Contents lists available at ScienceDirect
Journal of Experimental Social Psychology
journal homepage: www.elsevier.com/locate/jesp
Smartphone use undermines enjoyment of face-to-face social interactions
Ryan J. Dwyer
, Kostadin Kushlev
, Elizabeth W. Dunn
Department of Psychology, The University of British Columbia, 2136 West Mall, Vancouver, BC V6T 1Z4, Canada
Department of Psychology, University of Virginia, 485 McCormick Road Gilmer Hall, Room 102, Charlottesville, VA 22903, United States
Using a ﬁeld experiment and experience sampling, we found the ﬁrst evidence that phone use may undermine
the enjoyment people derive from real world social interactions. In Study 1, we recruited over 300 community
members and students to share a meal at a restaurant with friends or family. Participants were randomly as-
signed to keep their phones on the table or to put their phones away during the meal. When phones were present
(vs. absent), participants felt more distracted, which reduced how much they enjoyed spending time with their
friends/family. We found consistent results using experience sampling in Study 2; during in-person interactions,
participants felt more distracted and reported lower enjoyment if they used their phones than if they did not.
This research suggests that despite their ability to connect us to others across the globe, phones may undermine
the beneﬁts we derive from interacting with those across the table.
Decades of research on human happiness points to one central con-
clusion: Engaging in positive social interactions is critical for well-being
(Baumeister & Leary, 1995; Epley & Schroeder, 2014; Kahneman, Krueger,
Schkade, Schwarz, & Stone, 2004; Sandstrom & Dunn, 2014). But the
current technological revolution may be altering how and when we derive
these beneﬁts. Smartphones enable us to connect with friends and family
throughout the day, potentially allowing us to reap the beneﬁts of social
interactions even when we are alone. Could these devices—with their
ability to connect us with anyone, anywhere—distract us from enjoying
interactions with the people sitting right next to us?
In a recent Pew study, almost 90% of cell phone owners reported using
their phones during their most recent social activity (Pew Research Center,
2015). Multi-tasking by using phones may be a major source of distraction
in daily life, leaving people unable to concentrate fully on their primary
activity. For example, using phones while driving is comparable to driving
drunk (Strayer, Drews, & Crouch, 2006), using phones in the classroom
has been shown to impede learning (Wood et al., 2012), and frequent
notiﬁcations via phones can increase symptoms of inattention associated
with ADHD (Kushlev, Proulx, & Dunn, 2016). Theoretically, distraction
should also reduce the ability to derive pleasure from positive experiences
(Brown & Ryan, 2003; Quoidbach, Berry, Hansenne, & Mikolajczak,
2010). Several studies lend support to this contention (e.g.,
Csikszentmihalyi, 1990; LeBel & Dubé, 2001), although this idea has not
been tested directly. In addition to increasing distraction, phones may
compromise the beneﬁts of social interactions by increasing perceived
opportunity costs; texting a romantic partner during lunch with friends or
other things they want or need to be doing. Thus, researchers have the-
their immediate social environment, potentially decreasing enjoyment of
social interactions (Przybylski & Weinstein, 2012; Srivastava, 2005). In
sum, by increasing feelings of distraction or perceived opportunity costs,
smartphone use may undermine the emotional beneﬁts people derive from
It is also possible, however, that phone use could play a positive role in
social interactions. When a conversation lags or turns to dull topics,
smartphones could provide reliable access to an array of brief engaging
activities. Researchers have theorized that engaging one's attention with
desired stimuli should decrease boredom, speed the passage of time, and
promote a sense of agency (for a review, see Eastwood, Frischen, Fenske, &
Smilek, 2012). By allowing us to engage our attention with an array of
stimuli on demand, therefore, phones may decrease boredom, make time
pass more quickly, and give us a greater sense of control.
There is abundant speculation about the possible eﬀects of phones
on social interactions (Turkle, 2012, 2015), but research examining
how phone use shapes the beneﬁts people derive from social interac-
tions is in its infancy. Using correlational analyses, recent studies have
documented a negative relationship between the presence of phones
Received 14 March 2017; Received in revised form 15 October 2017; Accepted 15 October 2017
This work was supported by the Social Sciences and Humanities Research Council of Canada grant H08-02739 awarded to Elizabeth Dunn.
E-mail address: email@example.com (R.J. Dwyer).
Journal of Experimental Social Psychology xxx (xxxx) xxx–xxx
0022-1031/ © 2017 Elsevier Inc. All rights reserved.
Please cite this article as: Dwyer, R., Journal of Experimental Social Psychology (2017), https://doi.org/10.1016/j.jesp.2017.10.007
and the quality of social interactions (Brown, Manago, & Trimble, 2016;
Misra, Cheng, Genevie, & Yuan, 2014; Rotondi, Stanca, & Tomasuolo,
2017), and these studies are supported by similar ﬁndings in the lab
(Przybylski & Weinstein, 2012; Vanden Abeele, Antheunis, & Schouten,
2016). However, no research has experimentally manipulated phone
use in the real world, and research has yet to document the psycholo-
gical mechanisms underlying the eﬀects of phone use on the rewards
derived from social interactions.
Thus, in Study 1, we conducted a ﬁeld experiment in which we
manipulated phone use during a central social activity—sharing a meal
out with friends and family. In Study 2, we used experience sampling to
capture the relationship between phone use and well-being across a
wider range of social contexts over the course of a week. In both stu-
dies, we examined whether and how phone use aﬀects the social and
emotional beneﬁts people reap during in-person social interactions. In
line with current best practices, we pre-registered both studies, and we
report all measures, conditions, and exclusions, as well as how sample
sizes were determined.
1. Study 1
1.1.1. Pre-registered power analysis
Based on a pilot study, we estimated an eﬀect size of d = 0.4; using
G*Power3, we calculated that we would need a sample size of N = 200
for 80% power, which we pre-registered on the Open Science
Framework (OSF) at http://tinyurl.com/hwmo9t6. Given the high costs
of this study, we planned and pre-registered three sequential analyses
(at N = 100, 200, and 300); this technique allows for interim analyses
by adjusting the critical alpha to control overall Type 1 error (see
Lakens, 2014 for a primer on sequential analysis). At each analysis
point, data collection can be stopped if results are signiﬁcant at the
adjusted alpha level. The results of our interim analyses led us to
continue collecting data until we reached N = 300, with a pre-speci-
ﬁed, adjusted alpha-level of 0.0278.
Because sessions were scheduled ahead of time, we slightly sur-
passed our target sample size, with 304 participants (64% females,
ages = 19–69, M = 29.9 years, SD = 10.6). The sample included both
university students (34%) and adults from the Vancouver, BC commu-
nity (66%). An additional 2 participants did not have usable data due to
a technical error that occurred while completing the survey. We re-
quired that all participants own a smartphone, ostensibly so that they
could receive study-related reminders and survey questions.
1.1.3. Procedure and measures
Participants were invited to complete a “study investigating people's
experience dining out with friends.”Groups of 3–5 friends or family
members participated in the study at a local café. After providing
consent, each group was randomly assigned to the phone or phoneless
condition. Speciﬁcally, to manipulate phone use without revealing the
purpose of the study, we told participants in the phone condition that
they would be asked to answer a survey question after ordering their
food, and that the RA would text them this question; to ensure that they
received the survey, they were told to set their phone on the table with
the ringer or vibration on. In the phoneless condition, participants were
also told that they would answer a survey question, which would be
handed to them on paper; these participants were then instructed to
turn their phones on silent and place them in a container on the table.
To support our cover story, we asked participants to rate how they were
feeling that day on a scale from 0 to 100 via text (phone condition) or
paper (phoneless condition). Participants then ate their meal together
without further interruption by the experimenter.
After their meal, all participants were given iPads to complete a
questionnaire (thereby maximizing the privacy of their responses; for
complete survey see http://tinyurl.com/hwmo9t6). This questionnaire
included our key measures of social connectedness, aﬀect, opportunity
costs, interest/enjoyment, distraction, perceived control, time percep-
tion, and boredom, in that order (see Table 1 for details on all mea-
sures). Next, participants were asked to answer questions about their
overall amount of phone use during the session (providing a manip-
ulation check); we also included exploratory questions about the nature
of their phone use (e.g., text messaging, social media, photos). Finally,
participants were asked to indicate the nature of their relationship to
each other person in the group (e.g., spouse, sibling, friend), and to
provide demographics. After completing this survey, participants were
asked to provide feedback about the study and to report how many
notiﬁcations they received on their phones. In exchange for partici-
pating, each participant received up to $20 to spend toward their
group's total bill. All sessions were videotaped using a small camera
(GoPro Hero 4) positioned so that it was visible but unobtrusive. To
minimize any potential for experimenter bias, research assistants were
kept blind to our hypotheses. During the meal, research assistants sat at
a separate table without observing participants.
1.1.4. Manipulation checks
We asked participants to report “During the dining experience
today, how often did you use your mobile phone?”from Not at all (1) to
Constantly (7). To capture phone use compared to participants' normal
behavior, we asked “How frequently did you use your phone as com-
pared to how you would have normally used your phone in a restaurant
with your friends/family?”from Much less (−3) to Much more (3). To
Sample questionnaire items from Study 1.
Measure αSample items Source
Social connectedness .77 I felt close to people. Lee, Draper, & Lee, 2001
Aﬀect: valence (mood) .87 Pleasant. Schimmack & Grob, 2000
Aﬀect: tense arousal .72 Jittery. Schimmack & Grob, 2000
Aﬀect: energetic .90 Awake. Schimmack & Grob, 2000
Interest and enjoyment .69 I enjoyed this experience very much. Ryan, 1982
Perceived control .61 I felt I had control. Bernstein & Claypool, 2012
Boredom .83 I felt bored. Fahlman et al., 2013
Time perception .88 Time was dragging on. Fahlman et al., 2013
Opportunity cost –Did you feel there were other things you wanted or needed to be doing? Kushlev, 2011
Distraction .54 I was easily distracted. Feldman, Hayes, Kumar, Greeson, & Laurenceau, 2007
All items were measured on a scale from 1 (not at all) to 7 (very much), except social connectedness, perceived control, boredom, and time perception, which were measured on a scale
from 1 (Strongly disagree) to 7 (Strongly agree).
R.J. Dwyer et al. Journal of Experimental Social Psychology xxx (xxxx) xxx–xxx
provide an objective behavioral measure, the amount of time partici-
pants spent interacting with their phones was assessed by two coders,
and then divided by the total duration of the video. Coders showed high
agreement (α= .95) and thus we averaged their ratings.
1.1.5. Pre-registered hypotheses
We pre-registered our main hypotheses at the Open Science
Framework (http://tinyurl.com/hwmo9t6). Speciﬁcally, we predicted
that phone use would undermine the beneﬁts people derived from
sharing a meal with others, such that participants in the phone (vs.
phoneless) condition would experience less social connectedness, more
tense arousal, less pleasant aﬀect, and less interest/enjoyment (see
Table 1). We expected that these negative eﬀects would be mediated by
distraction or by perceived opportunity costs. We also anticipated that
phone use would provide beneﬁts by giving people a greater sense of
perceived control, reducing boredom, and making time seem to move
faster. We expected that the positive and negative eﬀects of being as-
signed to the phone condition would be ampliﬁed for individuals who
used their phones heavily during the meal.
1.2.1. Analytic strategy
Because all participants within a group were assigned to the same
condition, we employed multilevel modeling (MLM), which accounts
for non-independence. In these data, participants (level 1) are nested
within groups (level 2). We used SPSS 24 to run all reported mixed
models. For all of the models, we used maximum likelihood estimation
and treated predictors as ﬁxed eﬀects, allowing only the intercept to
vary as a random eﬀect at the person level. The reported ﬁxed eﬀects
below represent the eﬀect of Condition (level-2 variable) on partici-
pants' feelings and behavior (level-1 variables).
1.2.2. Manipulation checks
Conﬁrming the success of our manipulations, participants in the
phone condition reported using their phones more during the study
(M = 2.21, SD = 1.31) than those in the phoneless condition
(M = 1.07, SD = 0.31), b= 1.16, t(73.45) = 8.79, p< .001.
Participants in the phone condition reported using their phones slightly
less than normal (M = −0.63, SD = 1.24), while participants in the
phoneless condition reported using their phones much less than normal
(M = −1.79, SD = 1.44), creating a signiﬁcant diﬀerence between
conditions, b= 1.16, t(72.02) = 7.31, p< .001. Additionally, coders'
objective ratings conﬁrmed that participants in the phone condition
used their phones for a greater percentage of time during the sessions
(M = 11%, SD = 8%) compared to those in the phoneless condition
(M = 1%, SD = 2%), b= 0.11, t(79.31) = 10.02, p< .001.
1.2.3. Pre-registered hypotheses
Consistent with our pre-registered hypothesis, participants in the
phone condition reported signiﬁcantly lower interest and enjoyment
than those in the phoneless condition, b=−0.37, t(82.75) = −2.55,
p= .013 (see Tables 2 and 3). Participants in the phone (vs. phoneless)
condition also reported feeling more distracted, b= 0.46, t(79.08)
= 3.14, p= .002. In contrast to our original prediction—but consistent
with the other negative eﬀects of phone use—participants reported
marginally higher boredom, b= 0.28, t(77.06) = 2.15, p= .035. Si-
milarly, participants in the phone (vs. phoneless) condition reported
worse subjective experience on all of our other dependent variables,
although these eﬀects were not signiﬁcant.
Our dependent variables
were signiﬁcantly correlated, r= .38, which inﬂates the family-wise
Type I error. We accounted for this by calculating the average eﬀect
across all our outcome variables. We found that phone use had a small
negative eﬀect on well-being, d=−0.31, bootstrapped 95% CI
We conducted all mediation analyses using Rockwood and Hayes'
(2017) MLmed macro for SPSS with robust standard errors (REML es-
timation). We estimated all nonredundant parameters for a 2-1-1
mediational model, including ﬁxed and random intercepts, ﬁxed ef-
fects, and the random eﬀect between mediator and outcome (the
random eﬀects for condition were redundant). Only between-group
eﬀects were estimated—there is no within-group eﬀect for condition,
which is a Level 2 (group-level) variable. Covariances between random
intercepts and eﬀects are also not modelled in 2-1-1 mediational
models. Although we originally predicted that distraction and oppor-
tunity costs might mediate the eﬀects, condition only aﬀected distrac-
tion, b= 0.46, p= .003. In turn, distraction signiﬁcantly predicted
interest/enjoyment, b=−0.29, p= .008. Thus, we examined whether
distraction mediated the eﬀect of condition on interest and enjoyment.
Indeed, we found an indirect eﬀect of condition on interest/enjoyment
through distraction, b=−0.13, Z = −1.99, p= .046, which was
marginal by our more stringent alpha level. Condition was no longer a
signiﬁcant predictor of interest and enjoyment after controlling for
distraction, b=−0.23, p= .121, suggesting that the negative eﬀect of
phone use on interest/enjoyment was due in part to people being dis-
tracted by their phones.
Our manipulation also had signiﬁcant indirect eﬀects, via distrac-
tion, on nearly all of our dependent measures; that is, the presence of
phones led people to feel more distracted, which in turn was associated
with less positive aﬀect valence, more tense arousal, less energetic
arousal, less control, and more boredom (see Table 4).
In the real-world setting of a café, we found that people enjoyed a
Means and standard deviations for measures in Study 1.
Phone (n= 152) Phoneless (n= 152)
Dependent variables M (SD) M (SD)
Social connectedness 5.78 (0.82) 5.82 (0.79)
Valence 5.99 (0.89) 6.12 (0.82)
Tense arousal 2.65 (0.97) 2.55 (0.88)
Energetic arousal 4.8 (1.23) 5.05 (1.2)
Interest/enjoyment 4.98 (1.17) 5.36 (1.02)
Control 4.85 (0.84) 4.97 (0.94)
Boredom 2.21 (1.19) 1.93 (1.02)
Time perception 2.04 (1.17) 1.85 (1.1)
Mediator variables M (SD) M (SD)
Opportunity cost 2.84 (1.66) 2.92 (1.63)
Distraction 2.84 (1.2) 2.38 (1.03)
Summary of ﬁxed eﬀects for MLM analyses in Study 1.
DV bdft p
Social connectedness −0.03 80.52 −0.31 .758
Valence −0.12 79.72 −1.14 .257
Tense arousal 0.1 83.01 0.9 .369
Energetic arousal −0.25 79.96 −1.64 .106
Interest/enjoyment −0.37 82.75 −2.55 .013
Control −0.11 81.66 −1 .323
Boredom 0.28 77.06 2.15 .035
Time perception 0.18 75.17 1.31 .194
Note. See Table S-1 in SOM for full model statistics.
We also predicted that the negative eﬀects of phone use would be ampliﬁed for
participants who exhibited the highest levels of phone use. We did not ﬁnd consistent
support for this hypothesis (see SOM).
R.J. Dwyer et al. Journal of Experimental Social Psychology xxx (xxxx) xxx–xxx
meal with their friends less when phones were present than when
phones were put away. They also felt more distracted when phones
were present (vs. absent), which had negative downstream con-
sequences for their broader subjective experience (e.g., more tense
arousal and boredom). Taken together, these results provide initial
evidence that phone use may undermine some of the beneﬁts people
derive from the central social experience of sharing a meal. However,
given that the eﬀects were somewhat weak statistically and that we
examined only one social setting, we conducted a second study ex-
amining phone use, distraction, and well-being across diverse social
interactions. By using a within-subjects design and collecting data
across numerous time points, we sought to enhance power to detect the
eﬀects initially observed in our between-subjects experiment.
2. Study 2: experience sampling
In Study 2, we surveyed people 5 times a day for one week, asking
them to report how they had been feeling and what they had been
doing over the past 15 min.
We preregistered our hypotheses and analysis plan on the Open
Science Framework (http://tinyurl.com/z7xe43d). Based on Study 1,
we predicted that during social interactions, people would feel more
distracted and would experience less enjoyment when they were using
smartphones compared to when they were not using smartphones. In
addition, consistent with Study 1, we expected that phone use would
have detrimental indirect eﬀects (via distraction) on social connected-
ness, aﬀect valence, boredom, and time perception. We preregistered to
collect at least 100 usable participants with replacement. Participants
were considered unusable for the analyses if they had no episodes when
they interacted with others in person.
Students at a large public university in the U.S. who owned smart-
phones participated for course credit. In total, we obtained usable data
from 123 participants (M
= 18.6, 69% women; see SOM for further
details). The participants responded to a total of 3008 ESM surveys
(68.6% of surveys sent), of which 1244 were episodes when they were
interacting with others face-to-face.
2.1.3. Procedure and measures
Participants completed a demographic survey, and on the following
day, they began to receive links to brief ESM surveys sent via text
message using the service surveysignal.com. For seven days, partici-
pants received ﬁve prompts per day between 9 am and 9 pm. Within
this 12-hour period, the prompts were sent randomly within ﬁve
equally divided intervals. Each message instructed participants to re-
spond as quickly as they safely could. The links expired after 1 h, and
each survey was administered at least 2 h after the previous prompt.
At the top of each survey, participants were instructed that all the
questions pertained to their experience “over the last 15 min before com-
pleting the survey.”Participants then completed brief versions (adapted for
ESM) of selected measures from Study 1. Speciﬁcally, participants reported
whether they had enjoyed their experience in the last 15 min and whether
they would describe their experience as very interesting (1 –Not true;7–
Very true). We averaged these two responses to form our composite mea-
sure of interest/enjoyment (α=.80).Aﬀect valence, social connectedness,
distraction, boredom, and time perception were each measured on 7-point
scales using single items adapted from Study 1; for the full survey, see
At the end of the survey, participants were asked what they had been
doing in the past 15 min, including how much they had used their
smartphones (1 –Not at all;7–Constantly). As preregistered based on the
ﬁndings in Study 1, this measure was transformed into a dichotomous
measure of whether or not people used their phones during the episode. To
dichotomize phone use, we coded as 0 episodes when participants re-
ported using their phones ‘notatall’, and coded any use as 1 (i.e., 2–7
were coded as 1). Participants also indicated whether or not they had
“socialized in person/face-to-face,”among a number of other common
daily activities (e.g., eating/drinking, working/studying).
2.2.1. Analytic strategy
We pre-registered our analysis plan on OSF; see http://tinyurl.com/
z7xe43d. Because each person had multiple episodes, we used multi-
level modeling (MLM) to estimate the eﬀects. For all of the models, we
used SPSS 24, specifying restricted maximum likelihood estimation and
treating predictors as ﬁxed eﬀects, allowing only the intercept to vary
as a random eﬀect at the person level.
Mediational and indirect eﬀects of phone use through distraction in Study 1.
Outcome variable Path bPath cPath c′a∗b95% CI Sobel Z
Social connectedness −0.22
−0.03 0.07 −0.10
[−0.21, −0.02] −2.13
−0.12 0.06 −0.18
[−0.32, −0.06] −2.67
Tense arousal 0.44
0.10 −0.10 0.21
[0.07, 0.36] 2.73
Energetic arousal −0.38
−0.25 −0.08 −0.17
[−0.34, −0.05] −2.24
[−0.29, −0.03] −1.99
−0.11 0.03 −0.14
[−0.28, −0.04] −2.35
[0.07, 0.38] 2.65
Time perception 0.23
0.18 0.07 0.11 [0.01, 0.24] 1.75
Note. The eﬀect of condition on distraction (path a) is signiﬁcant for all DV's, b= 0.46, p= .003. Path brepresents the eﬀect of distraction on each outcome variable. Path crepresents
the total eﬀect of phone use on each outcome variable. Path c′represents the direct eﬀect of phone use on each outcome variable after accounting for the indirect eﬀect through
distraction. Conﬁdence intervals were estimated using 10,000 Monte Carlo stimulations.
There was little variability in the slope from the mediator to outcome variable, so the model did not converge. Thus, we did not estimate the random M to Y slope for the Control
There was little variability in the Y intercept, so the model did not converge when it the random Y intercept was included in the model. Thus, the model used for the Boredom variable
included a ﬁxed Y intercept only.
Estimating the random component of each eﬀect did not substantively change the size
of the ﬁxed component or the signiﬁcance level.
R.J. Dwyer et al. Journal of Experimental Social Psychology xxx (xxxx) xxx–xxx
2.2.2. Eﬀects of phone use
As predicted, during episodes that included face-to-face social in-
teractions (N = 1244), people reported feeling more distracted when
they used their smartphones than when they did not, b= 0.95,
SE = 0.11, t(91.98) = 8.23 p< .001 (see Table 5 for means). In ad-
dition, during these episodes, participants experienced less interest and
enjoyment when they used smartphones than when they did not,
b=−0.41, SE = 0.09, t(1239.60) = −4.44, p< .001. Although we
did not predict direct eﬀects of phone use on the other outcomes, we
found that people also reported worse aﬀect, b=−0.31, SE = 0.09, t
(1237.09) = −3.46, p= .001, felt less socially connected, b=−0.33,
SE = 0.09, t(1234.80) = −3.87, p< .001, more bored, b= 0.53,
SE = 0.10, t(1240.14) = 5.25, p< .001, and perceived time to be
moving slower, b= 0.34, SE = 0.11, t(1239.77) = 3.20, p= .001 (see
Table S-3 for exhaustive list of model parameters for all reported
models). In additional exploratory analyses, we used the continuous
phone use variable as the predictor (1 –Not at all;7–Constantly) and
found the same eﬀects across all outcomes (all p's < .001). As in Study
1, our dependent variables were signiﬁcantly correlated, r= .50,
which inﬂates family-wise Type 1 error. Calculating the average eﬀect
across our variables, we found that social interactions with phone use
were associated with a small decrease in well-being compared to social
interactions without phone use, d=−0.23, bootstrapped 95% CI
2.2.3. Mediation analyses
As preregistered, we used person-level means for the mediation
analyses. For distraction and all outcome variables, we ﬁrst aggregated
each participant's scores separately for episodes with versus without
smartphones. We then predicted the diﬀerence scores between the
means of the outcome variables (e.g. interest/enjoyment) when people
were using versus not using their phones from the diﬀerence and sum
scores of distraction (i.e., the mediator). As in Study 1, we conducted all
mediation analyses using Rockwood and Hayes' (2017) MLmed macro
for SPSS. All random eﬀects were estimated, within- and between-
persons eﬀects were deconﬂated, and covariances were estimated ex-
cept for the covariance between the intercepts of the mediator and
outcome variables. Because we were interested how the same people
felt when using their phones or not, we focus on the within-subjects
eﬀects below (see Table S-4 for between-subject eﬀects). We examined
whether distraction could explain the eﬀect of phone use on our out-
We found support for the mediating role of distraction
across all outcomes (see Table 6). To the extent people felt more dis-
tracted when they used their phones, they reported feeling less enjoy-
ment, worse aﬀect, less socially connected, more bored, and slower
3. General discussion
Using a ﬁeld experiment and intensive experience sampling, we
found the ﬁrst evidence that phone use undermines the enjoyment
people derive from real world social interactions. In Study 1, phone use
caused individuals to feel distracted, which reduced how much they
enjoyed sharing a meal with friends at a local café. In Study 2, we
obtained similar results by asking people to report what they had been
doing and feeling during the past 15 min. When they had been engaging
in face-to-face interactions, they felt more distracted and reported
lower enjoyment if they had been using their smartphones than if they
had not. Phone use also had indirect negative eﬀects, via distraction, on
other well-being outcomes; in both studies, phone use predicted
distraction, which in turn predicted greater boredom and worse overall
Going beyond previous research, we found these eﬀects in natur-
alistic contexts by conducting a ﬁeld experiment in a café and by using
experience sampling to capture a variety of real-world social situations.
These results held across two distinct samples, including both students
and community members in Western Canada (Study 1), and students in
the Southern United States (Study 2). It is especially notable that we
observed negative eﬀects of phone use among university students. This
generation has grown up with mobile technology, and some have raised
the possibility that young people might therefore be relatively adept at
multi-tasking in real world contexts (Foehr, 2006). This idea is parti-
cularly compelling in the context of extended social interactions, such
as sharing a meal with friends, given that natural lulls in conversation
might aﬀord the ability to attend to one's phone without any detectable
cost. Yet, our ﬁndings suggest that even the moderate levels of phone
use we observed are suﬃcient to create feelings of distraction that
undermine the emotional rewards of social interaction.
These results build on other recent work showing that phone use is
associated with less positive social impressions (Vanden Abeele et al.,
2016), lower interaction quality with friends (Brown et al., 2016; Misra
et al., 2014), and lower relationship satisfaction with a romantic
partner (Roberts & David, 2016). We found that the negative eﬀects of
phone use were mediated by distraction, but not opportunity costs,
suggesting that phone use prevents individuals from fully engaging in
the present moment. Contrary to our original prediction, phones
slightly increased boredom. While Eastwood et al. (2012) argue that
interesting stimuli can decrease boredom, they also theorize that rela-
tively subtle sources of distraction may lead people to misattribute their
feelings of inattention to being bored with the situation. For example,
people report greater boredom when a TV plays quietly in another room
than when it is blaring (Damrad-Frye & Laird, 1989). Thus, people may
Means for episodes with face to face interactions with phone use versus no phone use in
Study 2 (N = 1244).
With phone use M(SD) No phone use M(SD)
Distraction 3.45 (0.91) 2.42 (1.26)
Interest/enjoyment 4.78 (1.01) 5.33 (1.25)
Aﬀect valence 5.28 (0.98) 5.77 (1.11)
Connectedness 5.39 (0.97) 5.92 (1.03)
Boredom 2.97 (0.97) 2.28 (1.13)
Time perception 3.11 (1.11) 2.70 (1.53)
Note. In total, 120 people had episodes with phone use, whereas 96 people had episodes
without phone use. This could explain why the standard deviations for episodes with
phone use are smaller than those for episodes with no phone use.
Within-subjects indirect eﬀects of phone use through distraction in Study 2.
Path cPath c′a∗b95% CI Sobel Z
Aﬀect valence −0.34
[0.21, 0.51] 4.79
Time perception 0.33
[0.11, 0.47] 3.24
Note. Path crepresents the total eﬀect of phone use on each outcome variable. Path c′
represents the direct eﬀect of phone use on each outcome variable; that is, the remaining
eﬀect of phone use after accounting for the indirect eﬀect through distraction. Conﬁdence
intervals were estimated using 10,000 Monte Carlo stimulations.
p< .001 (signiﬁcance based on Sobel's Z test of
The sample used for the mediation analyses was smaller than the total sample because
these analyses require people to have episodes both with and without reported phone use.
The number of participants who had both episodes with reported distraction was N = 93.
As indicated by the total eﬀects in Table 6, the eﬀects of phone use on the outcome
variables mirrored those from the MLM analyses reported above.
R.J. Dwyer et al. Journal of Experimental Social Psychology xxx (xxxx) xxx–xxx
anticipate that phones will reduce boredom (as we did), but by pro-
viding a subtle source of distraction, phones may ironically increase
Still, it is important to note that the observed eﬀects in both studies,
as shown by the meta-analyses, were relatively small. Based on the
eﬀect size (d = −0.31) observed across variables in Study 1, future
experiments should include at least 165 participants per condition to
detect reliable eﬀects. As this high required sample size illustrates,
phones may have minimal negative eﬀects on individual social inter-
actions; however, these small eﬀects are likely consequential over time.
McDaniel and Coyne (2014) argue that small, frequent interruptions
from phones can compound into relationship conﬂict and lower life
satisfaction. Our study provides a window into one underlying process
through which phone use may chip away at life satisfaction: phones
may undermine the enjoyment derived from face-to-face social inter-
Of course, everyday life is riddled with other sources of distraction,
such as newspapers and television, but phones diﬀer from these earlier
forms of information technology in two critical ways: Phones provide
access to a virtually inﬁnite array of potential diversions, while being so
portable that they are almost always with us, enabling them to easily
pervade our social interactions. For example, a recent observational
study found that caregivers exhibited a high degree of absorption in
their phones while sharing a meal with children at fast food restaurants
(Radesky, Kistin, Augustyn, & Silverstein, 2014). Theoretically, then,
smartphones provide an ideal proxy for studying how the increasing
pervasiveness of our digital activities is interacting with fundamental
human activities (e.g., sharing a meal).
Our studies have several limitations. Because participants in Study 1
were assigned to the phone or phoneless condition in groups, it is un-
clear whether the eﬀects were caused by the individual's phone use, the
phone use of others in the group, or an interaction between individual
and group phone use. For example, Sally might experience reduced
enjoyment if her dining companions were using their phones while she
was not. Thus, an interesting open question is how individual and group
phone use interact. It is also possible that some participants guessed the
purpose of our studies, and thus responded according to their lay beliefs
about phone use. In both studies, however, we ensured our interest in
phone use was minimally salient by embedding any mention of phones
within broader study instructions or questionnaire items.
Whereas Study 1 used an experimental design, Study 2 relied on a
correlational design and thus may have captured bi-directional eﬀects
between phone use and participants' feelings; for example, phone use
may have increased distraction, but feeling distracted could also have
promoted phone use. Another limitation of Study 2 is that we may have
inadvertently increased participants' typical phone use by about ﬁve
instances per day by asking participants to use their phones to complete
surveys about phone use. According to a Deloitte (2016) report, how-
ever, 18–24 year olds use their phones 82 times a day on average. Thus,
the survey method likely only increased phone use by around 6%.
Given the small magnitude of this increase, we think it is fairly unlikely
to have fundamentally altered the relationship between phone use and
our well-being measures. Finally, although we have framed these re-
sults as showing negative eﬀects of phone use, it would be equally valid
to describe these results as showing positive eﬀects of putting phones
away. That is, it is possible that putting phones away increases atten-
tion, which in turn increases well-being outcomes. Given the ubiquity
of phones in everyday life, we think either causal pathway is important
and the main principle is the same: enjoyment is higher during social
interactions without phones.
Although we found that phone use negatively aﬀected enjoyment of
social interactions, phones should have positive eﬀects in situations where
distraction is desirable. Indeed, phone use has been found to reduce the
need for anesthesia during minor surgery (Guillory, Hancock, & Woodruﬀ,
2015). By aﬀecting distraction, phones may also have a wide range of
other psychological consequences. Distraction has been linked to reduced
memory (Craik, Govoni, Naveh-Benjamin, & Anderson, 1996), increased
stress (Mark, Gudith, & Klocke, 2008), and reduced self-control (Shiv &
Fedorikhin, 1999), pointing to the value of testing the eﬀects of phone use
on these outcomes.
Our research may also serve as a model for future experimental
studies, given that the literature on phones and well-being has dis-
proportionately utilized correlational designs (Brown et al., 2016;
Kushlev & Heintzelman, in press; Misra et al., 2014; Rotondi et al.,
2017). Several other new studies have manipulated the presence of
phones during social interactions and cognitive tasks (Allred &
Crowley, 2017; Avelar, 2015; Lyngs, 2017), but these studies have re-
lied on small samples (cell sizes ≤25) and have produced mixed re-
sults, highlighting the need for well-powered, conﬁrmatory research in
Research on the cognitive eﬀects of distraction have led govern-
ments to enact policy changes restricting phone use while driving
(World Health Organization, 2011), and many course instructors have
implemented analogous policies in their classrooms (Hammer et al.,
2010). In a similar vein, our research highlights the need for change in
social norms surrounding phone use in social interactions. In particular,
this work reveals how phones can distract us from engaging with people
in our immediate environment. Despite their ability to connect us to
others across the globe, phones may undermine the beneﬁts we derive
from interacting with those across the table.
All authors contributed to the study design. Data collection, analysis
and interpretation for Study 1 was performed by R. J. Dwyer under the
supervision of E. W. Dunn and K. Kushlev. Data collection, analysis and
interpretation for Study 2 was performed by K. Kushlev. R. J. Dwyer
drafted the manuscript, and E. W. Dunn and K. Kushlev provided cri-
tical revisions. All authors approved the ﬁnal version of the manuscript
All data have been made publicly available via the Open Science
Framework. The data for Study 1 can be viewed at http://tinyurl.com/
hwmo9t6, and the data for Study 2 can be viewed at http://ti-
We thank the following research assistants for their instrumental
help conducting these studies: Camille Hunt, Charlotte Freitag, Kara
Lee, Nishi Sumant, Shauna Moore, Zoe Bethune, Tara Arvan, Briana
Nolletti, Remy Panikkar, Luan Wei, Bahja Ammari, and Allison Yang.
This research protocol was approved by The University of British
Columbia Behavioral Research Ethics Board.
Appendix A. Supplementary materials
Supplementary data to this article can be found online at https://
Allred, R. J., & Crowley, J. P. (2017). The “mere presence”hypothesis: Investigating the
nonverbal eﬀects of cell-phone presence on conversation satisfaction. Communication
Studies, 68(1), 22–36. http://dx.doi.org/10.1080/10510974.2016.1241292.
Avelar, D. (2015). Tell me why you're texting! Eﬀect of cell phone use on interpersonal in-
teractions. Franklin and Marshall College. (Master's Thesis, Retrieved from) https://
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal
attachments as a fundamental human motivation. Psychological Bulletin, 117(3),
Bernstein, M. J., & Claypool, H. M. (2012). Not all social exclusions are created equal:
R.J. Dwyer et al. Journal of Experimental Social Psychology xxx (xxxx) xxx–xxx
Emotional distress following social exclusion is moderated by exclusion paradigm.
Social Inﬂuence, 7(2), 113–130. http://dx.doi.org/10.1080/15534510.2012.664326.
Brown, G., Manago, A. M., & Trimble, J. E. (2016). Tempted to text: College students
mobile phone use during a face-to-face interaction with a close friend. Emerging
Adulthood, 2–5. http://dx.doi.org/10.1177/2167696816630086.
Brown, K. W., & Ryan, R. M. (2003). The beneﬁts of being present: Mindfulness and its
role in psychological well-being. Journal of Personality and Social Psychology, 84(4),
Craik, F. I., Govoni, R., Naveh-Benjamin, M., & Anderson, N. D. (1996). The eﬀects of
divided attention on encoding and retrieval processes in human memory. Journal of
Experimental Psychology. General, 125(2), 159–180. http://dx.doi.org/10.1037/0096-
Csikszentmihalyi, M. (1990). Flow: The psychology of optimal experience. New York, NY:
Harper and Row.
Damrad-Frye, R., & Laird, J. D. (1989). The experience of boredom: The role of the self-
perception of attention. Journal of Personality and Social Psychology, 57(2), 315–320.
Deloitte (2016). 2016 global mobile consumer survey: US edition (Retrieved from)
Eastwood, J. D., Frischen, A., Fenske, M. J., & Smilek, D. (2012). The unengaged mind:
Deﬁning boredom in terms of attention. Perspectives on Psychological Science, 7(5),
Epley, N., & Schroeder, J. (2014). Mistakenly seeking solitude. Journal of Experimental
Psychology: General, 143(5), 1980–1999. http://dx.doi.org/10.1037/a0037323.
Fahlman, S. A., Mercer-Lynn, K. B., Flora, D. B., & Eastwood, J. D. (2013). Development
and validation of the multidimensional state boredom scale. Assessment, 20(1),
Feldman, G., Hayes, A., Kumar, S., Greeson, J., & Laurenceau, J. P. (2007). Mindfulness
and emotion regulation: The development and initial validation of the cognitive and
aﬀective mindfulness scale-revised (CAMS-R). Journal of Psychopathology and
Behavioral Assessment, 29(3), 177–190.
Foehr, U. G. (2006). Media multitasking among American youth: Prevalence, predictors, and
pairings. The Henry J. Kaiser Family Foundation. (Retrieved from) http://www.kﬀ.
Guillory, J. E., Hancock, J. T., & Woodruﬀ, C. (2015). Text messaging reduces analgesic
text messaging reduces analgesic requirements during surgery. Pain Medicine, 16,
Hammer, R., Ronen, M., Sharon, A., Lankry, T., Huberman, Y., & Zamtsov, V. (2010).
Mobile culture in college lectures: Instructors' and students' perspectives.
Interdisciplinary Journal of E-Learning & Learning Objects, 6, 293–304 (http://doi.org/
Kahneman, D., Krueger, A. B., Schkade, D., Schwarz, N., & Stone, A. A. (2004). A survey
method for characterizing daily life experience: The day reconstruction method.
Science, 306(5702), 1776–1780. http://dx.doi.org/10.1126/science.1103572.
Kushlev, K. (2011). Exploring parental well-being: Is childcare associated with parental well-
being and what factors can enhance it. University of British Columbia. (Master's Thesis,
Retrieved from) https://open.library.ubc.ca/cIRcle/collections/24/items/1.
Kushlev, K., & Heintzelman, S. J. (2017). Put the Phone Down. Social Psychological and
Personality Science. Advanced online publication..http://dx.doi.org/10.1177/
Kushlev, K., Proulx, J., & Dunn, E. W. (2016). “Silence your phones”: Smartphone noti-
ﬁcations increase inattention and hyperactivity symptoms. Proceedings of CHI, 2016.
Lakens, D. (2014). Performing high-powered studies eﬃciently with sequential analyses.
European Journal of Social Psychology, 701–710. http://dx.doi.org/10.1002/ejsp.
LeBel, J. L., & Dubé, L. (2001). The impact of sensory knowledge and attentional focus on
pleasure and on behavioral responses to hedonic stimuli. 13th Annual American
Psychological Society Convention.
Lee, R. M., Draper, M., & Lee, S. (2001). Social connectedness, dysfunctional interpersonal
behaviors, and psychological distress: Testing a mediator model. Journal of Counseling
Psychology, 48(3), 310–318. http://dx.doi.org/10.1037/0022-0126.96.36.1990.
Lyngs, U. (2017). It's more fun with my phone: A replication study of cell phone presence
and task performance. Proceedings of the 2017 CHI conference extended abstracts on
human factors in computing systems (pp. 136–141). New York, NY, USA: ACM. http://
Mark, G., Gudith, D., & Klocke, U. (2008, April). The cost of interrupted work: More speed
and stress. Proceedings of the SIGCHI conference on human factors in computing systems
(pp. 107–110). ACM.
McDaniel, B. T., & Coyne, S. M. (2014). “Technoference”:The interference of technology
in couple relationships and implications for women's personal and relational well-
being. Psychology of Popular Media Culture, 5(1), http://dx.doi.org/10.1037/
Misra, S., Cheng, L., Genevie, J., & Yuan, M. (2014). The iPhone eﬀect: The quality of in-
person social interactions in the presence of mobile devices. Environment and
Behavior, 1–24. http://dx.doi.org/10.1177/0013916514539755.
Americans' views on mobile etiquette. Pew Research Center. (Retrieved from) http://www.
Przybylski, A. K., & Weinstein, N. (2012). Can you connect with me now? How the
presence of mobile communication technology inﬂuences face-to-face conversation
quality. Journal of Social and Personal Relationships, 30(3), 237–246. http://dx.doi.
Quoidbach, J., Berry, E. V., Hansenne, M., & Mikolajczak, M. (2010). Positive emotion
regulation and well-being: Comparing the impact of eight savoring and dampening
strategies. Personality and Individual Diﬀerences, 49(5), 368–373. http://dx.doi.org/
Radesky, J. S., Kistin, C. J., Augustyn, M., & Silverstein, M. (2014). Patterns of mobile
device use by caregivers and children during meals in fast food restaurants. Pediatrics,
133(4), e843–e849. http://dx.doi.org/10.1542/peds.2013-3703.
Roberts, J. A., & David, M. E. (2016). My life has become a major distraction from my cell
phone: Partner phubbing and relationship satisfaction among romantic partners.
Computers in Human Behavior, 54, 134–141. http://dx.doi.org/10.1016/j.chb.2015.
Rockwood, N. J., & Hayes, A. F. (2017, May). MLmed: An SPSS macro for multilevel
mediation and conditional process analysis. Poster presented at the annual meeting of
the Association of Psychological Science (APS), Boston, MA.
Rotondi, V., Stanca, L., & Tomasuolo, M. (2017). Connecting alone: Smartphone use,
quality of social interactions and well-being (No. 357). DEMS working paper series
(Available at SSRN) https://ssrn.com/abstract=2893027.
Ryan, R. M. (1982). Control and information in the intrapersonal sphere: An extension of
cognitive evaluation theory. Journal of Personality and Social Psychology, 42(3),
Sandstrom, G. M., & Dunn, E. W. (2014). Social interactions and well-being: The sur-
prising power of weak ties. Personality and Social Psychology Bulletin, 40(7), 910–922.
Schimmack, U., & Grob, A. (2000). Dimensional models of core aﬀect: A quantitative
comparison by means of structural equation modelling. European Journal of
Personality, 14, 325–345.
Shiv, B., & Fedorikhin, A. (1999). Heart and mind in conﬂict: The interplay of aﬀect and
cognition in consumer decision making. Journal of Consumer Research, 26(3),
Srivastava, L. (2005). Mobile phones and the evolution of social behaviour. Behaviour &
Information Technology, 24(2), 111–129. http://dx.doi.org/10.1080/
Strayer, D. L., Drews, F. A., & Crouch, D. J. (2006). A comparison of the cell phone driver
and the drunk driver. Human Factors, 48(2), 381–391. http://dx.doi.org/10.1518/
Turkle, S. (2012). Alone together: Why we expect more from technology and less from each
other. New York, NY: Basic Books.
Turkle, S. (2015). Reclaiming conversation: The power of talk in a digital age. Penguin
Vanden Abeele, M. M. P., Antheunis, M. L., & Schouten, A. P. (2016). The eﬀect of mobile
messaging during a conversation on impression formation and interaction quality.
Computers in Human Behavior, 62, 562–569. http://dx.doi.org/10.1016/j.chb.2016.
Wood, E., Zivcakova, L., Gentile, P., Archer, K., De Pasquale, D., & Nosko, A. (2012).
Examining the impact of oﬀ-task multi-tasking with technology on real-time class-
room learning. Computers & Education, 58(1), 365–374. http://dx.doi.org/10.1016/j.
World Health Organization (2011). Mobile phone use: A growing problem of driver distrac-
tion. World Health Organization. (Retrieved from) http://www.who.int/violence_
R.J. Dwyer et al. Journal of Experimental Social Psychology xxx (xxxx) xxx–xxx