Distress tolerance and mindfulness mediate relations between depression and anxiety sensitivity with problematic smartphone use

Accepted Manuscript Distress tolerance and mindfulness mediate relations between depression and anxiety sensitivity with problematic smartphone use

Jon D. Elhai, Jason C. Levine, Kelsey D. O’Brien, Cherie Armour PII:

S0747-5632(18)30131-6

DOI:

10.1016/j.chb.2018.03.026

Reference:

CHB 5430

To appear in:

Computers in Human Behavior

Received Date:

05 January 2018

Revised Date:

02 March 2018

Accepted Date:

16 March 2018

Please cite this article as: Jon D. Elhai, Jason C. Levine, Kelsey D. O’Brien, Cherie Armour, Distress tolerance and mindfulness mediate relations between depression and anxiety sensitivity with problematic smartphone use, Computers in Human Behavior (2018), doi: 10.1016/j.chb. 2018.03.026

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

Distress tolerance and mindfulness mediate relations between depression and anxiety sensitivity with problematic smartphone use Jon D. Elhaia,b Jason C. Levinea,b Kelsey D. O’Briena Cherie Armourc a Department

of Psychology, University of Toledo, 2801 W. Bancroft Street, Toledo, OH, 43606, USA b Department of Psychiatry, University of Toledo, 3000 Arlington Ave., Toledo, OH, 43614, USA c Psychology Research Institute, Ulster University, Coleraine Campus, Cromore Road, Coleraine, Northern Ireland, UK Reprints from this paper can be addressed to Jon Elhai, PhD, through his website: www.jon-elhai.com For the Journal Editor and Production Staff: For fastest correspondence from this journal about this submission, email Jon Elhai at: [email protected] Declarations of interests: none This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

ACCEPTED MANUSCRIPT 1

Distress tolerance and mindfulness mediate relations between depression and anxiety sensitivity with problematic smartphone use

ACCEPTED MANUSCRIPT 2

Abstract Excessive, problematic smartphone use (PSU) has demonstrated relationships with depression and anxiety severity across studies. However, less is known about psychopathology-related variables that may mediate relations between depression/anxiety with PSU – especially variables involving emotional regulation processes. We recruited 261 college students for a repeated-measures web survey, administered self-report measures of depression, anxiety sensitivity, distress tolerance, mindfulness, smartphone use frequency, and PSU; one month later, participants completed these measures again. We tested a model where depression severity and anxiety sensitivity predicted distress tolerance and mindfulness, in turn predicting smartphone use frequency, and one-month PSU severity, adjusting for age and sex. Distress tolerance and mindfulness were inversely associated with levels of PSU. Distress tolerance mediated relations between anxiety sensitivity and levels of PSU. Mindfulness mediated relations between both depression and anxiety sensitivity with PSU severity. Results are discussed in the context of emotion regulation theory and compensatory internet use theory, with clinical implications for emotion regulation skills training and mindfulness in offsetting PSU.

Keywords: Depression; Anxiety sensitivity; Distress tolerance; mindfulness; Smartphone addiction

ACCEPTED MANUSCRIPT 3

1. Introduction Excessive, problematic smartphone use (PSU) has deleterious consequences in society. PSU is associated with traffic and pedestrian accidents (Cazzulino, Burke, Muller, Arbogast, & Upperman, 2014; Thompson, Rivara, Ayyagari, & Ebel, 2013), and unfavorable health consequences to the user (Demirci, Akgonul, & Akpinar, 2015; Shan et al., 2013; Xie, Szeto, Dai, & Madeleine, 2016). Additionally, PSU is associated with psychopathology - most notably, depression and anxiety severity (reviewed in Elhai, Dvorak, Levine, & Hall, 2017). However, few PSU studies have examined psychopathology constructs beyond depression and anxiety, and most studies on PSU have relied on cross-sectional data. PSU is a construct involving the excessive use of a smartphone, with symptoms observed in substance use disorders (e.g., tolerance, withdrawal when separated from one’s phone), and corresponding functional impairment from overuse (Billieux, Maurage, Lopez-Fernandez, Kuss, & Griffiths, 2015). Numerous studies have examined PSU severity in relation to psychopathology constructs. Across the literature, the most widely studied psychopathology-related constructs in association with PSU severity include depression, anxiety, stress, and low self-esteem symptoms (reviewed in Elhai, Dvorak, et al., 2017). PSU severity is most consistently related to severity of depression, with moderate effects (e.g., Demirci et al., 2015; Lu et al., 2011; Smetaniuk, 2014), and anxiety, with mild effects (e.g., Demirci et al., 2015; Elhai, Dvorak, et al., 2017; R. Kim, Lee, & Choi, 2015; Lee, Chang, Lin, & Cheng, 2014). Only in recent years have studies broadened their focus from depression and anxiety to explore other psychopathology-related constructs connected with PSU. This

ACCEPTED MANUSCRIPT 4

line of inquiry is important, because despite the relatively consistent effects for depression and anxiety severity with PSU, numerous studies have instead found small or inverse relationships (reviewed in Elhai, Dvorak, et al., 2017). Therefore, other psychopathology-related constructs may help to explain relationships between depression/anxiety with PSU. Recently, several psychopathology-related constructs have been supported as mediators between levels of depression and/or anxiety severity with PSU. These mediating variables include decreased behavioral activation (Elhai, Levine, Dvorak, & Hall, 2016), rumination (Elhai, Tiamiyu, & Weeks, in press), emotion dysregulation (Elhai et al., 2016; Elhai, Tiamiyu, Weeks, et al., in press), low self-control (Cho, Kim, & Park, 2017), fear of missing out (Wolniewicz, Tiamiyu, Weeks, & Elhai, in press), and proneness to boredom (Elhai, Vasquez, Lustgarten, Levine, & Hall, in press). In the present paper, we tested additional psychopathology-related constructs that have not been examined for relations with PSU. First, instead of measuring anxiety symptoms, we assessed anxiety sensitivity. Anxiety sensitivity is the fear of arousalbased sensations, and that these sensations will cause social or physical consequences (Reiss & McNally, 1985). Anxiety sensitivity serves to amplify and intensify anxiety symptoms (reviewed in Taylor, 2014), and is associated with anxiety disorders (Olatunji & Wolitzky-Taylor, 2009). Anxiety sensitivity correlates with addictive disorders - alcohol use disorder in particular (DeMartini & Carey, 2011) - but has not been investigated in relation to PSU or other problematic technology use (PTU). We also assessed two constructs related to regulation of negative emotion that have not been previously tested for relations with PSU: distress tolerance and

ACCEPTED MANUSCRIPT 5

mindfulness. We assessed these two variables as mediators between both depression and anxiety sensitivity with PSU. Distress tolerance is commonly defined as the perceived ability to endure negative emotion (Leyro, Zvolensky, & Bernstein, 2010). This construct consists of several appraisals of experiencing negative emotion, including tolerability, perception of acceptability, attentional interference, and regulation of the emotion (Simons & Gaher, 2005). Distress tolerance can also influence specific regulation strategies used to manage negative emotion (Simons & Gaher, 2005). Distress tolerance inversely correlates with negative affectivity and depressive symptoms (Leyro et al., 2010), as well as anxiety and anxiety sensitivity (Keough, Riccardi, Timpano, Mitchell, & Schmidt, 2010). It is inversely related to addictive behaviors - primarily substance-related disorders (reviewed in Leyro et al., 2010). Skues, Williams, Oldmeadow, and Wise (2016) found distress tolerance inversely related to problematic internet use (PIU). While PSU has not been explored in relation to distress tolerance, of relevance, emotion regulation has been explored. Specifically, two studies found adaptive emotion regulation inversely related to PSU (Elhai et al., 2016; Elhai, Tiamiyu, Weeks, et al., in press). Also relevant to emotion regulation is mindfulness. Mindfulness involves being open, attentive to and aware of the present moment (Bishop et al., 2004). Mindfulness can offset negative emotion and psychopathology because being aware and attentive lets one focus on his/her basic emotional needs (Brown & Ryan, 2003). Mindfulness has also been conceptualized as a way to regulate emotion, specifically through engagement rather than suppression (Chambers, Gullone, & Allen, 2009). Meta-

ACCEPTED MANUSCRIPT 6

analyses of clinical trials demonstrate that mindfulness interventions reduce depression and anxiety disorder symptoms (Hedman-Lagerlof, Hedman-Lagerlof, & Ost, in press; Spijkerman, Pots, & Bohlmeijer, 2016). Some evidence suggests that mindfulness is inversely associated with addictive disorders, including substance use (Bowen & Enkema, 2014; Fernandez, Wood, Stein, & Rossi, 2010). Because emotional avoidance strategies such as rumination and emotional dysregulation are related to PSU (Elhai et al., 2016; Elhai, Tiamiyu, & Weeks, in press; Elhai, Tiamiyu, Weeks, et al., in press), on the contrary, increased mindful awareness should be related to decreased PSU. Also mindfulness is reduced when an individual engages in automatic or habitual behavior (Deci & Ryan, 1980), such as is the case with habitual engagement in automatic, phone-checking behavior (Oulasvirta, Rattenbury, Ma, & Raita, 2012; van Deursen, Bolle, Hegner, & Kommers, 2015). In fact, Arslan (2017) found mindfulness inversely related to PIU; but mindfulness has not been studied in relation to PSU. 1.1. Purpose Our purpose was to a) test both distress tolerance and mindfulness in relation to levels of PSU, and b) test distress tolerance and mindfulness as mediators in the relationship between both depression and anxiety sensitivity with PSU. We tested our research questions in a sample of college students measured at two time-points, separated by a one-month interval, with PSU severity as the dependent variable measured at the second time-point. Most studies on PSU have used strictly cross-sectional correlational designs, collecting data measured at only one timepoint (reviewed in Elhai, Dvorak, et al., 2017). Thus, the advantage of our design

ACCEPTED MANUSCRIPT 7

was the ability to test psychopathology-related variables and smartphone use frequency predicting later, short-term (after one month) PSU severity. This issue is significant, because numerous (albeit cross-sectional) studies on PSU include smartphone use frequency as an intervening variable between psychopathology and PSU (Elhai, Levine, Dvorak, & Hall, 2017; Elhai, Tiamiyu, & Weeks, in press; J. Kim, Seo, & David, 2015; van Deursen et al., 2015). Yet, increased and habitual smartphone use can grow into PSU (Oulasvirta et al., 2012; van Deursen et al., 2015), and thus assessing PSU severity at a subsequent time-point to smartphone use frequency appears warranted. 1.2. Theory Compensatory internet use theory (CIUT) (Kardefelt-Winther, 2014) is relevant in conceptualizing our research questions. CIUT attempts to understand PIU based on psychological variables. CIUT theorizes that adverse life events and stressors motivate some people to excessively use internet media as a means to cope with their negative emotion. Thus, PIU is conceptualized in CIUT as a compensatory, regulation strategy for managing negative emotion and psychopathology. Several studies have discovered empirical support for CIUT in explaining PSU (Elhai & Contractor, 2018; Elhai, Tiamiyu, & Weeks, in press; Long et al., 2016; Wang, Wang, Gaskin, & Wang, 2015; ZhitomirskyGeffet & Blau, 2016). CIUT would conceptualize depression and anxiety sensitivity, in our research model, as driving PSU. CIUT would also conceptualize people with increased distress tolerance and mindfulness as having less motivation to engage in PSU.

ACCEPTED MANUSCRIPT 8

Theory on emotion regulation is relevant to this paper. Gross (1998) conceptualized emotion regulation as a regulatory process people use to adapt to their environment. Gross (1998) discussed how people can use adaptive or maladaptive emotion regulation strategies. Adaptive emotion regulation includes focusing on and processing negative emotion, while maladaptive regulation involves suppressing emotion. Gross (1998) discussed behavioral addictions such as substance use as one (albeit maladaptive) method of regulating emotion. In fact, adaptive emotion regulation strategies can buffer psychopathology, including behavioral addictions (Aldao, NolenHoeksema, & Schweizer, 2010; Weiss, Sullivan, & Tull, 2015). In our study, distress tolerance and mindfulness, both of which are relevant to emotion regulation, should buffer the impact of depressive and anxious psychopathology on PSU. Prior work has found that more adaptive emotion regulation is associated with decreased PSU (Elhai, Tiamiyu, Weeks, et al., in press), and mediates relations between depression severity and PSU (Elhai et al., 2016). 1.3. Research Model Figure 1 shows our research model. All variables were assessed at a baseline time of measurement, and approximately one month later. In Figure 1, we analyzed baseline measured variables, but analyzed PSU severity from the subsequent one-month time-point. Depression and anxiety sensitivity were predicted to correlate (inversely) with both distress tolerance and mindfulness. Distress tolerance and mindfulness were predicted to correlate (inversely) with smartphone use frequency, and with PSU severity one month later. Finally, age and sex served as covariates of PSU in the model, as PSU is associated with younger age (Lu et al., 2011;

ACCEPTED MANUSCRIPT 9

van Deursen et al., 2015) and female sex (Jeong, Kim, Yum, & Hwang, 2016; Wang et al., 2015). 1.4. Hypotheses Distress tolerance involves adaptive tolerance of negative emotion (Leyro et al., 2010). According to CIUT (Kardefelt-Winther, 2014), PIU and PSU are methods of coping with negative emotion. Therefore, people higher in distress tolerance should express a decreased need for managing negative emotion that would drive PSU. Distress tolerance is inversely associated with addictive disorders (Leyro et al., 2010), and was found to inversely correlate with PIU (Skues et al., 2016). H1.

Distress tolerance should inversely predict levels of PSU after one month.

Mindfulness buffers and regulates negative emotion (Brown & Ryan, 2003; Chambers et al., 2009). So as with distress tolerance, CIUT would predict that people engaging in more mindfulness strategies should have less need for PSU as a means of regulating negative emotion. Also, mindfulness involves thoughtful awareness and less automatized behavior, whereas the automatized and habitual behavior of checking one’s smartphone is more characteristic of individuals engaging in PSU (Oulasvirta et al., 2012). Additionally, Arslan (2017) discovered mindfulness was inversely associated with PIU. H2.

Mindfulness should inversely predict PSU severity after one month.

Distress tolerance involves enduring and regulating negative emotion (Leyro et al., 2010). Consistent with emotion regulation theory (Gross, 1998), because distress tolerance buffers the impact of anxious/depressive psychopathology, and itself is inversely associated with behavioral addictions (Leyro et al., 2010), it should mediate

ACCEPTED MANUSCRIPT 10

relations between depression and anxiety sensitivity with PSU. In fact, prior studies have demonstrated that adaptive emotion regulation mediates relations between severity of depression and PSU (Elhai et al., 2016). H3.

Distress tolerance should mediate relations between both depression

(H3a) and anxiety sensitivity (H3b) with PSU severity. Mindfulness buffers and regulates negative emotion (Brown & Ryan, 2003; Chambers et al., 2009). Consistent with emotion regulation theory (Gross, 1998), because mindfulness buffers the impact of depressive/anxious symptoms (Brown & Ryan, 2003; Chambers et al., 2009), and with evidence that mindfulness is inversely associated with behavioral addictions (Bowen & Enkema, 2014; Fernandez et al., 2010), it should mediate relations between depression and anxiety sensitivity with PSU. In fact, Arslan (2017) found that mindfulness inversely correlated with PIU, and mindfulness mediated relations between adverse stressors and PIU. H4.

Mindfulness should mediate relations between both depression (H4a) and

anxiety sensitivity (H4b) with PSU severity. 2. Method 2.1. Participants and Procedure In 2017, we recruited participants from the research pool of the psychology department at a large, Midwestern University. The study was listed on the department’s Sona Systems web portal of available studies for enrollment in exchange for required course research points. We obtained institutional review board approval, and advertised the study as a two-part study involving a 30-minute web survey inquiring about “use of electronic devices, and your emotions,” and a follow-up survey about one month later.

ACCEPTED MANUSCRIPT 11

Individuals enrolling were routed to an online consent statement, (and for those consenting) subsequently routed to the web survey (Time 1 data). A total of 302 participants enrolled in the study. Each participant was contacted by email approximately one month after web survey completion, instructed to complete the entire web survey again (Time 2 data). The interval of one month was used so that we had two full months after the semester started to recruit participants for the Time 1 assessment, while allowing them to complete the Time 2 assessment one month later, in that same 3.5-month semester. We used a set of non-meaningful questions to match a participant’s Time 1 and Time 2 data, such as shoe size and first four letters of the student’s mother’s maiden name. Of 302 participants, 35 individuals did not participate in Time 2’s survey. Another 6 individuals’ Time 2 data could not be matched with their Time 1 data because we could not locate common responses to these questions between the datasets. The remaining 261 participants served as the effective sample. Among the effective sample, the average time delay between Time 1 and Time 2 survey dates was 34.08 days (SD = 6.74)1. Average age was 19.73 years (SD = 3.52). Roughly three-quarters of participants were women (n = 200, 76.9%). A majority were freshman (n = 150, 57.7%) or sophomores (n = 73, 28.1%). Most participants identified as Caucasian (n = 214, 82.0%), with representation from African American (n = 31, 11.9%), Asian (n = 14, 5.4%), and Hispanic (n = 14, 5.4%) backgrounds (these

1

Because of scheduling errors and contact difficulties for the one-month (Time 2) follow-up assessment with participants, the time-lag between Time 1 and Time 2 assessments ranged from 13 to 54 days. For premature Time 2 data, aside from one participant completing Time 2’s assessment 17 days early, 17 other participants (7% of the sample) completed Time 2 between 5 to 7 days early. For overdue Time 2 data, 45 participants (17%) completed Time 2’s assessment between 10 to 20 days late, with 2 participants (<1%) completing it between 21 to 24 days late.

ACCEPTED MANUSCRIPT 12

designations were not mutually exclusive). Employment status was reported as parttime for a slight majority (n = 140, 53.6%), full-time (n = 24, 9.2%), or unemployed (n = 97, 37.2%). 2.2. Instruments We first queried demographics, including age, sex, college class, race, ethnicity, and employment status. The following surveys were administered online at both Time 1 and Time 2. However, as stated above, in order to model the first timepoint’s variables in predicting later, short-term PSU severity, we analyzed PSU only using the second time-point’s data. Item responses on each scale were summed to form total scores (reverse scoring items when appropriate). 2.2.1. Smartphone use frequency. We administered the scale by Elhai et al. (2016), querying frequency of using 11 smartphone features. Response options range from “1 = Never” to “6 = Very often.” The features queried included “Voice/video calls (making and receiving),” “Texting/instant messaging (sending and receiving),” “Email (sending and receiving),” “Social networking sites,” “Internet/websites,” “Music/podcasts/radio,” “Games,” “Taking pictures or videos,” “Watching video/TV/movies,” “Reading books/magazines,” and “Maps/navigation.” Elhai et al. (2016) found adequate internal consistency for the scale. The present sample’s Cronbach alpha was .78. 2.2.2. Smartphone Addiction Scale-Short Version (SAS-SV). We used Kwon, Cho, and Yang’s (2013) SAS-SV to measure current PSU. The SAS-SV is a 10-item scale, with response options ranging from “1 = Strongly disagree” to “6 = Strongly agree.” We reworded several items into a first-person voice for greater accessibility and

ACCEPTED MANUSCRIPT 13

consistency for participants, such as rewording “Missing planned work due to smartphone use” to “I missed planned work due to smartphone use,” as done previously with this scale (Duke & Montag, 2017). Items measure smartphone use-related health and social impairment, withdrawal and tolerance. Coefficient alpha is adequate (Kwon et al., 2013). Scores demonstrate relations with other measures of PIU and PSU (Kwon et al., 2013), and with smartphone use frequency measured by self-report (LopezFernandez, 2017) and objectively (Elhai, Tiamiyu, Weeks, et al., in press). Cronbach’s alpha was .91 in our sample. 2.2.3. Patient Health Questionnaire–9 (PHQ–9). We administered the PHQ–9 (Spitzer, Kroenke, Williams, & the Patient Health Questionnaire Primary Care Study Group, 1999), a nine-item scale of depression symptoms experienced over the past two weeks. Response options range from “0 = Not at all” to “3 = Nearly every day.” Internal consistency is adequate (Manea, Gilbody, & McMillan, 2015), with convergent validity, and diagnostic validity for detecting depression (Manea et al., 2015). Cronbach’s alpha was .89 in our sample. 2.2.4. Anxiety Sensitivity Index-3 (ASI–3). The ASI-3 (Taylor et al., 2007) is a shorter (18-item) version of the original ASI (Peterson & Reiss, 1992). The ASI-3 has response options ranging from “0 = Very little” to “4 = Very much.” Internal consistency is good (Taylor et al., 2007), scores converge with other anxiety measures (Osman et al., 2010; Wheaton, Deacon, McGrath, Berman, & Abramowitz, 2012), and discriminate between anxious and non-anxious participants (Taylor et al., 2007). Cronbach’s alpha was .93 in our sample.

ACCEPTED MANUSCRIPT 14

2.2.5. Distress Tolerance Scale (DTS). The DTS (Simons & Gaher, 2005) is a 15-item scale of perceived distress tolerance, with response options ranging from “1 = Strongly agree” to “5 = Strongly disagree.” Internal consistency is good (Simons & Gaher, 2005), and scores inversely correlate with measures of negative affect and substance use (reviewed in Leyro et al., 2010). Cronbach’s alpha was .93 in our sample. 2.2.6. Mindful Attention Awareness Scale (MAAS). The MAAS (Brown & Ryan, 2003) is a 15-item scale of mindfulness, with response options ranging from “1 = Almost always” to “6 = Almost never.” Internal consistency is adequate (Brown & Ryan, 2003), with convergent validity against scales measuring attentional awareness, selfcontrol, and positive focus (Black, Sussman, Johnson, & Milam, 2012; Brown & Ryan, 2003; Osman, Lamis, Bagge, Freedenthal, & Barnes, 2016). Cronbach’s alpha was .93 in our sample. 2.3. Analysis Among the effective sample, there were nominal amounts of missing item-level data - typically about 5-10% of participants missed 1–2 items on each scale. We used maximum likelihood estimation with the expectation maximization algorithm (Graham, 2009) to estimate missing item-level data, separately for each scale, before summing scale items to produce scale scores. We first conducted descriptive statistics for the primary scale scores. No nonnormality was evidenced, with the largest value (in absolute size) of 1.3 for skewness, and 1.5 for kurtosis. We also created a correlation matrix for our scale scores and covariates.

ACCEPTED MANUSCRIPT 15

We computed a measurement model for the SAS-SV - as the primary dependent variable - using confirmatory factor analysis (CFA), adjusting for age and sex. We specified a one-factor model based on the measure’s 10 items, with support for a unidimensional model found previously (Elhai, Tiamiyu, & Weeks, in press; LopezFernandez, 2017). We treated items as ordinal, and consequently estimated a polychoric covariance matrix, using weighted least squares estimation with a mean- and variance-adjusted chi-square, and estimated factor loadings using probit regression (DiStefano & Morgan, 2014). We scaled the latent factor by fixing its factor variance to 1, freely estimating all factor loadings. We fixed all residual error covariances to zero. Fit indices reported are the comparative fit index (CFI), Tucker-Lewis Index (TLI), and root mean square error of approximation (RMSEA), with CFI and TLI > .89 and RMSEA < .09 typically indicating adequate fit (Hu & Bentler, 1999). We tested the model in Figure 1 using structural equation modeling (SEM)2. As we only had 261 participants, we only modeled the SAS-SV as a latent variable (from Time 2 data); all other psychological constructs were treated as observed scale scores (from Time 1), to preserve statistical power. We used the same estimation approach for SEM as for CFA. We report standardized path estimates. The path from distress tolerance to PSU severity tests Hypothesis 1. The path from mindfulness to PSU tests Hypothesis 2.

2

We tested variations of this model, such as separately removing distress tolerance, mindfulness, depression and anxiety sensitivity. However, model variations resulted in non-convergence, likely because our sample size was not large enough to converge with model variations that resulted in decreased fit.

ACCEPTED MANUSCRIPT 16

We assessed mediation using indirect effect testing by computing cross-products of direct path coefficients. We used the Delta method for estimating the standard error of a given indirect effect. Dividing the indirect effect’s path coefficient by its Deltaestimated standard error produces a z-test for assessing statistical significance of the mediation effect. Because indirect effect estimates obtained this way are not normally distributed on a sampling distribution, we implemented 1000 bootstrapped replications for more accurate estimates. Such indirect testing procedures are discussed elsewhere (MacKinnon, 2008). We tested distress tolerance as a mediator between depression and PSU severity (H3a); we re-computed this mediation test, substituting anxiety sensitivity as the predictor (H3b). We assessed mindfulness as a mediator between depression and PSU severity (H4a); we reconducted this analysis by substituting anxiety sensitivity as the predictor (H4b). 3. Results Descriptive statistics for scale scores appear in Table 1. Bivariate correlations among primary variables are also displayed in Table 1, demonstrating that PSU severity (using the SAS-SV) was moderately, inversely correlated with both distress tolerance (r = -.37, p < .001; supporting H1) and mindfulness (r = -.42, p < .001; supporting H2). Table 1 also indicates that all psychological scales correlated significantly with PSU severity. Additionally, while PSU severity was not correlated with sex (r = .10, p > .05), it inversely correlated with age (r = -.13, p < .05). Table 1 also shows that depression, anxiety sensitivity, distress tolerance and mindfulness evidenced large intercorrelations with each other, in the expected direction.

ACCEPTED MANUSCRIPT 17

The SAS-SV measurement model demonstrated some evidence for adequate fit, robust 2(53, N = 261) = 211.12, CFI = .96, TLI = .95, RMSEA = .11 (90% CI: .09 to .12). Fit index values for CFI, TLI, and RMSEA were mostly consistent with benchmarks for adequate fit (Hu & Bentler, 1999). Table 2 displays standardized factor loadings. Loadings were uniformly high, with the smallest loading of .68 for the SAS-SV’s item 9. The SEM model from Figure 1 fit well based on established benchmarks (Hu & Bentler, 1999), 2(113, N = 261) = 290.09, CFI = .96, TLI = .95, RMSEA = .08 (90% CI: .07 to .09). Figure 2 displays standardized parameter estimates from the model. Figure 2’s path coefficients demonstrate that (after adjusting for smartphone use frequency, age and sex) distress tolerance ( = -.20, SE = .10, p < .05) and mindfulness ( = -.39, SE = .10, p < .001) (both measured at Time 1) inversely predicted PSU severity (Time 2), supporting H1 and H2, respectively. Although not hypothesized, distress tolerance ( = -.11, SE = .09, p > .05) and mindfulness ( = .01, SE = .09, p > .05) were not significantly associated with smartphone use frequency (Figure 2). Figure 2 also shows that smartphone use frequency at Time 1 predicted PSU severity at Time 2 ( = .14, SE = .06, p < .05), after adjusting for age and sex. Finally, severity of depression and anxiety sensitivity were inversely associated with both distress tolerance and mindfulness (Figure 2). Mediation results are displayed in Table 3, with p values displayed for statistical significance of indirect effects. Distress tolerance did not significantly mediate relations between severity of depression and PSU ( = .09, SE = .07, p > .05), thus failing to support H3a. However, distress tolerance mediated relations between

ACCEPTED MANUSCRIPT 18

anxiety sensitivity and PSU severity ( = .06, SE = .03, p < .05), supporting H3b. Mindfulness mediated relations between depression and PSU severity ( = .22, SE = .09, p < .05), supporting H4a. Mindfulness also mediated relations between anxiety sensitivity and PSU severity ( = .10, SE = .05, p < .05), supporting H4b.3 4. Discussion In the present study, we aimed to test two psychopathology-related variables involving emotion regulation, previously unexplored for relations with PSU – distress tolerance and mindfulness. We also assessed the extent to which distress tolerance and mindfulness mediated relations between both depression and anxiety sensitivity with PSU severity measured one month later. Our measurement of PSU about one month after assessing psychopathology and smartphone use frequency is notable, as most prior PSU research has been cross-sectional (reviewed in Elhai, Dvorak, et al., 2017), thus failing to conceptualize increased smartphone use frequency as progressing to PSU (Oulasvirta et al., 2012; van Deursen et al., 2015). We found support for Hypothesis 1 - distress tolerance inversely predicted levels of PSU one month later. Additionally, we found support for Hypothesis 2 – mindfulness inversely predicted PSU levels one month later. These findings fit with CIUT (KardefeltWinther, 2014), which posits that negative emotion and psychopathology drive excessive internet use. Distress tolerance and mindfulness, however, involve adaptive strategies for regulating negative emotion (Chambers et al., 2009; Leyro et al., 2010).

3

Table 3’s indirect effect coefficients () are positive, whereas the relevant direct effect coefficients from Figure 2 are negative. This is because an indirect effect coefficient is based on the cross-product of two direct coefficients, and multiplying two negative values results in a positive value.

ACCEPTED MANUSCRIPT 19

Thus, those individuals with more adaptive methods of regulating emotion should have less motivation for engaging in PIU/PSU (Kardefelt-Winther, 2014). In fact, better emotion regulation is associated with decreased PSU severity (Elhai et al., 2016; Elhai, Tiamiyu, Weeks, et al., in press). And relatedly, better distress tolerance relates to decreased PIU (Arslan, 2017). Our aims also included exploring distress tolerance and mindfulness as mediators between both severity of depression and anxiety sensitivity with PSU. Distress tolerance and mindfulness are relevant to emotion regulation, and prior work has revealed that emotion regulation mediates relations between depression and PSU severity (Elhai et al., 2016). We found that distress tolerance mediated relations between anxiety sensitivity (H3b) with levels of PSU measured after one month. This finding is consistent with emotion regulation theory (Gross, 1998) whereby emotion regulation is conceptualized and found to buffer the impact of depression and anxiety on behavioral addictions (Leyro et al., 2010). However, distress tolerance did not mediate relations between severity of depression and PSU. Thus, anxiety sensitivity may not solely account for PSU, but distress tolerance may serve as an intermediary variable in this relationship. We also discovered that mindfulness mediated relations between both severity of depression and anxiety sensitivity with PSU measured after one month. These findings also fit with emotion regulation theory (Gross, 1998) in that mindfulness can buffer the impact of anxious and depressive psychopathology on behavioral addictions. Mindfulness is also inversely related to other behavioral addictions (Bowen & Enkema, 2014; Fernandez et al., 2010). With mindfulness being an emotion regulation strategy,

ACCEPTED MANUSCRIPT 20

and emotion regulation mediating relations between depression and PSU severity (Elhai et al., 2016), the significant mediating effect of mindfulness here is not surprising. Findings also fit with CIUT – specifically, because of the psychopathology-buffering impact of emotion regulation, people using such regulation strategies should have less need for PIU. We should also note that PSU severity was correlated inversely with age in bivariate analyses, supporting prior findings (Lu et al., 2011; van Deursen et al., 2015). However, in our SEM model, age no longer remained a significant predictor of PSU. We did not find female sex related to PSU severity in bivariate or SEM models. Sex has been found related to PSU severity in some studies (Jeong et al., 2016; Wang et al., 2015), but not in others (Elhai et al., 2016; Elhai, Levine, et al., 2017; Lopez-Fernandez, 2017; van Deursen et al., 2015). Results have clinical implications in managing excessive technology use. Intervention strategies aimed at decreasing excessive technology use could include techniques to improve emotion regulation strategies, such as mindfulness-based cognitive behavioral therapies (Gu, Strauss, Bond, & Cavanagh, 2015; Hofmann, Sawyer, Witt, & Oh, 2010) and emotion regulation interventions (Mennin & Farach, 2007). The aims of intervention could include increasing attentional awareness, and thus decreasing automatic phone-checking behaviors associated with PSU (Oulasvirta et al., 2012), as well as the processing rather than avoidance of emotion (Gross, 1998). Limitations of this study include that smartphone use and PSU were measured via self-report methods, and validity of self-reported amount of smartphone use has limitations (Andrews, Ellis, Shaw, & Piwek, 2015; Elhai, Tiamiyu, Weeks, et al., in

ACCEPTED MANUSCRIPT 21

press). Additionally, our use of college students has challenges to generalizability. In fact, our sample lacks wide diversity, as the majority of participants were white women. Our sample had only 261 participants, though this sample size is within the ballpark for numerous other studies on PSU (reviewed in Elhai, Dvorak, et al., 2017). Furthermore, we used self-administered assessments of depression and anxiety symptoms, rather than structured diagnostic interviews for diagnosing mood and anxiety disorders. In fact, mean depression and anxiety sensitivity scores were at relatively subclinical levels in this sample. Finally, the use of self-reported mindfulness measures, such as the MAAS, has been criticized (Van Dam et al., 2018). Notwithstanding these limitations, this study presents several advantages and innovations over prior studies examining psychopathology in relation to PSU. These advantages involve use of previously unexplored variables for relations with PSU, including anxiety sensitivity, distress tolerance and mindfulness. Also, the two-wave, repeated measures design is innovative in studying effects of psychopathology from one time-point to levels of PSU at a subsequent time-point, rarely done in this research area. 5. Conclusions Variables involving emotion regulation may be important in understanding how some people with depression or anxiety symptoms engage in excessive use of a smartphone. Such emotion regulation variables analyzed in the present study were distress tolerance and mindfulness, previously unexplored for associations with PSU. Interventions aimed at improving emotion regulation skills, such as enhancing distress tolerance and mindful awareness, may offset the risk of

ACCEPTED MANUSCRIPT 22

excessively engaging in smartphone use to alleviate negative emotion. Future research should explore additional mediating variables between psychopathology and PSU that may serve as risk or resilience factors for excessive technology use.

ACCEPTED MANUSCRIPT 23

Disclosure Section No competing financial interests exist.

ACCEPTED MANUSCRIPT 24

References

Aldao, A., Nolen-Hoeksema, S., & Schweizer, S. (2010). Emotion-regulation strategies across psychopathology: A meta-analytic review. Clinical Psychology Review, 30(2), 217-237. doi: 10.1016/j.cpr.2009.11.004 Andrews, S., Ellis, D. A., Shaw, H., & Piwek, L. (2015). Beyond self-report: Tools to compare estimated and real-world smartphone use. PLoS One, 10(10), e0139004. doi: 10.1371/journal.pone.0139004 Arslan, G. (2017). Psychological maltreatment, forgiveness, mindfulness, and internet addiction among young adults: A study of mediation effect. Computers in Human Behavior, 72, 57-66. doi: 10.1016/j.chb.2017.02.037 Billieux, J., Maurage, P., Lopez-Fernandez, O., Kuss, D. J., & Griffiths, M. D. (2015). Can disordered mobile phone use be considered a behavioral addiction? An update on current evidence and a comprehensive model for future research. Current Addiction Reports, 2(2), 156-162. doi: 10.1007/s40429-015-0054-y Bishop, S. R., Lau, M., Shapiro, S., Carlson, L., Anderson, N. D., Carmody, J., Segal, Z. V., Abbey, S., Speca, M., Velting, D., & Devins, G. (2004). Mindfulness: A proposed operational definition. Clinical Psychology: Science and Practice, 11(3), 230-241. doi: 10.1093/clipsy/bph077

ACCEPTED MANUSCRIPT 25

Black, D. S., Sussman, S., Johnson, C. A., & Milam, J. (2012). Psychometric assessment of the Mindful Attention Awareness Scale (MAAS) among Chinese adolescents. Assessment, 19(1), 42-52. doi: 10.1177/1073191111415365 Bowen, S., & Enkema, M. C. (2014). Relationship between dispositional mindfulness and substance use: Findings from a clinical sample. Addictive Behaviors, 39(3), 532-537. doi: 10.1016/j.addbeh.2013.10.026 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), 822-848. doi: 10.1037/0022-3514.84.4.822 Cazzulino, F., Burke, R. V., Muller, V., Arbogast, H., & Upperman, J. S. (2014). Cell phones and young drivers: A systematic review regarding the association between psychological factors and prevention. Traffic Injury Prevention, 15(3), 234-242. doi: 10.1080/15389588.2013.822075 Chambers, R., Gullone, E., & Allen, N. B. (2009). Mindful emotion regulation: An integrative review. Clinical Psychology Review, 29(6), 560-572. doi: 10.1016/j.cpr.2009.06.005 Cho, H. Y., Kim, D. J., & Park, J. W. (2017). Stress and adult smartphone addiction: Mediation by self-control, neuroticism, and extraversion. Stress and Health, 33(5), 624-630. doi: 10.1002/smi.2749

ACCEPTED MANUSCRIPT 26

Deci, E. L., & Ryan, R. M. (1980). Self-determination Theory: When mind mediates behavior. The Journal of Mind and Behavior, 1(1), 33-43. DeMartini, K. S., & Carey, K. B. (2011). The role of anxiety sensitivity and drinking motives in predicting alcohol use: A critical review. Clinical Psychology Review, 31(1), 169-177. doi: 10.1016/j.cpr.2010.10.001 Demirci, K., Akgonul, M., & Akpinar, A. (2015). Relationship of smartphone use severity with sleep quality, depression, and anxiety in university students. Journal of Behavioral Addictions, 4(2), 85-92. doi: 10.1556/2006.4.2015.010 DiStefano, C., & Morgan, G. B. (2014). A comparison of diagonal weighted least squares robust estimation techniques for ordinal data. Structural Equation Modeling: A Multidisciplinary Journal, 21(3), 1-14. doi: 10.1080/10705511.2014.915373 Duke, E., & Montag, C. (2017). Smartphone addiction, daily interruptions and selfreported productivity. Addictive Behaviors Reports, 6, 90-95. doi: 10.1016/j.abrep.2017.07.002 Elhai, J. D., & Contractor, A. A. (2018). Examining latent classes of smartphone users: Relations with psychopathology and problematic smartphone use. Computers in Human Behavior, 82, 159-166. doi: 10.1016/j.chb.2018.01.010 Elhai, J. D., Dvorak, R. D., Levine, J. C., & Hall, B. J. (2017). Problematic smartphone use: A conceptual overview and systematic review of relations with anxiety and

ACCEPTED MANUSCRIPT 27

depression psychopathology. Journal of Affective Disorders, 207, 251-259. doi: 10.1016/j.jad.2016.08.030 Elhai, J. D., Levine, J. C., Dvorak, R. D., & Hall, B. J. (2016). Fear of missing out, need for touch, anxiety and depression are related to problematic smartphone use. Computers in Human Behavior, 63, 509-516. doi: 10.1016/j.chb.2016.05.079 Elhai, J. D., Levine, J. C., Dvorak, R. D., & Hall, B. J. (2017). Non-social features of smartphone use are most related to depression, anxiety and problematic smartphone use. Computers in Human Behavior, 69, 75-82. doi: 10.1016/j.chb.2016.12.023 Elhai, J. D., Tiamiyu, M. F., & Weeks, J. W. (in press). Depression and social anxiety in relation to problematic smartphone use: The prominent role of rumination. Internet Research. Elhai, J. D., Tiamiyu, M. F., Weeks, J. W., Levine, J. C., Picard, K. J., & Hall, B. J. (in press). Depression and emotion regulation predict objective smartphone use measured over one week. Personality and Individual Differences. doi: 10.1016/j.paid.2017.04.051 Elhai, J. D., Vasquez, J. K., Lustgarten, S. D., Levine, J. C., & Hall, B. J. (in press). Proneness to boredom mediates relationships between problematic smartphone use with depression and anxiety severity. Social Science Computer Review. doi: 10.1177/0894439317741087

ACCEPTED MANUSCRIPT 28

Fernandez, A. C., Wood, M. D., Stein, L. A., & Rossi, J. S. (2010). Measuring mindfulness and examining its relationship with alcohol use and negative consequences. Psychology of Addictive Behaviors, 24(4), 608-616. doi: 10.1037/a0021742 Graham, J. W. (2009). Missing data analysis: Making it work in the real world. Annual Review of Psychology, 60, 549-576. doi: 10.1146/annurev.psych.58.110405.085530 Gross, J. J. (1998). The emerging field of emotion regulation: An integrative review. Review of General Psychology, 2(3), 271-299. doi: 10.1037/1089-2680.2.3.271 Gu, J., Strauss, C., Bond, R., & Cavanagh, K. (2015). How do mindfulness-based cognitive therapy and mindfulness-based stress reduction improve mental health and wellbeing? A systematic review and meta-analysis of mediation studies. Clinical Psychology Review, 37, 1-12. doi: 10.1016/j.cpr.2015.01.006 Hedman-Lagerlof, M., Hedman-Lagerlof, E., & Ost, L. G. (in press). The empirical support for mindfulness-based interventions for common psychiatric disorders: A systematic review and meta-analysis. Psychological Medicine. doi: 10.1017/S0033291718000259 Hofmann, S. G., Sawyer, A. T., Witt, A. A., & Oh, D. (2010). The effect of mindfulnessbased therapy on anxiety and depression: A meta-analytic review. Journal of Consulting and Clinical Psychology, 78(2), 169-183. doi: 10.1037/a0018555

ACCEPTED MANUSCRIPT 29

Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6(1), 1-55. doi: 10.1080/10705519909540118 Jeong, S.-H., Kim, H., Yum, J.-Y., & Hwang, Y. (2016). What type of content are smartphone users addicted to?: SNS vs. games. Computers in Human Behavior, 54, 10-17. doi: 10.1016/j.chb.2015.07.035 Kardefelt-Winther, D. (2014). A conceptual and methodological critique of internet addiction research: Towards a model of compensatory internet use. Computers in Human Behavior, 31, 351-354. doi: 10.1016/j.chb.2013.10.059 Keough, M. E., Riccardi, C. J., Timpano, K. R., Mitchell, M. A., & Schmidt, N. B. (2010). Anxiety symptomatology: The association with distress tolerance and anxiety sensitivity. Behavior Therapy, 41(4), 567-574. doi: 10.1016/j.beth.2010.04.002 Kim, J., Seo, M., & David, P. (2015). Alleviating depression only to become problematic mobile phone users: Can face-to-face communication be the antidote? Computers in Human Behavior, 51(Part A), 440-447. doi: 10.1016/j.chb.2015.05.030 Kim, R., Lee, K. J., & Choi, Y. J. (2015). Mobile phone overuse among elementary school students in Korea: Factors associated with mobile phone use as a behavior addiction. Journal of Addictions Nursing, 26(2), 81-85. doi: 10.1097/JAN.0000000000000074

ACCEPTED MANUSCRIPT 30

Kwon, M., Kim, D. J., Cho, H., & Yang, S. (2013). The smartphone addiction scale: Development and validation of a short version for adolescents. PLoS One, 8(12), e83558. doi: 10.1371/journal.pone.0083558 Lee, Y.-K., Chang, C.-T., Lin, Y., & Cheng, Z.-H. (2014). The dark side of smartphone usage: Psychological traits, compulsive behavior and technostress. Computers in Human Behavior, 31, 373-383. doi: 10.1016/j.chb.2013.10.047 Leyro, T. M., Zvolensky, M. J., & Bernstein, A. (2010). Distress tolerance and psychopathological symptoms and disorders: A review of the empirical literature among adults. Psychological Bulletin, 136(4), 576-600. doi: 10.1037/a0019712 Long, J., Liu, T. Q., Liao, Y. H., Qi, C., He, H. Y., Chen, S. B., & Billieux, J. (2016). Prevalence and correlates of problematic smartphone use in a large random sample of Chinese undergraduates. BMC Psychiatry, 16, 408. doi: 10.1186/s12888-016-1083-3 Lopez-Fernandez, O. (2017). Short version of the Smartphone Addiction Scale adapted to Spanish and French: Towards a cross-cultural research in problematic mobile phone use. Addictive Behaviors, 64, 275-280. doi: 10.1016/j.addbeh.2015.11.013 Lu, X., Watanabe, J., Liu, Q., Uji, M., Shono, M., & Kitamura, T. (2011). Internet and mobile phone text-messaging dependency: Factor structure and correlation with dysphoric mood among Japanese adults. Computers in Human Behavior, 27(5), 1702-1709. doi: 10.1016/j.chb.2011.02.009

ACCEPTED MANUSCRIPT 31

MacKinnon, D. P. (2008). Introduction to statistical mediation analysis. New York, New York: Routledge Academic. Manea, L., Gilbody, S., & McMillan, D. (2015). A diagnostic meta-analysis of the Patient Health Questionnaire-9 (PHQ-9) algorithm scoring method as a screen for depression. General Hospital Psychiatry, 37(1), 67-75. doi: 10.1016/j.genhosppsych.2014.09.009 Mennin, D. S., & Farach, F. (2007). Emotion and evolving treatments for adult psychopathology. Clinical Psychology: Science and Practice, 14(3), 329-352. doi: 10.1111/j.1468-2850.2007.00094.x Olatunji, B. O., & Wolitzky-Taylor, K. B. (2009). Anxiety sensitivity and the anxiety disorders: A meta-analytic review and synthesis. Psychological Bulletin, 135(6), 974-999. doi: 10.1037/a0017428 Osman, A., Gutierrez, P. M., Smith, K., Fang, Q., Lozano, G., & Devine, A. (2010). The Anxiety Sensitivity Index-3: Analyses of dimensions, reliability estimates, and correlates in nonclinical samples. Journal of Personality Assessment, 92(1), 4552. doi: 10.1080/00223890903379332 Osman, A., Lamis, D. A., Bagge, C. L., Freedenthal, S., & Barnes, S. M. (2016). The Mindful Attention Awareness Scale: Further Examination of Dimensionality, Reliability, and Concurrent Validity Estimates. Journal of Personality Assessment, 98(2), 189-199. doi: 10.1080/00223891.2015.1095761

ACCEPTED MANUSCRIPT 32

Oulasvirta, A., Rattenbury, T., Ma, L., & Raita, E. (2012). Habits make smartphone use more pervasive. Personal and Ubiquitous Computing, 16(1), 105-114. doi: 10.1007/s00779-011-0412-2 Peterson, R. A., & Reiss, S. (1992). Anxiety Sensitivity Index revised manual. Worthington, Ohio: IDS Publishing. Reiss, S., & McNally, R. J. (1985). Expectancy model of fear. In S. Reiss & R. R. Bootzin (Eds.), Theoretical issues in behavior therapy (pp. 107-121). San Diego, CA: Academic Press. Shan, Z., Deng, G., Li, J., Li, Y., Zhang, Y., & Zhao, Q. (2013). Correlational analysis of neck/shoulder pain and low back pain with the use of digital products, physical activity and psychological status among adolescents in Shanghai. PLoS One, 8(10), e78109. doi: 10.1371/journal.pone.0078109 Simons, J., & Gaher, R. M. (2005). The Distress Tolerance Scale: Development and validation of a self-report measure. Motivation and Emotion, 29(2), 83–102. doi: 10.1007/s11031-005-7955-3 Skues, J., Williams, B., Oldmeadow, J., & Wise, L. (2016). The effects of boredom, loneliness, and distress tolerance on problem internet use among university students. International Journal of Mental Health and Addiction, 14(2), 167-180. doi: 10.1007/s11469-015-9568-8

ACCEPTED MANUSCRIPT 33

Smetaniuk, P. (2014). A preliminary investigation into the prevalence and prediction of problematic cell phone use. Journal of Behavioral Addictions, 3(1), 41-53. doi: 10.1556/JBA.3.2014.004 Spijkerman, M. P., Pots, W. T., & Bohlmeijer, E. T. (2016). Effectiveness of online mindfulness-based interventions in improving mental health: A review and metaanalysis of randomised controlled trials. Clinical Psychology Review, 45, 102114. doi: 10.1016/j.cpr.2016.03.009 Spitzer, R. L., Kroenke, K., Williams, J. B. W., & the Patient Health Questionnaire Primary Care Study Group. (1999). Validation and utility of a self-report version of the PRIME-MD: The PHQ primary care study. JAMA, 282(18), 1737-1744. doi: 10.1001/jama.282.18.1737 Taylor, S. (2014). Anxiety sensitivity: Theory, research, and treatment of the fear of anxiety. London, England: Routledge. Taylor, S., Cox, B., Heimberg, R. G., Roth Ledley, D., Holaway, R. M., Stewart, S. H., Eng, W., Arrindell, W. A., Zvolensky, M. J., Deacon, B. J., Abramowitz, J. S., Sandin, B., Coles, M. E., Daly, E., Bouvard, M., & Cardenas, S. J. (2007). Robust dimensions of anxiety sensitivity: Development and initial validation of the Anxiety Sensitivity Index-3. Psychological Assessment, 19(2), 176-188. doi: 10.1037/1040-3590.19.2.176

ACCEPTED MANUSCRIPT 34

Thompson, L. L., Rivara, F. P., Ayyagari, R. C., & Ebel, B. E. (2013). Impact of social and technological distraction on pedestrian crossing behaviour: An observational study. Injury Prevention, 19(4), 232-237. doi: 10.1136/injuryprev-2012-040601 Van Dam, N. T., van Vugt, M. K., Vago, D., Schmalzl, L., Saron, C. D., Olendzki, A., Meissner, T., Lazar, S. W., Kerr, C. E., Gorchov, J., Fox, K. C. R., Field, B. A., Britton, W. B., Brefczynski-Lewis, J. A., & Meyer, D. E. (2018). Mind the hype: A critical evaluation and prescriptive agenda for research on mindfulness and meditation. Perspectives on Psychological Science, 13(1), 36-61. doi: 10.1177/1745691617709589 van Deursen, A. J. A. M., Bolle, C. L., Hegner, S. M., & Kommers, P. A. M. (2015). Modeling habitual and addictive smartphone behavior: The role of smartphone usage types, emotional intelligence, social stress, self-regulation, age, and gender. Computers in Human Behavior, 45, 411-420. doi: 10.1016/j.chb.2014.12.039 Wang, J.-L., Wang, H.-Z., Gaskin, J., & Wang, L.-H. (2015). The role of stress and motivation in problematic smartphone use among college students. Computers in Human Behavior, 53, 181-188. doi: 10.1016/j.chb.2015.07.005 Weiss, N. H., Sullivan, T. P., & Tull, M. T. (2015). Explicating the role of emotion dysregulation in risky behaviors: A review and synthesis of the literature with directions for future research and clinical practice. Current Opinion in Psychology, 3, 22-29. doi: 10.1016/j.copsyc.2015.01.013

ACCEPTED MANUSCRIPT 35

Wheaton, M. G., Deacon, B. J., McGrath, P. B., Berman, N. C., & Abramowitz, J. S. (2012). Dimensions of anxiety sensitivity in the anxiety disorders: Evaluation of the ASI-3. Journal of Anxiety Disorders, 26(3), 401-408. doi: 10.1016/j.janxdis.2012.01.002 Wolniewicz, C. A., Tiamiyu, M. F., Weeks, J. W., & Elhai, J. D. (in press). Problematic smartphone use and relations with negative affect, fear of missing out, and fear of negative and positive evaluation. Psychiatry Research. doi: 10.1016/j.psychres.2017.09.058 Xie, Y., Szeto, G. P., Dai, J., & Madeleine, P. (2016). A comparison of muscle activity in using touchscreen smartphone among young people with and without chronic neck-shoulder pain. Ergonomics, 59(1), 61-72. doi: 10.1080/00140139.2015.1056237 Zhitomirsky-Geffet, M., & Blau, M. (2016). Cross-generational analysis of predictive factors of addictive behavior in smartphone usage. Computers in Human Behavior, 64, 682-693. doi: 10.1016/j.chb.2016.07.061

ACCEPTED MANUSCRIPT 36

Figure 1. Hypothesized model. Notes: PSU=Problematic smartphone use. Circles represent latent variables; squares represent observed variables. For visual clarity, the latent PSU variable’s observed items are not pictured.

AC C

EP

TE

D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

37

Table 1. Correlations among primary variables, and descriptive statistics.

Variable 1. Sex 2. Age 3. Depression 4. ASI-3 5. DTS 6. MAAS 7. SUF 8. SAS-SV

M N/A 19.73 6.05 17.09 48.92 57.47 49.63 26.31

SD N/A 3.52 5.58 14.52 12.74 16.08 7.50 10.35

1 -.06 .11 .10 -.23*** -.16** .21** .10

2

3

4

5

6

7

8

-.07 .05 -.01 .09 -.07 -.13*

.59*** -.51*** -.57*** .09 .46***

-.53*** -.55*** .07 .34***

.49*** -.15* -.37***

-.07 -.42***

.19**

-

Note: ASI-3=Anxiety Sensitivity Index-3; DTS=Distress Tolerance Scale; MAAS=Mindful Awareness Attention Scale; SUF=Smartphone Use Frequency; SAS-SV=Smartphone Addiction Scale-Short Version at Time 2. * p < .05, ** p < .01, *** p < .001

38

Table 2. Standardized factor loadings for the problematic smartphone use measurement model. SAS-SV Item 1. I miss planned work due to smartphone use

Standardized Loading .72

2. I have a hard time concentrating in class, while doing assignments, or while working due to smartphone use 3. I feel pain in the wrists or at the back of the neck while using a smartphone

.72

4. I won’t be able to stand not having a smartphone

.81

5. I feel impatient and fretful when I am not holding my smartphone

.89

6. I have my smartphone in my mind even when I am not using it

.86

7. I will never give up using my smartphone even when my daily life is already greatly affected by it 8. I constantly check my smartphone so as not to miss conversations between other people on Twitter or Facebook 9. I use my smartphone longer than I had intended

.85 .73

10. The people around me tell me that I use my smartphone too much

.72

.73

.68

ACCEPTED MANUSCRIPT 39

Figure 2. Hypothesized model with standardized path coefficients. Notes: PSU=Problematic smartphone use. Circles represent latent variables; squares represent observed variables. For visual clarity, the latent PSU variable’s observed items are not pictured. Factor loadings for the PSU factor are displayed in Table 2. * p < .05, ** p < .01 *** p < .001

AC C

EP

TE

D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

40

Table 3. Mediation results, with standardized estimates displayed. Indirect Effect H3a: Depression->Distress Tolerance->PSU H3b: Anxiety Sensitivity->Distress Tolerance->PSU H4a: Depression->Mindfulness->PSU H4b: Anxiety Sensitivity->Mindfulness->PSU

 .09 .06 .22 .10

SE .07 .03 .09 .05

z 1.33 2.07 2.49 2.02

p .19 .04 .01 .04

ACCEPTED MANUSCRIPT

Highlights     

Distress tolerance inversely predicted problematic smartphone use (PSU). Mindfulness inversely predicted PSU. Mindfulness mediated relations between depression and PSU severity. Mindfulness mediated relations between anxiety sensitivity and PSU severity. Distress tolerance mediated relations between anxiety sensitivity and PSU severity.