ArticlePDF Available

Antecedents and consequences of problematic smartphone use: A systematic literature review of an emerging research area

Authors:

Abstract and Figures

This article provides a systematic review of existing research on problematic smartphone use (PSU) to guide other researchers in search of relevant studies, and to propose areas for future research. In total, 293 studies were analyzed leading to the development of an overview model in the field of PSU, presenting findings on demographic factors, explanations for smartphone use and why this use becomes problematic, consequences of PSU, and how such use can be corrected. In addition, we considered in which contexts, with which methods, and with which theoretical lenses this stream of research has been studied to date. Smartphone use is most often explained by the smartphone design, and users' emotional health and their ability to control smartphone use. Our review suggests that people who are young, female, and highly educated are more prone to PSU. Emotional health issues are the most frequently identified consequence of PSU. Strategies for correcting PSU fall into three categories: information-enhancing, capacity-enhancing, and behavior reinforcement strategies. The studies on PSU are most often conducted using quantitative surveys with university and college participants considering their personal smartphone use. Whereas a variety of theoretical frameworks have been adopted to investigate PSU, they are often related to identifying factors explaining use and problematic use, and more seldom to analyze the findings. A future research agenda for PSU is proposed consisting of seven key research questions which can be investigated by researchers going forward.
Content may be subject to copyright.
Computers in Human Behavior 114 (2021) 106414
Available online 20 May 2020
0747-5632/© 2020 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Review
Antecedents and consequences of problematic smartphone use: A
systematic literature review of an emerging research area
Peter Andr´
e Busch
a
,
*
, Stephen McCarthy
b
a
Department of Information Systems, University of Agder, Kristiansand, Norway P.O. Box 422, 4604, Kristiansand, Norway
b
Department of Business Information Systems, University College Cork, Ireland
ARTICLE INFO
Keywords:
Mobile phone
Technology addiction
Smartphone addiction
Internet addiction
Nomophobia
ABSTRACT
This article provides a systematic review of existing research on problematic smartphone use (PSU) to guide
other researchers in search of relevant studies, and to propose areas for future research. In total, 293 studies were
analyzed leading to the development of an overview model in the eld of PSU, presenting ndings on de-
mographic factors, explanations for smartphone use and why this use becomes problematic, consequences of
PSU, and how such use can be corrected. In addition, we considered in which contexts, with which methods, and
with which theoretical lenses this stream of research has been studied to date. Smartphone use is most often
explained by the smartphone design, and usersemotional health and their ability to control smartphone use. Our
review suggests that people who are young, female, and highly educated are more prone to PSU. Emotional
health issues are the most frequently identied consequence of PSU. Strategies for correcting PSU fall into three
categories: information-enhancing, capacity-enhancing, and behavior reinforcement strategies. The studies on
PSU are most often conducted using quantitative surveys with university and college participants considering
their personal smartphone use. Whereas a variety of theoretical frameworks have been adopted to investigate
PSU, they are often related to identifying factors explaining use and problematic use, and more seldom to analyze
the ndings. A future research agenda for PSU is proposed consisting of seven key research questions which can
be investigated by researchers going forward.
1. Introduction
Information systems (IS) research is often conducted under the
premise that technology use is positive for society leading to innova-
tion, development, and value creation (Turel, Serenko, & Giles, 2011).
However, we continue to see examples of technology use which do not
create a better world (Turel et al., 2011; Walsham, 2012). For example,
cases abound of where private details about individuals are exposed,
fundamentalist groups are enabled to contact and recruit vulnerable
people, and people have become overly attached to technologies. These
unfortunate use of technology can have long-term and severe impacts on
the quality of life for users of technology and their peers. Despite the
notable practical and theoretical implications of problematic technology
use, research efforts to date have been more focused on studying the
positive sides of technology (Chen, Liu, et al., 2017).
The smartphone is one notable type of technology which can have
negative use consequences. Smartphone use becomes problematic when
users have difculties controlling their use and as a result suffers
impaired daily functioning (Ezoe et al., 2009; Horwood & Anglim,
2018). These effects were recognized as early as 2006 when Americans
popularized the term « CrackBerry » to describe the addictive nature of
the BlackBerry smartphone (Turel, Serenko, & Bontis, 2008). Problem-
atic smartphone use (PSU) may lead to various unfortunate conse-
quences such as the lack of sleep (Lapointe, Boudreau-Pinsonneault, &
Vaghe, 2013), family conicts (Turel et al., 2008), and the experience
of imagined phone signals (Tanis, Beukeboom, Hartmann, & Vermeulen,
2015). It could also lead to more serious consequences such as
dangerous driving (Soror, Steelman, & Limayem, 2012), depression
(Harwood, Dooley, Scott, & Joiner, 2014), and anxiety (Hartanto &
Yang, 2016).
PSU, therefore, concerns us not only on the individual level but also
more broadly on an organizational and societal level. The smartphone
has become one of the most widespread and inuential technological
innovations that we as a society immerse ourselves in and offers a
When the phone was tied with a wirehumans were free(unknown source).
* Corresponding author.
E-mail addresses: peter.a.busch@uia.no (P.A. Busch), stephen.mccarthy@ucc.ie (S. McCarthy).
Contents lists available at ScienceDirect
Computers in Human Behavior
journal homepage: http://www.elsevier.com/locate/comphumbeh
https://doi.org/10.1016/j.chb.2020.106414
Received 29 July 2019; Received in revised form 24 February 2020; Accepted 4 May 2020
Computers in Human Behavior 114 (2021) 106414
2
computing platform with greater portability than many other devices
such as laptops and tablets (Barnes, Pressey, & Scornavacca, 2019;
Bernroider, Krumay, & Margiol, 2014). Smartphones are widely used in
various contexts such as in personal life, in encounters with private and
public organizations, and in work life. As a result, the number of active
mobile subscriptions are now estimated to exceed the total world pop-
ulation (Konok, Pog´
any, & Mikl´
osi, 2017). Because smartphones are
easily accessible, and their use socially acceptable, the actual frequency
and duration required for smartphone use to become problematic are
disputed (Kim & Koh, 2018). Despite an increasing body of research
within this area, important questions remain about normative smart-
phone usage, and antecedents and consequences of PSU.
Given the potential for unfortunate consequences, researchers must
create societal awareness about PSU, suggest normative smartphone
use, and recommend how unfortunate consequences can be avoided and
how problematic usage can be corrected to return to a healthy level
(Mahapatra, 2019; Turel et al., 2008).
This article presents a systematic review of studies which have
looked at PSU to date, summarizes what we know about the phenome-
non, and suggests areas for future research. The literature review is
guided by the following research questions:
1.1. RQ1. What factors explain smartphone use?
Technology use has often been explored through various technology
acceptance and continuance models mostly underpinned by rational
explanations such as perceived usefulness, perceived ease of use, and
facilitating conditions. We are interested in knowing why people use
smartphones.
1.2. RQ2. What are the antecedents of PSU?
Whereas smartphone use can be highly rational, engaging in prob-
lematic use may not. After initial adoption, researchers focus on why
people continue to use the smartphone to the extent that the use is
embedded or routinized in users lives (Davazdahemami, Hammer, &
Soror, 2016). These repetitive usage behaviors are dened as habit in
the IS literature and can have negative consequences on the lives of the
users (Davazdahemami et al., 2016; Turel et al., 2011). We are inter-
ested in explanations for why smartphone use can become problematic.
1.3. RQ3. Which demographic groups are most prone to PSU?
Over the past four decades, the personal computer has become
publicly available for many people worldwide. Over time, other tech-
nological artifacts such as the smartphone, have become widespread in
use, and for some, are now considered necessary to conduct everyday
tasks. Adolescents today have not experienced a society without these
artifacts and take them for granted. We are interested in knowledge
about how factors such as age, gender, education, and contextual use
impact proneness to PSU.
1.4. RQ4. What are the consequences of PSU?
Whereas outcomes of technology use are often considered favorable
expecting gains in terms of productivity and quality (Bruzzi & Joia,
2015), studies of PSU have shown that excessive smartphone use can
have severe effects on mental health and well-being (e.g., Samaha &
Hawi, 2016). We are interested in learning about the various conse-
quences of PSU identied in the literature.
1.5. RQ5. What strategies are used to correct PSU?
Knowing about potential ill consequences of PSU, it is important to
create awareness and gain knowledge about unhealthy use for it to be
corrected and brought back to normal use levels. We seek to understand
what measures can be taken to help people suffering from PSU.
This review presented in this article has analyzed 293 articles about
PSU published between 2008 and 2019. To identify studies, we searched
Scopus, Web of Science, and PsycInfo for high-end journals and con-
ference proceedings across multiple disciplinary elds such as psychol-
ogy, sociology, medicine, and computer science. The remainder of this
article is organized in the following manner. First, we describe our un-
derstanding of PSU derived from the studies included in our review.
Thereafter, we present our methodology for searching, identifying, and
analyzing the studies. Then, we lay out the main characteristics of the
studies before we answer our research questions. The literature review
ends with a discussion of potential areas for future research and a
summary of our main ndings and discussion points.
2. Problematic smartphone use
PSU, also often referred to as smartphone addiction (e.g., Sun, Liu, &
Yu, 2019) or nomophobia (no mobile phone phobia) (e.g., Tams,
Legoux, & L´
eger, 2018), has garnered increased attention from re-
searchers and public health practitioners in recent years. However,
given the relatively recent emergence of PSU as an area of research,
denitions on the concept are still evolving (Horwood & Anglim, 2018;
Nahas, Hlais, Saberian, & Antoun, 2018). PSU is broadly dened as a
compulsive pattern of smartphone usage which can result in negative
consequences that impair the daily functioning of the user (Ezoe et al.,
2009; Horwood & Anglim, 2018; Lepp, Li, & Barkley, 2016; Shin & Dey,
2013). Compulsive use refers to an uncontrollable overuse characterized
by maladaptive dependency (Chen, Liu, et al., 2017) and a tendency to
use the smartphone without being separated from it (Cho & Lee, 2017).
Negative consequences refer to symptoms such as withdrawal, and
impeded user productivity, social relationships, physical health, or
emotional well-being in daily life (Horwood & Anglim, 2018; Shin &
Dey, 2013).
Literature has also discussed PSU in relation to the setting in which
the smartphone is used (Lepp et al., 2016; Shin & Dey, 2013; Soror et al.,
2012). For instance, PSU in the bedroom during normal hours of sleep
has been linked to poor sleep quality and sleep disorders (Bernroider
et al., 2014), while PSU in the classroom has been linked to procrasti-
nation (Rozgonjuk, Kattago, & T¨
aht, 2018). Furthermore, dangerous
smartphone use has been put forward as a specic type of PSU, where
usage of the smartphone places the user or other individuals at risk of
injury (Soror et al., 2012; Steelman, Soror, Limayem, & Worrell, 2012).
Dangerous smartphone use can result in road trafc injuries from
smartphone use while driving, as well as pedestrian collisions and falls
from smartphone use while walking (Chang et al., 2019; Soror et al.,
2012).
A number of scholars have conceptualized PSU as an addiction, one
which is non-chemical and behavioral in nature (Billieux, 2012;
Contractor, Frankfurt, Weiss, & Elhai, 2017; Enez Darcin et al., 2016;
G¨
okçearslan, Mumcu, Has¸laman, & Çevik, 2016). Davazdahemami et al.
(2016) further differentiate between addiction to a mobile phone and
addiction through a mobile phone (i.e., where the user becomes addic-
ted to a mobile phone application). Some researchers have described
smartphone addiction as analogous to other forms of addiction such as
gambling or internet addiction given that it may similarly lead to an
uncontrolled psychological dependency on use, craving, withdrawal
symptoms, and anxiety when not available (Bian & Leung, 2015; Jun,
2016; Seo, Kim, & David, 2015).
While the concepts of ‘smartphone addictionand ‘PSUare often
used interchangeably (Nahas et al., 2018), some scholars assert that use
of the term addiction in the context of PSU is controversial (cf. Seo et al.,
2015). It is argued that problematic use should not necessarily be
considered an addiction prima facie as it may also derive from other
sources such as obsessive compulsive checking (Steelman et al., 2012),
loneliness (Kim, 2017, 2018), anxiety (Contractor et al., 2017), or un-
resolved real-life problems (Wang, Wang, Gaskin, & Wang, 2015). For
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
3
instance, Steelman et al. (2012) note that while addiction motivates
repeated behaviors in order to achieve pleasure, obsessive compulsive
checking in contrast aims to reduce anxiety and distress. Steelman et al.
(2012) also point out that research on the inuence of addiction on
mobile phone use has only been able to explain part of the phenomenon
with R2 values ranging from 10 to 27% (cf. Beranuy, Oberst, Carbonell,
& Chamarro, 2009; Leung, 2008). Whereas the term ‘addictionshould
be used with caution, PSU shares characteristics similar to other
addictive behaviors, which is why our review on PSU includes literature
discussing such behaviors.
In this paper, PSU is dened as the recurrent craving to use a
smartphone in a way that is difcult to control and leads to impaired
daily functioning (adapted from Ezoe et al., 2009; Horwood & Anglim,
2018). The remainder of this paper is dedicated to presenting a sys-
tematic review of literature on PSU. In particular, the main focus of the
literature review is to investigate potential antecedents of the recurrent
craving of smartphone use and how users daily functioning is impaired
as a result. However, it should be noted that our denition of PSU does
not consider the inappropriate or illicit usage of smartphones e.g.,
sending unwarranted nude photos without the recipients consent;
contacting and recruiting vulnerable people for fundamentalist groups;
or disseminating private or falsied details about another individual.
While these forms of smartphone use are indeed problematic, our de-
nition aligns with the majority of papers in our sample by solely focusing
on instances where smartphone use results in recurrent cravings and
impaired daily functioning.
3. Methodology
This study follows the guidelines for systematic literature reviews
presented by Kitchenham (2004), and has been conducted in seven
steps: (1) development of a review protocol, (2) identication of
research, (3) selection of relevant studies, (4) supplementary searches,
(5) quality assessment of studies, (6) data extraction, and (7) data syn-
thesis. The rst step was to make a plan for executing the review. The
protocol was peer reviewed prior to the study and changes were made to
the protocol based on the provided feedback. The protocol claried
several aspects of the review: study rationale, search engines, outlets,
research questions, search strings and identication of studies, study
selection criteria, and data extraction. The protocol functioned as a
detailed manual to ensure rigor in the review process. The protocol is
provided as supplementary material to this review.
Given the relatively newfound interest in the phenomenon of PSU,
we aimed for exhaustiveness in our search. Hence, we included as many
articles from journal and conference outlets as possible. In particular, we
completed comprehensive searches on Scopus, Web of Science, and
PsycInfo databases. To identify literature, we rst sought to identify
appropriate search words for our review. These were selected based on
several initial literature searches where titles, abstract, and denitions
were read. The initial investigations showed that research on PSU
adopted a variety of terms, and therefore, several search strings were
necessary to capture as much of the extant literature as possible. During
the initial screening of the articles, additional search terms were iden-
tied and used. The terms (a) smartphone, (b) ‘mobile phone, and (c)
‘cell phonewere the strings we combined with the terms ‘problematic
use, addiction, dependence, overuse, obsessive, and disorder. In the
database, titles, abstracts, and keywords were searched to ensure rele-
vancy. Our argument is that if the selected search strings were not
mentioned in these elds, the article was most likely not relevant for our
study since our search strings could appear in passing in some articles
without constituting the main focus in them. The initial search identied
788 journal and conference articles.
In the third step, non-relevant studies were excluded. Exclusion was
based on seven criteria described in more detail in the review protocol
(number of excluded articles in parentheses): duplicates (165), acces-
sibility (56), research-in-progress articles (21), recurrence of study (18),
articles written in a non-English language (10), non-research articles (8),
and articles with anonymous author (1). After exclusion, we assessed the
remaining articles based on three criteria described in more detail in the
review protocol (number of articles not matching the inclusion criteria
in parentheses): not pertinent to our research questions (115), articles
mainly focusing on developing measurement instruments (94), and not
focusing on smartphones (7).
We thereafter engaged in backward and forward reference searching
as the fourth step. These techniques involve identifying research cited in
an already identied article (backward; we build on literature reviews
by peers) and research citing an already identied article (forward; we
include research by peers who showed interest in the same articles as
us). The search for literature ended in identifying a total of 293 articles.
In the fth step, we conducted a quality assessment of all the articles
in our dataset. To guide the assessment, we adapted the validated
quality assessment criteria and scoring measures developed by Beecham
et al. (2008), detailed in the review protocol. In particular, we adapted
the ‘response options for scoring slightly by adding a ‘moderate
response option for the rst, third, and sixth assessment criteria. This
provided us with a more nuanced scoring system to aid the quality
assessment of our extensive literature review. Following Beecham et al.
(2008), we used the scoring as a heuristic to assess quality rather than as
a basis to reject a study. We rst examined a sample of 15 articles and
compared results to ensure scoring consistency. We then independently
reviewed the remainder of the articles in our set of articles to generate a
score for each article. The data was normalized by recording the per-
centage score. Overall, the majority of articles in the sample were rated
as goodor very good by the authors.
In the sixth step, we examined each article independently to identify
core characteristics of PSU research and answers to our research ques-
tions. A list of the identied articles was made listing their core focus
and research questions, theoretical frameworks, methodological ap-
proaches, and conclusions. The title, abstract, introduction and
conclusion for each article were read. Other parts of an article were read
if necessary to extract relevant data. To assure consistency in the data
extraction, two reviewers worked through all the coded articles and
discussed problematic issues together until discrepancies were resolved.
Appendix A lists all the reviewed articles with their individual core
focus, methodology, and sample.
The nal step consisted of synthesizing the literature to answer our
research questions. The synthesis was based on our data extraction and
in-depth reading of the articles if necessary. To understand antecedents
of smartphone use (RQ1) and PSU (RQ2), we rst derived all factors
from the data extraction. Each researcher analyzed the list to (1) remove
recurring factors and (2) categorize them. The categorization was
demanding since it required researchers to read articles in-depth to
identify and/or assess whether the factors explained smartphone use or
PSU. The categorization was discussed between the researchers until
agreement around the coding process was achieved. To gain high-level
insights into the demographic groups which are most prone to PSU
(RQ3), we categorized studies based on the demographic factors age,
gender, educational level, and occupation. Finally, we investigated the
associated consequences of PSU (RQ4) and strategies that can be used to
correct PSU (RQ5). This process followed the same procedures as earlier
where we (1) removed recurring consequences and strategies, (2)
categorized them, and (3) discussed the categorization until agreement
was achieved between the researchers.
4. Descriptives
The research interest in PSU is relatively newfound. Our literature
search resulted in a list of 293 articles, the earliest article identied
being a study from 2008. Computers in Human Behavior was the most
favored journal and Oulasvirta, Rattenbury, Ma, and Raita (2012) the
most cited article (according to Google Scholar). The number of pub-
lished articles and citations varies by each year. To illustrate the
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
4
research interest in PSU, we calculated a research interest factor (RIF) by
dividing the number of Google Scholar citations the reviewed articles
received each year by the accumulated number of citable articles for
each year (i.e., as soon as they became available online). Thus, the RIF
shows how many citations each citable article in our review receives
each year on average. For example, 218 out of the 293 articles in our
review were citable in 2018 receiving 2946 citations resulting in a
RIF =13.51. Incorrect and undated citations were removed before
calculating the RIF. The procedures to calculate the RIF are described in
more detail in the review protocol. Fig. 1 depicts the research interest in
PSU in the years 20082019.
PSU has received attention from researchers all over the world (46
different countries representing all parts of the world). Researchers from
South Korea, the United States, and China are the most active in this
stream of research. Their afliation was used to associate them with a
research discipline showing that research within PSU is characterized by
multidisciplinary collaborations involving IS, psychology, health, media
and communication, management, education, and psychiatry. The
research background of the authors inuences the context in which the
studies are conducted. Whereas other research disciplines have sug-
gested that certain geographical areas are overrepresented such as the
US/Anglo-centric perspective in public administration research, our
review shows that PSU has been studied in different geographical areas
with data from several countries, of which the United States and China
were the most frequently studied. Smartphone use may differ based on
geographical factors such as the digitalization rate in the country
(Barnes et al., 2019), pressures to perform in business (Seo et al., 2015),
and the level of parental control (Chang et al., 2019). Thus, the extent to
which smartphone use becomes problematic can vary among countries.
A wide variety of theoretical frameworks have been adopted to
investigate PSU. Based on our analysis, we found that more than 25
different theoretical lenses were used in the articles. However, the the-
ories are often related to identifying factors explaining use and prob-
lematic use, and more seldom to analyze the ndings. Also, there was
frequently no specic theoretical lens employed. Considering all the
lenses, theoretical frames looking at cognitive aspects were most
frequent. Of the reviewed articles, the following theories were most
frequently used: compensatory internet use theory (e.g., Elhai &
Contractor, 2018; Elhai, Levine, Dvorak, & Hall, 2016; Hong et al., 2019;
Rozgonjuk, Levine, Hall, & Elhai, 2018; Wang et al., 2015), extended
self-theory (e.g., Clayton, Leshner, & Almond, 2015; Hartanto & Yang,
2016), the functionalist perspective (e.g., Chen, Liu, et al., 2017; Zhang,
Chen, Zhao, & Lee, 2014a), uses and gratications theory (e.g., Elhai &
Contractor, 2018; Kim, Park, Lee, Ko, & Lee, 2019; Mei, Xu, Gao, Ren, &
Li, 2018; Zhitomirsky-Geffet & Blau, 2016), social cognitive theory (e.g.,
Chen, Liu, et al., 2017; Kim, Park, Lee, et al., 2019), and attachment
theory (e.g., Eichenberg, Schott, & Schroiff, 2019; Li & Hao, 2019;
Zhang, Tan, & Lei, 2019).
Only a few studies applied (combinations of) qualitative methods
using interviews, experiments, focus groups, cluster analysis, and anal-
ysis of documents. Six of the studies used mixed methods approaches
applying both qualitative and quantitative methods. The rest of the
studies applied quantitative research methods. During our review of the
studies, we observed that several quantitative studies did not provide
readers with the research model and results of the hypotheses. We
encourage authors to do so, since a study becomes easier to understand
and the main ndings are more easily identied. Many of the studies
applied well-dened and validated measurement scales. However, we
would like to point out the value of including the measurement in-
struments where items have been developed or adapted for other re-
searchers to reuse them and/or replicate studies. Appendix B lists
measurement instruments used to measure PSU in the quantitative
studies (for additional instruments, see e.g., Billieux (2012), De-Sola
Guti´
errez, Rodríguez de Fonseca, and Rubio (2017), and (Harris, Regan,
Schueler, & Fields, 2020). Several of the studies measured other con-
structs as well, which are not included in Appendix B. Furthermore,
reliability is stated in the appendix using values from Cronbachs Alpha
tests. However, some of the studies used the newer composite reliability
(CR) to claim measurement reliability.
The samples were biased in terms of types of informants where
university students comprised the majority of them. It is debatable
whether students are representative of the general population (Peterson,
2001) and smartphone users in general. For example, people working
and receiving salaries are faced with tougher requirements to perform
than students, children do not have the same cognitive capabilities as
adults, and elderly people are less used to smartphones because they
have not grown up with them. Thus, researchers should exercise caution
when attempting to extend ndings from using student subjects to
non-student populations (Peterson, 2001).
The number of respondents across the quantitative studies ranged
from about a hundred to nearly ten thousand. The studies reported a
slight majority of female respondents. Some of the studies (e.g., Chot-
pitayasunondh & Douglas, 2016; Contractor et al., 2017; Elhai et al.,
2016; Seo et al., 2015; Steelman et al., 2012; Tanis et al., 2015) recruited
their respondents from Amazons Mechanical Turk (MTurk); a crowd-
sourcing internet marketplace where work can be posted for pay
(Buhrmester, Kwang, & Gosling, 2011). The service was originally
intended for internal purposes, but has evolved to be open for both
people requesting work to be done and people with various backgrounds
completing this work (Landers & Behrend, 2015). Social science re-
searchers have used MTurk since at least 2009 to recruit participants for
a variety of topics and research designs (Buhrmester et al., 2011;
Landers & Behrend, 2015). The included studies in this literature review
used MTurk exclusively for surveys. Sampling based on MTurk is
debatable since it could (1) facilitate repeated completions of different
forms for different studies, (2) raise concerns over the participants
commitment due to low compensation, (3) create selection bias because
participants can choose not to complete the forms, and (4) be
non-representative of working populations (Landers & Behrend, 2015).
Whereas these concerns are not without merit, we consider MTurk as
a promising option for researchers studying PSU. It could address the
problems with severe oversampling of university students and recruit
participants from developing countries (e.g., African countries which
are underrepresented in the reviewed studies), from different organi-
zational contexts (the reviewed studies mainly considers smartphone
use in general, i.e., both work-related and private use), and with
different educational backgrounds (e.g., other users than those who are
highly educated). Studies in our review reporting monetary compensa-
tion paid MTurk respondents modestly (ranging from 20 cents to $1.5).
The consistency in terms of the applied research methods imply that
research on PSU has reached some agreement, and that researchers vary
their perspectives based on explanations for PSU, consequences, medi-
ators, and moderators. The various research endeavors within PSU are
presented in Fig. 2.
Fig. 2 can also be mapped to the aforementioned research questions
Fig. 1. Research interest in PSU (20082019).
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
5
in this paper, which aims to provide a broad overview of PSU studies to
date in literature. Firstly, factors explaining smartphone use map to
RQ1. Factors explaining PSU map to RQ2, and actors and settings map to
RQ3. Consequences relate to RQ4. And lastly, evaluation relates to RQ5.
Next, we describe research on PSU based on the perspectives repre-
sented in Fig. 2.
5. Research on PSU
In this section, we outline the different strands of research on PSU as
depicted in Fig. 2 and represented in our ve broad research questions.
Based on an analysis of the ndings from the systematic literature re-
view, the authors coded factors associated with smartphone use, ante-
cedents of PSU, and consequences of PSU into six overarching
categories: emotional health, physical health, control, professional
performance, social performance, and technology features. The factors,
antecedents, and consequences are henceforth classied as ‘sub-
categories in the proceeding tables. This classication was useful for
making sense of the multitude of factors identied across the research
questions and exploring overarching themes.
5.1. What factors explain smartphone use?
The categories of smartphone use investigated across all reviewed
studies were largely consistent. Commonly investigated categories of
smartphone use included: voice-calls, text, social media, instant
messaging, email, information seeking (i.e., news headlines), web surf-
ing, playing games, music/video streaming, taking photos/videos,
functional apps (i.e., maps, calendar, clock, memo/note taking), and
educational apps (i.e., online learning, school websites). More recently,
studies have begun to increasingly study the relationship between spe-
cic types of apps available on their smartphone and an individuals
level of usage (Lee, Kim, & Choi, 2017; Prasad et al., 2018; Rozgonjuk,
Kattago, et al., 2018); this contrast previous approaches which sought to
study smartphone use as a more holistic phenomenon, irrespective of the
app used. However, the growing capabilities of smartphones mean they
are no longer used solely for communication, and users can now utilize
dedicated apps for other purposes such as gaming, web surng,
gambling etc. Positive associations have been found between smart-
phone usage and the availability of different features such as voice-calls,
text, social media, email, instant messenger, video, gaming, functional
apps etc. (Davazdahemami et al., 2016; Lapointe et al., 2013; Van
Deursen, Bolle, Hegner, & Kommers, 2015). In particular, No¨
e et al.
(2019) found that the social app Snapchat was associated with the
highest levels of smartphone use, an app widely used by teenagers.
Many studies included in our review investigate both antecedents of
smartphone usage and PSU in tandem, at times using the two terms
interchangeably, or focusing on instances of high levels of smartphone
usage. However, for the purposes of RQ1, we have focused on the factors
explicitly associated with smartphone usage (regardless of the level of
usage), rather than factors explicitly associated with PSU. A distinction
should also be made between studies which adopted ‘self-reported
measures of smartphone usage, and studies which adopted ‘actual
measures of usage. While most studies included in our systematic review
asked individuals to self-report their level of smartphone use, more
recent studies have adopted objective measures of ‘real useusing ap-
plications that monitor actual time spent on the smartphone; e.g., the
duration and frequency of calls made using the smartphone (No¨
e et al.,
2019; Yook, Park, Choi, Kim, & Choi, 2019). These studies suggest a
discrepancy between self-reported assessment of mobile phone use and
real use activity (Yook et al., 2019). Differentiations have also been
made between screen time minutes and phone screen unlocking, sug-
gesting different causal links for each (Rozgonjuk, Kattago, et al., 2018).
This has implications for categorizing different levels of smartphone
usage (e.g., light vs. heavy usage) across users, and understanding
smartphone usage as an antecedent of smartphone addiction (Carbonell
et al., 2012; Lee, Ahn, Choi, & Choi, 2014).
Based on our systematic literature review of RQ1, factors related to
Fig. 2. Research on PSU.
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
6
emotional health constituted one of the primary reasons for smartphone
use, suggesting that an individuals mental wellbeing may affect the
degree to which they engage in smartphone usage. Smartphone use is
often motivated by an individuals desire to experience an emotional lift
and regulate or alleviate their mood through texting, email, and social
media (Kim, Seo, & David, 2015; Shin & Lee, 2015; Zhang, Chen, Zhao,
& Lee, 2014b; Zhang, Chen, Zhao, & Lee, 2014a). In particular, users
may be more likely to engage in smartphone usage in order to alleviate
negative moods such as depression and spend more time on communi-
cation activities via the smartphone to distract from their feelings (Elhai
et al., 2016; Kim, Seo, et al., 2015). Nevertheless, the results are
inconclusive as to whether depression constitutes an antecedent of
smartphone usage (Nahas et al., 2018). In addition, individuals experi-
encing loneliness may be more likely to engage in smartphone usage to
contact others in order to alleviate negative feelings and gain assurance
from friends, family, or their partner (Bian & Leung, 2015; Kim, 2017;
Lapointe et al., 2013). Lonely individuals may also become more
reluctant to engage in face-to-face interaction, preferring
smartphone-mediated communication instead (Kim, 2017).
Other studies have shown a positive association between an in-
dividuals desire to alleviate boredom and smartphone usage (Fullwood,
Quinn, Kaye, & Redding, 2017; Lapointe et al., 2013) and an in-
dividuals desire for instant gratication and smartphone use (Rozgon-
juk, Kattago, et al., 2018; aZhang et al., 2014; 2014b). These factors can
in turn impede professional performance both within an academic and
work setting, leading to distraction and decreased engagement with the
task at hand. Habituation has been found to be a key driver of smart-
phone usage here, as conditioned patterns of usage tend to be repeated
by individuals over time (Fullwood et al., 2017; Oulasvirta et al., 2012;
Soror et al., 2012). Deciencies in self-regulation and control can further
contribute to higher levels of mobile phone use (e.g., Lee, Lee, et al.,
2014; Soror et al., 2012), in turn affecting an individuals ability to alter
their habits of smartphone use.
The main factor explaining smartphone use within the social per-
formance category was the personality characteristics of individuals
(Horwood & Anglim, 2018; Lapointe et al., 2013; Panda & Jain, 2018);
in particular, studies employing the big ve personality traitsor
Five-Factor Model suggest a positive relationship between extraversion
and an individuals levels of smartphone usage, as extraverts are more
likely to use their smartphone to socialize (Bian & Leung, 2015;
Horwood & Anglim, 2018; Lapointe et al., 2013). Meanwhile,
negative relationships have been found between openness to experience
and PSU (Kita & Luria, 2018), as well as conscientiousness (Prasad et al.,
2018). Results on the impact of personality on smartphone use are not
consistent however, with some studies nding no relationship between
social extraversion and mobile phone usage behavior (Herrero, Torres,
Vivas, & Urue˜
na, 2019; Hong, Chiu, & Huang, 2012).
The most frequently identied moderator of smartphone use was age
(Anshari et al., 2016; Elhai & Contractor, 2018; Nahas et al., 2018; Van
Deursen et al., 2015). Studies have found that smartphone usage varied
depending on which generation (i.e., X, Y, or Z) the participants
belonged to. Most studies suggest that younger generations tend to use
smartphones more heavily than older generations; nevertheless, Nahas
et al. (2018) caution against this generalization by asserting that older
adults still engage in considerable levels of smartphone use, which
warrants further investigation. Gender was also identied as a moder-
ator of smartphone use in a number of papers (Anshari et al., 2016;
Davazdahemami et al., 2016; Elhai & Contractor, 2018; Lee, 2015; Lee,
Lee, et al., 2014; Van Deursen et al., 2015; Volkmer & Lermer, 2019).
Research on gender differences suggests that males have a more
instrumental view of smartphone use, whereas females use the smart-
phone more to facilitate social interaction (Anshari, Alas, & Sulaiman,
2019; Chotpitayasunondh & Douglas, 2016; Hong et al., 2012; Lee, Lee,
et al., 2014; Van Deursen et al., 2015). Women may also have a higher
chance of developing habitual smartphone behavior (Van Deursen et al.,
2015). Nevertheless, Salehan and Negahban (2013) found that the
moderating inuence of gender on smartphone use was inconclusive. In
addition, the papers showed that smartphone use can be moderated by
context (Fullwood et al., 2017) and time spent on the smartphone
(Panda & Jain, 2018).
5.2. What are the antecedents of PSU?
Table 1 presents our literature review results for the antecedents of
PSU. The most commonly referenced cut-off point for determining when
smartphone use becomes problematic was based on the work of Kwon,
Lee, et al. (2013) and their Smartphone Addiction Scale (Chotpitaya-
sunondh & Douglas, 2016; Contractor et al., 2017; Enez; Enez Darcin
et al., 2016; Hartanto & Yang, 2016). Kwon, Lee, et al. (2013) suggest a
cut-off score of >31 for males and >33 for females, with higher scores
predicting a higher risk of PSU. Meanwhile, a cut-off score of 160 has
been suggested for the original MPPUS-27 and an extrapolated cut-off
score of 59 for MPPUS-10 (Nahas et al., 2018). Rozgonjuk, Kattago,
et al. (2018) and Rozgonjuk, Levine, et al. (2018) also found that PSU
was positively associated with an individuals average minutes of screen
time over a week, which ranged from a minimum of 46.571 min over a
week, to a maximum of 608.143 min over a week. In contrast, they
found that average phone screen unlocks over a week was not associated
with PSU. However, overall, there is a lack of consistency in the cut-off
scores for determining when smartphone use becomes PSU, and there is
no unanimously agreed cut-off score to determine PSU (F.-C. Chang
et al., 2019; Nahas et al., 2018).
In terms of antecedents, control appears to be central to our under-
standing of how PSU emerges, with a number of studies suggesting that
deciencies in an individuals ability to self-regulate their smartphone
use can lead to problematic habitual behaviors over time. Indeed, PSU is
commonly dened as a poorly controlled occupation with the smart-
phone or usage behavior marked by a loss of control (Chang et al., 2019;
Jeong, Kim, Yum, & Hwang, 2016; Roberts, Yaya, & Manolis, 2014).
Individuals with poor self-control may be more likely to respond to
notications as soon as they appear, potentially creating a habitual
dependence on their smartphone (increased use frequency and uncon-
trolled frequent checking) (Berger, Wyss, & Knoch, 2018). Both social
and process related usage of the smartphone may also increase the risk
of PSU through the development of habitual behaviors (Van Deursen
et al., 2015). These habitual usage patterns can in turn lead to users
repeatedly engaging in content consumption through the smart-phone,
especially when patterns are positively reinforced (Kwon, So, Han, &
Oh, 2016; Lee, Lee, et al., 2014). Self-control can also mediate the
relationship between stress and PSU, as individuals with poor
self-control use their smartphone to deal with stress (Cho, Kim, & Park,
2017; Heo & Lee, 2018; Liu et al., 2018). However, in contrast, Kant-
hawongs, Jabutay, Upalanala, and Kanthawongs (2016) did not nd a
signicant relationship between a respondents self-regulation and PSU.
In addition, our systematic review suggests that PSU often arises in
tandem with emotional health issues such as depression, anxiety, anger,
and stress. Smartphone use can act as an avoidance strategy to distract
from negative emotional experiences, potentially leading to the devel-
opment of problematic usage trends. For instance, individuals experi-
encing anxiety may become dependent on their smartphone by regularly
communicating (calls and messaging) with others, or engage in sensa-
tion seeking through entertainment to deal with their negative
emotional state (Lopez-Fernandez, M¨
annikk¨
o, K¨
a¨
ari¨
ainen, Grifths, &
Kuss, 2018; Wang et al., 2018). The effect may also be cumulative as Jun
(2016) found a consistently increasing severity of PSU and depressive
symptoms among respondents over a three-year period, and noted that
the relationship between the two was bidirectional.
However, the link between both depression and anxiety sensitivity
with PSU may be mediated by an individuals mindfulness i.e., their
ability to consciously present and aware of what is being experienced in
the moment (Elhai, Levine, et al., 2018; Yang, Zhou, Liu, & Fan, 2019).
Meanwhile, positive associations were found between loneliness and
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
7
Table 1
Antecedents of PSU.
Category Subcategory References
Control Habit, checking smartphone Davazdahemami et al. (2016); Lee, U., Lee, J., Ko, M., Lee, C., Kim, Y., Yang, S.,
et al. (2014); Xie et al. (2018);
Jilisha et al. (2019)
Intolerance for uncertainty, need for compulsive control, dysfunctional
impulsivities, propensity to risk
Cho and Lee (2017); Rozgonjuk et al. (2019); Rho et al. (2019); b)Herrero,
Torres, et al. (2019) and Herrero, Urue˜
na, et al. (2019)
Self-control, parental control, self-efcacy in exercising control, regulation,
mood regulation, emotional lift, emotional gain, self-expressive benets,
self-directedness
Berger et al. (2018); Chen, Liu, et al. (2017) and Chen, Zhang, et al. (2017);
Chotpitayasunondh and Douglas (2016); Davazdahemami et al. (2016);
G¨
okçearslan et al. (2016); Jeong et al. (2016); Kwon et al. (2016); Lee et al.,
2014b; Shin and Lee (2015); Vaghe et al. (2017); Van Deursen et al. (2015);
Zhang et al. (2014b); Zhitomirsky-Geffet and Blau (2016); Lachmann et al.
(2019); Kim, Park, Lee, et al. (2019); Rho et al. (2019); Servidio (2019); Lee and
Kim (2018); Lee and Ogbolu (2018); Arpaci (2019); Ayar et al. (2017); Tang
et al. (2017); Lee, Chang, Lin, and Cheng (2017); Kanthawongs et al. (2016);
Roberts et al. (2015); Kim, Kim, Kim, et al. (2015); Chen, Liu, et al. (2017) and
Chen, Zhang, et al. (2017); Yang et al. (2016); Kim, Min, Min, Lee, Yoo. (2018);
Kim, M., Kim, H., Kim, K., Ju, S., Choi, J., and Yu, M. (2015); Yildiz (2017);
Mitchell and Hussain (2018); Pivetta et al. (2019)
Threats to smartphone users sense of agency, obsessed Marchant and ODonohoe (2019); Vaghe et al. (2017)
Use frequency (overuse, smartphone usage, time spent on mobile, process
and social oriented smartphone usage, daily usage time, average minutes of
screen time over a week), forward-looking mindset (Extent of consumption
inertia), use states (time availability), duration of ownership
Aljomaa et al. (2016); Elhai, Levine, OBrien and Armour (2018); Elhai,
Vasquez, Lustgarten, Levine, and Hall (2018); G¨
okçearslan et al. (2016); Jeong
et al. (2016); Khang et al. (2013); Kwon et al. (2016); Lee et al., 2014b;
Rozgonjuk, Kattago, et al. (2018) and Rozgonjuk, Levine, et al. (2018); Salehan
and Negahban (2013);
C. Shin and Dey (2013); Van Deursen et al. (2015); Zhitomirsky-Geffet and Blau
(2016); Kim, Park, Lee, et al. (2019); Tunc- Arnavut and Nuri (2018); Aktürk
et al. (2018); Arnavut & Nuri, 2018; Lee and Kim (2018); Nayak (2018);
Alhazmi et al. (2018); Ibrahim, Baharoon et al. (2018); Lopez- Fernandez et al.
(2018); Cha and Seo (2018); Xie et al. (2018) Guazzini et al. (2019); Mahmoodi
et al. (2018); Jilisha et al. (2019); Ayar et al. (2017); Bae (2017); Swar and
Hameed (2017); Liu et al. (2016); Alosaimi et al. (2016); Ding et al. (2016); Lee,
Seo, and Choi (2016); Cho and Lee (2016); Lee, Lee, and Lee (2016); Haug et al.
(2015); Choi et al. (2015); Olufadi (2015); Demirci et al. (2015); Karada˘
g et al.
(2015); Tossell et al. (2014); Lee,
Arnavut and Nuri (2018); Carbonell et al. (2012); Boumosleh and Jaalouk
(2017); Kim et al. (2016); Durak
(2018); Chou and Chou (2019); Tras
¸ and ¨
Oztemel (2019);S¨
ozbilir and Dursun
(2018)
Emotional health Addiction-proneness, insecure attachment style, reinforcement rewards Sapacz et al. (2016), Eichenberg et al. (2019)
Anger, hostility, need frustration Dey et al. (2019); Fırat et al. (2018); Li and Lin (2018); Gugliandolo et al.
(2019); Kim, Kim, Kim, et al. (2015); Lee, Sung, et al. (2018); Kim et al., 2015b
Anxiety (social anxiety, psychological-social, attachment anxiety),
somatization
Aljomaa et al. (2016); Elhai, Levine, et al. (2018) and Elhai, Vasquez, et al.
(2018); Elhai, Levine, et al. (2018) and Elhai, Vasquez, et al. (2018); Enez
Darcin et al. (2016); Han et al. (2017); Hong et al. (2012); Kim and Koh (2018);
Lapointe et al. (2013); Lu et al. (2011); Sapacz et al. (2016); Vaghe et al.
(2017); Van Deursen et al. (2015); Rho et al. (2019); Guazzini et al. (2019); You
et al. (2019); Dey et al. (2019); Fırat et al. (2018); Lopez-Fernandez et al.
(2018); Liu et al. (2019); Yuchang et al. (2017); Aker et al. (2017); De-Sola et al.
(2017); Lee (2015); Mok et al. (2014); Boumosleh and Jaalouk (2017); Ayar
et al. (2018)
Conscientiousness Lian and You (2017); Lee (2015); Pivetta et al. (2019); Mosalanejad et al.
(2019); Lee, J., Chung, Y., Kim, S., Kim, J., Shin, I., Yoon, J., et al. (2019)
Depression, less optimistic than others Chang et al. (2019); Elhai, Levine, et al. (2018) and Elhai, Vasquez, et al. (2018);
Elhai, Levine, et al. (2018) and Elhai, Vasquez, et al. (2018); Jun (2016); Kim,
Seo, et al. (2015); Lu et al. (2011); Rozgonjuk, Kattago, et al. (2018) and
Rozgonjuk, Levine, et al. (2018); Vaghe et al. (2017); Lu, Xu et al. (2019); Rho
et al. (2019); Dey et al. (2019); Lee and Ogbolu (2018); Aker et al. (2017);
De-Sola et al. (2017); Kim, Kim, Kim, et al. (2015); Kim et al., 2015b; Boumosleh
and Jaalouk (2017); Mitchell and Hussain (2018); Chiang et al. (2019); Kim,
J.-H., Seo, M., and David, P. (2015)
Empathy Lachmann et al. (2018)
Escapism motivation, dissociative experiences, alexithymia (lack of
emotional awareness)
Wang et al. (2015); De Pasquale et al. (2019); Mei et al. (2018); Gao, Zhang et al.
(2018); Hao et al. (2019)
Intensifying emotional investment in the human-smartphone assemblage (e.
g., identity as an ‘iPhone user)
Marchant and ODonohoe (2019)
Impatience Aren et al. (2018)
Loneliness Bian and Leung (2015); Enez Darcin et al. (2016); Jeong et al. (2016); Kim
(2018); Lapointe et al. (2013); Mahapatra (2019); Durak (2018); Taghizadeh
et al. (2019); Mosalanejad et al. (2019)
Negative parenting style, child neglect, psychological abuse, parental
phubbing, mothersabusive parenting, parental attachment, parent-child
communication, parental control
Lian et al. (2016); Sun et al. (2019), Xie et al. (2019); Jahng (2019); Zhang et al.
(2019); Emirtekin et al. (2019); Li and Hao (2019); Lee and Kim (2018); Kwak
et al. (2018); Gugliandolo et al. (2019); Lee and Lee (2017); Bae (2015); Lee,
Sung, et al. (2018); Kim, Jun, et al. (2018) and Kim, Min, et al. (2018)
Mindfulness Elhai, Levine, et al. (2018) and Elhai, Vasquez, et al. (2018); Volkmer and
Lermer (2019)
(continued on next page)
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
8
Table 1 (continued )
Category Subcategory References
Personality distortion, disturbances, narcissism Cho and Lee (2017); Xie et al. (2018); Pearson and Hussain (2017)
Stress, Post-traumatic Stress Disorder (PTSD), psychological distress,
distress tolerance
Beranuy et al. (2009); Chiu (2014); Contractor et al. (2017); Elhai, Levine, et al.
(2018) and Elhai, Vasquez, et al. (2018); Jeong et al. (2016); Lapointe et al.
(2013); Liu et al. (2018); Samaha and Hawi (2016); Volungis et al. (2019);
G¨
okçearslan et al. (2018); Lachmann et al. (2018); Lopez-Fernandez et al.
(2018); Cho et al. (2017); Kuang-Tsan and Fu-Yuan (2017); Long et al. (2016);
Mosalanejad et al. (2019)
Well-being, emotional stability, negative emotions, neuroticism, mental
state, emotional suppression, withdrawal
Volkmer and Lermer (2019); Kim, Park, Lee, et al. (2019), Rozgonjuk and Elhai
(2019); Lachmann et al. (2018); Xie et al. (2018); Gao et al. (2017); Roberts
et al. (2015); Lee (2015); Mok et al. (2014); Yildiz (2017); Aren et al. (2018);
Pivetta et al. (2019); Hana et al. (2019)
Physical health Individuals health status, physical status, drinking, smoking, cannabis use (Aljomaa et al., 2016; Boumosleh & Jaalouk, 2017; Choi et al., 2015; Chung
et al., 2018; De-Sola et al., 2017; Dey et al., 2019; Haug et al., 2015; H.-J.; Kim,
Min, Kim, & Min, 2017; J.; Kim, Park, Lee, et al., 2019)
Over-exercise, perception og being overweight (Ergun & Guzel, 2019; Lu et al., 2019)
Physical predisposition Konok et al. (2017)
Sleep quality (Aker et al., 2017; Chung et al., 2018; Mahmoodi et al., 2018; Randler et al.,
2016)
Preconditions Domestic violence, characteristics of the family (e.g., alcoholism) Kim, Jun, et al. (2018)
Characteristics of the father (e.g., educational level) and the mother (e.g.,
income)
(Beison & Rademacher, 2016; Long et al., 2016; Taghizadeh et al., 2019)
Professional
performance
Pastime Chen, Liu, et al., 2017; Zhang et al. (2014b)
Academic performance, perceived academic competence (Aktürk et al., 2018; Boumosleh & Jaalouk, 2017; Chang et al., 2019; Chung
et al., 2018; Coban & Gundogmus, 2019; Ibrahim et al., 2018; Samaha & Hawi,
2016; Tunc-Aksan & Akbay, 2019)
Attention Decit Hyperactivity Disorder (Dey et al., 2019; Kim, Park, Lee, et al., 2019; Seo et al., 2015; Kim, S.-G., Park,
J., Kim, H.-T., Pan, Z., Lee, Y., & McIntyre, R. S. (2019a))
Boredom (Elhai, Vasquez, Lustgarten, Levine, & Hall, 2018; Lapointe et al., 2013; Vaghe,
Lapointe, & Boudreau-Pinsonneault, 2017)
Cyberloang, trait procrastination (G¨
okçearslan et al., 2016; G¨
okçearslan et al., 2018; Rozgonjuk, Kattago, et al.,
2018)
Workload context, dependence on smartphone for work Kim, Park, Lee, et al. (2019); Li and Lin (2018); Boumosleh and Jaalouk (2017)
Social
performance
Personality (e.g., social extraversion)/self-traits, personality beliefs Hong et al. (2012); Horwood and Anglim (2018); Khang et al. (2013); Lapointe
et al. (2013); Vaghe et al. (2017); Zhitomirsky-Geffet and Blau (2016), Balta
et al. (2019); Lachmann et al. (2019); Li and Lin (2019); Direkt¨
or and Nuri
(2019); De Pasquale et al. (2019); Volungis et al. (2019); Cocorad˘
a et al. (2018);
Olivencia-Carri´
on, Ferri-García, Rueda, and L´
opez-Torrecillas (2018);
Olivencia-Carri´
on et al. (2018); Lachmann et al. (2018)
Conformity, need for approval, social environment pressure to use a
smartphone, need to belong
(Arpaci, 2019; Chen, Liu, et al., 2017; Vaghe, Lapointe, &
Boudreau-Pinsonneault, 2017; Wang et al., 2017; Zhang, Chen, Zhao, & Lee,
2014b; Zhitomirsky-Geffet & Blau, 2016)
Daily-life disturbance Aren et al. (2018)
Evolving identities Marchant and ODonohoe (2019)
FoMO, fear of rejection, abandonment, avoidant attachment, envy (Arpaci, 2019; Chotpitayasunondh & Douglas, 2016; Elhai et al., 2016; Kim &
Koh, 2018; Lapointe et al., 2013; Tras¸; ¨
Oztemel, 2019; Tunc-Aksan & Akbay,
2019; Vaghe, Lapointe, & Boudreau-Pinsonneault, 2017; Wang, P., Wang, X.,
Nie, J., Zeng, P., Liu, K., Wang, J., et al. (2019))
Isoloation, interference with daily life, shame (Arpaci et al., 2017; Cho & Lee, 2017; Shim, 2019a, 2019b)
Romantic relationships Kuang-Tsan and Fu-Yuan (2017)
Shyness, social liquidity (i.e., the ease with which one can establish
interpersonal relationships), introversion, self-esteem, social assurance,
social self-efcacy, interpersonal sensitivity, relational maladjustment,
adult attachment, attachment to friends
Bian and Leung (2015); Chiu (2014); Han et al. (2017); Hong et al. (2012);
Hong et al. (2019); Kim and Koh (2018); Kwon et al. (2016); Seo et al. (2015);
You et al. (2019); Kim and Jahng (2019); Fırat et al. (2018); Kwak et al. (2018);
Yuchang et al. (2017); Kim, Cho, and Kim (2017); Lee and Lee (2017); Roberts
et al. (2015); Lee, Sung, et al. (2018)
Social relationships, relationship with teacher, association with classmates,
social support, social involvement, phubbing, cyber friendship
Chen, Liu, et al., 2017; Marchant and ODonohoe (2019); Lu, Xu et al. (2019); a)
Herrero, Torres, et al. (2019) and Herrero, Urue˜
na, et al. (2019); Guazzini et al.
(2019); b)Herrero, Torres, et al. (2019) and Herrero, Urue˜
na, et al. (2019);
Guazzini et al. (2019); ayak (2018); Lee, Kim, et al. (2018), Ihm (2018); Wang
et al. (2017); Aker et al. (2017); Ayar et al. (2017); Lian and You (2017); Wang
et al. (2017); Kim, Jun, et al. (2018) and Kim, Min, et al. (2018); Aren et al.
(2018)
Socialization practices, need for immediate connection Kim (2018); Marchant and ODonohoe (2019); Jilisha et al. (2019); Coban and
Gundogmus (2019)
Victim of bullying (cyber and traditional) Li and Lin (2019); Lee, J.-I., Yen, C.-F., Hsiao, R. C., and Hu, H.-F. (2019)
Technology
features
Preoccupation, cognitive absorption, engagement Aljomaa et al. (2016); Barnes et al. (2019); Fan et al. (2017)
Breakdowns as material and immaterial components failed to interact as
expected
Marchant and ODonohoe (2019)
Mobile use motives, user needs, entertainment motivation (Khang et al., 2013; Wang et al., 2015; Zhitomirsky-Geffet & Blau, 2016)
Need for touch Elhai et al. (2016)
Ownership of a smartphone, unlimited mobile data (Chang et al., 2019; Chen & Pai, 2018; Ibrahim et al., 2018)
Personalization of components and capacities, role of smartphone in life (Arnavut & Nuri, 2018; Marchant & ODonohoe, 2019)
Satisfaction, perceived enjoyment, mobile ow, sensation seeking Chen, Liu, et al., 2017; Fan et al. (2017); Kwon et al. (2016); Shin and Lee
(2015); Zhang et al. (2014b); a)Herrero, Torres, et al. (2019) and Herrero,
Urue˜
na, et al. (2019); Wang, Lei, Wang, Nei et al. (2018)
Self-reported internet safety literacy Chang et al. (2019); Wang, J., Wang, P., Yang, X., Zhang, G., Wang, X., Zhao, F.,
et al. (2019)
(continued on next page)
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
9
PSU, where individuals experiencing loneliness tend to rely more on
their smartphone to connect with others, in turn leading to problematic
behaviors (Bian & Leung, 2015; Enez; Enez Darcin et al., 2016; Kim,
2018; Lapointe et al., 2013; Mahapatra, 2019). Finally, emotional pain
inicted from parents has also been shown to increase the likelihood of
PSU as the smartphone may be used as a means to avoid dealing with
past traumas such as child neglect and psychological abuse from parents
(Lian, You, Huang, & Yang, 2016; Sun et al., 2019; Xie, Chen, Zhu, & He,
2019).
Social relationships have been identied as another antecedent of
PSU by a number of studies. Individuals with large social networks may
rely on smartphones more than others to maintain relationships, making
them susceptible to PSU over time. Nevertheless, this link has not been
consistently demonstrated and some studies did not nd any signicant
connection between PSU and social connectedness or social relation-
ships (Chen, Liu, et al., 2017; Ihm, 2018; Sapacz, Rockman, & Clark,
2016). Similarly, Li and Lin (2019) found that a dependence on smart-
phones for communication has no inuence on PSU. However, a sig-
nicant relationship was found between self-esteem and PSU, suggesting
that PSU may be affected by an individuals condence in their own
worth or abilities (Hong et al., 2012; Kim & Koh, 2018). Individuals
suffering from low self-esteem may experience a strong need for social
assurance by contacting friends, partners, and family, using the smart-
phone as a means of multi-communication (Seo et al., 2015). This in turn
can make them more at risk of developing a reliance on their mobile
phone, potentially leading to problematic usage. Similarly, shyness may
moderate PSU through social anxiety for adolescents with a higher level
of relatedness need satisfaction from the smartphone (Hong et al.,
2012). Taken together, this suggests that self-esteem, shyness, and social
anxiety may be confounding antecedents of PSU (Kim, Cho, et al., 2017).
A number of studies have utilized the Five-Factor Model to investigate
the impact of personality on PSU as well. In general, they nd that in-
dividuals with the personality trait of extroversion may be susceptible to
PSU given their heightened motivation to engage in frequent commu-
nication using the smartphone to form and maintain relationships (Hong
et al., 2012; Li & Lin, 2019; Panda & Jain, 2018). Individuals with the
trait of neuroticism may also develop an excessive dependence on the
smartphone due to social anxiety and their need to constantly seek
reassurance from peers through smartphone mediated communication
(Horwood & Anglim, 2018; Li & Lin, 2019). The relationship between
other personality traits (e.g., agreeableness and conscientiousness) and
PSU was generally negative but inconclusive overall (Lee, 2015; Lee,
Chung, et al., 2019; Panda & Jain, 2018).
Technology features of smartphones have been identied as a further
antecedent of PSU. Features and characteristics related to the smart-
phone design such as ease of use, speed, portability, and accessibility
may contribute to PSU (Aljomaa, Qudah, Albursan, Bakhiet, & Abdul-
jabbar, 2016; Kwon et al., 2016; Lapointe et al., 2013; Shin & Lee, 2015;
Vaghe, Lapointe, & Boudreau-Pinsonneault, 2017). The availability of
applications such as social networking services (SNS) and instant
messaging (IM) was also identied as an antecedent of PSU (Chang et al.,
2019; Nahas et al., 2018; Zhitomirsky-Geffet & Blau, 2016), as was
smartphone/tablet gaming and other forms of entertainment (Chang
et al., 2019; Jeong et al., 2016; Nahas et al., 2018; Zhitomirsky-Geffet &
Blau, 2016). Some studies suggest that individuals who use their
smartphones as a pastime without any productive interest are more
likely susceptible to PSU (Chen, Liu, et al., 2017; aZhang et al., 2014).
However, more recent studies have offered contradictory evidence
around the impact of mobile gaming on PSU, suggesting that individuals
that regularly use mobile gaming are not necessarily at a higher risk of
addiction to their smartphone or problematic use (Lopez-Fernandez
et al., 2018). Mobile game addiction was also shown to have no signif-
icant impact on studentsacademic performance (Fabito et al., 2018).
Nevertheless, professional performance may itself constitute an
antecedent of PSU, and students with below average academic perfor-
mance were more at risk of PSU (Chang et al., 2019; Samaha & Hawi,
2016). Rozgonjuk, Kattago, et al. (2018) and Rozgonjuk, Levine, et al.
(2018), looking at a student cohort, found that the relationship between
procrastination and PSU may be mediated by social media use in lec-
tures, diverting attention from the lecturer.
5.3. Which demographic groups are most prone to PSU?
Our review suggests that a number of demographic groups may be
prone to PSU. Firstly, age has been identied as a predictor of PSU by
previous research (e.g., Billieux, 2012; De-Sola, Talledo, Rubio, & de
Fonseca, 2017). Most studies found that younger age predicted higher
levels of PSU (Aljomaa et al., 2016; Anshari et al., 2016; Elhai, Levine,
et al., 2018; Hong et al., 2019; Lu et al., 2011; Nahas et al., 2018;
Rozgonjuk, Kattago, et al., 2018; Rozgonjuk, Kattago, et al., 2018; Tanis
et al., 2015; Van Deursen et al., 2015), with smartphone users in the
adolescent age group most at risk (e.g., Chang et al., 2019; Kim & Koh,
2018). Yet other studies looked at differences between generations (Ahn
& Jung, 2016; Gentina, Tang, & Dancoine, 2018; Zhitomirsky-Geffet &
Blau, 2016). Findings suggest that, although usage remains the same,
Generation Y (the middle generation) has the highest addictive behavior
rate (Kim, 2017; Zhitomirsky-Geffet & Blau, 2016), followed by Gen-
eration Z, and Gener-ation X has the lowest addictive behavior level
(Zhitomirsky-Geffet & Blau, 2016). However, while older generations
may feel comfortable living without their smartphones, younger gen-
erations do not, and some of the young generation even agreed that they
would rather give up their breakfast than give up their phones (Anshari
et al., 2016). Differences may also be explained by digital nativity since
the young generation has been exposed to smartphones at a younger age,
increasing their reliance on the technology over time (Ahn & Jung,
2016; Anshari et al., 2016; Boumosleh & Jaalouk, 2017; Wang, Sigerson,
& Cheng, 2019; Wang, H.-Y., Sigerson, L., and Cheng, C. (2019)).
Despite this, digital natives may be better able to understand PSU as
actual users of smartphones, compared to non-digital natives who can
only recognize it as outside observers (Ahn & Jung, 2016).
However, results are inconclusive as some studies did not nd a
relationship between age and PSU (Barnes et al., 2019; Clayton et al.,
2015; Elhai & Contractor, 2018; Elhai, Levine, et al., 2018; Enez; Enez
Darcin et al., 2016; Lapointe et al., 2013; Panda & Jain, 2018; Roz-
gonjuk, Kattago, et al., 2018; Vaghe, Lapointe, &
Boudreau-Pinsonneault, 2017). Reasons for this may be attributed to the
fact that there was not much variety with respect to age in many of the
samples since respondents mostly were university students (e.g., Elhai,
Levine, et al., 2018; Vaghe et al., 2017). Additional explanations have
also been studied nding different predictors of PSU when other factors
Table 1 (continued )
Category Subcategory References
Smartphone use classes, type of smartphone use, various mobile apps, social
networking sites and instant messaging, gaming, internet addiction/app
addiction, task contex
Chang et al. (2019); Chotpitayasunondh and Douglas (2016); Davazdahemami
et al. (2016); Elhai and Contractor (2018; Jeong et al. (2016); Lee, S.-J., Lee, C.,
and Lee, C. (2016); Nahas et al. (2018); Zhitomirsky-Geffet and Blau (2016);
Kim, Park, Lee, et al. (2019); Mahmoodi et al. (2018)
Technological dimensions, smartphone characteristics, design (ease of use,
speed, useful, efcient, convenient, portable, easily accessible), materialism
Aljomaa et al. (2016); Kwon et al. (2016); Lapointe et al. (2013); Shin and Lee
(2015); Vaghe et al. (2017); Chen, Zhang, Xiang et al. (2019); Chen and Pai
(2018); Yoon Y.W. with Lee, Chang, et al. (2014); Long et al. (2019); Lee, Y.-K.,
Chang, C.-T., Lin, Y., & Cheng, Z.-H. (2014)
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
10
were included in addition to age (Beison & Rademacher, 2016; Elhai,
Levine, et al., 2018; Horwood & Anglim, 2018; Randler et al., 2016).
These ndings suggest that although determining high-risk groups based
on age offers an efcient approach, there may exist atypical cases which
can be neglected such as high-risk individuals from age groups deemed
to be low risk in general (Wang, Sigerson, et al., 2019).
Several of the studies in our review were also interested in gender
differences in PSU. Whereas the highest risk group of PSU is yet to be
discovered, most extant research seems to suggest that females are more
prone to PSU than men (Arpaci, 2019; Beranuy et al., 2009; Billieux,
2012; De-Sola et al., 2017; Elhai, Levine, et al., 2018; Harwood et al.,
2014; Hong et al., 2019; Horwood & Anglim, 2018; Jeong et al., 2016;
Kruger & Djerf, 2017; Lee, 2015; Lee, Chang, Lin, & Cheng, 2014;
Nayak, 2018; Randler et al., 2016; Seo et al., 2015; Wang et al., 2015).
Using different typications of smartphone users, researchers found that
females were highly over-represented in the ‘addictscategory (Lapointe
et al., 2013; Vaghe et al., 2017), and also among ‘fanaticsand highly
engaged users (Vaghe et al., 2017). However, yet other studies found
that men were more prone to PSU than women (Aljomaa et al., 2016;
Jilisha, Venkatachalam, Menon, & Olickal, 2019; Kwon et al., 2016; Lu
et al., 2011).
Gender-specic interventions have consequently been proposed as a
means of tackling PSU (Lee, S.-Y., Lee, D., Nam, C. R., Kim, D. Y., Park,
S., Kwon, J.-G., et al. (2018); Lee & Kim, 2018; Mahmoodi et al., 2018).
For instance, females may be more likely to use their smartphone while
driving than males (Anshari et al., 2016), while males may be more
likely to experience cyberbullying as a consequence of PSU (Qudah
et al., 2019). Nevertheless, results are inconclusive as our review
revealed that many studies were not able to identify any correlation
between gender and PSU (Barnes et al., 2019; Chang et al., 2019;
Clayton et al., 2015; Elhai & Contractor, 2018; Elhai, Levine, et al.,
2018; Hadlington, 2015; Kwon et al., 2016; Panda & Jain, 2018; Roz-
gonjuk, Kattago, et al., 2018; Rozgonjuk, Kattago, et al., 2018; Salehan
& Negahban, 2013; Van Deursen et al., 2015; Zhitomirsky-Geffet & Blau,
2016). As a nal note on gender differences, researchers have raised
concerns about genders moderating effects since gender differences
usually are related to culture (Chen, Liu, et al., 2017).
In terms of occupation, none of the reviewed studies concluded on
how different occupations engaged in PSU. Whereas most of the studies
used students as respondents, they were not included in the studies due
to their occupation, but rather based on their age and out of convenience
for the researchers. Some of the studies investigated how different
educational levels inuenced proneness to PSU. Whereas early research
showed that educated people were more likely to become engaged in
problematic technology use, additional studies found no specic
distinction between different demographic groups. Our review revealed
inconsistent ndings suggesting that both those with lower formal ed-
ucation (Kwon et al., 2016; Shin & Dey, 2013) and higher formal edu-
cation (Barnes et al., 2019) tend to exhibit greater control over their
smartphone use. A few studies did not identify any relationship between
these socio-demographic factors and PSU (Ayar et al., 2017; Enez; Enez
Darcin et al., 2016; Vaghe et al., 2017).
5.4. What are the consequences of PSU?
Table 2 presents an overview of the consequences of PSU derived
from our literature review. The overview shows that the researched
consequences of PSU vary from those which are less serious such as
negative emotions and being less resilient to distractions (Chen et al.,
2016; Hadlington, 2015; Vaghe et al., 2017). Whereas most studies are
conducted basing their ndings on self-perceptions of PSU, other studies
use more objective measures such as an app. The use of self-perceptions
is disputed since smartphone users may misjudge their use. While Park
(2019) found that smartphone users got angry when they were unable to
use their devices but still denied that they were addicted to the smart-
phone, Dharmadhikari, Harshe, and Bhide (2019) found the high
self-awareness among students of their PSU promising. Whereas the
reviewed articles study a variety of consequences, none of them have
focused on differences based on the type of smartphone (see Table 3).
PSU is found to affect emotional health more than physical health (e.
g., Panda & Jain, 2018). In addition, research on the impact on physical
health is more scarce than on emotional health. Studies investigating
emotional health problems have looked at a variety of consequences
such as loneliness, self-esteem, anger, and anxiety. Studies in this cate-
gory resemble those of other categories; namely that they are also
studied as antecedents and to some extent also as mediators and mod-
erators. The fragmented focus in these studies suggests that the conse-
quences of PSU are not clear.
To illustrate these claims, we will point out some ambiguities in the
literature. Anxiety is much researched as a consequence of PSU (e.g.,
Clayton et al., 2015; Elhai et al., 2016; Hartanto & Yang, 2016; Hawi &
Samaha, 2017; Park, 2019; Rozgonjuk, Kattago, et al., 2018; Sapacz
et al., 2016). Several studies found that smartphone separation, being in
the risk of becoming a problematic smartphone user, and PSU led to
increased anxiety (e.g., Clayton et al., 2015; Elhai, Rozgonjuk, Algh-
raibeh, et al., 2019; Hartanto & Yang, 2016). However, studying
smartphone usage and involvement, Harwood et al. (2014) found that
neither higher involvement nor usage were associated with higher levels
of anxiety. Furthermore, Nayak (2018) found that PSU had hardly any
effect on females while males experienced severe problems such as
feeling anxious, neglecting work, and losing control of themselves.
These ambiguities certainly call for more research, and in particular, to
establish the conditions under which PSU leads to increased anxiety
levels in smartphone users.
Likewise, studies focusing on depression as a result of PSU come to
contradictory conclusions. Whereas some studies conclude that PSU
does not lead to depression (e.g., Elhai et al., 2016; Rozgonjuk, Kattago,
et al., 2018), other studies come to the opposite conclusion claiming that
PSU is associated with higher levels of depression. Furthermore, some
studies raise concerns about these ndings, stating that the positive
correlation between smartphone addiction and depression is alarming
(Alhassan et al., 2018, p. 7). The ambiguous results indicate that PSU
can have several effects on smartphone users based on who they are (e.
g., gender, age, personality), their background (e.g., culture), and work
situation (e.g., managers, politicians). Furthermore, both mediators and
moderators could potentially explain ambigu ousndings. Harwood
et al. (2014) suggest that it is the nature of attachment a user has with
their smartphone (i.e., thinking about the phone and keeping it close for
constant checking) that is predictive of depression, rather than the
extent of use. Such an angle has merit since high smartphone use does
not necessarily lead to negative consequences.
The reviewed studies suggest that PSU can result in various types of
physical health problems. In particular, studies seem to agree that
impaired sleep quality is a prominent consequence of PSU, which sub-
sequently, can lead to other problems such as hypertension and affect
growth and emotional stability (Dharmadhikari et al., 2019; Haripriya,
Samuel, & Megha, 2019; Liu et al., 2017; Panda & Jain, 2018). Kim,
Kim, and Jee (2015) found that people indicative of PSU behavior were
less likely to walk for each day and that their body composition, such as
muscle mass and fat mass, was signicantly different from those who
were not indicative of PSU behavior. However, the relationship between
PSU and exercise is less researched suggesting further research efforts.
PSU can lead to a loss of control of important aspects of life. One
prominent consequence of PSU is insufcient time management and
frequent smartphone use (Hong et al., 2012; Lee, Lee, et al., 2014;
Rozgonjuk, Kattago, et al., 2018; Sapacz et al., 2016; Steelman & Soror,
2017; Vaghe et al., 2017). Findings further suggest that smartphone
overuse can lead to monetary overspending and thus nancial problems
(Chen et al., 2016; Soror et al., 2012). This overuse may be explained by
inated beliefs about the utility of the smartphone with regard to the
levels of perceived enjoyment and perceived usefulness (Bernroider
et al., 2014) and lower mindfulness scores (Volkmer & Lermer, 2019).
P.A. Busch and S. McCarthy
Computers in Human Behavior 114 (2021) 106414
11
Table 2
Consequences of PSU.
Category Impact(s) References
Control Constant checking, self-control, relapse Li and Lin (2019); Rho et al. (2019); Sok et al. (2019), Csibi et al. (2019); Baggio et al.
(2018); Nayak (2018); Bhavana and Vijayalakshmi (2019); Yang et al. (2019a, 2019b)
Decision-making ability Tang et al. (2017)
Impact on nancial situation Chen et al. (2016); Soror et al. (2012)
Impaired judgment, denial Bernroider et al. (2014); Park (2019); Volkmer and Lermer (2019); Bragazzi et al. (2019)
Instant rewards, dysfunctional impulsivities, self-distraction,
maladjustment
Anshari et al. (2019); Rho et al. (2019); Bragazzi et al. (2019); Grant et al. (2019); Jo
et al. (2018); Swar and Hameed (2017); Lee, Seo, et al. (2016)
Phantom experiences Kruger and Djerf (2017); Tanis et al. (2015); Li and Lin (2019); Mangot et al. (2018)
Time management Hong et al. (2012); Lee, Lee, et al. (2014); Rozgonjuk, Kattago, et al. (2018) and
Rozgonjuk, Levine, et al. (2018); Sapacz et al. (2016); Steelman and Soror (2017);
Vaghe et al. (2017)
Emotional health
problems
Anger, dysfunctional attitudes, venting Park (2019), Serin et al. (2019); Bragazzi et al. (2019); Elhai, Rozgonjuk, Yildirim,
Alghraibeh, and Alafnan (2019), Lee, Chang, Cheng, and Lin (2018)
Anxiety, academic anxiety Clayton et al. (2015); Elhai et al. (2016); Hartanto and Yang (2016); Harwood et al.
(2014); Hawi and Samaha (2017); Park (2019); Rozgonjuk, Kattago, et al. (2018) and
Rozgonjuk, Levine, et al. (2018); Sapacz et al. (2016), Anshari et al. (2019); aYang,
Asbury, et al. (2019) and Yang, Zhou, et al. (2019), 2019b); Rho et al. (2019); Yang,
Asbury, et al. (2019) and Yang, Zhou, et al. (2019); Kim, Park, Lee, et al. (2019); Lee,
Chung et al. (2019); Grant et al. (2019); Elhai, Rozgonjuk, Alghraibeh, and Yang (2019);
Nayak (2018); Lee, Chang, and Cheng (2018); Selçuk and Ayhan (2019)
Depression, suicidal ideation, thought problems Chen et al. (2016); Elhai et al. (2016); Harwood et al. (2014); Lu et al. (2011);
Rozgonjuk, Kattago, et al. (2018) and Rozgonjuk, Levine, et al. (2018); Wang, P., Liu, S.,
Zhao, M., Yang, X., Zhang, G., Chu, X., et al. (2019c); Wang, Liu et al. (2019); Rho et al.
(2019); Yang, Asbury, et al. (2019) <