Digitally Connected, Socially Disconnected: The Effects of Relying on Technology Rather Than Other People

Article (PDF Available)inComputers in Human Behavior 76 · July 2017with 1,044 Reads
DOI: 10.1016/j.chb.2017.07.001
In less than a decade, smartphones have transformed how, when, and where people access information. We propose that turning to technology for information may lead individuals to miss out on opportunities to cultivate feelings of social connection. Testing this hypothesis, we asked participants to find an unfamiliar building and randomly assigned them to solve this everyday problem either with or without their smartphones. Compared to those who could not rely on technology, participants who used their smartphones found the building more easily but ended up feeling less socially connected. Although having access to smartphones improved participants’ mood by making their task easier, this beneficial effect was diminished by the costs to social connection. Our findings provide the first experimental evidence that the benefits of pervasive connectivity may be undercut when technology supplants social interactions.
Figures - uploaded by Kostadin Kushlev
Author content
All content in this area was uploaded by Kostadin Kushlev
Full length article
Digitally connected, socially disconnected: The effects of relying on
technology rather than other people
Kostadin Kushlev
, Jason D.E. Proulx, Elizabeth W. Dunn
Department of Psychology, University of British Columbia, 2136 West Mall, V6T 1Z4, Vancouver, BC, Canada
article info
Article history:
Received 10 June 2016
Received in revised form
20 April 2017
Accepted 1 July 2017
Available online 3 July 2017
Human-computer interaction
Social behavior
Ubiquitous computing
Pervasive connectivity
In less than a decade, smartphones have transformed how, when, and where people access information.
We propose that turning to technology for information may lead individuals to miss out on opportunities
to cultivate feelings of social connection. Testing this hypothesis, we asked participants to nd an un-
familiar building and randomly assigned them to solve this everyday problem either with or without
their smartphones. Compared to those who could not rely on technology, participants who used their
smartphones found the building more easily but ended up feeling less socially connected. Although
having access to smartphones improved participantsmood by making their task easier, this benecial
effect was diminished by the costs to social connection. Our ndings provide the rst experimental
evidence that the benets of pervasive connectivity may be undercut when technology supplants social
©2017 Elsevier Ltd. All rights reserved.
1. Introduction
Smartphones are the rst thing many Americans report reach-
ing for when they wake up in the morningdbeating out coffee or
even their own signicant others, according to a recent survey
(Braun Research Inc, 2015). In the same nationally representative
survey, almost half of Americans admitted that they could not go a
day without their smartphones. Smartphones provide unprece-
dented access to information, enabling individuals to harness the
full resources of the Internet from anywhere. But could this
omnipresent access to information carry unforeseen consequences
for the fabric of social life?
Smartphones represent a new branch in the evolution of infor-
mation technology because of two dening characteristics. First,
unlike many other computing devices, smartphones are portable
and constantly accessible, pervading people's daily lives (Pew
Research Center, 2015). Second, unlike other portable sources of
informationdfrom simple cell phones to newspapers and
mapsdsmartphones provide connectivity to limitless information
on-demand, enabling people to solve a wide variety of everyday
problems. It is this pervasive connectivity that theoretically sets
smartphones apart from any preceding information tool. There is a
great deal of public debate (e.g., Schwartz, 2015; Turkle, 2011, 2015),
but a dearth of rigorous experimental research on the effects of this
emerging technological revolution for social and emotional well-
Most existing research relevant to the effects of phones on well-
being has focused on apps that are explicitly designed to enable
people to connect with others through messaging and social media
(e.g., Guillory, Hancock, Woodruff, &Keilman, 2015; Hall &Baym,
2012; Pielot, Church, &de Oliveira, 2014; Pollet, Roberts, &
Dunbar, 2011; Valkenburg &Peter, 2007). In contrast, very little
research has explored whether the use of smartphones for
information-seeking (e.g., search engines, Google Maps, Apple
Maps) might also inuence social outcomes and emotional well-
being. And the few existing studies rely on correlational methods,
which cannot establish causality (e.g., Kushlev &Proulx, 2016). In
the present research, we used experimental methods to investigate
how relying on smartphones for information would shape social
and emotional well-being. We propose that by enabling people to
rely on technology for information anywhere, smartphones may
obviate the need for people to rely on each other, thereby leading
them to miss out on opportunities to foster a sense of
*Corresponding author. Present address: Department of Psychology, University
of Virginia, 102 Gilmer Hall, PO Box 40040 0, Charlottesville, VA, USA.
E-mail addresses: (K. Kushlev),
(J.D.E. Proulx), (E.W. Dunn).
Contents lists available at ScienceDirect
Computers in Human Behavior
journal homepage:
0747-5632/©2017 Elsevier Ltd. All rights reserved.
Computers in Human Behavior 76 (2017) 68e74
connectedness. We explored this idea by asking participants to
solve an everyday problem: nding an unfamiliar building either
with or without their smartphones. We chose this particular task
because people rank nding directions amongst the most indis-
pensable functions of smartphones (Pew Research Center, 2015).
2. Theoretical background and hypotheses
According to the principle of least effort (Ferrero, 1894), organ-
isms tend to seek the easiest way to achieve a given outcome.
Applying this principle to information-seeking behavior, Mann
(1990) argued that people would tend to rely on the most conve-
nient available method of obtaining information. And due to their
portability and connectivity to the Internet, smartphones are
nothing if not convenient. Indeed, conveniencewas the most
frequently mentioned word amongst U.S. poll respondents asked to
describe what they like about their phones (Pew Research Center,
2012). According to the principle of least effort, then, smartphone
users should be less likely to rely on other methods of seeking in-
formation if they can easily obtain information from their phones.
After all, why turn to a friendly stranger for directions to a caf
when Google Maps is just a nger swipe away? Thus, we hypoth-
esize that when people have access to their phones, they will be less
likely to rely on other human beings to obtain information, such as
getting directions while looking for an unfamiliar address (Hy-
pothesis 1).
To the extent that individuals rely on technology rather than
other people for information, they may miss out on opportunities
to satisfy fundamental human needs. Although different motivation
theories differ in their specication of basic human needs, social
needs feature in virtually all existing models of human motivation
(Baumeister &Leary, 1995; Kenrick, Griskevicius, Neuberg, &
Schaller, 2010; Maslow, 1943; Ryan &Deci, 2000; Ryff, 1989). In
his classic pyramid of needs, for example, Abraham Maslow (1943)
theorized that social needs for connection and belonging are
fundamental for human ourishingdsecondary only to basic sur-
vival needs for water, food, and safety. In a recent reformulation of
Maslow's classic pyramid, evolutionary theorists have kept social
needs at this central place within the hierarchy of human needs
(Kenrick et al., 2010). Furthermore, according to self-determination
theory (Ryan &Deci, 2000), a sense of relatedness to others is one of
only three universal psychological needs that are essential for hu-
man ourishing. Similarly, Baumeister and Leary (1995) integrated
decades of psychological research to place the need to belong
amongst the most fundamental human motivations.
Past research has primarily focused on social interactions with
strong ties (e.g., family, friends) in satisfying people's need for social
connectedness (e.g., Mehl, Vazire, Holleran, &Clark, 2010; Reis,
Sheldon, Gable, Roscoe, &Ryan, 2000; Vittengl &Holt, 1998;
Wheeler, Reis, &Nezlek, 1983). Recent research suggests, howev-
er, that even seemingly trivial interactions with strangers and ac-
quaintances can play a surprisingly important role in shaping
feelings of social connection (Sandstrom &Dunn, 2014a, 2014b).
For example, participants who were randomly assigned to have a
brief conversation with the barista at Starbucks left the coffee shop
with a greater sense of belonging compared to participants who
were assigned to conduct the same transaction as efciently as
possible (Sandstrom &Dunn, 2014a). Thus, if people who have
access to their phones are less likely to talk to other people while
searching for a building (Hypothesis 1), we hypothesize that they
will also experience a lower sense of social connectedness than
those who do not have access to their phones (Hypothesis 2).
This detrimental effect of phones on social connectedness, in
turn, should have downstream negative consequences for mood. A
great deal of research suggests that feeling socially connected is
important for emotional well-being (Cacioppo et al., 2006;
Lyubomirsky &Boehm, 2010; Myers &Diener, 1995; Reis et al.,
2000; Sandstrom &Dunn, 2014a). To the extent that phones lead
individuals to miss out on opportunities to cultivate a sense of
connection (Hypothesis 2), we hypothesize that people relying on
their phones for directions will experience lower emotional well-
beingdpotentially undercutting the emotional benets of conve-
nience that technology affords (Hypothesis 3).
3. Study 1
3.1. Overview
In Study 1, we examined the consequences of relying on phones
when looking for a building. We randomly assigned participants to
a condition in which they could rely on their phones (phone con-
dition) or a condition in which they could not rely on their phones
(phoneless condition).
3.2. Pre-registered hypotheses
We preregistered three central hypotheses on the Open Science
Framework (see
First, we expected that participants would be less likely to rely on
other people if they could use their phones (Hypothesis 1). Second,
we predicted that participants would feel less socially connected in
the phone condition than the phoneless condition (Hypothesis 2).
Finally, given the central importance of feeling sociallyconnected for
emotional well-being, we also predicted that participants in the
phone condition would report less positive mood compared to those
in the phoneless condition (Hypothesis 3).
3.3. Study 1 materials and methods
3.3.1. Participants and power
We pre-registered power analyses on the Open Science Frame-
work: Based on the most
closely related previous research (Sandstrom &Dunn, 2014a), we
expected a large effect size of d¼.80, necessitating a minimum
sample size of 84 participants for 95% power. We expected that
some participants might not be able to follow instructions due to
common phone issues (e.g., poor reception). To ensure a minimum
power of 95% given our assumed effect size, therefore, we recruited
approximately 15% more participants (N ¼98). Six participants who
were instructed to use their phones failed to do so because of various
issues (e.g., no Internet access), and were therefore excluded, leav-
ing a nal sample of 92 participants (Median age ¼19.50, Age Range:
17e42; 84% women).
All participants were University of British Columbia students
who completed the study for partial course credit. The study was
advertised on the UBC Psychology Department's Participant Pool
website and could be taken by any student enrolled in a psychology
class regardless of major. After arriving at the lab, all participants
provided informed consent. The informed consent and all study
procedures were approved by UBC's Ethics Board.
3.3.2. Procedure
Participants came to our lab individually. We asked if they knew
where various buildings on campus were located and sent them to
nd a building that was unfamiliar to them. We randomly assigned
participants to complete this task by using their smartphones
(phone condition) or without using their smartphones (phoneless
condition). Aside from this constraint, participants in both condi-
tions were free to use any strategies they wished to nd the
building, such as asking others for directions, using campus signs/
K. Kushlev et al. / Computers in Human Behavior 76 (2017) 68e74 69
maps, or simply wandering around. All participants were instructed
to leave their belongings in the lab, but participants in the phone
condition were told to keep their smartphones. All buildings were a
10e12 min walk from our lab; participants were instructed to re-
turn the lab if they did not nd the building within 30 min. To keep
track of time, participants were given basic wristwatches. When
participants found their building or returned to the lab, they
completed a survey containing our dependent measures. Seven
participants (2 in the phone condition and 5 in the phoneless con-
dition) did not nd their building within the allotted time; they
were included in the analyses.
3.3.3. Measures Convenience. We assumed that smartphones would be
useful in locating the building. To test this assumption, we asked
participants to report how difcult it was for them to locate the
building (from 0enot at all to 6every much). Pre-registered measures. To test Hypothesis 1, we asked
participants to indicate how many people they talked to in person
while looking for the building (from 0eNone to 4e4 or more). To
test Hypothesis 2, we assessed social connectedness with eight
items from the Social Connectedness Scale-Revised (Lee, Draper, &
Lee, 2001); these items were selected to assess people's general
sense of connectedness to other people. To test Hypothesis 3, we
measured mood using Schimmack and Grob's (2000) six-item
affect valence subscale, which captures the extent to which in-
dividuals are feeling pleasant vs. unpleasant (see Table 1 for details
and reliability of all measures). Exploratory measures. Our survey also included explor-
atory measures of tense and energetic arousal (Schimmack &Grob,
2000), trust (using items from the General Social Survey), sense of
community (using the Brief Sense of Community Scale; Peterson,
Speer, &McMillan, 2008), feelings of agency and communion
(adapted from Abele &Wojciszke, 2007), and self-sufciency (see
Table 1 for details of all measures). Finally, we also included a
measure of prosocial behavior by dropping pens on the ground
while participants were completing the survey and recording
whether they offered help (van Baaren, Holland, Kawakami, &van
Knippenberg, 2004).
3.4. Results
3.4.1. Convenience
Consistent with our assumption that phones would make the
task easier, participants who relied on their phones found it less
difcult to locate the buildings, t(89) ¼4.46, p<.001.
All means,
standard deviations, and effect sizes for these and subsequent an-
alyses are provided in Table 2.
3.4.2. Pre-registered hypotheses
Consistent with Hypothesis 1, participants who used their
phones talked to fewer people to obtain directions than partici-
pants who could not depend on their phones, t(65.97) ¼9.32,
Indeed, in the phone condition, over 80% of participants
searched for the building without ever talking to anyone else,
whereas less than 10% did so in the phoneless condition. Conrming
Hypothesis 2, we found that people who relied on their phones felt
less socially connected than those who left their phones in the lab,
t(89) ¼2.10, p¼.04. Contrary to Hypothesis 3, we found no sig-
nicant difference in mood (i.e., affect valence) between partici-
pants who relied on their phones and those who did not,
t(90) ¼.47, p¼.64 (see Table 2).
3.4.3. Countervailing effects of technology on mood
Given the well-established role of social connectedness in
emotional well-being (e.g., Baumeister &Leary,1995), it is puzzling
that we found a condition effect on connectedness but not mood. To
illuminate this surprising nding, we next explored whether the
negative downstream consequences of lost social connection for
mood might have been offset by positive downstream conse-
quences of convenience. To test this mediational hypothesis, we
used the PROCESS macro on SPSS21, which uses bootstrapping for
constructing condence intervals for the effects and thus provides
less biased tests of statistical signicance (Hayes, 2013). In a
mediational model using bootstrapping with 50,000 samples, we
entered social connectedness and task difculty as simultaneous
mediators of the condition effect on mood. We found that relying
on phones had both a negative effect on mood through lower social
connectedness and a positive effect on mood through reduced task
difculty (Fig. 1).
3.4.4. Exploratory measures
Consistent with the negative effect of phone use on social
connectedness, we found that participants in the phone condition
felt less trusting toward others than participants in the phoneless
condition, t(89) ¼2.25, p¼.03 (Table 2). We found no other
signicant main effects of condition on our exploratory measures,
p's >.220.
4. Study 2
4.1. Overview &pre-registered hypotheses
The results of Study 1 suggest that when people rely on tech-
nology rather than each other to solve an everyday problem, they
may miss out on opportunities to cultivate a sense of social
connection. In Study 2, we conducted a direct replication with a
larger sample. We expected to replicate the signicant effects of
condition on social interactions, social connectedness, and trust
that we observed in Study 1, and we preregistered these hypotheses
on OSF (see
4.2. Study 2 materials and method
4.2.1. Participants and power
Based on the effect size for social connectedness in Study 1
(d¼.44), we planned to recruit 220 participants for 90% power.
These power analyses were pre-registered on the Open Science
Framework (OSF) at
Because data collection proceeded more slowly than expected and
the study was conducted in the fall when weather conditions in
Canada get progressively less amenable to conducting outdoor
studies, we added a termination date of October 16, 2015, with the
goal of achieving at least 80% power; this corresponded to a min-
imum sample of 166 participants, which we registered on OSF
( By October 16, we had
recruited a total of 189 participants. We again excluded participants
who failed to comply with the experimental instructions; three
participants were instructed to use their phones, but failed to do so,
while four participants refused to leave their phones in the lab
when asked, leaving a nal sample of 182 participants (Median
To test differences between conditions, we conducted between-subjects t-tests
in SPSS21.
Degrees of freedom and the t-test value for some tests were adjusted because
the Levene's test for equality of variance indicated unequal variances between
experimental groups.
K. Kushlev et al. / Computers in Human Behavior 76 (2017) 68e7470
age ¼20, Range: 17e29; 68% women).
4.2.2. Procedure and measures
The recruitment procedures and experimental design were
identical to Study 1: Participants were asked to nd an unfamiliar
building on campus either by using their phones (phone condition)
or without using their phones (phoneless condition). All of those
who relied on their phones located their assigned buildings,
whereas eleven participants (12%) who left their phones in the lab
failed to locate the building. These participants were included in
the analyses. At the end of the study, participants completed the
same questionnaire as in Study 1 (see Table 1 for details).
Additionally, we measured the interest/enjoyment people experi-
enced when searching for the building (Ryan, 1982) and assessed
the usefulness of phones more objectively by recording the time it
took participants to nd the building. To simplify the procedure, we
eliminated our exploratory measure of prosocial behavior.
4.3. Results
4.3.1. Convenience
Consistent with Study 1, participants in the phone (vs. phoneless)
condition found searching for the buildings to be less difcult,
t(155.46) ¼8.19, p<.001 (see Table 2 for descriptives, effect sizes,
Table 1
Measure Cronbach's
Study 1; Study 2
Operationalization Response Options
Social Connectedness .86; .84 8 I felt close to people.
I felt distant from people. (R)
I didn't feel related to people. (R)
I felt like an outsider. (R)
I saw myself as a loner. (R)
I was in tune with the world.
I saw people as friendly and approachable.
I felt disconnected from the world around me. (R)
1eStrongly disagree
7eStrongly agree
Sense of Community .94; .94 6 I can get what I need from the UBC community.
The UBC community helps me fulll my needs.
I have a good bond with others in the UBC community.
I belong to the UBC community.
I feel like a member of the UBC community.
I feel connected to the UBC community.
1eStrongly disagree
7eStrongly agree
Trust .60; .71 3 Generally speaking, would you say that most people can be trusted, or that you cannot
be too careful in dealing with people?
1eYou cannot be too
careful in dealing with
2eMost people can be
How much do you trust strangers? 1eCannot be trusted at all
5eCan be trusted a lot
If you lost a wallet or purse that contained two hundred dollars, how likely is it to be
returned with the money in it, if it was found by a stranger?
1enot at all likely
2esomewhat likely
3every likely
Affect: Valence (Mood) .92; .87 6 (pleasant þgood þpositive) e(unpleasant þbad þnegative) 0enot at all; 6every much
Affect: Tense Arousal .90; .84 6 (calm þrelaxed þat rest) e(tense þjittery þrestless) 0enot at all; 6every much
Affect: Energetic Arousal .87; .80 6 (awake þwakeful þalert) e(tired þdrowsy þsleepy) 0enot at all; 6every much
Self-sufciency .84; .58 2 Self-sufcient; independent 0enot at all; 6every much
Agency .79; .75 6 active; competent; self-condent; dynamic; assertive; efcient 0enot at all; 6every much
Communion .79; .78 6 friendly; empathetic; likable; understanding; helpful; reliable 0enot at all; 6every much
Notes. We computed an overall trust composite by rst standardizing each of the three items and then computing the mean of the resulting z-scores.
Table 2
Effects of relying on smartphones for information.
Study 1 Study 2 Meta-Analysis
M (SD)
M (SD)
Cohen's d[95% CI] Phone
M (SD)
M (SD)
Cohen's d[95% CI] Cohen's d[95% CI]
Difculty 1.54 (1.63) 3.18 (1.86) e.95 [e1.30; .59]
1.15 (1.31) 3.13 (1.86) 1.24 [e1.47; 1.01]
1.14 [e1.39; .88]
# Social Interactions .29 (.68) 2.36 (1.30) 2.00 [e2.22; 1.78]
.30 (.82) 2.29 (1.25) 1.90 [e2.05; 1.75]
1.93 [e2.22; 1.64]
Connectedness 4.55 (1.18) 5.03 (1.00) e.44 [e.66; .22]
4.62 (.95) 5.02 (.98) e.42 [e.56; .28]
e.43 [e.67; .19]
Mood 3.03 (2.26) 2.80 (2.42) .10 [e.38; .57] 3.33 (1.53) 2.74 (1.99) .33 [.08; .59]
.25 [.01; .49]
Tense Arousal 1.18 (2.60) e.67 (2.77) e.19 [.73; .35] 2.01 (2.07) e.75 (2.40) e.57 [.89; .24]
e.44 [e.68; .20]
Energetic Arousal 2.57 (2.05) 2.83 (2.18) e.13 [.56; .30] 2.19 (2.02) 2.64 (2.06) e.22 [e.52; .07] e.19 [e.43; .05]
Trust e.17 (.78) .17 (.66) e.48 [e.63; .33]
.002 (.72) e.003 (.87) .01 [e.11; .12] e.15 [e.39; .09]
Sense of Community 4.63 (1.20) 4.96 (1.38) e.26 [e.52; .01] 4.81 (1.16) 4.84 (1.86) e.02 [e.19; .15] e.10 [e.34 .14]
Agency 3.83 (1.06) 3.93 (.89) e.10 [e.30; .10] 3.78 (.88) 3.90 (.95) e.13 [e.26; 0] e.12 [e.36; .12]
Communion 3.26 (.91) 3.46 (1.00) e.21 [e.40; .01] 3.36 (1.04) 3.34 (.82) .02 [e.12; .15] e.06 [e.30; .18]
Self-Sufciency 4.64 (1.05) 4.39 (1.28) .22 [e.02; .45] 4.56 (.92) 4.17 (1.24) .36 [.21; .52]
.31 [.08; .55]
Prosocial Behavior 48.9% 53.5% NA NA NA NA NA
Interest/Enjoyment NA NA NA 4.77 (1.16) 4.90 (1.20) e.11[e.28; .06] NA
Notes. The scores for affect vary from 6toþ6 because composite scores are calculated by subtracting the reverse items of each scale from the other items (e.g., feeling bad is
subtracted from feeling good). Meta-analytic effects were calculated with a xed effect method using Cumming's ESCI software for conducting meta-analyses of Cohen's
d based on two independent groups (Cumming, 2013). Percentages for prosocial behavior indicate percent of people who offered help with picking up the pens.
indicates a
signicant effect at two-tailed
¼.05 or lower. The raw data for both studies can be obtained at
K. Kushlev et al. / Computers in Human Behavior 76 (2017) 68e74 71
and condence intervals). Participants relying on their phones also
located the buildings faster (M¼12.28 min, SD ¼4.32) than those
who left their phones at the lab (M¼16.44 min, SD ¼7.31),
t(122.29) ¼4.44, p<.001, Cohen's d¼.71, 95%CI [e1.01; .39].
4.3.2. Pre-registered hypotheses
We expected to replicate the signicant effects of condition on
social interactions, social connectedness, and trust that we
observed in Study 1. As in Study 1, participants who used their
phones talked to fewer people to obtain directions than partici-
pants who could not depend on their phones, t(153.94) ¼12.67,
p<.001. Again, over 80% of people in the phone condition searched
for the building without ever talking to anybody, while less than
10% did so in the phoneless condition. Replicating our central
nding from Study 1, we found that people in the phone condition
felt less socially connected than those in the phoneless condition,
t(180) ¼2.80, p¼.006. There was, however, no effect of condition
on trust, t(172.67) ¼.05, p¼.96 (Table 2).
4.3.3. Countervailing effects of technology on mood
Participants in the phone condition reported more positive mood
compared to those in the phoneless condition, t(167.14) ¼2.21,
p¼.03 (Table 2). Still, using mediational analyses through PROCESS
for SPSS21 (Hayes, 2013), we explored whether the social costs
associated with relying on phones may have limited the emotional
benets that phones conferred. Replicating the ndings of Study 1,
we found that relying on phones had both a negative effect on mood
through lower social connectedness and a positive effect on mood
through reduced difculty (see Fig. 2). After accounting for the role
of social connectedness and difculty, condition did not signi-
cantly predict mood.
4.3.4. Exploratory measures
We found no signicant effects of condition on our exploratory
measures of agency, communion, sense of community, energetic
arousal, or interest/enjoyment p's >.14 (see Table 2). Compared to
those in the phoneless condition, participants who could rely on
their phones felt less tense arousal, t(180) ¼3.80, p<.001, and
more self-sufcient, t(163.71) ¼2.43, p¼.02.
4.3.5. Meta-analysis of studies 1 and 2
Next, we conducted a meta-analysis across Studies 1 and 2 (see
Table 2). This meta-analysis conrmed that participants in the
phone (vs. phoneless) condition were less likely to interact with
other people and ended up feeling less socially connected. At the
same time, relying on phones made it much easier to nd the
building, resulting in a small net positive effect on mood across the
two studies. Out of our exploratory measures, we found two sig-
nicant meta-analytic effects of condition, whereby people who
relied on their phones felt less tense and more self-sufcient.
5. General discussion
In an initial study and a larger direct replication, we found the
rst experimental evidence that relying on smartphones for infor-
mation may compromise opportunities for social connection.
Compared to people who were not allowed to use their phones to
nd a building, those who could rely on their phones talked to
fewer people and ended up feeling less socially connected. Of
course, phones also conferred an important benet by reducing the
difculty of this everyday task, with positive downstream conse-
quences for participantsoverall mood. This benecial effect on
mood, however, was partially undercut by the insidious effect of
phone use on social connection.
5.1. Implications
Social connection has earned a prominent place in the pantheon
of essential psychological needs (e.g., Baumeister &Leary, 1995;
Ryan &Deci, 2000), but theory and research have traditionally
focused on the role of close relationships in promoting this basic
human need (e.g., Nelson, Kushlev, &Lyubomirsky, 2013; Reis et al.,
2000; Ryff, 1989). Our ndings add to a growing body of research
documenting the surprising power of casual interactions with ac-
quaintances and strangers to make people feel more socially con-
nected (Sandstrom &Dunn, 2014a &2014b; Wesselmann, Cardoso,
Slater, &Williams, 2012). Interestingly, both theory (e.g., media
richness theory, media naturalness theory; Daft &Lengel, 1986;
Kock, 2004) and research (e.g., Kross et al., 2013; Park et al., 2016;
Verduyn, Ybarra, R
esibois, Jonides, &Kross, 2017) suggest that
computer-mediated communication does not provide an
Fig. 1. Indirect effects of relying on phones on emotional well-being through social
connectedness and difculty of nding the building (Study 1).Notes. All b's repre-
sent unstandardized regression coefcients obtained through bootstrapping using
50,000 resamples (Hayes, 2013). The range in brackets represents the 95% condence
interval of the indirect effect.
Fig. 2. Indirect effects of relying on phones on emotional well-being through so-
cial connectedness and difculty of nding the building (Study 2).Notes. All b's
represent unstandardized regression coefcients obtained through bootstrapping us-
ing 50,000 resamples (Hayes, 2013). The range in brackets represents the 95% con-
dence interval of the indirect effect.
K. Kushlev et al. / Computers in Human Behavior 76 (2017) 68e7472
equivalent substitute for real, face-to-face social interactions. Going
beyond past research, our ndings demonstrate that when tech-
nology supplants even trivial face-to-face social interactions, peo-
ple can miss out on opportunities to satisfy their basic need for
This nding is important given that new technologies are
increasingly poised to replace casual social interactions. In 2016, for
example, Starbucks added a feature to its app that enables cus-
tomers to place orders through their smartphones, rather than
waiting in line and placing an order with the barista. While this
new method of obtaining one's morning dose of caffeine is more
convenient and efcient, it potentially obviates the need to talk to
any actual people. Our ndings suggest that this might lead cus-
tomers to miss out on an easy opportunity to cultivate a sense of
connection while getting their coffee. By appreciating the value of
social connection, organizations and designers can consider how to
deploy new technologies in ways that are sensitive to psychological
needs. For example, although Starbucks could maximize conve-
nience by placing mobile orders near the entrancedallowing cus-
tomers to grab their drinks and go without talking to anyonedour
research points to the value of placing the drinks near a friendly
barista who could briey greet customers, thereby balancing con-
venience with connection.
More broadly, our research underscores the importance of
considering what might be lost in the immediate (i.e., non-digital)
social environment when people engage in digital activities. Recent
research has shed light on the psychological consequences of
engaging in specic activities from Facebook to email (e.g., Kross
et al., 2013; Kushlev &Dunn, 2015; Park et al., 2016), but almost
no research has examined how these activities affect and are
affected by the immediate social context. Checking Facebook dur-
ing a business trip, for example, might make a working mother feel
more connected to her family and friends back homedbut check-
ing Facebook during a family dinner might produce the opposite
effect. Thus, future research should examine when digital behav-
iordfrom information consumption to computer-mediated com-
municationdcomplements or interferes with the psychological
benets people can gain from their nondigital environment. This
research goal is particularly pressing given that improvements in
sensor technology may soon enable smartphones to automatically
adapt their settings according to the social context (e.g., disabling
notications during family meals).
5.2. Limitations and future directions
We rst address two potential limitations of our methodology,
and then consider several key questions that remain to be
addressed by future research. First, given the intense cultural
speculation surrounding smartphones, our studies could have been
subject to possible demand characteristics: Participants in the
phone condition may have reported lower social connection due to
their beliefs about the detrimental effects of technology. But when
we presented a separate sample of 102 participants with a
description of the task faced by participants in either the phone
condition or the phoneless condition, no differences emerged in
how socially connected they expected to feel.
These ndings cast
doubt on the possibility that our main ndings could be explained
by demand characteristics. Second, although we only included
participants who reported following our instructions to use (or not
use) their phones, we lacked an objective measure of compliance.
Of course, because we locked participants' phones away in the
phoneless condition, we can be certain that they did not access
their own phones. Still, it is conceivable that participants could
have borrowed someone else's phone, but such noncompliance
would have only diluted our effects.
The present research has several additional limitations,
providing important directions for future research. First, our reli-
ance on convenience samples of younger users limits our conclu-
sions mostly to people who grew up with smartphones. Future
research should examine whether people who grew up in a world
without smartphones would be more likely to seek information
from strangers even when they have access to their phones. Second,
our conclusions are limited to situations in which people are
looking for concrete information, such as the location of a building.
It would be interesting to explore whether people are more likely to
rely on others when looking for more subjective information, such
as recommendations for restaurants, caf
es, or bars. Third, in
contrast to our other measures, our measure of trust showed low
internal consistency (perhaps due to the small number of items).
This might help to explain our inconsistent results on trust across
the two studies; it would therefore be worthwhile for future
research to include longer measures of trust. Fourth, we did not
include personality measures such as extraversion, which could
have moderated the effects of experimental condition. Interest-
ingly, recent research shows that even introverts benet from
acting in extraverted ways (e.g., Fleeson, Malanos, &Achille, 2002;
Sandstrom &Dunn, 2014b), but that introverts may underestimate
these benets (Zelenski et al., 2013). This points to the hypothesis
that introverts may be especially inclined to relyon their phones for
information, making them especially vulnerable to declines in so-
cial connectedness due to missed opportunities to engage with
Finally, in interpreting our results, it is important to note that we
intentionally conducted these studies in a context where people
could safely and effectively navigate the situation either by relying
on technology or other people. We would not expect our results to
extend to situations in which other people are unavailable or un-
willing to help. That said, recent research suggests that people may
often underestimate strangerswillingness to engage in casual so-
cial interactions. Far from the idyllic setting of a college campus,
people taking public transit in Chicago were instructed to try
talking to a stranger on their commute (Epley &Schroeder, 2014).
Although participants predicted that less than half of their fellow
commuters would be willing to talk to them, no one actually re-
ported getting snubbed. Chatting with strangers also provided an
emotional boost that people failed to foresee. Likewise, Flynn and
Lake (2008) found that individuals systematically underestimated
how willing strangers would be to help them in response to a va-
riety of in-person requests for information or other assistance.
Because seeking information from phones (vs. other people)
eliminates the perceived risk of social rejection, individuals may be
overly inclined to rely on technologydthereby missing out on op-
portunities for social connection across a broad range of contexts.
6. Conclusion
Over 100 years ago, French philosopher Guillaume Ferrero
postulated the Principle of Least Effort: Organisms tend to seek the
easiest way to achieve the greatest outcome (Ferrero, 1894). Our
ndings provide evidence for the social costs of the Principle of
Least Effort. By easily accessing information on smartphones, peo-
ple may forgo opportunities to foster a sense of connection through
casual social interactions.
Funding: This work was supported by the Social Sciences and
Data available at:
K. Kushlev et al. / Computers in Human Behavior 76 (2017) 68e74 73
Humanities Research Council [grant number H08e02739] awarded
to Elizabeth Dunn.
Abele, A. E., & Wojciszke, B. (20 07). Agency and communion from the perspective of
self versus others. Journal of Personality and Social Psychology, 93(5), 751e763.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 117,
Braun Research, Inc. (2015). Bank of America: Trends in consumer mobility report.
Retrieved on March 2, 2016 from
Cacioppo, J. T., Hawkley, L. C., Ernst, J. M., Burleson, M., Bernston, G. G., Nouriani, B.,
et al. (2006). Loneliness within a nomological net: An evolutionary perspective.
Journal of Research in Personality, 40, 1054e1085.
Cumming, G. (2013). Understanding the new statistics: Effect sizes, condence in-
tervals, and meta-analysis. Routledge.
Daft, R. L., & Lengel, R. H. (1986). Organizational information requirements, media
richness and structural design. Management Science, 32(5), 554e571.
Epley, N., & Schroeder, J. (2014). Mistakenly seeking solitude. Journal of Experimental
Psychology. General, 143(5), 1980e1999.
Ferrero, G. (1894). Linertie mentale et la loi du moindre effort. Revue Philosophique
de la France et de l
Etranger, 37,169e182.
Fleeson, W., Malanos, A. B., & Achille, N. M. (2002). An intraindividual process
approach to the relationship between extraversion and positive affect: Is acting
extraverted as goodas being extraverted? Journal of Personality and Social
Psychology, 83(6), 1409e1422.
Flynn, F. J., & Lake, V. K. B. (2008). If you need help, just ask: Underestimating
compliance with direct requests for help. Journal of Personality and Social Psy-
chology, 95(1), 128e143.
Guillory, J. E., Hancock, J. T., Woodruff, C., & Keilman, J. (2015). Text messaging re-
duces analgesic requirements during surgery. Pain Medicine (Malden, Mass.),
16(4), 667e672.
Hall, J. A., & Baym, N. K. (2012). Calling and texting (too much): Mobile maintenance
expectations, (over)dependence, entrapment, and friendship satisfaction. New
Media &Society, 14(2), 316e331.
Hayes, A. (2013). Introduction to mediation, moderation, and conditional process
analysis: A regression-based approach. New York, NY: Guilford Press.
Kenrick, D. T., Griskevicius, V., Neuberg, S. L., & Schaller, M. (2010). Renovating the
pyramid of Needs: Contemporary extensions built upon ancient foundations.
Perspectives on Psychological Science, 5(3), 292e314.
Kock, N. (2004). The psychobiological model: Towards a new theory of computer-
mediated communication based on Darwinian evolution. Organization Science,
15(3), 327e348.
Kross, E., Verduyn, P., Demiralp, E., Park, J., Lee, D. S., Lin, N., et al. (2013). Facebook
use predicts declines in subjective well-being in young adults. PLoS One, 8(8),
e69841. .
Kushlev, K., & Dunn, E. W. (2015). Checking email less frequently reduces stress.
Computers in Human Behavior, 43, 220e228.
Kushlev, K., & Proulx, J. (2016). The social costs of ubiquitous information:
Consuming information on mobile phones is associated with lower trust. PLoS
Lee, R. M., Draper, M., & Lee, S. (2001). Social connectedness, dysfunctional Inter-
personal behaviors, and psychological distress: Testing a mediator model.
Journal of Counseling Psychology, 48,310e318.
Lyubomirsky, S., & Boehm, J. K. (2010). Human motives, happiness, and the puzzle of
parenthood: Commentary on Kenrick et al. (2010). Perspectives on Psychological
Science, 5(3), 327e334.
Mann, T. (1990). A guide to library research methods. Oxford: Oxford University Press.
Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50,
Mehl, M. R., Vazire, S., Holleran, S. E., & Clark, C. S. (2010). Eavesdropping on
happiness. Psychological Science, 21, 539e541.
Myers, D. G., & Diener, E. (1995). Who is happy? Psychological Science, 6,10e19 .
Nelson, S. K., Kushlev, K., & Lyubomirsky, S. (2013). The pains and pleasures of
parenting: When, why, and how is parenthood associated with more or less
well-being? Psychological Bulletin, 140(3), 846e895.
Park, N., Lee, S., & Chung, J. E. (2016). Uses of cellphone texting: An integration of
motivations, usage patterns, and psychological outcomes. Computers in Human
Behavior, 62,712e719.
Park, J., Lee, D. S., Shablack, H., Verduyn, P., Deldin, P., Ybarra, O., et al. (2016). When
perceptions defy reality: The relationships between depression and actual and
perceived Facebook social support. Journal of Affective Disorders, 200,37e44.
Peterson, N. A., Speer, P. W., & McMillan, D. W. (2008). Validation of a brief sense of
community scale: Conrmation of the principal theory of sense of community.
Journal of Community Psychology, 36(1), 61e73.
Pew Research Center. (2012). The best (and worst) of mobile connectivity. Retrieved
on May 22, 2015 from
Pew Research Center. (2015). The Smartphone difference. Retrieved on May 22, 2015
Pielot, M., Church, K., & de Oliveira, R. (2014). An in-situ study of mobile phone
notications. In Proceedings of the 16th international conference on human-
computer interaction with mobile devices &services eMobileHCI '14 (pp.
233e242). New York, NY: ACM Press.
Pollet, T. V., Roberts, S. G., & Dunbar, R. I. (2011). Use of social network sites and
instant messaging does not lead to increased ofine social network size, or to
emotionally closer relationships with ofine network members. Cyberpsychol-
ogy Behavior and Social Networking, 14(4), 253e258.
Reis, H. T., Sheldon, K. M., Gable, S. L., Roscoe, J., & Ryan, R. M. (2000). Daily well-
being: The role of autonomy, competence, and relatedness. Personality and
Social Psychology Bulletin, 26,419e435.
Ryan, R. M. (1982). Control and information in the intrapersonal sphere: An
extension of cognitive evaluation theory. Journal of Personality and Social Psy-
chology, 43, 450e461.
Ryan, R. M., & Deci, E. L. (20 00). Self-determination theory and the facilitation of
intrinsic motivation, social development, and well-being. American Psychologist,
55(1), 68e78.
Ryff, C. D. (1989). Happiness is everything, or is it? Explorations on the meaning of
psychological well-being. Journal of Personality and Social Psychology, 57,
Sandstrom, G. M., & Dunn, E. W. (2014a). Is Efciency overrated? Minimal social
interactions lead to belonging and positive affect. Social Psychological and Per-
sonality Science, 5(4), 437e442.
Sandstrom, G. M., & Dunn, E. W. (2014b). Social interactions and well-being: The
surprising power of weak ties. Personality and Social Psychology Bulletin, 40,
Schimmack, U., & Grob, A. (2000). Dimensional models of core affect: A quantitative
comparison by means of structural equation modeling. European Journal of
Personality, 14, 325e345.
Schwartz, T.(2015, November 28). Addictedto distraction.Retrievedon Dec 10,2015 from
Turkle, S. (2011). Alone together: Why we expect more from technology and less from
each other. Old Saybrook, CN: Tantor Media.
Turkle, S. (2015). Reclaiming conversation: The power of talk in a digital age.New
York, NY: Penguin Press.
Valkenburg, P. M., & Peter, J. (2007). Online communication and adolescent well-
being: Testing the stimulation versus the displacement hypothesis. Journal of
Computer-Mediated Communication, 12(4), 1169e118 2 .
Van Baaren, R. B., Holland, R. W., Kawakami, K., & Van Knippenberg, A. (2004).
Mimicry and prosocial behavior. Psychological Science, 15(1), 71e74 .
Verduyn, O., Ybarra, O., R
esibois, M., Jonides, J., & Kross, E. (2017). Do social network
sites enhance or undermine subjective well-being: A critical review. Social Is-
sues and Policy Review (in press).
Vittengl, J. R., & Holt, C. S. (1998). Positive and negative affect in social interactions
as a function of partner familiarity, quality of communication, and social anx-
iety. Journal of Social &Clinical Psychology, 17,196e208.
Wesselmann, E. D., Cardoso, F. D., Slater, S., & Williams, K. D. (2012). To be looked at
as though air: Civil attention matters. Psychological Science, 23,166e168 .
Wheeler, L., Reis, H., & Nezlek, J. B. (1983). Loneliness, social interaction, and sex
roles. Journal of Personality and Social Psychology, 45, 943e953.
Zelenski, J. M., Whelan, D. C., Nealis, L. J., Besner, C. M., Santoro, M. S., & Wynn, J. E.
(2013). Personality and affective forecasting: Trait introverts underpredict the
hedonic benets of acting extraverted. Journal of Personality and Social Psy-
chology, 104(6), 1092e110 8 .
K. Kushlev et al. / Computers in Human Behavior 76 (2017) 68e7474
396.43 KB
  • Article
    Full-text available
    Previous research on associations between screen time and psychological well-being among children and adolescents has been conflicting, leading some researchers to question the limits on screen time suggested by physician organizations. We examined a large (n = 40,337) national random sample of 2- to 17-year-old children and adolescents in the U.S. in 2016 that included comprehensive measures of screen time (including cell phones, computers, electronic devices, electronic games, and TV) and an array of psychological well-being measures. After 1 h/day of use, more hours of daily screen time were associated with lower psychological well-being, including less curiosity, lower self-control, more distractibility, more difficulty making friends, less emotional stability, being more difficult to care for, and inability to finish tasks. Among 14- to 17-year-olds, high users of screens (7+ h/day vs. low users of 1 h/day) were more than twice as likely to ever have been diagnosed with depression (RR 2.39, 95% CI 1.54, 3.70), ever diagnosed with anxiety (RR 2.26, CI 1.59, 3.22), treated by a mental health professional (RR 2.22, CI 1.62, 3.03) or have taken medication for a psychological or behavioral issue (RR 2.99, CI 1.94, 4.62) in the last 12 months. Moderate use of screens (4 h/day) was also associated with lower psychological well-being. Non-users and low users of screens generally did not differ in well-being. Associations between screen time and lower psychological well-being were larger among adolescents than younger children.
  • Article
    Adolescents spend a substantial and increasing amount of time using digital media (smartphones, computers, social media, gaming, Internet), but existing studies do not agree on whether time spent on digital media is associated with lower psychological well-being (including happiness, general well-being, and indicators of low well-being such as depression, suicidal ideation, and suicide attempts). Across three large surveys of adolescents in two countries (n = 221,096), light users (<1 h a day) of digital media reported substantially higher psychological well-being than heavy users (5+ hours a day). Datasets initially presented as supporting opposite conclusions produced similar effect sizes when analyzed using the same strategy. Heavy users (vs. light) of digital media were 48% to 171% more likely to be unhappy, to be in low in well-being, or to have suicide risk factors such as depression, suicidal ideation, or past suicide attempts. Heavy users (vs. light) were twice as likely to report having attempted suicide. Light users (rather than non- or moderate users) were highest in well-being, and for most digital media use the largest drop in well-being occurred between moderate use and heavy use. The limitations of using percent variance explained as a gauge of practical impact are discussed.
  • Article
    Full-text available
    In the U.S., 95% of smartphone users admit to having used their smartphones during their latest social gathering. Although smartphones are designed to connect us with others, such smartphone use may create a source of distraction that disconnects us from the people in our immediate social environment. Focusing on one fundamental social relationship—between parents and their children—we examined whether smartphones made parents feel distracted, thereby undermining key benefits parents reap when spending time with their children. Ina field experiment at a science museum (Study 1), we randomly assigned parents to use their phones frequently or infrequently. Frequent phone use led parents to feel more distracted, which in turn impaired feelings of social connection and the meaning that parents derived when spending time with their children. In an additional weeklong diary study (Study 2), we found further evidence that smartphones can distract parents from reaping a sense of social connection when spending time with their children. These studies suggest that being constantly connected to the Internet may carry subtle costs for the fabric of social life.
  • Chapter
    Full-text available
    Media technology—from mass media to social media and from video gaming to computer-mediated communication—plays an increasingly central role in people’s lives. Due to exponential increases in computing power, people now carry incredibly powerful computers—their smartphones—everywhere they go. This ever-greater access to media technology is generating an ever-greater conflict between media activities and the unmediated activities critical for psychological well-being—from our face-to-face conversations and family time to our down time and work lives. What are the costs and benefits of people’s modern media technology use for psychological well-being? Using a complementarity-interference (CI) framework, I review research to illuminate key psychological processes (i.e., mediators) and conditions (i.e., moderators) of the relationship between media technology and psychological well-being. Based on the existing evidence, I propose an initial theoretical CI model of the effects of media technology on psychological well-being. I use this CI model to outline important directions for future research, providing guidelines for an integrated, theoretically informed research on media technology.
  • Article
    Full-text available
    Social network sites are ubiquitous and now constitute a common tool people use to interact with one another in daily life. Here we review the consequences of interacting with social network sites for subjective well-being—that is, how people feel moment-to-moment and how satisfied they are with their lives. We begin by clarifying the constructs that we focus on in this review: social network sites and subjective well-being. Next, we review the literature that explains how these constructs are related. This research reveals: (a) negative relationships between passively using social network sites and subjective well-being, and (b) positive relationships between actively using social network sites and subjective well-being, with the former relationship being more robust than the latter. Specifically, passively using social network sites provokes social comparisons and envy, which have negative downstream consequences for subjective well-being. In contrast, when active usage of social network sites predicts subjective well-being, it seems to do so by creating social capital and stimulating feelings of social connectedness. We conclude by discussing the policy implications of this work. © 2017 The Society for the Psychological Study of Social Issues
  • Article
    Full-text available
    In an age already saturated with information, the ongoing revolution in mobile computing has expanded the realm of immediate information access far beyond our homes and offices. In addition to changing where people can access information, mobile computing has changed what information people access—from finding specific directions to a restaurant to exploring nearby businesses when on the go. Does this ability to instantly gratify our information needs anytime and anywhere have any bearing on how much we trust those around us—from neighbors to strangers? Using data from a large nationally representative survey (World Values Survey: Wave 6), we found that the more people relied on their mobile phones for information, the less they trusted strangers, neighbors and people from other religions and nationalities. In contrast, obtaining information through any other method—including TV, radio, newspapers, and even the Internet more broadly—predicted higher trust in those groups. Mobile information had no bearing on how much people trusted close others, such as their family. Although causality cannot be inferred, these findings provide an intriguing first glimpse into the possible unforeseen costs of convenient information access for the social lubricant of society—our sense of trust in one another.
  • Article
    This study suggests an integrated model that explains the associations among motivations for using cellphone texting, usage patterns, and psychological consequences. Using data from an online survey (N = 335), the study identified motivations of communication with strong ties and weak ties, which were found to be associated with different usage patterns of cellphone texting. Further, time spent on cellphone texting was negatively associated with relationship satisfaction, while the number of text messages sent and received was associated with reduced feelings of loneliness through higher levels of perceived intimacy and relationship satisfaction.
  • Conference Paper
    Full-text available
    Notifications on mobile phones alert users about new messages, emails, social network updates, and other events. However, little is understood about the nature and effect of such notifications on the daily lives of mobile users. We report from a one-week, in-situ study involving 15 mobile phones users, where we collected real-world notifications through a smartphone logging application alongside subjective perceptions of those notifications through an online diary. We found that our participants had to deal with 63.5 notifications on average per day, mostly from messengers and email. Whether the phone is in silent mode or not, notifications were typically viewed within minutes. Social pressure in personal communication was amongst the main reasons given. While an increasing number of notifications was associated with an increase in negative emotions, receiving more messages and social network updates also made our participants feel more connected with others. Our findings imply that avoiding interruptions from notifications may be viable for professional communication, while in personal communication, approaches should focus on managing expectations.
  • Article
    When we buy our daily cup of coffee, sometimes we engage in a social interaction with the barista, and sometimes we are in a rush. Every day we have opportunities to transform potentially impersonal, instrumental exchanges into genuine social interactions, and the happiness literature suggests that we may reap benefits by doing so; in other words, treating a service provider like we would an acquaintance (i.e., weak tie) might make us happier. In the current study, people who had a social interaction with a barista (i.e., smiled, made eye contact, and had a brief conversation) experienced more positive affect than people who were as efficient as possible. Further, we found initial evidence that these effects were mediated by feelings of belonging. These results suggest that, although people are often reluctant to have a genuine social interaction with a stranger, they are happier when they treat a stranger like a weak tie.