are counter to those found in flow
and absorption studies which generally find highly positive
experiences [1, 3] which suggests that our Focus state is
distinct from a state of flow or absorption. We found that
activities that demand high engagement and high challenge
can in fact also involve stress, as well as happiness.
Our framework can be used by others to assess engagement
and challenge in work activities. Our results are based on
repeated responses of participants over a period of five
days, roughly 40 hours. Though Engagement and Challenge
have been validated as separate dimensions capturing
experience, as a first step our results suggest "situational
validity", or the internal logic  of the framework. The
probes occurred in a variety of contexts and times yet on
the whole seemed to capture what is intuitive.
Our study shows how different attentional states vary in the
workplace according to context: type of online activity,
time of day, and even day of the week. We had expected to
find higher focus reports at the start of the workday and
after a mid-day break, but various contextual factors and
individual differences could explain this. A future study
could target analyzing attentional states at key times to test
whether breaks can replenish attentional resources .
Our work extends the multitasking literature which has so
far been agnostic about attention before a distraction. It is
possible that if people are doing boring or rote work, they
might be more easily distracted, and thus susceptible to
interruptions. Our FB result (RQ3) is consistent with this
idea: people are not focused when they use FB. Similarly,
Internet surfing and window switching are both associated
with the Bored state, activities we think of as interruptions.
Thus, our work raises the question: it may not be the
interruptions that break focus; it may be that lack of focus
comes first, leading to susceptibility to interruptions.
The results of this study suggest that people may gradually
move into a Focused state (see Figure 3). Activities that are
more personal and less critical, e.g., Facebook and personal
email, may allow people to slowly ease into a more
engaging and productive state when they more heavily use
Productivity Applications, as we found.
How can these results be used in practice? We provide a
methodology for assessing people's attentional states as
focused, bored, or doing rote work, which can be applied in
a range of studies, for example in examining the effects of
tool adoption or organizational interventions. Our results
provide basic and valuable information about workplace
behavior that can lead to further studies on how to promote
focus. Further, our results can help address a long-standing
question in the domain of interruption and multitasking:
when are opportune moments to interrupt? We would
propose not interrupting users when they are in the focus
state unless the topic of interruption is of high priority or is
highly related to the work in focus. Finally, we believe that
our results can be used to inform the design of workplace
tools so they can promote more focus during use.
Our participants were highly educated information workers
(all had at least a Bachelor's degree). We must be cautious
in generalizing our results to a broader sample. However,
our sample of 32 participants is more than double that of
other work observation studies, e.g., [4, 7]. We feel that
data collected for five full days per person enabled us to
analyze the variability of attentional states across a range of
contexts. Also, the experience sampling methodology can
interrupt participants. We did carefully instruct participants
not to reflect their annoyance of the probe in their rating.
We realize that our attentional state labels may not reflect
the true construct of what we were measuring. Quite
literally, participants were rating how engaged and
challenged they felt at that moment. It is therefore more
accurate to consider the quadrants in our framework in
terms of these dimensions than the referent labels we used.
Studying focused attention in the workplace provides a
counterpoint to the study of digital media distractions. We
hope that our study can lead to comprehensive approaches
to studying digital media use and effects in the workplace.
This material is based upon work supported by the NSF
under grant #1218705. We thank Duncan Brumby and
Alfred Kobsa for their valuable comments.
1. Agarwal, R. and Karahanna, E. Time flies when you're
having fun: Cognitive absorption and beliefs about
information technology usage. MIS Quaterly, 2000, 24
2. Begole, J.B., Tang, J.C., Smith, R.B. and Yankelovich,
N., Work rhythms: analyzing visualizations of
awareness histories of distributed groups. in
Proceedings CSCW'02, (2002), 334-343.
3. Csikszentmihalyi, M. Flow: The Psychology of Optimal
Experience. . Harper & Row, NY, 1990.
4. Czerwinski, M., Horvitz, E. and Wilhite, S., A diary
study of task switching and interruptions. in
Proceedings CHI'04, (2004), 175-182.
5. Dane, E. Paying attention to mindfulness and its effect
on task performance in the workplace. Journal of
Management, 2011, 37 (4). 997-1018.
6. Danziger, S., Levav, J. and Avnaim-Pessoa, L.
Extraneous factors in judicial decisions. Proc of the
Nat'l Academy of Sciences, 2011, 108 (17). 6889-6892.
7. Gonzalez, V.M. and Mark, G., "Constant, Constant,
Multi-tasking Craziness”: Managing Multiple Working
Spheres. Proceedings CHI'04, (2004), 113-120.
8. Gross, S., Semmer, N.K., Meier, L.L., Kalin, W.,
Jacobshagen, N. and Tschan, F. The effects of positive
events at work on after-work fatigue: They matter most
in face of adversity. Journal of Applied Psychology,
2011, 96 (3). 654-664.
9. Hektner, J., Schmidt, J. and Csikszentmihalyi, M.
Experience Sampling Method: Measuring the quality of
everyday life. Sage, Thousand Oaks, CA, 2007.
10. Iqbal, S.T. and Horvitz, E., Disruption and Recovery of
Computing Tasks: Field Study, Analysis and
Directions. in Proceedings of CHI'07, (2007), 677-686.
11. Karlson, A.K., Iqbal, S.T., Meyers, B., Lee, K., Ramos,
G. and Tang, J., Mobile Task Flow in Context: A
Screenshot Study of Smartphone Usage. Proceedings of
CHI'10, (2010), 2009-2018.
12. Lefevre, J. Flow and the quality of experience during
work and leisure. in Czikszentmihalyi, M. and
Czikszentmihalyi, I. eds. Optimal Experience:
Psychological Studies of Flow in Consciousness,
Cambridge University Press, Cambridge, 1988.
13. Macey, W. and Schneider, B. The meaning of employee
engagement. Indust and Organ'l Psych, 2008, 1. 3-30.
14. Mark, G., Iqbal, S., Czerwinski, M., and Johns, P.
(2014). Capturing the Mood: Facebook and Face-to-
Face Encounters in the Workplace. Proceedings of
CSCW'14, Baltimore, MD, ACM Press..
15. Mark, G., Voida, S. and Cardello, A. "A Pace Not
Dictated by Electrons": An Empirical Study of Work
Without Email Proc. of CHI'12, (2012), 555-564.
16. Massimini, F. and Carli, M. The systematic assessment
of flow in everyday experience. in Czikszentmihalyi,
M. and Czikszentmihalyi, I. eds. Optimal Experience:
Psychological Studies of Flow in Consciousness,
Cambridge University Press, Cambridge, 1988.
17. Posner, J., Russell, J.A. and Peterson, B.S. The
circumplex model of affect: An integrative approach to
affective neuroscience, cognitive development, and
psychopathology. Development and Psychopathology,
2005, 17. 715-734.
18. Russell, J.A. A circumplex model of affect. J of
Personality and Soc Psychology, 1980, 39. 1161-1178.
19. Schallberger, U. The influence of personality
characteristics on self reports of working conditions.
Zeit. f. Experimentelle Psych, 1995, 42 (1). 111-131.
20. Schaufeli, W., Salanova, M., Gonzalez-Rom, V. and
Bakker, A. The measurement of engagement and
burnout: A two sample confirmatory factor analytic
approach. Journal of Happiness studies, 2002, 3. 71-92.
21. Scherer, K.R., Banziger, T. and Roesch, E. A Blueprint
for Affective Computing: A sourcebook and manual.
Oxford University Press, 2010.
22. Stone, A.A., Hedges, S.M., Neale, J.M. and Satin, M.S.
Prospective and crosssectional mood reports offer no
evidence of a “Blue Monday” phenomenon. Journal of
Personality and Social Psychology, 1985, 49. 129-134.
23. Tice, D., Baumister, R., Shmuei, D. and Muraven, M.
Restoring the self: Positive affect helps improve self-
regulation following ego depletion. Journal of
Experimental Social Psychology, 2007, 43 (3). 379-384.
24. Tschan, F., Banziger, T. and Roesch, E. Discrete
emotions in interaction with superiors: Some are more
role-related, some are more relationship related. in
Rigotti, T., Korek, S. and Otto, K. eds. Gesund mit und
ohne Arbeit (S. 289 – 304), Pabst Science Publishers,
Lengerich, 2010, 289-304.
25. Webster, J. and Ho, H. Audience engagement in multi-
media presentations. Data Base for the Advancement in
Information Systems, 1997, 28 (2). 63-77.
26. Weick, K. and Sutcliffe, K. Mindfulness and the quality
of organizational attention. Org Sci, 2006, 17. 514-524.