Content uploaded by Kostadin Kushlev
Author content
All content in this area was uploaded by Kostadin Kushlev on Dec 08, 2017
Content may be subject to copyright.
“Silence Your Phones”: Smartphone Notifications Increase
Inattention and Hyperactivity Symptoms
Kostadin Kushlev
University of Virginia
Charlottesville, USA
kushlevk@gmail.com
Jason Proulx
University of British Columbia
Vancouver, Canada
j.proulx19@gmail.com
Elizabeth W. Dunn
University of British Columbia
Vancouver, Canada
edunn@psych.ubc.ca
ABSTRACT
As smartphones increasingly pervade our daily lives, people
are ever more interrupted by alerts and notifications. Using
both correlational and experimental methods, we explored
whether such interruptions might be causing inattention and
hyperactivity—symptoms associated with Attention Deficit
Hyperactivity Disorder (ADHD)—even in people not
clinically diagnosed with ADHD. We recruited a sample of
221 participants from the general population. For one week,
participants were assigned to maximize phone interruptions
by keeping notification alerts on and their phones within
their reach/sight. During another week, participants were
assigned to minimize phone interruptions by keeping alerts
off and their phones away. Participants reported higher
levels of inattention and hyperactivity when alerts were on
than when alerts were off. Higher levels of inattention in
turn predicted lower productivity and psychological well-
being. These findings highlight some of the costs of
ubiquitous connectivity and suggest how people can reduce
these costs simply by adjusting existing phone settings.
Author Keywords
Smartphones; interruptions; pervasive computing;
multitasking; experimental study; attention management;
subjective well-being; psychology
ACM Classification Keywords
H.1.2. [Information Systems]: User/Machine Systems—
Human Factors; H.1.2. [Information Systems]:
User/Machine Systems—Human Information Processing;
H.5.2 [Information Systems]: User Interfaces—Auditory
Feedback.
INTRODUCTION
Ken is a married man, a father of four, and the founder and
CEO of a digital ad agency. And he is struggling. In his
early 40’s, Ken noticed a significant decline in his ability to
manage tasks, and gaps in his memory and performance.
His desk had gotten increasingly cluttered, and when his
employees started leaving the company frustrated by his
inability to manage it, Ken decided to seek help. He
wondered whether he might have an early-onset dementia
or a thyroid dysfunction. What he did not expect is that his
doctor would diagnose him with Attention Deficit
Hyperactivity Disorder (ADHD). Ken wondered whether
the recent exacerbation of his symptoms might be due to his
increasingly stimulated digital environment.
The real story of Ken [43] is an example of how factors in
our contemporary environment might exacerbate ADHD
symptoms. Because ADHD is a neurodevelopmental
disorder, however, a diagnosis cannot be based only on
current symptoms, but also on a lifetime history of such
symptoms. Ken’s story, therefore does not suggest that
digital distractions can cause ADHD. What his story does
suggest, however, is that our increasingly pervasive digital
technology may be causing inattention and hyperactivity—
not only in people already suffering from ADHD, but also
in the general population. In a recent survey, for example,
one out of every two American smartphone owners reported
having been distracted by their phones over the preceding
week [31].
When Steve Jobs introduced the first smartphone in 2007,
he promised: “This will change everything”; and he was
right. Today, we can access the Internet virtually anytime
and anywhere. And we do just that. The vast majority of
smartphone owners use their phones not only for text
messaging and calling, but also to check their email, keep
up with their social media feeds, or simply to browse the
Internet [31]. Of course, people could access these
functions on their stationary computers even before they
could carry these powerful computers in their pockets.
What smartphones have changed, however, is where and
when people access those functions. In a recent poll of
smartphone users in the US, for example, almost all (95%)
admitted to using their smartphones during their most
recent social gathering, including to read incoming
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for 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 the author(s) must be
honored. Abstracting with credit is permitted. To copy otherwise, or
republish, to post on servers or to redistribute to lists, requires prior specific
permission and/or a fee. Request permissions from
Permissions@acm.org.
CHI'16, May 07 - 12, 2016, San Jose, CA, USA
Copyright is held by the author(s). Publication rights licensed to ACM.
ACM 978-1-4503-3362-7/16/05…$15.00
DOI: http://dx.doi.org/10.1145/2858036.2858359
messages or emails or simply to check whether they had
received any alerts [34]. In other polls, seven out of ten
American smartphone owners were found to use their
phones while working [31], and one in ten admitted to
checking their phone even during sex [17]. According to
some estimates people reach for their phones more than one
hundred times a day [30] and spend close to two hours a
day using their phones [12, 42].
Might the deluge of smartphone alerts and notifications be
contributing to an increase in symptoms of ADHD in the
general population? According to the Diagnostic and
Statistical Manual (DSM-V)—the official guide for
diagnosing mental disorders of the American Psychiatric
Association—ADHD is characterized by two sets of
symptoms: inattention and hyperactivity/impulsivity [4].
Inattention includes symptoms such as having difficulty
focusing on one task at a time, being easily distracted, and
getting easily bored when trying to focus. Hyperactivity is
characterized by symptoms such as fidgeting, having
trouble sitting still, and having difficulty doing quiet tasks
and activities; in adults, hyperactivity also manifests as a
sense of restlessness [22]. Smartphones may contribute to
both sets of symptoms in the general population by serving
as a readily available source of distraction (causing
inattention) and of virtually unlimited array of alternative
activities (causing hyperactivity). These effects could be
exacerbated by the inflow of new notifications, which can
both interrupt ongoing activities and present alternative
activities [11; 14]. We propose, therefore, that the
interruptions caused by smartphone notifications—from
text messages and social media updates to emails and
calendar reminders—could cause inattention and
hyperactivity even in the general population.
Notifications, Interruptions, and Disruptions
A wealth of basic research and theory documents the toll of
frequent task interruptions on attention and cognition.
According to the time-based resource sharing model of
attention [6], the very act of switching between tasks—even
briefly—requires cognitive effort above and beyond the
effort required to complete the tasks themselves. Switching
attention thus increases cognitive load [6, 28]. And
according to load theory of attention [24], the higher the
cognitive load, the more susceptible people are to new
distractors [36, 26]. To the extent that phone notifications
draw users’ attention away from other ongoing activities,
phones may increase cognitive load. And by making people
more prone to distractions, increased cognitive load may in
turn make people—even in the general population—suffer
from inattention and hyperactivity.
Beyond theory and basic research in cognitive psychology,
a large body of work in HCI documents the detrimental
effects of digital interruptions, particularly during tasks that
require attention [3, 5, 10]. In field experiments, for
example, digital interruptions at work (e.g., email alerts)
have been associated with feeling more distracted, stressed,
and anxious [23, 29]. And experiments in a controlled lab
environment have revealed that notifications are disruptive
in part because they interrupt people at random times. In
one particularly well-designed lab experiment, participants
completing common work tasks (e.g., editing text) felt more
annoyed and frustrated, experienced greater time pressure,
and expended more mental effort when they were
interrupted by notifications at random than when
uninterrupted [1]. In contrast, people suffered no
consequences from being interrupted (i.e., no added
frustration, time pressure, and mental effort) when they
were interrupted at opportune times in between work tasks.
Not all digital interruptions, then, are born equal: When
people are interrupted by the random receipt of a new
notification, they may be particularly likely to incur
attentional and emotional costs.
An emerging body of literature shows that people often
attend immediately to notifications received on their phones
[8, 32, 37]. This suggests that phone notifications often
interrupt other activities at random times, making phones
particularly disruptive [1]. Phones’ power to interrupt their
users is no doubt in part due to their omnipresence
throughout our daily activities. Indeed, people attend to new
notifications quickly regardless of the alert mode on their
phones (e.g., ring vs. vibrate vs. silent) [8, 32]. Still, people
are more likely to attend immediately to, for example, new
text messages when their phones are in vibrate or ring mode
than when they are in silent mode [8]. Above and beyond
phones’ omnipresence in our lives, then, the rings, dings,
and vibrations accompanying new notifications may levy
users’ attention further by interrupting other tasks at
random times. In fact, the very sound of a new alert has
been shown to disrupt ongoing tasks. In a recent study, the
mere buzz of a new notification was enough to hurt
performance on a cognitively demanding task even when
people did not attend to the notification (e.g., checked the
text message announced by the notification) [41]. Yet,
people typically keep their phone alerts on—about 9 in 10
notifications are received in ring and vibrate modes [32]. To
the extent that people normally attend immediately to phone
notifications and random interruptions tax attention, phone
notifications may cause even people not diagnosed with
ADHD to suffer from inattention and hyperactivity.
Study Overview
We recruited a sample of university students from the
general student population. During an initial study session,
we assessed individual differences between participants in
inattention and hyperactivity while also measuring how
frequently people felt interrupted by their phone
notifications. We thus explored correlationally whether
feeling interrupted by phone notifications was associated
with inattention and hyperactivity. Because ADHD
symptoms have been associated with poorer performance at
work/school and with relationship and social skills
problems [13, 15, 22], we also included exploratory
measures of productivity and psychological well-being.
While correlational analyses provide the most naturalistic
way to examine the association of phone interruptions with
inattention and hyperactivity, such analyses preclude causal
conclusions. Accordingly, we ran a two-week within-
subjects experiment in which we manipulated how
frequently participants felt interrupted by their
smartphones. For one week, we assigned participants to
maximize phone interruptions by keeping notification alerts
on and their phones within their sight and/or reach. During
another week, we assigned the same participants to
minimize phone interruptions by keeping alerts off and their
phones away. By employing this within-subjects design, we
were able to examine whether phone interruptions had
causal effects on inattention and hyperactivity.
METHOD
Participants
Two-hundred and twenty-one undergraduate students (Mage
= 19.89; 73.76% female) from the University of British
Columbia completed the study in exchange for course
credit. Importantly, students were recruited from the
university’s general participant pool, rather than from a
pool of students diagnosed with ADHD. We calculated our
sample size based on a priori analyses with 80%
probability of detecting small effects of d = .20, α = .05,
two-tailed
1
. Because we were interested in manipulating
how frequently people felt interrupted by their smartphones,
we prescreened participants for smartphone ownership; all
participants thus had a smartphone
2
.
Procedure
Participants attended an introductory session in the lab in
groups of up to 10 students. Before assigning them to
experimental condition, we assessed our main measures of
interest at baseline, including phone interruptions,
inattention and hyperactivity, as well as productivity and
psychological well-being (see Measures). Participants were
then assigned to condition using a randomized
counterbalanced within-subjects design. Half the
participants were assigned to minimize interruptions for one
week and then to maximize interruptions for a second week.
For the other half of the participants, the order of these
instructions was reversed.
In the Do-Not-Interrupt condition, participants configured
their smartphones to Do-Not-Disturb settings, disabling
auditory (e.g., ring), tactile (i.e., vibration) and visual alerts
(i.e., LED flashes). In order to prevent participants from
1
The power analyses were registered on Open Science
Framework: http://tinyurl.com/ofx3jjy.
2
Because we did not want people to get around our
manipulation of smartphone interruptions by using other
portable devices, we also prescreened people for how much
they used tablets. People who reported using a tablet often
or all the time in prescreening were not eligible to
participate.
compensating by monitoring incoming notifications on their
screens, we additionally asked them to keep their phones
out of sight (e.g., in their bags, pockets).
In the Interrupt condition, participants configured their
phones to enable auditory, tactile, and/or visual alerts.
Additionally, participants kept their phones within their
sight and/or reach. In situations where their phones may be
disruptive to others (e.g., in class), participants in this
condition were instructed to use common sense and to
switch off audible alerts if necessary. Even in such
situations, however, participants were instructed to have
their phones at least on vibrate mode.
All participants within a single introductory session were
assigned to the same condition. All participants were
guided through how to configure their phones in silent,
ring, and vibrate settings. Additionally, each participant was
given a sheet detailing how to configure his or her phone to
the assigned settings; the sheet contained information about
the most common smartphones and operating systems (e.g.,
iPhone, Android).
After the introductory session, participants were sent daily
online surveys in the evening on each of the next six days.
During the first day of the second week, participants were
sent a prompt via email to change their phone settings as
shown on their instructions sheets. Then, for each of the
following six days, participants again received links to
complete daily surveys. In the first five surveys of each
week, we assessed how frequently people felt interrupted
by their phones during a specified one-hour period of the
day. This measure served as our manipulation check (see
Measures)
3
. On the seventh and last day of each week,
participants received a longer survey asking them to report
how they felt over the past week. This survey contained our
main variables of interest, including inattention and
hyperactivity, productivity, and psychological well-being.
At the end of the two weeks, participants returned to the lab
to be debriefed in person.
Measures
Smartphone Interruptions
At baseline, we assessed how frequently participants
normally felt interrupted by their phones (from 0—not at
all to 6—constantly) when engaging in 14 common daily
activities (e.g., attending class, reading for pleasure,
commuting, or eating
4
). Interruptions were defined as any
time participants’ attention shifted from their current
3
These daily questionnaires included other questions
beyond the main scope of the present investigation of the
effects of phone interruptions on inattention and
hyperactivity. The full questionnaire is available here:
http://tinyurl.com/nfsux27.
4
The full version of the baseline survey, including a list of
all activities is available here: http://tinyurl.com/nfsux27
activities to their phones because of a notification. We
asked people to report interruptions separately for these 14
activities in order to cue people to think about their
behavior during specific activities, making their estimates
more accurate. We averaged people’s responses across all
14 activities to form a single measure of phone
interruptions at baseline.
In order to assess the successfulness of the manipulation,
we also asked people to report daily how frequently they
felt interrupted by their phones (0–not at all; 6–constantly).
We used a self-report measure of interruptions because no
available technology can objectively assess how frequently
people feel interrupted by their phones. Indeed, people’s
subjective sense of being interrupted should depend not
only on how quickly people attend to notifications but also
on what they are currently doing (e.g., trying to pay
attention in class versus mindlessly commuting). Still, we
used past research to design our self-report measure to be as
accurate as possible. Specifically, people have been shown
to provide more accurate information about their behavior
when asked what they were doing during specific periods of
time in the day [e.g., 21]. Accordingly, participants were
first prompted to remember what they were doing during a
one-hour period of time on each day, and then to report how
frequently they felt interrupted by their phones during this
period. The specific time period participants were asked
about varied by day. To represent the range of activities
people engage in during different times of the day, we first
selected every other possible one-hour period, starting at 10
am and ending at 7 pm (i.e., 10 am to 11 am, 12 pm to 1
pm, and so forth). We then randomly assigned one of these
time periods for each day of the week.
Inattention and Hyperactivity
To measure inattention and hyperactivity (both at baseline
and at the end of each week of the study), participants rated
the extent to which they experienced 18 symptoms over the
past week (1–never; 2–rarely; 3–sometimes; 4–often).
These 18 items were based on the criteria of the Diagnostic
and Statistical Manual – V (DSM-V) for diagnosing ADHD
[4, 16]. Nine items assessed inattention (e.g., How often
were you easily distracted by external stimuli, like
something in your environment or unrelated thoughts?); the
other nine items assessed hyperactivity (e.g., How often did
you have difficulty waiting your turn, such as while waiting
in line?). We averaged each set of nine items to form our
measures of inattention (α = .82) and hyperactivity (α =
.79
5
) at baseline and for each of the two experimental
weeks.
Productivity
We measured participants’ productivity at baseline and at
the end of each week of the study with three self-report
5
All alphas are reported as averages of the alphas of the
three measures of each construct (i.e., baseline, week 1, and
week 2)
items used in previous research [23]. We asked participants
the extent to which in the previous week they: (1) felt that
they got the things done at work or school that were
important to them; (2) were satisfied with what they had
accomplished at school; and (3) felt a sense of
accomplishment from school (α = .88).
Psychological Well-Being
To measure well-being broadly, we assessed several
constructs theorized as essential components of
psychological well-being. In particular, Ryff [36] has
argued for six components of psychological well-being,
including a sense of autonomy, environmental mastery,
self-acceptance, personal growth, purpose and meaning,
and relatedness with others. To reduce the burden on
participants from completing measures of all these aspects
of psychological well-being, we assessed only four.
Specifically, we did not assess people’s sense of personal
growth and self-acceptance because we deemed it unlikely
that these aspects of psychological well-being would be
related to inattention and hyperactivity. In contrast, based
on past research [13, 15, 22], we theorized that inattention
and hyperactivity might have downstream consequences for
people’s sense of mastery over their environment,
autonomy over their actions, relatedness to others, as well
as how meaningful they found their daily lives.
To measure participants’ environmental mastery, we used
the short version of the environmental mastery
questionnaire [35], which includes three items (α = .73). To
assess mastery during each specific week of the study, we
adapted the items so that people reported their sense of
mastery over the past week. Participants provided their
responses using a scale from 1–strongly disagree to 7–
strongly agree.
Although Ryff [36] has developed scales for all aspects of
psychological well-being, Ryff’s scales of the other aspects
were not adaptable for measuring the constructs over a one-
week period (rather than in their life in general).
Accordingly, we adapted measures from other commonly
used scales to assess the other aspects of psychological
well-being. To measure participants’ sense of relatedness
with others, we adapted 5 items from Social Connectedness
Scale [27]. Participants reported, for example, how distant
they felt from others and how disconnected they felt from
the world around them (both reversed scored). Participants
used a scale from, 1–strongly disagree to 7–strongly agree.
This adapted scale demonstrated high reliability (α = .94).
To measure participants’ sense of autonomy, we adapted
five items from the Perceived Choice Scale [38, 39] (α =
.85). For each item, participants were given two statements
(A and B) and asked to respond on a scale from 1–only A
feels true to 5–only B feels true. For example, participants
were asked to decide between A—I felt like I always chose
the things I did and B—I sometimes felt that I was not really
choosing the things I did.
To measure participants’ meaning in life, we used five
items of the Meaning In Life Questionnaire—Presence
Subscale [40] (α = .89). Participants reported their
agreement with each item on a scale from 1–strongly
disagree to 7–strongly agree.
At baseline, we additionally included a broader measure of
life satisfaction in general: Taking all things together, how
satisfied are you with your life these days? (0–not at all
satisfied; 6–very satisfied) [7].
RESULTS
Are smartphone interruptions associated with inattention
and hyperactivity? To explore this question, we first ran a
series of correlational analyses using people’s self-reports
at baseline. We then examined this question causally by
comparing whether people reported higher levels of
inattention and hyperactivity when they were randomly
assigned to maximize (versus minimize) interruptions.
Correlational Analyses at Baseline
As seen in Table 1, people who reported more phone
interruptions at baseline also reported higher levels of both
inattention (r = .31, p < .001) and hyperactivity (r = .30, p <
.001). Consistent with past research, we also found that
inattention and hyperactivity were negatively associated
with our measures of self-reported productivity and
psychological well-being (see Table 1). Notably, inattention
was more strongly associated with these outcomes than was
hyperactivity. In addition, people who reported feeling
interrupted by their phones more frequently felt less
productive, experienced a lower sense of environmental
mastery, and also reported marginally lower sense of
meaning in life.
To explore whether inattention and hyperactivity mediate
the relationship between smartphone interruptions and our
measures of productivity and psychological well-being, we
ran a series of mediational analyses using both
bootstrapping with 50,000 replications [18] and Sobel’s test
of mediation. Inattention mediated the effect of
interruptions on productivity, indirect effect = .15, 95% CI
[.23; .08], Sobel’s Z = 3.78, p < .001, leaving a
nonsignificant direct effect of interruptions on productivity,
b = .02, 95% CI [.17; .12]. Inattention also mediated the
effect of interruptions on environmental mastery, indirect
effect = .17, 95% CI [.26; .09], Sobel’s Z = 4.18, p <
.001, leaving a nonsignificant direct effect, b = .02, 95%
CI [.15; .11]. Finally, inattention explained the effect of
interruptions on meaning in life, indirect effect = .11, 95%
CI [.19; .05], Sobel’s Z = 3.02, p < .01, leaving a
nonsignificant direct effect, b = .04, 95% CI [.21; .14]. In
short, inattention explained the negative associations
between phone interruptions and our measures of
productivity and psychological well-being.
In contrast to the strong mediating role of inattention,
hyperactivity played a less important role in explaining the
negative associations of phone interruptions with
productivity and psychological well-being. Specifically,
hyperactivity did not mediate the association of phone
interruptions to productivity, indirect effect = .02, 95% CI
[.07; .03], Sobel’s Z = .71, p = .48 or meaning in life,
Interruptions
Inattention
Hyperactivity
Productivity
Environmental
Mastery
Social Connectedness
Perceived Choice
Meaning in Life
Life Satisfaction
Interruptions
Inattention
.31***
Hyperactivity
.30***
.41***
Productivity
.14*
.41***
.09
Environmental Mastery
.18**
.52***
.27***
.34***
Social Connectedness
.01
.26***
.27***
0.13*
.29***
Perceived Choice
.09
.36***
.13†
.25***
.35***
.24***
Meaning in Life
.11†
.28***
.11
.36***
.49***
.35***
.38***
Life Satisfaction
.05
.41***
.15*
.42***
.64***
.46***
.39***
.50***
Table 1: Table of correlations between baseline measures. *** p < .001; ** p < .01; * p < .05; † = p < .10
indirect effect = .03, 95% CI [.10; .02], Sobel’s Z =
1.15, p = .25. And although hyperactivity mediated the
effect of interruptions on environmental mastery, indirect
effect = .08, 95% CI [.15; .03], Sobel’s Z = 2.79, p <
.01, this indirect effect was weaker than the indirect effect
of inattention.
The above mediational analyses are correlational; it is,
therefore, possible that inattention may make people more
susceptible to feeling interrupted by their phones (rather
than that phone interruptions cause inattention). If this
alternative direction of causality describes the data, phone
interruptions should mediate the negative associations of
inattention with productivity and psychological well-being.
In contrast to this possibility, phone interruptions did not
mediate the associations of inattention (b’s = .01 to .02,
p’s > .68) with our measures of productivity and
psychological well-being. Thus, our correlational analyses
at baseline are most consistent with a single causal path:
Phone interruptions impair productivity and psychological
well-being to the extent that interruptions increase
inattention.
Experimental Results
To directly explore whether phone interruptions can cause
inattention and hyperactivity
6
, we next present the effects of
our experimental manipulation.
Manipulation Check
Confirming the success of our manipulation, we found that
people felt more frequently interrupted in the Interrupt
condition (M = 2.64, SD = 1.04) than in the Do-Not-
Interrupt condition (M = 2.03, SD = .92), t(210) = 8.06, p <
.001. Interestingly, people felt less interrupted by their
phones in the Do-Not-Interrupt condition than normal, as
assessed at baseline (M = 2.75, SD = 96), t(210) = 10.20, p
< .001; in contrast, people felt no more interrupted in the
interrupt condition than at baseline, t(210) = 1.56, p = .119.
Main Effects
Did maximizing phone interruptions for a week result in
reporting higher levels of inattention and hyperactivity? A
paired-samples t-test comparing inattention symptoms
between each of the two weeks revealed that participants
indeed experienced higher levels of inattention in the
Interrupt condition (M = 2.38, SD = .58) than in the Do-
Not-Interrupt condition (M = 2.27, SD = .56), t(189) = 3.02,
p = .003, d = .44. Mirroring this effect on inattention,
people also reported higher levels of hyperactivity in the
Interrupt condition (M = 2.11, SD = .59) than in the Do-
Not-Interrupt condition (M = 2.01, SD = .55), t(189) =
3.12, p = .002, d = .45. Notably, both effects were of
6
Note that degrees of freedom vary slightly between tests
due to failure by some participants to complete some
measures.
medium size according to established statistical guidelines
in the behavioral sciences [8].
We found no significant main effects of our manipulation
on our measures of productivity and psychological well-
being, t’s < .89, p’s > .37. It is, of course, hardly surprising
that our minimal manipulation did not produce significant
differences in these outcomes within the frame of one week.
Still, our correlational findings showed that phone
interruptions were negatively associated with productivity
and psychological well-being, as well as that these negative
associations were explained by inattention. Accordingly, we
next examined whether our manipulation of phone
interruptions had downstream consequences for
productivity and well-being to the extent people suffered
greater inattention when maximizing (vs. minimizing)
interruptions.
Indirect Effects
To explore the indirect effects of our manipulation on
productivity and psychological well-being through
inattention, we followed recommendations by Judd and
colleagues [20] for conducting mediation analyses with
repeated measures. As shown in Eq. (1) below, in each case
we predicted the difference scores in the outcome variables
(Y) from the sum and difference scores of inattention (X).
The regression coefficient of the difference scores of
inattention (controlling for its sum scores) is the measure of
the indirect effects of the manipulation on productivity and
well-being through inattention [20]. As shown in Figure 1,
to the extent that people experienced greater inattention in
the Interrupt (vs. Do-Not-Interrupt) condition, they felt less
productive and less socially connected, reported less
mastery over their environment and lower meaning in life,
and perceived less choice over their actions.
Ydiff = b1Xsum + b2Xdiff + 𝜀, where
(a) Ydiff = Ydo-not-interrupt – Yinterrupt
(b) Xdiff = X do-not-interrupt – Xinterrupt
(c) Xsum = X do-not-interrupt + Xinterrupt
Figure 1. Indirect effects of manipulating smartphone
interruptions on psychological well-being via inattention
symptoms. Numbers are unstandardized regression
coefficients.
DISCUSSION
We provide the first evidence that interruptions due to
phone notifications can cause inattention and hyperactivity
in the general population. When people maximized phone
interruptions by switching alerts on and keeping their
phones within reach, they felt higher levels of inattention
and hyperactivity than when they minimized phone
interruptions. In line with these experimental findings,
correlational analyses at baseline also indicated that people
who felt interrupted by their phones more frequently
experienced higher levels of inattention and hyperactivity.
In both the correlational and experimental analyses, the
effects of phone interruptions on inattention had
downstream consequences for self-reported productivity
and psychological well-being. By triangulating on
correlational and experimental methods, therefore, we
demonstrated that phone interruptions might cause
symptoms of ADHD in the general population. Our
findings also suggest, however, that people could mitigate
these negative effects by making simple modifications to
the notification settings of their phones.
Despite our promising findings, we caution that keeping
notifications at bay may not work equally well for
everybody trying to reduce phone interruptions. Preliminary
qualitative research suggests that for some, disabling alerts
may produce anxiety over missing notifications [33]. Such
anxiety may drive some to self-interrupt more, resulting in a
net increase in interruptions. In one study, for example,
information workers disabled their email notifications for
one week; while some workers were able to reduce their
email checking compared to baseline, others checked their
email more to avoid missing important messages [19].
Future research should, therefore, examine what personality
characteristics (e.g., neuroticism) and what situational
factors (e.g., job type, duration of intervention) determine
when and for whom switching off notifications leads to net
benefits.
To manipulate phone interruptions in the present research,
we asked people to switch alerts on versus off, as well as to
keep their phones within sight versus out of sight.
Assuming that the old adage—out of sight, out of mind—
holds true, our manipulation may have influenced not only
external interruptions but also self-interruptions. While past
research suggests that the unpredictable timing of external
notifications may be especially disruptive to users [1],
future research should examine whether self-interruptions
contribute to inattention and hyperactivity above and
beyond external interruptions. Indeed, even when using
silent mode, people normally attend to new phone
notifications within 6 minutes [8], suggesting that self-
interruptions may account for a large proportion of the
interruptions in people’s daily activities [cf. 11].
For the purposes of the present research, we used self-report
measures to detect differences in how frequently people felt
interrupted by their phones. Because people have been
shown to underestimate the frequency of their smartphone
use [42], the particular values reported by participants
should be interpreted with caution. We note, however, that
self-report measures were better suited for the purposes of
the present investigation than available objective measures.
Indeed, past research using objective measures has captured
only how quickly people attend to incoming notifications,
rather than whether people feel interrupted by these
notifications given their current context and activities [8;
32]. Using these objective measures, researchers have
shown only small—but statistically significant—differences
between how quickly people attend to new notifications in
silent versus ring/vibrate modes [8]. By relying on a
subjective measure, the present research suggests that even
such minor differences in attentiveness to new notifications
may make users feel more interrupted by their phones.
By using a subjective measure of phone interruptions, we
were able to hone in specifically on how frequently
people’s attention was drawn to their phone due to a
notification. But our measure thus captured only external
rather than self-initiated interruptions. In contrast, objective
measures in past research have captured both external and
self-interruptions (i.e., any time a person attends to a
notification [8; 32]). It is possible, therefore, that a
subjective measure assessing both external and self-initiated
interruptions may have produced a smaller difference
between conditions—similar to the available objective
measures. Even such a subjective measure, however, would
not be conceptually equivalent to the available objective
measures. Such measures can accurately assess how
frequently users attend to notifications; but they cannot
adequately capture how frequently users are interrupted
because these measures ignore people’s current goals and
activities. Attending to a notification may be interruptive
when a person is in a meeting or class, but not when she is
on the bus or bored at home. The present research thus
suggests that subjective measures of interruptions are not
simply coarser, less precise versions of the available
objective measures; rather, subjective measures can be
useful additional tools for studying the psychological
effects of our ubiquitous digital devices. We invite future
research to use both subjective and objective measures in
order to disentangle how the feeling of being interrupted by
notifications relates to the objective frequency of attending
to notifications.
In order to conduct the first systematic investigation of the
effects of phone interruptions, we included a wide range of
outcome measures, which included inattention and
hyperactivity, but also many other constructs. Given this
exploratory approach, it is possible that the significant
effects we observed are simply an artifact of the large
number of statistical tests we conducted. Importantly,
however, we found consistent effects on inattention and
hyperactivity both in our correlational analyses at baseline
and in our two-week field experiment. This converging
evidence of our correlational and experimental analyses
make less likely the possibility that our findings are a mere
statistical fluke. Still, the present study should be seen as
laying the groundwork for further confirmatory research.
Inherent in our within-subjects design lies another
limitation of the present study: Participants were aware of
the two conditions within our manipulation. To the extent
that people might have theories about how their phones
affect them, people may have reported how they thought
they should feel rather than how they actually felt. But we
found the same associations of phone interruptions with
inattention and hyperactivity using our baseline measures,
which were administered before condition assignment.
Viewed in the context of our correlational findings,
therefore, our experimental findings are unlikely to be
completely explained by participants’ lay theories.
We found no direct effects of our manipulation on our
measures of productivity and psychological well-being. It
is, of course, hardly surprising that keeping notifications at
bay for a week did not produce direct effects on, for
example, how meaningful people found their lives. Still, to
the extent that phone interruptions influenced inattention,
they had downstream consequences for productivity and
well-being. Future research should explore whether these
indirect effects may materialize into direct effects if people
minimize phone interruptions over periods of time longer
than a week (e.g., one month, an entire semester).
We found that smartphone interruptions were associated
with self-reported symptoms of ADHD in a sample drawn
from the general population. But our findings do not in any
way suggest that smartphone notifications can cause
ADHD, which is a neurodevelopmental disorder with
complex neurological and early developmental causes [4].
But as with any other disorder, symptoms represent a
continuum from the normal to the pathological. Our
findings thus simply show that smartphone interruptions
may be driving individuals from the general population to
feel somewhat more inattentive and hyperactive. The
present research can thus be seen only as an initial
demonstration of the potential role of smartphones in
exacerbating symptoms associated with ADHD.
Just as we do not suggest that phone interruptions can cause
ADHD, we by no means claim that reducing phone
interruptions can treat ADHD. Indeed, our findings are
based on a sample drawn from the general student
population, rather than from a population diagnosed with
ADHD. What our findings do suggest is that being
constantly interrupted by alerts and notifications may be
contributing towards an increasingly problematic deficit of
attention in our digitally connected society. In an age of
incessant digital stimulation, we hope that our findings
would spur further research on how people can reclaim
control over their attention.
REFERENCES
1. Pitor D. Adamczyk and Brian P. Bailey. 2004. If not
now when?: The effects of interruption at different
moments within task execution. In Proceedings of the
SIGCHI Conference on Human Factors in Computing
Systems 6, 1: 271–278.
http://doi.org/10.1145/985692.985727
2. Lenard A. Adler, L., Thomas Spencer, Stephen V.
Faraone, Ronald C. Kessler, Mary J. Howes, Joseph
Biederman, and Kristina Secnik. 2006. Validity of pilot
adult ADHD Self- Report Scale (ASRS) to rate adult
ADHD symptoms. Annals of Clinical Psychiatry. 18,
3: 145–148.
3. Rachel F. Adler, and Raquel Benbunan-Fich. 2002.
Juggling on a high wire: Multitasking effects on
performance. International Journal of Human-
Computer Studies. 70, 2: 156–168.
4. American Psychiatric Association. 2013. Diagnostic
and Statistical Manual of Mental Disorders (Fifth ed.).
Arlington, VA: American Psychiatric Publishing.
5. Anja Baethge and Thomas Rigotti. 2013. Interruptions
to workflow: Their relationship with irritation and
satisfaction with performance, and the mediating roles
of time pressure and mental demands. Work & Stress.
27, 1: 43–63.
6. Pierre Barrouillet, Sophie Bernardin, and Valérie
Camos. 2004. Time constraints and resource sharing in
adults' working memory spans. Journal of
Experimental Psychology: General. 133, 1: 83–100.
7. Christian Bjørnskov. 2008. How comparable are the
Gallup World Poll life satisfaction data? Journal of
Happiness Studies. 11, 1: 41–60.
8. Yung-Ju Chang, John C Tang. 2015. Investigating
mobile users’ ringer mode usage and attentiveness and
responsiveness to communication. In Proceedings of
the SIGCHI Conference on Human-Computer
Interaction with Mobile Devices and Services
(MobileHCI '15): 6–15.
http://dx.doi.org/10.1145/2785830.2785852
9. Jacob Cohen. 1992. A power primer. Psychological
Bulletin. 112, 1: 155–159.
10. Edward Cutrell, Mary Czerwinski, and Eric Horvitz.
2001. Notification, disruption, and memory: Effects of
messaging interruptions on memory and performance.
In Proceedings of the SIGCHI Conference on Human
Computer Interaction (Interact '01), 1999: 263–269.
11. Mary Czerwinski, Eric Horvitz, and Susan Wilhite.
2004. A diary study of task switching and
interruptions. In Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems
(CHI ’04) 6, 1: 175–182.
http://doi.org/10.1145/985692.985715
12. Tilman Dingler and Martin Pielot. 2015. I’ll be there
for you: Quantifying attentiveness towards mobile
messaging. In Proceedings of the SIGCHI Conference
on Human-Computer Interaction with Mobile Devices
and Services: 1–5. Retrieved December 17th, 2015 from
http://pielot.org/pubs/Dingler2015-MobileHCI-
Attentiveness.pdf
13. Colleen L. Dobie, W. Brooks Donald, M. Hanson, C.
Heim, John Huxsahl, and et al. 2012. Diagnosis and
management of attention deficit hyperactivity disorder
in primary care for school-age children and
adolescents. Institute for Clinical Systems
Improvement.
14. Denzil Ferreira, Jorge Gonçalves, Vassilis Kostakos,
Louise Barkhuus, and Anind K. Dey. 2014. Contextual
experience sampling of mobile application micro-
usage. In Proceedings of SIGCHI Conference on
Human-computer interaction with Mobile Devices and
Services (MobileHCI’14): 91-100.
http://doi.org/10.1145/2628363.2628367
15. Julie P. Gentile, Rafay Atiq, and Paulette M. Gillig.
2006. Adult ADHD: Diagnosis, differential diagnosis,
and medication management. Psychiatry (Edgmont). 3,
8: 25–30.
16. James M. Grohol. 2013. Attention Deficit Disorder
(ADD/ADHD) Test. Retrieved September 09, 2015
from http://psychcentral.com/cgi-
bin/addquiz.cgi#sthash.LxTPdK6V.dpuf
17. Harris Interactive. 2013. Where do you take your
phone? Retrieved September 10, 2015 from
http://www.zdnet.com/article/phone-sex-using-our-
smartphones-from-the-shower-to-the-sack/
18. Andrew Hayes. 2013. Introduction to Mediation,
Moderation, and Conditional Process Analysis: A
Regression-Based Approach. Guilford Press.
19. Shamsi T. Iqbal and Eric Horvitz. 2010. Notifications
and awareness: A field study of alert usage and
preferences. In Proceedings of the 2010 ACM
Conference on Computer Supported Cooperative Work
(CSCW ’10): 27–30.
http://doi.org/10.1145/1718918.1718926
20. Charles M. Judd, David A. Kenny, and Gary H.
McClelland. 2001. Estimating and testing mediation
and moderation in within-subject designs.
Psychological Methods. 6, 2: 115-134.
21. Daniel Kahneman, Alan B. Krueger, David A.
Schkade, Norbert Schwarz, and Arthur A. Stone. 2004.
A survey method for characterizing daily life
experience: The day reconstruction method. Science.
306: 1776–1780.
22. Sandra J. J. Kooij, Susanne Bejerot, Andrew
Blackwell, and et al. 2010. European consensus
statement on diagnosis and treatment of adult ADHD:
The European Network Adult ADHD. BMC
Psychiatry. 10, 1: 67.
23. Kostadin Kushlev and Elizabeth W. Dunn. 2015.
Checking email less frequently reduces stress.
Computers in Human Behavior. 43: 220–228.
24. Nilli Lavie. 2010. Attention, distraction, and cognitive
control under load. Current Directions in
Psychological Science. 19, 3: 143–148.
25. Nilli Lavie, N. and Jan W. De Fockert. 2005. The role
of working memory in attentional capture.
Psychonomic Bulletin & Review. 12, 4: 669–674.
26. Nilli Lavie, Aleksandra Hirst, Jan W. de Fockert, and
Essi Viding. 2004. Load theory of selective attention
and cognitive control. Journal of Experimental
Psychology: General. 133, 3: 339–354.
27. Richard M. Lee, Matthew Draper, and Sujin Lee. 2001.
Social connectedness, dysfunctional interpersonal
behaviors, and psychological distress: Testing a
mediator model. Journal of Counseling Psychology.
48, 3: 310–318.
28. Baptist Liefooghe, Pierre Barrouillet, André
Vandierendonck, and Valérie Camos. 2008. Working
memory costs of task switching. Journal of
Experimental Psychology: Learning, Memory, and
Cognition. 34, 3: 478–494.
29. Gloria J. Mark, Stephen Voida, and Armand V.
Cardello. 2012. A pace not dictated by electrons: An
empirical study of work without email. In Proceedings
of the SIGCHI Conference on Human Factors in
Computing Systems (CHI ’12): 555–564.
http://dx.doi.org/10.1145/2207676.2207754
30. Mary Meeker and Liang Wu. 2013. Internet trends.
Retrieved September 10, 2015 from
http://www.kpcb.com/blog/2013-internet-trends
31. Pew Research Center. 2015. The Smartphone
difference. Retrieved May 22, 2015 from
http://www.pewinternet.org/2015/04/01/us-
smartphone-use-in-2015/
32. Martin Pielot, Karen Church, and Rodrigo de Oliveira.
2014. An in-situ study of mobile phone notifications.
In Proceedings of the SIGCHI Conference on Human-
Computer Interaction with Mobile Devices and
Services (MobileHCI ’14): 233–242.
http://doi.org/10.1145/2628363.2628364
33. Martin Pielot and Luz Rello. 2015. The Do Not Disturb
challenge. In Proceedings of the ACM Conference
Extended Abstracts on Human Factors in Computing
Systems (CHI EA ’15): 1761–1766.
http://doi.org/10.1145/2702613.2732704
34. Lee Rainie, and Kathryn Zickuhr. 2015. Americans’
views on mobile etiquette. Retrieved on September 10,
2015 from
http://www.pewinternet.org/2015/08/26/americans-
views-on-mobile-etiquette/
35. Carol D. Ryff. 1989. Happiness is everything, or is it?
Explorations on the meaning of psychological well-
being. Journal of Personality and Social Psychology.
57: 1069–1081
36. Carol D. Ryff, and Corey Lee M. Keyes. 1995. The
structure of psychological well-being revisited. Journal
of Personality and Social Psychology. 69: 719–727.
37. Alireza Sahami Shirazi, Niels Henze, Tilman Dingler,
Martin Pielot, Dominik Weber, and Albrecht Schmidt.
2014. Large-scale assessment of mobile notifications.
In Proceedings of the ACM Conference on Human
factors in Computing Systems (CHI ’14): 3055–3064.
http://doi.org/10.1145/2556288.2557189
38. Kennon M. Sheldon. 1995. Creativity and self-
determination in personality. Creativity Research
Journal. 8: 61-72.
39. Kennon M. Sheldon, Richard Ryan, and Harry T. Reis.
1996. What makes for a good day? Competence and
autonomy in the day and in the person. Personality and
Social Psychology Bulletin. 22: 1270-1279.
40. Michael F. Steger, Patricia Frazier, Shigehiro Oishi,
and Matthew Kaler. 2006. The meaning in life
questionnaire: Assessing the presence of and search for
meaning in life. Journal of Counseling Psychology. 53,
1: 80–93. doi:10.1037/0022-0167.53.1.80
41. Cary Stothart, Ainsley Mitchum, and Courtney
Yehnert. 2015. The attentional cost of receiving a cell
phone notification. Journal of Experimental
Psychology: Human Perception and Performance. 41,
4: 893–897. doi:10.1037/xhp0000100
42. WNYC. 2015. Bored and Brilliant by the numbers (29
Jan, 2015). Retrieved September 09, 2015 from
http://www.wnyc.org/story/bored-and-brilliant-
challenge-tracker/
43. WNYC. 2015. Yes, you are distracted. Is it ADHD?
(13 May, 2015). Retrieved September 09, 2015 from
http://www.wnyc.org/story/you-will-be-distracted-
while-you-listen-podcast/
