Mobile phone distraction while studying

Abstract

The mobile phone is a breakthrough advance for human communication. But with the plethora of choices available via smartphone, individuals who are deficient in self-regulation or with a propensity for addiction may face challenges in managing these choices strategically. To examine this potential dysfunctional aspect, we examined the effect of multitasking when studying or doing homework and found that both frequency and attention to texting and social media were positively related to mobile phone interference in life (MPIL). However, frequency of music use during study was not associated with MPIL, although allocated attention to music while studying was positively associated with MPIL. Ownership of a smartphone and the number of Facebook friends were positively associated with MPIL and women reported more MPIL than men.

Full text

new media & society
1 –19
© The Author(s) 2014
Reprints and permissions:
sagepub.co.uk/journalsPermissions.nav
DOI: 10.1177/1461444814531692
nms.sagepub.com
Mobile phone distraction
while studying
Prabu David
Washington State University, USA
Jung-Hyun Kim
Sogang University, Korea
Jared S Brickman, Weina Ran and
Christine M Curtis
Washington State University, USA
Abstract
The mobile phone is a breakthrough advance for human communication. But with
the plethora of choices available via smartphone, individuals who are deficient in self-
regulation or with a propensity for addiction may face challenges in managing these
choices strategically. To examine this potential dysfunctional aspect, we examined
the effect of multitasking when studying or doing homework and found that both
frequency and attention to texting and social media were positively related to mobile
phone interference in life (MPIL). However, frequency of music use during study
was not associated with MPIL, although allocated attention to music while studying
was positively associated with MPIL. Ownership of a smartphone and the number of
Facebook friends were positively associated with MPIL and women reported more
MPIL than men.
Keywords
Attention, deficient self-regulation, media use, mobile phone, multitasking,
smartphone, task switching
Corresponding author:
Jung-Hyun Kim, School of Communication, Sogang University, K328, 35 Baekbeom-ro (Sinsu-dong), Mapo-
gu, Seoul 121-742, Korea.
Email: junghyunk@sogang.ac.kr
531692NMS0010.1177/1461444814531692new media & societyDavid et al.
research-article2014
Article
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

2 new media & society
Introduction
Multitasking is pervasive in everyday life and is spurred in part by mobile technologies
such as smartphone that are widely used for work-related communication as well as
interpersonal interactions. These days, mobile phone can be used to listen to music and
play games, and users can download applications for activities such as online banking,
booking airline tickets, shopping, making vacation plans, or tracking diet and physical
activity. The versatility of mobile phone allows for seamless integration of work, play,
and social interaction and enriches life in many ways. However, constant use of the
mobile phone may also interfere with work. For example, an observational study of a
small sample of information workers found that task switching occurred every 3 minutes
and voluntary interruptions were as likely as external interruptions to cause switching
(González and Mark, 2004). In another study that employed biometrics and embedded
cameras during non-working time, task switching occurred 27 times per hour among
digital natives and 17 times per hour among those who grew up with older technologies
(Marci, 2012).
Multitasking is also common among students (Carrier et al., 2009; Ophir et al., 2009;
Pea et al., 2012; Rosen et al., 2013; Srivastava, 2013; Wang and Tchernev, 2012) and
associated with lower academic or test performance (Junco, 2012; Junco and Cotten,
2012; Wood et al., 2012). However, the emphasis on academic performance as the key
outcome fails to account for emotional or social functions of multitasking (Wang and
Tchernev, 2012) and the willingness among students to trade off performance on the
academic task for entertainment, emotional or social gains. Various features available via
mobile phone, including music, texting, social media, and games, offer a rich mix of
multitasking options to address these needs.
In this study, we examine three popular options available via mobile phone—music,
texting and social media—and their interference while studying or doing homework. A
recent investigation found that ongoing brief visual inspections of 30 seconds or less
spread throughout the day is typical among mobile phone users (Oulasvirta et al., 2012).
Such ongoing use of the mobile phone when coupled with deficient self-regulation can
evolve into an addiction, resulting in poor academic performance. Therefore, our pri-
mary aim is to examine the relationship between mobile phone-related multitasking
activities that interfere with studying and self-reported “loss of control behaviors” stem-
ming from deficient self-regulation (Kim & LaRose, 2004; LaRose et al., 2003).
Furthermore, given the paucity of established measures of multitasking, we develop and
test three measures of multitasking: frequency of bundled multitasking, frequency of
pairwise multitasking, and attention allocation within a multitasking bundle.
Literature review
Media multitasking and theoretical approaches
Related, yet different, theoretical approaches apply to multitasking and task switching,
two terms that are used interchangeably in the literature. Strictly speaking, multitasking
involves simultaneous involvement in two or more tasks without disengagement or a
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 3
temporary break from either task. For example, singing and playing a guitar or driving a
car and conversing with a friend co-occur in real time without a break in either task.
However, both in the real world and the academic literature, multitasking also refers to
task switching, which requires temporary disengagement from one task to attend to the
other. For example, texting while doing homework requires temporary halting of one to
attend to the other (task switching), whereas listening to music while studying can co-
occur without a break in either activity (multitasking).
Most youth participate in a bundle of activities that include multitasking and task
switching. For example, consider a common activity bundle that includes studying, lis-
tening to music, exchanging text messages, and updating Facebook. Although listening
to music can co-occur with studying, texting and social media require a temporary hold
on studying to free up cognitive and motor resources required for typing messages. To
account for such media behaviors that occur in bundles, frequency should be assessed for
clusters of activities, which is a departure from pairwise assessment currently reported in
the literature. One popular measure is the Media Multitasking Index (Ophir et al., 2009),
which consists of pairwise assessments of 12 types of media. Although this measure has
been shown to have good scale properties, it has two drawbacks: it does not address
attention to each task and multitasking is limited to two concurrent activities.
Instead of relying on a matrix of n × n media activities, which can cause respondent
fatigue, some authors have opted for a list of pairings that are strategically chosen for a
given research context (e.g. Jeong et al., 2010). While this approach reduces respondent
fatigue, it does not account for multitasking situations that involve more than two activi-
ties. Hence, in this study we examine multitasking involving common pairings of
activities while doing homework and compare this to common clusters of more than two
concurrent activities, such as studying, listening to music, and texting. Furthermore, we
examine frequency and attention as distinct measures of media multitasking.
Limited capacity model
To account for division of attention within a bundle of multitasking activities, com-
munication researchers have relied on the limited capacity theory of media processing,
which is based on the premise that human cognitive resource is finite and as the demand
on this resource increases, task performance will decrease (Basil, 1994; Lang, 2000).
For instance, watching a television soap opera while doing homework can hurt home-
work performance because of cognitive overload (Pool et al., 2000, 2003). Likewise,
loss in performance on a primary task has been noted for multitasking interference
from an audio advertisement (Voorveld, 2011), podcast (Srivastava, 2013), or televi-
sion news crawls (Bergen et al., 2005). Furthermore, when attention is divided during
media multitasking, messages become less persuasive perhaps because of limited
availability of cognitive resources for the more persuasive, but attention-demanding
central route processing (Jeong et al., 2010; Jeong and Hwang, 2012; Voorveld, 2011).
Interestingly, when music has been introduced as distraction, no significant drop in
performance has been in observed (Pool et al., 2000, 2003). A plausible explanation for
the benign effect of music on primary task performance may be that it can be ignored
as background noise.
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

4 new media & society
Threaded cognition model
In addition to the limited capacity model, threaded cognition (Salvucci and Taatgen,
2008, 2010) offers another suitable framework for the study of multitasking and task
switching. Three distinct pools of resources—cognitive, perceptual, and motor—are
identified in threaded cognition, and it is posited that these resources are assigned to vari-
ous tasks. Furthermore, each task is instantiated as a separate thread with options for
switching between threads. Managing threads and switching between threads is handled
by the procedural component of the cognitive resource (Salvucci and Taatgen, 2008).
Maintaining multiple threads and switching between threads, however, comes with a
cost. When a task thread is temporarily set aside to attend to another, inevitably delays
and errors occur (Altmann and Gray, 2008). The preparation cost associated with switch-
ing to a new task thread and the dissipation cost when a task is placed on hold can con-
tribute to loss in performance during task switching (Meiran et al., 2000). Nonetheless, a
moderate level of task switching is associated with increased productivity, although
errors build up proportionally, as well. The onus is on the user to determine the trade-off
between productivity and errors, which is likely dependent on the seriousness of the task
and tolerance for errors (Adler and Benbunan-Fich, 2012), though there is evidence that
errors can be minimized through practice (Dux et al., 2009).
In limited capacity and threaded cognition, attention is the key resource. Moreover, in
both approaches, attention is a limited or finite resource that explains performance dete-
rioration in multitasking. Therefore, in this study, we decided to examine attention to
different multitasking activites separately from frequency of multitasking. Furthermore,
to simulate the finite attentional capacity, a constant-sum estimation task was used to
constrain attention allotted to different activities to 100%.
Motivated cognition model
Lang (2006) has advanced a model of attentional resource allocation that is guided by
motivation. In Lang’s model, motivation is explained as a strategic activation of appeti-
tive and aversive systems. While the appetitive systems seek to maximize positive affect
through new experiences, the aversive system, which is built on the flight response,
seeks to avoid negative affect (Cacioppo and Berntson, 1994).
Applying motivated cognition to task switching can offer fruitful insights into why
distractions are irresistible during studying. For some students, homework is an inher-
ently boring or moderately aversive activity. On the other hand, an ongoing text exchange
with a friend can be an appetitive activity that can induce positive affect that offsets the
boredom of homework. Recent studies offer empirical evidence that such emotional and
social goals drive multitasking, rather than cognitive or performance goals (Wang and
Tchernev, 2012).
In short, multitasking or task switching can be examined simply as cognitive over-
load that interferes with a primary task. However, threaded cognition offers a more
nuanced perspective by describing each activity in a multitasking bundle as a separate
thread with its own demands for perceptual, cognitive, and motor resources. Motivated
cognition offers another extension by casting resource allocation among threads as a
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 5
volitional function, with appetitive activities like interactions with friends receiving
more attention than less appetitive or aversive activities like homework. Therefore,
despite self-reports that multitasking is driven by concerns about productivity and cog-
nitive needs, it appears that multitasking among students might be motivated by social
and emotional fulfillment (Wang and Tchernev, 2012). To account for such volitional
control of attention to multitasking activities, a measure of individual preference of
multitasking was introduced.
Deficient self-regulation and mobile phone interference
in life
The purpose of this study is to examine how multitasking while studying is associated
with deficient self-regulation (LaRose et al., 2003) behaviors in mobile phone use,
which is labeled in this study as mobile phone interference in life (MPIL). Unregulated
media use behaviors have been examined for media such as television (Griffiths,
1999) and Internet (Brenner, 1997) as an addiction (McIlwraith, 1998) or dependence
(Kubey, 1996). It has been suggested that unregulated media use is believed to be less
consequential than a drug or alcohol addiction. Still, unregulated media use behav-
iors, which are usually benign, can evolve over time into behaviors that are compul-
sive, uncontrolled and indulgent, and interfere with life (LaRose et al., 2003; Marlatt
et al., 1988).
Given that media addiction starts as a benign habit, we adopt a definition of addiction
as a “repetitive habit pattern that increases the risk of diverse and/or associated personal
or social problems” (Marlatt et al., 1988: 224). Frequent multitasking can become self-
reinforcing and eventually become a habit (Olson and Fazio, 2001; Wang and Tchernev,
2012), which is the repetition of behavior without active self-regulation. Habits that go
beyond healthy behavioral patterns can interfere with studying and other required activi-
ties of everyday life. In essence, we contend that deficient self-regulation of mobile
phone use during required activities can lead to interference in daily life.
Loss of volitional control to abstain or moderate mobile phone use can have long-term
effects and deserves the attention of communication scholars. For example, heavy mul-
titasking among youth has been associated with fewer interpersonal interactions and
lower well-being (Pea et al., 2012). Also, lasting cognitive effects have been identified
with multitasking. For instance, students who reported heavy multitasking were found to
have limited ability to ignore peripheral distractions, which led to poor performance on
the primary task (Ophir et al., 2009). Furthermore, there is mounting evidence that media
multitasking during studying is associated with shallower processing (Carr, 2010), poor
performance in the classroom (Junco, 2012; Rosen et al., 2013; Wood et al., 2012), and
lower grade point average (Junco, 2012), which can collectively hinder success and well-
being in life. In light of these deleterious effects on life, deficient self-regulation in
mobile phone use deserves scrutiny.
For the purpose of this article, we look at some common behavioral manifestations of
deficient self-regulation in mobile phone use, which include using mobile phone longer
than intended, inability to cut down on mobile phone use, and continued use of mobile
phone at the expense of not completing other required tasks. MPIL, as conceptualized in
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

6 new media & society
this article, is more than nuisance incursions into daily life via mobile phone, but the
inability to curb the use of the mobile phone despite one’s better judgment.
Interference from music, texting, and social media
As already noted, recent findings among youth suggest that music, texting, and social
networking are among the most common elements of multitasking activities (Moreno
et al., 2012), which also bears out in our sample of college students (see Tables 1 and 2)
and corroborated through a pilot study.1
While Facebook use and texting allow for social interaction, music use stems from
different motivations. Although music can be engaging and evoke strong emotional
responses, with practice it also can be ignored as background noise. In fact, music
Table 1. Minutes per day for studying, media, and communication activities (N = 992).
Mean SD
Studying 198.8 101.2
Communication and media activities
Face-to-face interaction 310.4 214.2
Texting 180.3 190.5
Music 160.7 142.0
Social media 133.9 112.0
Video 116.6 95.6
Browsing 81.1 86.4
Email 37.3 50.9
Voice 36.3 53.9
Video games 31.5 63.7
Books 31.4 47.5
Total 1119.6 523.4
Table 2. Frequency of engaging in other activities when studying (0 = Never, 100 = Always)
(N = 992).
Mean SD
Music 62.24 30.56
Texting 57.13 28.31
Social media 49.74 26.74
Face-to-face interaction 43.72 25.81
Browsing 37.24 26.34
Video 32.88 26.12
Email 24.96 22.77
Voice 20.12 21.30
Books 16.82 22.09
Video games 10.88 17.73
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 7
through headphones is often used as a buffer to block out ambient noises in a dorm room
or coffee shop. Texting and social media, on the other hand, are interactive and require
active engagement and have stronger potential for interfering with studying.
Findings suggest that music and music videos have little or no effect as distractions
while studying (Pool et al., 2000, 2003). Motivated cognition may offer an explanation
for the minimal harm from music. When music is used out of habit or to filter out ambi-
ent noise, the listener exercises volitional control to limit attention to music. Furthermore,
multitasking research shows that with practice individuals become adept with some
types of multitasking and listening to music as a secondary activity may be such an
acquired skill that requires very little attention. Based on the null findings in the extant
literature buttressed by motivated cognition, we predict that listening to music when
studying will not be significantly associated with MPIL.
Furthermore, motivated cognition suggests individuals pursue appetitive responses
(Lang, 2006), and Wang and Tchernev (2012) have shown that entertainment, social
and emotional needs are the primary motivations of multitasking among students. If
multitasking meets social and emotional needs, Facebook and texting are ideal plat-
forms to satiate these needs during studying, though from a threaded cognition per-
spective performance deterioration can be expected in turn resulting in a positive
correlation with MPIL.
Hypothesis 1 (H1): Facebook use during studying will be positively associated with
MPIL.
Hypothesis 2 (H2): Texting during studying will be positively associated with
MPIL.
Media multitasking and measurement
Measurement of media use is a challenging task because of reliance on self-reports
(Chaffee and Schleuder, 1986; Pinkleton and Austin, 2002), which suffer both from
motivational biases, such as the social desirability bias, and cognitive biases, such as
errors in recalling media use episodes within a prescribed period of time. The cognitive
complexity of retrospective self-reports is compounded by the pervasive nature of media
and media multitasking.
The complexity can be reduced to some extent by imposing context and boundary
conditions on retrospective recall. For example, it is much easier to respond to the ques-
tion, “How much do you text when studying?” than to the question “How much do you
text in a typical day?” In addition, time spent on texting is correlated and constrained by
other multitasking activities that occur within the context of a primary activity. If the
primary activity is studying or doing homework, multitasking can be examined within a
limited set of activities that co-occur while studying.
Until now, multitasking researchers have focused mainly on frequency as the key
measure of multitasking. Earlier research on television use suggests that attention is a
better predictor of media effects than frequency (Chaffee and Schleuder, 1986). Likewise,
attention to music might be a more effective measure than the frequency or duration that
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

8 new media & society
music is on. Furthermore, given the important role of attention in the theoretical frame-
works applied to media multitasking, attention deserves examination as a separate meas-
ure of multitasking, which led to the following research question.
Research Question 1 (RQ1): Is there a difference between frequency and attention as
measures of multitasking for listening to music, using Facebook, and texting while
studying?
Also, in keeping with the basic premise of limited capacity, participants were given a
constant-sum task to allocate 100% points of their attention to the different multitasking
activities within a bundle, which forces a compensatory allocation model. That is, higher
allocation of attention to one activity reduces attention resources available for other
activities in the bundle, and the constant-sum estimation task simulates this psychologi-
cal process.
Measurement models of multitasking should also consider individual preference for
multitasking as a covariate. Human factors researchers have differentiated between mon-
ochronicity, preference for doing one task at a time, and polychronicity, preference for
doing more than one task at a time (Zhang et al., 2005), and demonstrated that preference
for multitasking is related to multitasking performance (Poposki and Oswald, 2010).
These scales, however, were designed to examine multitasking and task switching in
work contexts. Therefore, a new scale that emphasizes media multitasking was devel-
oped by integrating constructs from earlier research. In addition, gender, relationship
status, number of friends, number of Facebook friends, and ownership of a smartphone
are included as covariates.
Method
Undergraduate students from different majors participated in the survey for extra credit.
The survey was approved by Institutional Review Board (IRB) and conducted during
November–December 2012. Students had 2 weeks to fill out the online survey. Links to
the survey were distributed by instructors and posted on course websites. In all, 1053
students started the survey. Participants whose communication and media-related times
were more than 3 SD (standard deviation) from the mean were coded as outliers. After
dropping outliers and incompletes, 992 participants were retained. Average age of par-
ticipants was 19.7 (SD = 1.9). The sample consisted of freshmen (29.4%), sophomores
(36.9%), juniors (18.3%), and seniors (15.5%). Females (60.2%) outnumbered males
(39.8%) and 75% of the sample was made up of Caucasians. Two out of five participants
(39.4%) reported that they were in a relationship. An overwhelming majority (85.4%)
reported owning a smartphone. Key variables used in the analysis are described below.
MPIL
Four items were used to create the MPIL scale: (1) “Use my mobile phone longer than I
intended,” (2) “Would be more productive if I didn’t use my mobile phone so much,” (3)
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 9
“I have tried to cut down the amount of time spent with my mobile phone, but failed,”
and (4) “Even when I have other things to do, I find myself saying ‘Just a few more min-
utes’ and continue to use my mobile phone.” These items, rated on 5-point scale (1 =
Never, 5 = Always), were significantly correlated and had good internal consistency (M
= 2.89, SD = 0.89, α = .81).
Time spent on activities
Time spent on diverse activities of daily life, including sleeping, working, exercising,
and studying, and activities involving communication or media were assessed.
Participants entered time for each activity in hours and minutes, which was transformed
to minutes before analysis. Summary of time spent on various activities that are relevant
to this article are summarized in Table 1.
This study was limited to the three most common task switching activities while stud-
ying or doing homework that were reported in a pilot study (see Footnote 1). Given the
lack of consensus on operational definitions, three measures were examined—frequency
of bundled multitasking, frequency of pairwise multitasking, and attention allocation to
activities within a bundle.
Frequency of bundled multitasking
Multitasking or task switching happens within a bundle or cluster of activities. Based on
a qualitative pilot study and the descriptive statistics presented in Table 1, the following
combinations or bundles of activities were identified and rated on a 5-point scale (1 =
Never, 5 = Always): How often do you engage in the following group of activities (1)
homework, music, social media; (2) homework, texting/instant messaging (IM), social
media; and (3) homework, music, texting/IM. This 3-item measure had acceptable reli-
ability and was averaged to create a composite score for bundled multitasking while
studying (M = 3.70, SD = 0.87, α = .81).
Frequency of pairwise multitasking
This measure included pairwise assessment of frequency of doing another activity when
studying or doing homework. Participants provided ratings on a 100-point frequency
scale (0 = Never, 100 = Always) for co-occurrence of the following activities: studying
and texting/IM (M = 56.5, SD = 28.2), studying and social media (M = 49.2, SD = 26.5),
and studying and music (M = 61.5, SD = 30.7). Pairwise assessment of a variety of other
media activities when studying was evaluated and a summary is presented in Table 2.
Attention allocation within a multitasking bundle
A third measure of multitasking focused on allocation of attention to various activities
within a bundle through a constant-sum estimation task. Participants were asked how
much attention they allocated to each activity in a multitasking bundle that included
studying, texting/IM, social media and music. Total attention allocated to the four tasks
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

10 new media & society
was constrained to 100%. Attention allocated to studying (M = 60.6, SD = 20.8) was
significantly higher than attention allocated to the sum of the other activities, including
texting/IM (M = 14.1, SD = 9.8), social media (M = 14.1, SD = 11.3), and music (M =
11.3, SD = 9.4) (see Table 3).
Multitasking preference
A new multitasking preference scale with 14 items was introduced (Appendix 1). All
items were rated on a 7-point scale (1 = Strongly disagree, 7 = Strongly agree).
Exploratory factor analysis revealed a single-factor structure. After reverse coding nega-
tively worded items, the scale had good reliability, and the items were averaged to create
a composite score for multitasking preference (M = 5.23, SD = 0.90, α = .88). Correlations
between the multitasking preference scale and the aforementioned measures can be
found in Table 4.
Results
Descriptive analysis of multitasking when studying
We begin with descriptive statistics of time spent on communication and media activi-
ties, followed by descriptive statistics for the measures of multitasking and an examina-
tion of correlations among these measures. Table 1 shows time in minutes spent on
texting (M = 180, SD = 190), music (M = 161, SD = 142), and social media (M = 134, SD
= 112), which adds to 8 hours when summed. Furthermore, when texting, social media,
music, and other media activities were added to face-to-face communication (M = 310,
SD = 214), total was more than 18 hours (M = 1120, SD = 523). When time spent on
studying (M = 199, SD = 101) and other activities of daily life, such as sleeping, attend-
ing classes, and working were taken into account, the total exceeded 24 hours.
Next, frequency of engaging in multitasking when studying was examined. Music
was the most frequent media activity that students combined with studying (Table 2). On
a 100-point scale (0 = Never, 100 = Always), frequency of music use was highest
Table 3. Distribution of percentage of attention between studying and other multitasking
activities (N = 992).
Studying (%) Music (%) Texting (%) Social media (%)
Studying + music 79 21 – –
Studying + texting 77 – 23 –
Studying + social media 74 – – 26
Studying + music + texting 69 13 18 –
Studying + music + social media 67 14 – 19
Studying + texting + social media 67 – 16 17
Studying + music + texting +
social media
60 12 14 14
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 11
(60, M = 62.24, SD = 30.56), followed by texting (M = 57.13, SD = 28.31) and social
media (M = 49.74, SD = 26.74). In addition to frequency, attention allocated to various
activities within a multitasking bundle was examined. Common multitasking scenarios
were generated in a pilot study, and allocation of attention within each scenario is pre-
sented in Table 3. Across scenarios, attention to studying ranged from 60% to 79% and
decreased as the number of activities in a multitasking bundle increased. For instance, in
the multitasking scenario that included listening to music, texting, and social media,
attention to studying was 60%. Whereas when music was the only concurrent activity
while studying, attention to studying was 79%.
In the next step, correlations among the three measures of multitasking—frequency of
bundled multitasking, frequency of pairwise multitasking, attention allocation within a
multitasking bundle—and multitasking preference were examined (Table 4). The fre-
quency of bundled multitasking was significantly correlated with both the frequency of
pairwise multitasking and attention within a multitasking bundle. However, the correla-
tions between bundled frequency and pairwise frequency for texting (r = .35), Facebook
(r = .34), and music (r = .43) were higher than correlations between bundled frequency
and attention to texting (r = .10), Facebook (r = .16), and music (r = .18) (see Column 2,
Table 4). Also, the correlations between pairwise frequency and attention to texting (r =
.31), social media (r = .35), and music (r = .27) were statistically significant. The highest
pairwise correlation was between frequency of texting and frequency of social media use
(r = .61) (Table 4).
Except for the frequency of listening to music when studying, all measures of multi-
tasking were significantly correlated with MPIL (see Column 1, Table 4). Multitasking
Table 4. Correlations among popular multitasking activities when studying and self-reported
mobile phone interference in life (MPIL).
1 2 3 4 5 6 7 8
1. MPIL –
2. B_Freq .26*** –
3. P_Freq_Text .23*** .35*** –
4. Atten_Text .28*** .10** .31*** –
5. P_Freq_FB .21*** .34*** .61*** .16*** –
6. Atten_FB .21*** .16** .18*** .33*** .35*** –
7. P_Freq_Music .03 .43*** .25*** −.07*.26*** .01 –
8. Atten_Music .09** .18*** .00 .10** .08*.13*** .18*** –
9. MT Pref Scale .11*** .30*** .16*** .03 .17*** .04 .27*** .04
M2.89 3.70 56.5 14.1 49.2 14.0 61.5 11.3
SD 0.89 0.87 28.2 9.8 26.5 11.3 30.7 9.4
MPIL: mobile phone interference in life, B_Freq: bundled frequency, P_Freq_Text: pairwise frequency for
texting, Atten_Text: attention allocated to texting, P_Freq_FB: pairwise frequency for Facebook and social
media, Atten_FB: attention to Facebook and social media, P_Freq_Music: pairwise frequency for music, At-
ten_Music: attention to music, MT Pref Scale: multitasking preference scale.
N varied from 894 to 992.
* p ≤ .05, ** p ≤ .01, *** p ≤ .001.
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

12 new media & society
preference was significantly correlated with multitasking frequency (both bundled and
pairwise), but not with attention to texting, Facebook, or music (see Row 9, Table 4).
MPIL and multitasking preference were also significantly correlated (r = .11).
Relationship between multitasking while studying and MPIL. Although respondents reported
that more than 60% of their attention was allocated to studying, the focus of this article
is the interference caused from the allocation of the remaining attention to music, texting,
and social media. Therefore, with self-reported MPIL as the outcome variable, three two-
step hierarchical regression models were examined.
In the first step, control variables and covariates were introduced (gender, relationship
status, smartphone ownership, number of Facebook friends, number of close friends, and
preference for multitasking). In the second step, one of the multitasking measures was
introduced and the relationship between that measure and MPIL was examined (see
Table 5).
Ownership of smartphone and the number of Facebook friends were significant pre-
dictors of MPIL in all three models. Relationship status and the number of non-Facebook
friends were not significant in any of the models, and gender was significant in two out
of the three models. The standardized coefficients for smartphone ownership were
between .23 and .25 (see Table 5), with higher MPIL among smartphone owners (M =
3.0, SD = 0.84) than non-owners (M = 2.27, SD = 0.93), and the difference was tested
with an unequal variances t-test, which was significant, t = 8.84, df = 190, p < .001.
The effect of gender, which was significant in two out of the three models, was exam-
ined further using a t-test. Females (M = 3.01, SD = 0.90) reported higher MPIL than
males (M = 2.71, SD = .84), t = 5.23, df = 988, p < .001. The number of Facebook friends
had a positive relationship with MPIL, with standardized coefficients ranging from .10
to .13. Preference for multitasking, a new scale introduced in this study, was a significant
predictor (β = 0.08) of MPIL only in Model 3 that was based on the attention measure,
but not in the other two models based on frequency measures.
After controlling for the variables explained above, bundled frequency was a signifi-
cant predictor of MPIL (β = 0.18) (Model 1) and pairwise frequency was significant
(Model 2) for social media (β = 0.11) and texting (β = 0.13), but not for music. Attention
(Model 3) to social media (β = 0.10), texting (β = 0.18), and music (β = 0.08) were sig-
nificant predictors of MPIL as well. In support of H1 and H2, findings based on pairwise
frequency (Model 2, Table 5) confirm that Facebook use and texting are positively cor-
related with MPIL and significant predictors after controlling for other variables. For
music, frequency was not a significant predictor, which is in line with findings reported
in the literature, but attention to music was a significant predictor of MPIL. RQ1 focused
on the difference between frequency and attention measures of media multitasking. The
findings suggest that bundled frequency, pairwise frequency, and attention are significant
predictors of MPIL for Facebook and texting (Models 1-3, Table 5). The exception was
music, for which frequency was not predictive, whereas attention was.
In a follow-up analysis, bundled frequency, pairwise frequency, and attention (predic-
tor variables from Columns 1–3 in Table 5) were entered simultaneously. While attention
continued to be significant for texting, Facebook, and music, pairwise frequencies were
no longer significant. Given that bundled frequency subsumed pairwise frequencies,
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 13
Table 5. Predicting mobile phone interference in life (MPIL) from measures of multitasking activities when studying or doing homework.
Model 1 Model 2 Model 3
B (SE) Std βB (SE) Std βB (SE) Std β
Gender −.16** (.06) −09** −.10 (.06) −.06 −.14* (.06) .08*
Relationship status −.03 (.05) −.02 −.09 (.05) −.06 −.04 (.05) −.03
Smartphone ownership .60*** (.08) .25*** .56*** (.08) .23*** .60*** (.07) .24***
No. of Facebook friends .01*** (.003) .11*** .01*** (.004) .13*** .01*** (.003) .12***
No. of non-Facebook friends .01 (.01) .04 .01 (.01) .04 .01 (.009) .04
Frequency of bundled multitasking .17*** (.03) .17***
Frequency of pairwise multitasking
Social media/studying .004** (.001) .11**
Texting/studying .004** (.001) .13**
Music/studying −001 (.001) −.05
Attention allocation
Social media .01*** (.002) .10**
Texting .02*** (.003) .18***
Music .01* (.003) .07*
Preference for multitasking .04 (.03) .04 .03 (.04) .03 .07* (.03) .08*
Adj R2.14 .14 .184
F F(6, 921) = 26.6 F(8, 758)= 17.1 F(9, 918)= 23.2
SE: standard error.
* p ≤ .05, ** p ≤ .01, *** p ≤ .001.
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

14 new media & society
these findings are predictable. Furthermore, when interactions between bundled fre-
quency and attention were added for texting, Facebook, and music, there was no signifi-
cant increase in R-square and the interaction terms were not significant. Subsequently, all
three interactions were tested with pairwise frequency and found to be not significant.
Because attention to different multitasking activities were measured using constant-sum
estimation, it is reasonable to assume that these measures were correlated and contribut-
ing to multicollinearity in the regression equation for attention (Model 3, Table 5).
Potential autocorrelation and multicollinearity were examined for the attention model
using the Durbin–Watson statistic (1.87) and average variance inflation factor (VIF) of
the predictor variables (1.2) as diagnostics. Based on these statistics (Durbin–Watson
tending toward 2; VIF tending toward 1), it is safe to assume that neither assumption was
violated (Field, 2013).
Discussion
The mobile phone is a breakthrough advance for human communication. But with
the plethora of choices available via mobile phone, individuals who are deficient in
self-regulation or with a propensity for addiction may face challenges in managing
these choices strategically. To examine this potential dysfunctional aspect, we exam-
ined the effect of media multitasking when studying within a convenience sample
and found that both frequency and attention to texting and social media were posi-
tively related to MPIL. However, frequency of music use during studying was not
associated with MPIL, although attention to music when studying was positively
associated with MPIL.
The difference between the effects of passive listening to music and active engage-
ment in texting or social media highlights a major shift in the intrusion of media in eve-
ryday life. Traditional media, such as radio, television, or music, which can be ignored as
background noise, are fundamentally different from human interactions via text mes-
sages or social media. These interactions need immediate and active participation that
can disrupt work, cause disorientation, and result in loss of efficiency on the primary task
of studying. Maintaining various conversation threads during studying is likely moti-
vated by the desire for human interaction. While ongoing text and Facebook conversa-
tions can alleviate boredom and even serve as a form of social support when one is
studying, overindulgence in social interactions and subsequent erosion in primary task
performance can have dire consequences on psychological well-being.
That media multitasking can be overwhelming is borne out in our findings that sug-
gest a dilation of the 24-hour day. When asked to offer free recall estimates of time spent
on various activities in a typical day, including media and communication activities,
participants’ estimates of time spent was approximately 39 hours in a 24-hour day. These
misperceptions or biases in estimating time spent on media-related activities can be
attributed to multitasking. Attending to various activities at the same time perhaps leads
to discordance between psychological time and physical time. While the perceived dila-
tion of time is interesting in itself, it poses a key methodological challenge to researchers
interested in the measurement of media use. The challenge is to design rigorous and
yet user-friendly approaches to capture media use habits that accommodate ongoing
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 15
multitasking and task switching. We attempted to address this challenge by employing a
number of methodological nuances.
First, measures of media use were tied to situational contexts to keep retrospective
self-reports of media use manageable. Second, since multitasking is carried out as a clus-
ter of activities, multitasking bundles were identified for the situational context of inter-
est, namely, studying. Third, frequency of media use and attention to an activity within a
multitasking bundle were examined separately. Fourth, participants were forced to allo-
cate a constant sum of 100% attention to various activities in a multitasking bundle,
which is in keeping with the theoretical model of limited capacity. Fifth, a new media
multitasking preference scale was introduced, which could be used in future research.
While the multitasking preference scale was significantly correlated with frequency
measures, it was not correlated with attention. Furthermore, while the correlation
between frequency and attention of each of the three activities (music, texting, and
Facebook) were significant, the correlations did not exceed r = 0.3, which suggests that
frequency and attention tap different aspects of the media multitasking experience.
Future work on measurement of multitasking should take into account both frequency
and attention within a multitasking bundle.
Another contribution of this article is the introduction of the constant-sum allocation
method to measure attention. When participants were forced to allocate a constant sum
of 100% attention to various activities in a multitasking bundle, attention to studying
decreased steadily, thus offering face validity to the measure. Further replication and
examination of the constant-sum approach to attention might be a fruitful direction for
future research on media multitasking measurement. Given the wide array of media mul-
titasking activities, we believe that a measurement approach that consists of bundles of
multitasking activities within a certain context offers much promise. This bundled
approach contributes not only to ecological validity, but eliminates pairwise estimates of
all conceivable media multitasking combinations, which can lead to respondent fatigue.
The multitasking preference scale offered paradoxical results in this study. Though
multitasking preference was positively correlated with the frequency of multitasking
activities, it was also positively correlated with MPIL. This finding suggests that with
increasing preference for multitasking, a small but significant corresponding increase in
perceived interference in life can be expected. Preference for multitasking and a concur-
rent perceived loss of control from multitasking might be emblematic of the inherent
trade-off forced on users by mobile phone and other mobile technologies.
A number of differences by demographic variables deserve mention. While women
reported higher MPIL than men and the number of Facebook friends was positively
related to MPIL, smartphone ownership made a difference. Given the variety of potential
distractions available through smartphones, it is not surprising that those with smart-
phones reported significantly higher MPIL than non-owners. Future studies are required
on costs and benefits of media multitasking in other contexts and with other populations
for a fuller understanding of the role of mobile media in everyday life.
The increasing penetration of the mobile phone raises interesting theoretical questions
as well for future research. With the burgeoning affordances available in new media
interfaces, what is the limit of human capacity for multitasking? More important perhaps
is a better understanding of the motivations for task switching and the underlying
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

16 new media & society
cognitive mechanisms used to coordinate media multitasking activities in real time.
These fundamental theoretical questions naturally lead to obvious practical questions
about gain and loss from multitasking.
A limitation of the study is reliance on a convenience sample of college students.
Furthermore, with self-report measures, there is potential for social desirability bias,
which could explain the greater than 60% attention to studying. Psychophysiological
measures are less vulnerable to such biases and can be pursued in future studies. Also,
findings reported in this study are correlational and causal claims cannot be made with-
out appropriate replications using experimental designs, which offer numerous avenues
for future research.
Funding
This research received no specific grant from any funding agency in the public, commercial, or
not-for-profit sectors.
Note
1. To determine common bundles of multitasking activities, a pretest was conducted with 21
undergraduates in a multimedia design course. Participants were asked to list activities they
typically engage in simultaneously when studying. The combination of music, texting, and
Facebook was reported to be the most common activity bundle during studying.
References
Adler RF and Benbunan-Fich R (2012) Juggling on a high wire: multitasking effects on perfor-
mance. International Journal of Human–Computer Studies 70(2): 156–168.
Altmann EM and Gray WD (2008) An integrated model of cognitive control in task switching.
Psychological Review 115(3): 602–639.
Basil MD (1994) Multiple resource theory I: application to television viewing. Communication
Research 21(2): 177–207.
Bergen L, Grimes T and Potter D (2005) How attention partitions itself during simultaneous mes-
sage presentations. Human Communication Research 31(3): 311–336.
Brenner V (1997) Psychology of computer use: XLVII. Parameters of Internet use, abuse and
addiction: the first 90 days of the Internet Usage Survey. Psychological Reports 80: 879–882.
Cacioppo JT and Berntson GG (1994) Relationship between attitudes and evaluative space.
Psychological Bulletin 115(3): 401–423.
Carr N (2010) The Shallows: What the Internet Is Doing to Our Brains. New York: W.W. Norton
& Company, Inc.
Carrier LM, Cheever NA, Rosen LD, et al. (2009) Multitasking across generations: multitask-
ing choices and difficulty ratings in three generations of Americans. Computers in Human
Behavior 25(2): 483–489.
Chaffee SH and Schleuder J (1986) Measurement and effects of attention to media news. Human
Communication Research 13: 76–107.
Dux PE, Tombu MN, Harrison S, et al. (2009) Training improves multitasking performance by
increasing the speed of information processing in human prefrontal cortex. Neuron 63(1):
127–138.
Field A (2013) Discovering Statistics Using IBM SPSS Statistics. Thousand Oaks, CA: Sage
Publications.
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 17
González VM and Mark G (2004, April) Constant, constant, multi-tasking craziness: managing
multiple working spheres. In: Proceedings of the SIGCHI conference on Human factors in
computing systems (pp. 113–120), ACM.
Griffiths M (1999) Internet addiction: Internet fuels other addictions. Student British Medical
Journal 7: 428–429.
Jeong S-H and Hwang Y (2012) Does multitasking increase or decrease persuasion? Effects of
multitasking on comprehension and counter-arguing. Journal of Communication 62: 571–587.
Jeong S-H, Hwang Y and Fishbein M (2010) Effects of exposure to sexual content in the media
on adolescent sexual behaviors: the moderating role of multitasking with media. Media
Psychology 13(3): 222–242.
Junco R (2012) In-class multitasking and academic performance. Computers in Human Behavior
28(6): 2236–2243.
Junco R and Cotten SR (2012) No A 4 U: the relationship between multitasking and academic
performance. Computers & Education 59(2): 505–514.
Kim J-H and LaRose R (2004) Interactive e-commerce: promoting consumer efficiency or impul-
sivity? Journal of Computer-Mediated Communication 10(1).
Kubey R (1996) Television dependence, diagnosis and prevention: with commentary on video
games, pornography, and media education. In: MacBeth T (ed.) Tuning into Young Viewers:
Social Science Perspectives on Television. Newbury Park, CA: SAGE, pp. 221–260.
Lang A (2000) The limited capacity model of mediated message processing. Journal of
Communication 50(1): 46–70.
Lang A (2006) Using the limited capacity model of motivated mediated message processing to
design effective cancer communication messages. Journal of Communication 56(Suppl. s1):
S57–S80.
LaRose R, Lin CA and Eastin MS (2003) Unregulated Internet usage: addiction, habit, or deficient
self-regulation? Media Psychology 5(3): 225–253.
McIlwraith RD (1998) “I’m addicted to television”: the personality, imagination and TV watch-
ing pattern of self-identified TV addicts. Journal of Broadcasting & Electronic Media 42:
371–386.
Marci C (2012) A (biometric) day in the life: engaging across media. Paper presented at Re: Think
2012, New York, 28 March.
Marlatt GA, Baer JS, Donovan DM, et al. (1988) Addictive behaviors: etiology and treatment.
Annual Review of Psychology 39(1): 223–252.
Meiran N, Chorev Z and Sapir A (2000) Component processes in task switching. Cognitive
Psychology 41(3): 211–253.
Moreno MA, Jelenchick L, Koff R, et al. (2012) Internet use and multitasking among older adoles-
cents: an experience sampling approach. Computers in Human Behavior 28(4): 1097–1102.
Olson MA and Fazio RH (2001) Implicit attitude formation through classical conditioning.
Psychological Science 12: 413–417.
Ophir E, Nass C and Wagner AD (2009) Cognitive control in media multitaskers. Proceedings of
the National Academy of Sciences of the United States of America 106(37): 15583–15587.
Oulasvirta A, Rattenbury T, Ma L, et al. (2012) Habits make smartphone use more pervasive.
Personal and Ubiquitous Computing 16(1): 105–114.
Pea R, Nass C, Meheula L, et al. (2012) Media use, face-to-face communication, media multitask-
ing, and social well-being among 8- to 12-year-old girls. Developmental Psychology 48(2):
327–336.
Pinkleton BE and Austin EW (2002) Exploring relationships among media use frequency, per-
ceived media importance and media satisfaction in political disaffection and efficacy. Mass
Communication and Society 5: 141–164.
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

18 new media & society
Pool MM, Koolstra CM and Voort TH (2003) The impact of background radio and television on
high school students’ homework performance. Journal of Communication 53(1): 74–87.
Pool MM, Van der Voort TH, Beentjes JW, et al. (2000) Background television as an inhibitor of
performance on easy and difficult homework assignments. Communication Research 27(3):
293–326.
Poposki EM and Oswald FL (2010) The multitasking preference inventory: toward an improved
measure of individual differences in polychronicity. Human Performance 27: : 247–264.
Rosen LD, Carrier ML and Cheever NA (2013) Facebook and texting made me do it: media-
induced task-switching while studying. Computers in Human Behavior 29(3): 948–958.
Salvucci DD and Taatgen NA (2008) Threaded cognition: an integrated theory of concurrent mul-
titasking. Psychological Review 115(1): 101–130.
Salvucci DD and Taatgen NA (2010) The Multitasking Mind. New York; Oxford: Oxford
University Press.
Srivastava J (2013) Media multitasking performance: role of message relevance and formatting
cues in online environments. Computers in Human Behavior 29(3): 888–895.
Voorveld HAM (2011) Media multitasking and the effectiveness of combining online and radio
advertising. Computers in Human Behavior 27(6): 2200–2206.
Wang Z and Tchernev JM (2012) The “Myth” of media multitasking: reciprocal dynamics of
media multitasking, personal needs, and gratifications. Journal of Communication 62(3):
493–513.
Wood E, Zivcakova L, Gentile P, et al. (2012) Examining the impact of off-task multi-tasking with
technology on real-time classroom learning. Computers & Education 58(1): 365–374.
Zhang Y, Goonetilleke RS, Plocher T, et al. (2005) Time-related behavior in multitasking situa-
tions. International Journal of Human-Computer Studies 62: 425–455.
Author biographies
Prabu David (PhD, University of North Carolina at Chapel Hill) is a Professor in the Edward R.
Murrow College of Communication at Washington State University. His research focuses on the
role of social and mobile media on health outcomes.
Jung-Hyun Kim (PhD, Michigan State University) is an Associate Professor at Sogang University,
Seoul, Korea. Her research interests include social and psychological impacts of new media, new
media and physical/psychological well-being, and social interaction and social influence in virtual
world.
Jared S Brickman is a graduate student in the Edward R. Murrow College of Communication at
Washington State University. His research interests include problematic use of new media, new
communication technologies, and the cognitive effects of viral media content.
Weina Ran (MA) is a doctoral candidate in the Edward R. Murrow College of Communication at
Washington State University. Her research interests include media effects on public health (e.g.
sexual health, drug abuse, and nutrition), entertainment education, and new technology, social
media, and multitasking.
Christine M Curtis (MA) is an academic advisor in the Edward R. Murrow College of
Communication at Washington State University. Her research interests include environmental and
crisis communication.
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from

David et al. 19
Appendix 1
Multitasking preference scale
1. I am more efficient when I am multitasking.
2. I try to multitask whenever possible.
3. I enjoy multitasking.
4. I am in a state of flow when multitasking.
5. I multitask out of habit.
6. Before multitasking I deliberately think about specific tasks that I can do
concurrently.
7. I lose track of time when multitasking.
8. I can do more through multitasking.
9. When I am on a computer or using my mobile phone, I am always drawn to do
more than one thing at a time.
10. I am distracted when I have to focus on only one task.
11. I find it difficult to do more than one task at a time.
12. I am bored when I am not multitasking.
13. I find it entertaining and enjoyable when multitasking.
14. I find it distracting to engage in different activities concurrently.
at Sogang University on February 11, 2015nms.sagepub.comDownloaded from