Problematic Smartphone Use Influences the Relationship Between Experiential Avoidance and Anxiety

Abstract

Experiential avoidance (EA), an unwillingness to stay in contact with unwanted inner experiences (e.g., emotions, thoughts, and bodily sensations), has been implicated in the development and maintenance of anxiety. Individuals with high levels of EA are more likely to employ maladaptive coping strategies (i.e., avoidance behaviors), which exacerbates emotional distress and anxious arousal. As smartphone ownership has become increasingly common in recent years, problematic smartphone use (PSU) has been suggested to serve as a “safety behavior” in situations in which individuals believe that they might experience emotional discomfort. That is, individuals experiencing emotional distress and/or anxious arousal may overengage in the use of technology to relieve emotional distress. As such, the purpose of this study was to examine PSU as a moderator of the relationship between EA and anxiety. Adult participants (N = 294) recruited through Amazon's Mechanical Turk (MTurk), an online labor market, completed an online battery of self-reported measures. Results indicated that the relationship between EA and anxiety became significantly stronger as PSU increased, thereby suggesting that PSU may exacerbate the effect of EA on anxiety. These findings are a first step toward the development of risk profiles that incorporate EA and PSU. Such risk profiles may be beneficial for early identification and intervention for individuals at high risk for the development of anxiety.

Introduction

Anxiety is one of the most prevalent classes of psychiatric problems in the developed world.^1–4 Given the sizeable emotional, social, and economic burden associated with anxiety, it is important to identify risk and resilience factors for this psychological problem.^5–7 Experiential avoidance (EA)—an unwillingness to stay in contact with unwanted inner experiences (e.g., emotions, thoughts, and bodily sensations)—is one such risk factor.⁸ EA and the maladaptive coping strategies associated with EA are maintained through negative reinforcement.^9,10 Specifically, avoidance of unwanted internal experiences may alleviate short-term emotional distress, but has been found to exacerbate emotional distress and subsequent anxiety pathology in the long term.

A strong association between EA and anxiety and related symptoms has been observed in both clinical and nonclinical samples.^11–14 Although the majority of studies examining the link between EA and anxiety have been cross-sectional, there are longitudinal studies that have provided evidence of temporal precedence.^15,16 For example, EA predicted increases in anxiety at both 1- and 2-year followup assessment sessions in a sample of community adults.¹⁷ Similarly, Spinhoven et al. found that EA predicted anxiety- and fear-related disorders up to 2 years later (e.g., generalized anxiety disorder and social anxiety disorder).¹⁶

Problematic smartphone use (PSU)—the excessive use of a smartphone despite negative functional consequences (e.g., daily life impairments and symptoms resembling those of behavioral addictions)—is an individual difference factor that may influence the degree to which EA contributes to anxiety.¹⁸ A growing literature has found that many smartphone users engage in PSU. PSU is associated with a number of safety and health concerns (e.g., texting while driving and sleep disturbances) and various forms of psychopathology, such as anxiety and depression.^19–21

Theories that focus on the relationship between technology use and psychopathology suggest that smartphone use may serve as a “safety behavior.”^20,22,23 That is, individuals experiencing negative emotions and/or anxious arousal may overengage with their smartphone to relieve emotional distress. Although the habitual use of technology to alleviate emotional distress may work in the short term, it likely maintains and exacerbates emotional distress over time by failing to provide the opportunity for new learning. For example, the individual fails to learn that (a) it is not only acceptable, but also important to occasionally experience emotional discomfort, (b) they can tolerate such emotional discomfort, and (c) predicted catastrophic outcomes do not occur when they do not use technology to avoid emotional discomfort.

Experimental studies have provided support of the temporal relationship between PSU and anxiety.^24,25 For example, Cheever et al. conducted a study in which undergraduate students, grouped by frequency of smartphone use (i.e., low daily users, moderate daily users, and high daily users), were either separated from, or retained, their smartphones during the study.²⁴ State anxiety was assessed at three separate time points including 10, 35, and 60 minutes after smartphone separation or retention. As predicted, only moderate and high daily smartphone users who had been separated from their smartphones reported experiencing significant increases in state anxiety over the course of the study.

As described, individuals with high levels of EA tend to be unwilling to stay in contact with their unwanted inner experiences (e.g., emotions and thoughts).⁸ PSU may be a method used to alleviate emotional distress in the short term, but may exacerbate it over prolonged periods. As such, the excessive use of a smartphone despite negative functional consequences (i.e., PSU) may be particularly problematic among those with relatively higher levels of EA. Among these individuals, PSU may increase the likelihood of developing clinically significant anxiety. As such, the purpose of this study was to examine PSU as a moderator of the relationship between EA and anxiety.

Methods

Participants and procedure

Adult participants who were fluent in English and between the ages of 18 and 64 years were recruited through Amazon's Mechanical Turk (MTurk). MTurk, an online labor market, is a reliable source of high-quality data with a participant pool that is more diverse than student and other Internet samples.^26–28 To ensure the quality of the data, only MTurk users with at least a 95 percent approval rating from requesters and who had completed 50 intelligence tasks in the past were eligible to participate. When accessing the MTurk portal, only eligible participants were able to see the recruitment advertisement and opt to participate in the study. To identify responses that were generated by “bots” (i.e., computer programs that complete online forms automatically),²⁹ participants were asked to provide text responses to open-ended questions. Participants who were identified as bots (n = 78) were excluded from further analysis. In addition, 17 participants who reported that they did not have regular access to a smartphone were removed from the sample. The final sample (N = 294; 52.7 percent female) had an average age of 37.05 years (standard deviation [SD] = 10.27, range = 21–64). Sample characteristics are presented in Table 1.

Table 1.

Sample Descriptive Characteristics

Characteristic
Gender (percent female)	52.7
Race (percent)
White/European American	75.5
Black/African American	16.3
Asian	5.1
American Indian/Alaska Native	1.4
Other	1.7
Ethnicity (percent)
Hispanic/Latino	11.6
Age (years)	37.05 (10.27, 21–64)

Note: N = 294. Values are reported as means along with standard deviation and range in parentheses, unless otherwise indicated.

Study procedures were approved by the local institutional review board. Participants completed the study using Qualtrics, a secure online survey platform. Self-report measures were presented in random order. After completion of self-report measures, participants were debriefed and compensated for their time.

Measures

Brief Experiential Avoidance Questionnaire

The Brief Experiential Avoidance Questionnaire (BEAQ) is a 15-item short form of the 62-item Multidimensional EA Questionnaire (MEAQ).^30,31 The BEAQ includes items that assess behavioral avoidance, distress aversion, distress endurance, suppression, procrastination, and repression/denial. Participants rate the extent to which they agree with each item on a 6-point scale (1 = strongly agree to 6 = strongly disagree), with higher scores indicating higher levels of EA. The BEAQ has demonstrated adequate psychometric properties in past research.³⁰ Internal consistency for the BEAQ in this study was adequate (α = 0.89, M = 50.48, SD = 14.39).

Depression Anxiety Stress Scales

The Anxiety Scale of the Depression Anxiety Stress scales (DASS-21) is composed of seven items (e.g., “I felt scared without any good reason”) that are rated on a 4-point scale (0 = Did not apply to me at all to 3 = Applied to me very much, or most of the time).³² Participants rate items based on the degree to which they applied to them over the past week. The DASS-21 has shown adequate psychometric properties in previous studies.^33,34 Internal consistency of the DASS-21 in this study was adequate (α = 0.92, M = 4.87, SD = 5.53).

Smartphone Addiction Scale-Short Version

The Smartphone Addiction Scale-Short Version (SAS-SV) was used in this study to assess PSU. The SAS-SV is composed of 10 items that are rated on a 6-point scale (1 = Strongly disagree to 6 = Strongly agree).³⁵ Participants rate each item based on the extent to which the item described their smartphone use. Item scores on the SAS-SV are summed, with higher total scores being indicative of PSU. Consistent with common conceptualizations of PSU, which suggest that differences in PSU reflect differences in degree, the total SAS-SV scores were used in a dimensional manner in this study.^35–38 The SAS-SV has demonstrated adequate psychometric properties in previous research.^35,39,40 Internal consistency in this study for the SAS-SV was adequate (α = 0.93, M = 25.08, SD = 12.65).

Data analytic plan

Regression analysis was used to test the proposed moderation effect. Both the predictor (i.e., BEAQ total) and moderator variables (i.e., SAS total) were mean centered and entered into the first step of the model.⁴¹ An interaction term was calculated as the product of the centered predictor and moderator variables and entered into the second step of the model. Significant interaction effects were explored using a simple slopes analysis.⁴¹ Simple slopes analysis tests the effects of the predictor variable (EA) on the outcome variable (anxiety) at high (+1 SD) and low (−1 SD) values of the moderator variable (PSU). As is standard practice, p < 0.05 was used as the threshold of significance for all analyses.^42,43

Results

In the first step of the regression model (Table 2), EA (p < 0.001) and PSU (p < 0.001) were significantly associated with DASS-21. In the second step of the model, the interaction term (EA by PSU) was significantly associated with DASS-21 (p = 0.02). Simple slopes analysis showed that the positive association between EA and DASS-21 was significantly stronger at higher (β = 0.48, p < 0.001), versus lower (β = 0.29, p < 0.001), levels of PSU (Fig. 1). The interaction effect was small in size (Cohen's f² = 0.02).^41,44

FIG. 1.

Moderating effect of problematic smartphone use on the relationship between experiential avoidance and anxiety.

Table 2.

Hierarchical Linear Regression with Anxiety Symptoms as the Outcome Variable

Predictor	Anxiety (DASS-21—Anxiety)
	ΔR²	S1	S2
	ΔR²	β	β
Step 1 (S1)	0.514
EA		0.355	0.383
PSU		0.477	0.424
Step 2 (S2)	0.009
EA × PSU			0.104

Note: N = 294. All values significant at p < 0.05 (two-tailed, alpha-corrected).

DASS-21, Depression Anxiety Stress Scales; EA, experiential avoidance; PSU, problematic smartphone use.

Discussion

These results suggest that engaging in PSU may exacerbate anxiety among those who have higher levels of EA. Moreover, individuals with this specific profile (i.e., high EA and high PSU) may have an elevated risk of developing anxiety. If individuals with this risk profile are able to be identified early on, possibly through institution-wide screenings (e.g., academic testing), intervention options can be made available before anxiety pathology develops or is exacerbated. A number of researchers have proposed educational curriculum as one form of early intervention aimed at targeting PSU.^45–48 Turel et al. found that instructional videos presented to a sample of undergraduate students were efficacious in improving participants views on Internet use reduction.⁴⁹ Similarly, behavioral reinforcement strategies for reducing smartphone use may be important to consider for individuals with this risk profile, as employing restricted smartphone use improves sleep, reduces anxiety, and improves overall well-being.⁵⁰ Furthermore, behavioral interventions aimed at reducing PSU could incorporate components targeting EA to facilitate engagement with new learning and maximize treatment effectiveness for anxiety symptom reduction.

Of note, some people consider PSU a behavioral addiction and, thus, opt to use the term “smartphone addiction” rather than PSU. Although individuals who engage in PSU may exhibit behaviors commonly observed in other behavioral addictions (e.g., continued use despite causing impairment in daily life), some have argued that it may not be appropriate to consider the excessive behaviors that comprise PSU as a construct of psychopathology in the typical behavioral addictions sense.^51,52 Instead, rather than being a behavioral addiction, PSU may serve as a safety behavior that alleviates distress in the short term, but ultimately drives and exacerbates anxiety and other forms of emotional distress over longer periods of time.²⁰ Importantly, PSU is associated with maladaptive behaviors and negative outcomes and, thus, should continue to be examined in relation to psychopathology to better understand the role that it plays in the development and maintenance of psychological problems.

This study is not without limitations. The cross-sectional study design precludes temporal inferences. Although theory suggests that the relationships between PSU and anxiety are bidirectional and mutually reinforcing,²⁰ the role that PSU might play in the development of anxiety over time has received relatively little empirical attention. It will be important to assess EA, PSU, and anxiety at multiple time points in future research to confirm temporal hypotheses and provide further support for this theory.

Although the DASS-21 Anxiety score was used in a manner consistent with a dimensional model of anxiety,⁵³ the present findings may not generalize to clinical samples. Replication of the results using a psychometrically supported clinical interview will be important for ensuring the generalizability of the findings for those with clinically significant anxiety. It is also worth noting that the frequency of smartphone use was not assessed in this study. According to theory, increased frequency and habitual use of one's smartphone may serve as a risk factor for the development of PSU.²⁰ As such, it may be important to include a measure of frequency of smartphone use in this line of research in the future to better understand the relationships among smartphone frequency of use, PSU, EA, and anxiety.

Two additional points are worth considering. First, although the limitations of retrospective power analysis have been well documented,⁵⁴ providing such information may be helpful for informing future research in this area. Based on data from this study, power analysis indicates power of 0.63 with our sample size of 294 and alpha set at 0.05. As such, we recommend using a larger sample in future research to ensure that the results of this study replicate. Based on the results of this study, a sample size of 434 participants would increase power to 0.80 to detect the interaction effect of primary interest when alpha is set at 0.05. Second, although our significant interaction effect was relatively small, the variance in anxiety symptoms accounted for by this effect is well within the range considered meaningful in Monte Carlo studies (i.e., at least 1 percent).⁵⁵

To our knowledge, this study is the first to provide evidence that PSU potentiates the effect of EA on anxiety. These findings provide an important first step in the identification of risk profiles for individuals who may be at higher risk for developing anxiety. If early screening procedures are established, those individuals at risk for developing anxiety can be identified and provided treatment options before symptoms worsen. The development of an early screening procedure and treatment designed for individuals with high levels of EA and PSU may be helpful in reducing anxiety pathology.

Footnotes

Authors' Contributions

J.Y.G. was responsible for the study conceptualization, data analysis and interpretation, and article preparation. J.R.B. was involved with study conceptualization, data collection, and article preparation.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

Bandelow

, Michaelis

. Epidemiology of anxiety disorders in the 21st century. Dialogues in Clinical Neuroscience, 2015; 17:327–335.

Simpson

, Neria

, Lewis-Fernández

, et al. (2010) Anxiety disorders: theory, research and clinical perspectives. Cambridge: Cambridge University Press.

Kessler

, Berglund

, Demler

, et al. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 2005; 62:593–602.

Vos

, Abajobir

, Abate

, et al. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. The Lancet, 2017; 390:1211–1259.

Hoffman

, Dukes

, Wittchen

H-U

. Human and economic burden of generalized anxiety disorder. Depression and Anxiety, 2008; 25:72–90.

Shirneshan

, Bailey

, Relyea

, et al. Incremental direct medical expenditures associated with anxiety disorders for the U.S. adult population: evidence from the Medical Expenditure Panel Survey. Journal of Anxiety Disorders, 2013; 27:720–727.

Gustavsson

, Svensson

, Jacobi

, et al. Cost of disorders of the brain in Europe 2010. European Neuropsychopharmacology, 2011; 21:718–779.

Hayes

, Wilson

, Gifford

, et al. Experiential avoidance and behavioral disorders: a functional dimensional approach to diagnosis and treatment. Journal of Consulting and Clinical Psychology, 1996; 64:1152–1168.

Abramowitz

, Moore

. An experimental analysis of hypochondriasis. Behaviour Research and Therapy, 2007; 45:413–424.

10.

Bardeen

. Short-term pain for long-term gain: the role of experiential avoidance in the relation between anxiety sensitivity and emotional distress. Journal of Anxiety Disorders, 2015; 30:113–119.

11.

Venta

, Sharp

, Hart

. The relation between anxiety disorder and experiential avoidance in inpatient adolescents. Psychological Assessment, 2012; 24:240–248.

12.

Kashdan

, Zvolensky

, McLeish

. Anxiety sensitivity and affect regulatory strategies: individual and interactive risk factors for anxiety-related symptoms. Journal of Anxiety Disorders, 2008; 22:429–440.

13.

Roemer

, Salters

, Raffa

, et al. Fear and avoidance of internal experiences in GAD: preliminary tests of a conceptual model. Cognitive Therapy and Research, 2005; 29:71–88.

14.

Tull

, Gratz

, Salters

, et al. The role of experiential avoidance in posttraumatic stress symptoms and symptoms of depression, anxiety, and somatization. The Journal of Nervous & Mental Disease, 2004; 192:754–761.

15.

Kumpula

, Orcutt

, Bardeen

, et al. Peritraumatic dissociation and experiential avoidance as prospective predictors of posttraumatic stress symptoms. Journal of Abnormal Psychology, 2011; 120:617–627.

16.

Spinhoven

, Drost

, de Rooij

, et al. A longitudinal study of experiential avoidance in emotional disorders. Behavior Therapy, 2014; 45:840–850.

17.

Moroz

, Dunkley

. Self-critical perfectionism, experiential avoidance, and depressive and anxious symptoms over two years: a three-wave longitudinal study. Behaviour Research and Therapy, 2019; 112:18–27.

18.

Billieux

, Maurage

, Lopez-Fernandez

, et al. 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, 2015; 2:156–162.

19.

Elhai

, Dvorak

, Levine

, et al. Problematic smartphone use: a conceptual overview and systematic review of relations with anxiety and depression psychopathology. Journal of Affective Disorders, 2017; 207:251–259.

20.

Elhai

, Levine

, Hall

. The relationship between anxiety symptom severity and problematic smartphone use: a review of the literature and conceptual frameworks. Journal of Anxiety Disorders, 2019; 62:45–52.

21.

Kim

J-H

, Seo

, David

. Alleviating depression only to become problematic mobile phone users: can face-to-face communication be the antidote?. Computers in Human Behavior, 2015; 51:440–447.

22.

Kardefelt-Winther

. A conceptual and methodological critique of internet addiction research: towards a model of compensatory internet use. Computers in Human Behavior, 2014; 31:351–354.

23.

Wolniewicz

, Tiamiyu

, Weeks

, et al. Problematic smartphone use and relations with negative affect, fear of missing out, and fear of negative and positive evaluation. Psychiatry Research, 2018; 262:618–623.

24.

Cheever

, Rosen

, Carrier

, et al. Out of sight is not out of mind: the impact of restricting wireless mobile device use on anxiety levels among low, moderate and high users. Computers in Human Behavior, 2014; 37:290–297.

25.

Clayton

, Leshner

, Almond

. The extended iSelf: the impact of iPhone separation on cognition, emotion, and physiology. Journal of Computer-Mediated Communication, 2015; 20:119–135.

26.

Behrend

, Sharek

, Meade

, et al. The viability of crowdsourcing for survey research. Behavior Research Methods, 2011; 43:800.

27.

Chandler

, Shapiro

. Conducting clinical research using crowdsourced convenience samples. Annual Review of Clinical Psychology, 2016; 12:53–81.

28.

Hauser

, Schwarz

. Attentive Turkers: MTurk participants perform better on online attention checks than do subject pool participants. Behavior Research Methods, 2016; 48:400–407.

29.

Yarrish

, Groshon

, Mitchell

, et al. Finding the signal in the noise: minimizing responses from bots and inattentive humans in online research. The Behavior Therapist, 2019; 42:235.

30.

Gámez

, Chmielewski

, Kotov

, et al. The Brief Experiential Avoidance Questionnaire: development and initial validation. Psychological Assessment, 2014; 26:35–45.

31.

Gámez

, Chmielewski

, Kotov

, et al. Development of a measure of experiential avoidance: the Multidimensional Experiential Avoidance Questionnaire. Psychological Assessment, 2011; 23:692–713.

32.

Lovibond

, Lovibond

. The structure of negative emotional states: comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behaviour Research and Therapy, 1995; 33:335–343.

33.

Antony

, Bieling

, Cox

, et al. Psychometric properties of the 42-item and 21-item versions of the Depression Anxiety Stress Scales in clinical groups and a community sample. Psychological Assessment, 1998; 10:176–181.

34.

Brown

, Chorpita

, Korotitsch

, et al. Psychometric properties of the Depression Anxiety Stress Scales (DASS) in clinical samples. Behaviour Research and Therapy, 1997; 35:79–89.

35.

Kwon

, Kim

D-J

, Cho

, et al. The Smartphone Addiction Scale: development and validation of a short version for adolescents. PLoS One, 2013; 8:e83558.

36.

Ching

, Yee

, Ramachandran

, et al. Validation of a Malay version of the Smartphone Addiction Scale among medical students in Malaysia. PLoS One, 2015; 10:e0139337.

37.

Horwood

, Anglim

. Personality and problematic smartphone use: a facet-level analysis using the Five Factor Model and HEXACO frameworks. Computers in Human Behavior, 2018; 85:349–359.

38.

Gugushvili

, Täht

, Rozgonjuk

, et al. Two dimensions of problematic smartphone use mediate the relationship between fear of missing out and emotional well-being. Cyberpsychology: Journal of Psychosocial Research on Cyberspace, 2020; 14:3.

39.

Luk

, Wang

, Shen

, et al. Short version of the Smartphone Addiction Scale in Chinese adults: psychometric properties, sociodemographic, and health behavioral correlates. Journal of Behavioral Addictions, 2018; 7:1157–1165.

40.

Samaha

, Hawi

. Relationships among smartphone addiction, stress, academic performance, and satisfaction with life. Computers in Human Behavior, 2016; 57:321–325.

41.

Aiken

, West

, Reno

. (1991) Multiple regression: testing and interpreting interactions. Thousand Oaks, CA: SAGE.

42.

Cowles

, Davis

. On the origins of the .05 level of statistical significance. American Psychologist, 1982; 37:553–558.

43.

Geher

, Hall

. (2014) Straightforward statistics: understanding the tools of research. Oxford: Oxford University Press.

44.

Cohen

. (1988) Statistical power analysis for the behavioral sciences. 2nd ed. Hillsdale, NJ: Earlbaum.

45.

Busch

, McCarthy

. Antecedents and consequences of problematic smartphone use: a systematic literature review of an emerging research area. Computers in Human Behavior, 2021; 114:106414.

46.

Lian

, You

, Huang

, et al. Who overuses Smartphones? Roles of virtues and parenting style in Smartphone addiction among Chinese college students. Computers in Human Behavior, 2016; 65:92–99.

47.

Liu

Q-Q

, Zhou

Z-K

, Yang

X-J

, et al. Mobile phone addiction and sleep quality among Chinese adolescents: a moderated mediation model. Computers in Human Behavior, 2017; 72:108–114.

48.

Mahapatra

. Smartphone addiction and associated consequences: role of loneliness and self-regulation. Behaviour & Information Technology, 2019; 38:833–844.

49.

Turel

, Mouttapa

, Donato

. Preventing problematic Internet use through video-based interventions: a theoretical model and empirical test. Behaviour & Information Technology, 2015; 34:349–362.

50.

Hughes

, Burke

. Sleeping with the frenemy: how restricting ‘bedroom use’ of smartphones impacts happiness and wellbeing. Computers in Human Behavior, 2018; 85:236–244.

51.

Griffiths

. Behavioural addiction and substance addiction should be defined by their similarities not their dissimilarities: commentaries. Addiction, 2017; 112:1718–1720.

52.

Rozgonjuk

, Levine

, Hall

, et al. The association between problematic smartphone use, depression and anxiety symptom severity, and objectively measured smartphone use over one week. Computers in Human Behavior, 2018; 87:10–17.

53.

Crawford

, Henry

, Crombie

, et al. Normative data for the HADS from a large non-clinical sample. The British Journal of Clinical Psychology, 2001; 40:429–434.

54.

Faul

, Erdfelder

, Lang

A-G

, et al. G*Power 3: a flexible statistical power analysis program for the social, behavioral, and biomedical sciences. Behavior Research Methods, 2007; 39:175–191.

55.

Evans

. A Monte Carlo study of the effects of correlated method variance in moderated multiple regression analysis. Organizational Behavior and Human Decision Processes, 1985; 36:305–323.