Explaining unplanned online media behaviors: Dual system theory models of impulsive use and swearing on social networking sites

Abstract

Unplanned (i.e. spontaneous) online behaviors such as impulsive use of social networking sites (SNS) and swearing on SNS are prevalent and can adversely affect users and society. Drawing on dual system theory, this article conceptualizes and empirically investigates the etiology of such behaviors. Results of two studies (n₁ = 295 SNS users, focuses on impulsive use; n₂ = 336 SNS users, focuses on swearing) show that both of these online behaviors are driven, in part, by cognitive-emotional preoccupation with the SNS and inhibited, in part, by cognitive-behavioral control over using the SNS. The inhibition effect is both direct and indirect, through the suppression of the cognitive-emotional preoccupation effects. The findings provide a theoretical lens of analysis through which impulsive and unitneded media use behaviors can be explained.

Keywords

Dark side of media use dual system theory impulsive use social media social networking sites swearing

Introduction

The use of social networking sites (SNSs) such as Facebook has substantially increased during the last decade (Smith, 2016). Given the potential influences of SNS technology on many important aspects of life, SNS use has attracted much research. Many studies point to various possible benefits of SNS use, including aiding the global fight against authoritarianism (Morozov, 2011), reducing social isolation and loneliness (Cattan et al., 2005), and alleviating boredom (Hanna et al., 2011). Nonetheless, it has recently become apparent that SNS use can also have several “dark side” facets, described as use patterns that can result in negative effects on people’s lives (Tarafdar et al., 2015), that people often willfully ignore (Morozov, 2011). These problematic use patterns can include unplanned SNS use behaviors, which are conducted at the spur of moment without much reflection on their consequences (e.g. checking SNS while driving or while doing other important things, just because a new message notification appeared; Turel and Bechara, 2016b). These behaviors can result in various negative effects on people’s lives, such as reduced wellbeing (Turel and Serenko, 2012; Turel et al., 2011a; Turel et al., 2011b), increased violence (Patton et al., 2014), elevated negative emotions (Turel, 2015; Turel, 2016), and diminished academic performance (Qahri-Saremi and Turel, 2016; Turel and Qahri-Saremi, 2016).¹ Thus, it is important to understand the etiology (i.e. root causes) of unplanned SNS use behaviors as a means to develop interventions for mitigating them and their harms.

Hence, this article seeks to theoretically develop and empirically test a model that can explain unplanned SNS use behaviors, with the hope that such a model can serve as a basis for understanding additional unplanned media use behaviors (e.g. texting while driving), the prevalence of which seem to have grown. The model is developed and tested in relation to two mainly unplanned, but common SNS use behaviors: impulsive SNS use (study 1) and swearing on the SNS (study 2). Impulsive SNS use captures spontaneous SNS use without planning to do so (Turel and Bechara, 2016a). This behavior is often performed automatically in response to internal (e.g. urge, memory, feeling) or external (e.g. cellphone beep, new message notification, seeing a computer) stimuli and without having a plan to use SNS. Swearing on the SNS involves the use of words that are typically considered to be profane by society (e.g. the “f-word”; see Ljung, 2011); some of these words may also be offensive, sexually, racially, or in other forms (e.g. the “n-word”), beyond being merely profane. This behavior is mostly unplanned, as most people do not plan to plug in swear words in their posts on SNS; they often do so spontaneously at the spur of the moment, and in retrospect, may regret the use of such words (Ljung, 2011; Rassin and Muris, 2005; Stone et al., 2011).²

It is important to study both of these behaviors because unplanned SNS use can be associated with various adverse phenomena, including deteriorated mental health (Pantic, 2014); misuse of social media, such as while driving (Turel and Bechara, 2016b) or during class time (Turel and Qahri-Saremi, 2016); flaming in online communications (Alonzo and Aiken, 2004; O’Sullivan and Flanagin, 2003); decrease in paralinguistic cues that set the tone of our social interactions (Hayes et al., 2016); cyberbullying; and aggressive online behaviors (Park et al., 2014; Van Cleemput et al., 2014; Vandebosch and Van Cleemput, 2008). Furthermore, swearing online can erode the civility of online communications (Greitemeyer et al., 2012) and promote retaliation in the form of online and offline aggression (Antoci et al., 2016; Rosner et al., 2016). For example, swearing online can be associated with trolling (Thacker and Griffiths, 2012), cyberbullying (Cao and Lin, 2015), and flaming³ (Thompsen and Foulger, 1996), all of which are undesirable aspects of social media use.

We contend that both of these behaviors can be explained using the dual system theory (DST) (Wood and Bechara, 2014), rooted in cognitive psychology and neuroscience research (Baumeister, 2002; Hofmann et al., 2009; Kahneman, 2011). According to DST, unplanned behaviors can be viewed as an outcome of a struggle between two separate cognitive systems. The first system (system 1, impulsive) produces an automatic drive (impulse) to engage in rewarding target behaviors (e.g. using SNS or swearing). The second system (system 2, reflective) mobilizes inhibition resources and tries to block or control the automatic and rewarding behaviors promoted by system 1, when they are deemed disadvantageous or in conflict with important life goals and self-standards. When system 1 is “strong” and system 2 is “weak,” people engage in unplanned and often disadvantageous behaviors (Turel and Qahri-Saremi, 2016). We accordingly theorize that unplanned SNS use behaviors are driven by system 1 and inhibited by system 2 (both directly and via suppression). We employ established broad conceptualizations of systems 1 and 2, which provide generalizable, theory-based, and overarching manifestations of the two systems. This approach can serve as a springboard for further research on spontaneous or less-planned social media use behaviors.

Theory and hypotheses

Many theories and models exploring user behaviors on online media focus on planned behaviors, that is, behaviors for which users consciously develop intentions based on calculative reflections on attitudes, consequences, feelings, abilities, and social norms. Examples include the Theory of Planned Behavior (Ajzen, 1991) and Protection Motivation Theory (Rogers, 1975). However, such theories do not predict well behaviors that less planned, for which people do not always develop intentions (Gibbons et al., 1998). For example, while users can certainly plan many instances of SNS use, there are other instances of SNS use that seem unplanned and irrational, such as using SNS while driving (why would a person rationally endanger own and others’ lives by using SNS while driving?). In such cases, alternative theories that explain unplanned and sometimes irrational behaviors can be more relevant. Such theories often take into account both the reflective-calculative path in one’s mind, in conjunction with the irrational-automatic path (Gibbons et al., 1998).

Consequently, a family of theories focusing on this dual-system perspective emerged (Evans, 2006); the term DST is used to describe the communal view of these theories and models. These theories used various and inconsistent terms to describe specific components or functions of the two systems (Evans, 2009), for example, automatic affective reactions versus self-control (Friese and Hofmann, 2009; Hofmann et al., 2009), use habit versus self-regulation (Soror et al., 2015), heuristic processing versus systematic processing (Chaiken, 1980), and automatic stereotyping versus suppression (Devine, 1989). In this study, rather than a-theoretically focusing on specific manifestations for these systems, we treat DST more holistically and theorize on and employ overarching manifestations of the two systems (systems 1 and 2). It should be noted that recent studies suggest that there is a third system in place, namely, a system that governs interoceptive awareness, which can alter the balance between systems 1 and 2 (Wood and Bechara, 2014). By focusing only on the dual system, without its extensions, this study provides an initial examination of the viability of DST to explain some media use behaviors that are less planned.

DST includes two core ideas. The first is that human behavior is driven by a tug-of-war between two cognitive systems that have different roles (Evans, 2008; Strack and Deutsch, 2004). The second idea is the assignment of specific roles and attributes to these two cognitive systems. The first system (impulsive, reactive; system 1) is an automatic cognitive system that generates impulsions to engage in a rewarding behavior (e.g. use an SNS). It records mental associations between stimuli, behaviors, and rewards. It creates expectations for dopamine release, which makes people feel good; the recorded associations therefore motivate immediate action and drive impulsions (Meshi et al., 2015). These associations are retrieved efficiently, quickly, and sometimes subconsciously at the mere exposure to a stimulus (e.g. a new SNS message notification). The second system (reflective, controlled, inhibition; system 2) reflects on the generated impulsion and tries to inhibit it, in light of other conflicting goals (e.g. avoid an impulsion to eat a high calorie food, if another life goal is to lose weight) (Turel and Bechara, 2016a). The DST systems have corresponding brain netwroks, which have been shown to be involved also in social media-related impulsions and self-control attempts (He et al., 2017a; He et al., 2017b, Turel et al., 2017; Turel et al., 2014).

Prior research has demonstrated the usefulness of DST in explaining various unplanned behaviors such as overeating and smoking (see Evans, 2008, for a review). These studies have also demonstrated that while unplanned behaviors may differ, they share similar etiology rooted in system 1 and system 2 activations (Evans, 2008). This etiology involves experiencing strong impulsions to engage in the behavior (i.e. system 1 is “strong”) and inability to exert enough inhibitory power to restrict or prevent the behavior (i.e. system 2 is “weak”), even though the behavior is judged, in retrospect, as disadvantageous or problematic (He et al., 2017a; He et al., 2017b; Turel et al., 2014). This refers to an “imbalance” between systems 1 and 2 that can arguably explain different unplanned and possibly problematic SNS use behaviors, such as its misuse in inappropriate situations, swearing, flaming, retaliation in cyberbullying, and trolling, because such behaviors are mainly unplanned and can stem from a strong drive to engage in the behavior (e.g. when one receives a rude message that bothers him or her) together with weak abilities to control this drive.

Prior studies have used a variety of different manifestations of the two DST systems (e.g. habit versus self-regulation, see Soror et al., 2015). Notwithstanding their contributions, this study takes a systematic and theory-driven perspective for capturing overarching manifestations of systems 1 and 2. According to this perspective, system 1 is manifested behaviorally in cognitive-emotional preoccupation with using the SNS and system 2 manifests in cognitive-behavioral control over using the SNS (Turel and Qahri-Saremi, 2016). Cognitive-emotional preoccupation with using the SNS captures intrusive thoughts and desires toward using the SNS (Collins and Lapp, 1992). Cognitive-behavioral control over using the SNS captures the person’s capacity to control, inhibit, or reduce SNS use thoughts and behaviors that are deemed to be rewarding and are consequently promoted by system 1, but conflict with other life goals or self and normative standards and hence require inhibition (Collins and Lapp, 1992).

The perspective depicting a tug-of-war between these two manifestations as the driver of problematic behaviors received much support across life domains (Hofmann et al., 2012). When the imbalance between the systems is excessive, people can develop deficits in decision making (He et al., 2017a; He et al., 2017b; Turel et al., 2014) that may be diagnosed as mental disorders on the spectrums of compulsive, addictive, and problematic behaviors. One notable application of this idea to media is the notion of Internet Gaming Disorder that is mentioned in the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5) as a future area for potential inclusion (American Psychiatric Association, 2013; Xu et al., 2012).

Applying the core ideas of DST to impulsive SNS use and swearing, we propose the research model depicted in Figure 1.

Figure 1.

Research model.

Drawing on DST, we first contend that unplanned SNS behaviors start with the activation of system 1. This system results in high cognitive-emotional preoccupation when the person has developed strong, oft-subconscious associations between stimuli (e.g. new friend request), the behavior (e.g. using the SNS), and expected rewards (e.g. release of dopamine in the brain) (He et al., 2017a; He et al., 2017b; Turel et al., 2014). The cognitive-emotional preoccupation is a potent driver of SNS use, because engaging in the behavior is a key direct approach to reducing unpleasant intrusive thoughts, such as bothering fear of missing out (i.e. pain avoidance), as well as increasing pleasant outcomes such as status gains and enjoyment (Turel and Qahri-Saremi, 2016). Extending these ideas to the context of unplanned SNS behaviors, we propose the following:

H1. Cognitive-emotional preoccupation with using an SNS is positively associated with (a) impulsive use of the SNS and (b) swearing on the SNS.

We next contend that once system 1 is engaged and generates cognitive-emotional preoccupation with using the SNS, system 2 considers these impulsions, their relation to other short-term and long-term goals, and decides whether it should and can inhibit the desire promoted by system 1 (Wood and Bechara, 2014). One way to block suboptimal action is to communicate with motor cortices and reduce the action potential (Noel et al., 2013). When system 2 fails to do so and consequently is unable to control impulsions, people engage in unplanned behaviors (He et al., 2017a; He et al., 2017b; Turel et al., 2014). In contrast, when one’s cognitive-behavioral control abilities are strong, he or she can resist temptations to engage in the rewarding behavior (He et al., 2017a; He et al., 2017b; Turel et al., 2014). Extending these ideas to the context of unplanned behaviors, we propose the following:

H2. Cognitive-behavioral control over using an SNS is negatively associated with (a) impulsive use of the SNS and (b) swearing on the SNS.

Finally, we propose that system 2 also moderates (suppresses, reduces) system 1 effects on unplanned behavior (Friese and Hofmann, 2009; Lindgren et al., 2014). This happens, because system 2 can reflect upon and control desires cognitively (Noel et al., 2013). Thus, people with strong cognitive-behavioral control abilities can suppress intrusive thoughts and cognitively convince themselves that the target behavior is not worthy of action (Wood and Bechara, 2014). This suppression effect has been demonstrated in various contexts, such as smoking (Collins and Lapp, 1992). Extending these findings to the context of unplanned SNS behaviors, we propose the following:

H3. Cognitive-behavioral control over using an SNS negatively moderates (suppresses) the positive association between cognitive-emotional preoccupation with using an SNS and (a) impulsive use of the SNS and (b) swearing on the SNS.

Study 1

This study focused on investigating impulsive use of the SNS. Two-wave data (t₁ and t₂, 1 week apart) were collected from 295 SNS users who were undergraduate students at a large North American university. The sample was split 50/50 between both genders. The average age was 23 years (18–60 years, standard deviation [SD] = 4.48); this sample included mostly young adults who are arguably frequent SNS users and who are more prone to unplanned SNS use behaviors, as compared to older individuals (Turel and Bechara, 2016a). Respondents had an average of 4.2 years of SNS experience (0.2–12 years, SD = 2.4) and were on average active on 3.5 (1–8, SD = 1.74) SNSs, with Facebook as the most prevalent SNS among participants (94%).

Measures: study 1

Measurement items were adapted from well-established and reliable scales (see Appendix 1). Cognitive-emotional preoccupation with using SNS and cognitive-behavioral control over using SNS were measured at t₁ using Collins and Lapp’s (1992) measurement inventory. This inventory was developed for problematic drinking and was adapted to the context of SNS use in this study. Cognitive-emotional preoccupation was operationalized as a second-order factor manifested in three first-order factors: (1) “emotion,” which refers to avoiding or alleviating negative emotions as a drive for using SNS; (2) “cognitive pre-occupation,” which captures persistent distracting thoughts about using SNS; and (3) “govern,” which refers to difficulty in controlling one’s use of SNS. Cognitive-behavioral control was operationalized as a second-order factor manifested in two first-order factors: (1) “concern,” which captures worries about and plans to reduce SNS usage and (2) “restrict,” which refers to the attempts to limit and inhibit using SNS.

Impulsive use of SNS was measured at t₂ and operationalized using impulsive SNS use measures (Turel and Bechara, 2016a). Specifically, we asked respondents to report their extent of use of their most-frequently used SNS during previous week (between t₁ and t₂) in terms of “spontaneity of use,” “impulsivity of use,” and “rashness/hastiness of use” (see Appendix 1).

Demographic variables, such as age and gender, as well as respondents’ level of perceived stress (an important control in DST studies, see Hofmann et al., 2008) using short form of perceived stress scale (Cohen et al., 1983), extent of SNS use experience, and the list of SNSs being used by them were captured for descriptive and control purposes.

Preliminary data analysis: study 1

In order to ensure the validity of all measurement items, a series of initial assessments were performed. First, the measures were presented to a group of 15 SNS users. Based on their feedback, minor adjustments were applied (item wording was slightly modified to ensure consistency). Next, the reliability and average variance extracted values were obtained to ensure the validity and reliability of the constructs (see Table 1); these suggested acceptable validity and reliability. Furthermore, serious skewness and kurtosis were not detected, with the indices within ±1. The correlations among the first-order constructs were moderate to high, which is expected among the components of a second-order factor, and were consistent with the correlations observed in Collins and Lapp (1992) from which the scale is originated (.18–.69).

Table 1.

Descriptive statistics, reliabilities, square root of AVEs (bold on diagonal), and correlations: study 1.

	Mean (SD)	Alpha	1	2	3	4	5	6	7	8	9
(1) Cognitive Preoccupation (t₁)	2.13 (1.13)	.86	.88
(2) Emotion (t₁)	2.7 (1.28)	.85	.53	.88
(3) Govern (t₁)	2.91 (1.22)	.83	.60	.47	.86
(4) Restrict (t₁)	2.79 (1.23)	.87	.54	.41	.50	.89
(5) Concern (t₁)	2.49 (1.25)	.87	.61	.48	.47	.64	.89
(6) Impulsive Use (t₂)	3.25 (1.26)	.80	.15	.33	.36	.09	.12	.85
(7) Perceived Stress (t₂)	2.5 (0.87)	.89	.23	.24	.23	.22	.27	.18	.86
(8) Age (t₁)	23.37 (4.48)	NA	−.04	−.06	−.13	−.10	−.10	−.08	−.08	NA
(9) Gender (t₁)	0.5 (0.5)	NA	.02	.05	.02	.08	.09	.04	.03	−.07	NA
(10) Years of SNS Use (t₁)	4.24 (2.35)	NA	.09	.03	−.05	−.04	.00	−.06	−.06	.07	−.01

SD: standard deviation; AVE: average variance extracted.

Hypothesis testing results: study 1

The results of structural equation modeling (SEM) analyses are depicted in Figure 2. The model exhibited good fit indices (see Figure 2). Estimated path coefficients and their significance levels demonstrate that all three hypotheses were supported (Figure 2). Consistent with H1, the results show that system 1, manifested in cognitive-emotional preoccupation with using SNS, has a significant positive effect on impulsive use of the SNS (.86, p < .001). Moreover, consistent with H2, system 2, manifested in cognitive-behavioral control over using SNS, has a negative direct effect on impulsive use of the SNS (−.64, p < .01). In addition, in line with H3, cognitive-behavioral control negatively moderates the effect of cognitive-emotional preoccupation on impulsive use of the SNS (−.35, p < .001). We have also controlled for the effects of age, gender, and perceived stress; none had a significant effect on impulsive SNS use. Together, dual-system mechanisms explained 30% of the variance in impulsive SNS use.

Figure 2.

Results of hypotheses testing: study 1.

Study 2

Study 2 focused on investigating swearing on the SNS. SNSs provide a fertile ground for the use of profane words and offensive language, as social online communications tend to be more aggressive and less formal, compared with face-to-face communications (Siegel, 2009). People in online environments feel less inhibited, do not see the immediate reactions of other offended people, and have a false sense of immunity (Turel and Bechara, 2017). They may hence use profanities that they may not use in face-to-face communications (Turel and Bechara, 2017). Indeed, 47% of Facebook walls contain swear words and 50% of Facebook users are comfortable with posting profanities, with the “f-word” being the most common profanity on Facebook (Davis, 2011).

This study also employed a two-wave data collection (t₁ and t₂, 1 week apart) from 336 SNS users from the same university. This sample included 48% women. The average age was 23 years (18–60 years, SD = 4.38). Respondents had an average of 4.2 years of SNS experience (0.2–12 years, SD = 2.3) and were on average active on 3.6 SNSs (1–8, SD = 1.77), with Facebook being the most prevalent SNS among participants (93.8%).

Measures: study 2

We operationalized Cognitive-emotional preoccupation with using SNS and cognitive-behavioral control over using SNS at t₁ with the same measures used in study 1. Swearing on the SNS was measured at t₂ and operationalized using an established measure of swearing (Gitter, 2010). Specifically, we asked respondents to report how often they swore or used offensive language on the SNS⁴ during the previous week (between t₁ and t₂) to “strengthen an argument,” “express anger,” “express negative emotions,” and “vent out” (see Appendix 1). Similar to study 1, participants’ age, gender, level of perceived stress, extent of SNS use experience, and the list of SNSs being used by them were also measured for control and descriptive purposes.

Preliminary data analysis: study 2

The same initial assessments employed in study 1 were employed here. The indices in study 2 (see Table 2) also indicated that the scales are reasonably valid and reliable. Similar to study 1, the correlations among the first-order constructs were moderate to high; this is expected among first-order constructs of a second-order factor and is consistent with the original Collins and Lapp’s (1992) measurement inventory.

Table 2.

Descriptive statistics, reliabilities, square root of AVEs (bold on diagonal), and correlations: study 2.

	Mean (SD)	Alpha	1	2	3	4	5	6	7	8	9
(1) Cognitive Preoccupation (t₁)	2.33 (1.29)	.88	.90
(2) Emotion (t₁)	2.86 (1.37)	.87	.59	.89
(3) Govern (t₁)	3.1 (1.35)	.86	.65	.54	.88
(4) Restrict (t₁)	2.98 (1.35)	.89	.58	.48	.54	.82
(5) Concern (t₁)	2.69 (1.34)	.88	.67	.56	.56	.68	.87
(6) Swearing on the SNS (t₂)	1.84 (1.31)	.94	.17	.24	.07	.03	.08	.92
(7) Perceived Stress (t₂)	2.56 (0.89)	.89	.23	.25	.24	.28	.29	.15	.87
(8) Age (t₁)	23.35 (4.38)	NA	−.02	−.04	−.11	−.07	−.08	−.02	−.07	NA
(9) Gender (t₁)	0.52 (0.5)	NA	.05	.08	.07	.10	.12	−.03	.09	−.04	NA
(10) Years of SNS Use (t₁)	4.22 (2.31)	NA	.07	.02	−.03	−.02	−.01	.05	−.03	.06	−.02

SD: standard deviation; AVE: average variance extracted.

Hypothesis testing results: study 2

Estimated path coefficients and their significance levels derived from SEM analysis show that all three hypotheses were supported (Figure 3). The model exhibits acceptable fit to the data (root mean square error of approximation [RMSEA] = .061, with 95% confidence interval of [0.055, 0.068]; standardized root mean square residual [SRMR] = .055; comparative fit index [CFI] = .94; Incremental Index of Fit [IFI] = .94; and Tucker–Lewis Index [TLI] = .93). Consistent with H1, cognitive-emotional preoccupation with using SNS (system 1) had a significant positive effect on swearing on the SNS (.74, p < 0.01). Moreover, consistent with H2, cognitive-behavioral control over using SNS (system 2) had a significant negative direct effect on swearing on the SNS (−.61, p < .05). Also, in line with H3, cognitive-behavioral control negatively moderated the effect of cognitive-emotional preoccupation on swearing on the SNS (−.24, p < .001). Similar to study 1, we controlled for the effects of age, gender, and perceived stress. Stress was a significant predictor of swearing on the SNS; high stress led to more swearing. Together, the model explained 16% of the variance in swearing on the SNS.

Figure 3.

Results of hypotheses testing: study 2.

Discussion

Drawing on DST, our findings explain the etiology of two unplanned online behaviors involving SNS, namely, impulsive use of SNS and swearing on SNS, both of which can have major personal and societal ramifications. This contributes to the emerging body of research on the “dark side of SNS use” (D’Arcy et al., 2014; Tarafdar et al., 2015; Tarafdar et al., 2013). By drawing on DST, which has been typically used for explaining problematic offline behaviors (e.g. problem drinking, impulsive eating, reckless driving, violence, and unprotected sex), our two studies managed to explain considerable variances in impulsive SNS use (30%) and swearing on the SNS (16%). Therefore, our model can be useful in explaining unplanned and possibly harming/risky media use behaviors (e.g. using cellphone while driving, online retaliation/flaming, taking and posting embarrassing pictures online, “sexting” [sending images with sexually suggesting content], and cyberbullying). As opposed to models that focus on planned, intentional behaviors, such as the Theory of Planned Behavior (Ajzen, 1991), models that rely on automatic reactions to cues and self-control strength may better explain behaviors that are relatively more unplanned and possibly risky (Gibbons et al., 1998). Future research should therefore explore extending our model to other instances in which media use is not well-thought-of and can be an automatic, unplanned behavior with little foresight.

Our findings specifically demonstrated that online behaviors that are mostly unplanned can be explained via a tug-of-war between system 1 and system 2 cognitive processes: strong impulsions generated by system 1 and weak inhibition of these preoccupying thoughts by system 2. As such, the findings point to possible similarities between the etiology of unplanned online and offline behaviors. Nonetheless, the prevalence of unplanned uses of social media, the scarcity of research on this family of behaviors, and the broad range of possible adverse consequences associated with these behaviors provide ripe opportunities for future research on these behaviors, their etiology, and consequences. Unplanned media use can become excessive and possibly meet addiction criteria (He et al., 2017a; He et al., 2017b; Turel and Serenko, 2012; Turel et al., 2014). Moreover, it can be associated with a myriad of mental problems, including attention deficit and hyperactivity disorder (ADHD), social phobia, depression, hostility (Yen et al., 2007), substance use (Benotsch et al., 2013), and conduct problems (Pujol et al., 2016). Therefore, it is important that future research extends our work and examines possible consequences and correlates of unplanned SNS use, such as mental and behavioral disorders, as well as interventions aimed at possibly reducing it.

It is interesting to consider the differences between the two studies, in terms of explained variance. The findings imply that DST is possibly more efficacious in explaining impulsive use than swearing. Specifically, while the DST was useful in both cases and fit the data similarly well, it is noteworthy that the model explained almost double the variance in impulsive SNS use (30%) than in swearing on SNS (16%). This can allude to at least two possible differences between these two behaviors. The first possible difference is in terms of planning (or lack thereof) to engage in these behaviors. The findings imply that swearing may be less unplanned as compared to impulsive SNS use. Therefore, future research can examine this proposition by contrasting the DST model with planned behavior models focusing on swearing as the outcome, or extending our model by including additional variables that can explain swearing behavior (e.g. mood, social norms). The second possible difference is in terms of the scope of the unplanned behavior and its triggers. It is reasonable to argue that impulsive use is a much broader behavior compared to swearing (by definition, use may involve reading and posting messages, whereas swearing is done in a subset of message posting instances). There may also be more triggers that drive impulsive use (e.g. new message notifications) than triggers driving swearing with little foresight (e.g. a message in which someone else used offensive language related to the recipient). These propositions should be examined in future research in order to better understand the boundaries of and the needed extensions of DST in various contexts.

From a practical standpoint, our findings suggest that media designers can help users reduce their unplanned online behavior instances by making it easier to detect and govern their behaviors. For example, they can provide warning notifications to users at login to increase awareness of instances of impulsive SNS checking. Case in point are applications that provide warnings to Facebook users and prevent automatic use of Facebook by asking users for the purpose of the visit before it passes them to their Facebook pages (e.g. Focustimeapp.com). Moreover, SNS administrators can develop applications that detect profanity in online messages while being typed (e.g. similar to Webpurify.com) and generate a “profanity warning” to the users. This would also help users’ system 2 to be activated, which can reduce the likelihood of the unplanned behavior. Based on our findings, users without deficits in their systems 1 and 2 may also help themselves by, for example, avoiding SNS use stimuli or learning to better govern their SNS use. For instance, they may switch off SNS notifications while driving to avoid stimulant cues. In very extreme cases (e.g., when impuslive use causes major life impairments), treatment via non-invaisve means such as tanscaranial brain stimulation, that can alter system 1 and system 2 activations (Zhao et al., 2017), may be considered. Future research should examine the efficacy of such intervention in reducing unplanned uses of SNS.

Conclusion

Unplanned online behaviors may be explained via the lens of analysis of DST. They are associated with a strong cognitive-emotional system that generates preoccupation and impulsions (system 1) and a weak cognitive system that controls and regulates cognitions and behaviors (system 2). This model can serve as a basis for further research on unplanned, less-rational media use behaviors.

Footnotes

Appendix 1

Table 3.

Measurement instruments used in study 1 and study 2.

Factor and reference	Time	Measurement items
Emotion (Turel and Qahri-Saremi, 2016)	t₁	Reflecting on your experience with using your most-frequently used social networking site (SNS), HOW OFTEN … (1 = never, 7 = always) • When you feel anxious, you crave to use this SNS? • When you feel lonely, you feel an urge to use this SNS? • Do you ever feel so nervous that you really need to use this SNS?
Cognitive Preoccupation (Turel and Qahri-Saremi, 2016)	t₁	Reflecting on your experience with using your most-frequently used SNS, HOW OFTEN … (1 = never, 7 = always) • Do you find yourself unable to stop thinking about using this SNS? • Is it hard to distract yourself from thinking about this SNS? • Do thoughts about using this SNS intrude into your daily activities?
Govern (Turel and Qahri-Saremi, 2016)	t₁	Reflecting on your experience with using your most-frequently used SNS: (1 = very little, 7 = very high) • How much effort does it take for you to keep your use of this SNS under control? • How much difficulty do you have controlling your use of this SNS? Reflecting on your use of this SNS, HOW OFTEN … (1 = never, 7 = always) • Do you find that once you start using this SNS it is difficult for you to stop?
Concern (Turel and Qahri-Saremi, 2016)	t₁	Reflecting on your experience with using your most-frequently used SNS, HOW OFTEN … (1 = never, 7 = always) • Does the sight of this SNS make you think about limiting your use of it? • Does seeing news about this SNS stimulate concerns about the need to limit your use of this SNS? • Does seeing other people using this SNS remind you of your efforts to control your use of this SNS?
Restrict (Turel and Qahri-Saremi, 2016)	t₁	Reflecting on your experience with using your most-frequently used SNS, HOW OFTEN … (1 = never, 7 = always) • Do you ever cut back on using this SNS in an attempt to change your use habits involving this SNS? • Do you attempt to cut down using this SNS? • Do feelings of guilt about too much use of this SNS help you to control your use of this SNS?
Impulsive Use of SNS (Turel and Bechara, 2016a)	t₂	Over the PREVIOUS WEEK, what was the extent of your use of your most-frequently used SNS … (1 = very low, 7 = very high) • … In terms of the spontaneity of my use? • … In terms of the impulsivity of my use? • … In terms of the rashness/hastiness of my use?
Swearing on SNS (Gitter, 2010)	t₂	Reflecting on your use of your most-frequently used SNS over the PREVIOUS WEEK, how often did you swear or use offensive language on the site … (1 = never to 7 = very often) • … to strengthen an argument • … to express your anger • … to express negative emotions • … to vent out
Perceived Stress (short form) (Cohen et al., 1983)	t₂	Please reflect on your life situation over the PREVIOUS WEEK and state how often you have felt like the statements below (1 = never, 5 = very often) In the LAST WEEK … • … how often have you felt that you were unable to control the important things in your life? • … how often have you felt unconfident about your ability to handle your personal problems? • … how often have you felt that things were NOT going your way? • … how often have you felt difficulties were piling up so high that you could not overcome them?

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author biographies

Ofir Turel is a professor of Information Systems and Decision Sciences at the College of Business and Economics, California State University, Fullerton, and a scholar in Residence at the Brain and Creativity Institute, Department of Psychology at the University of Southern California (USC). He has published over 80 journal articles in outlets such MIS Quarterly, Scientific Reports, Journal of Management Information Systems, MIT Sloan Management Review, Communications of the ACM, European Journal of Information Systems, Information Systems Journal, Information & Management, Psychological Reports, Journal of Information Systems, Behavior & Information Technology, Telecommunications Policy, Group Decision and Negotiation, Family Business Review, Communications in Statistics, Computers & Education, and Frontiers in Psychology. He is an Associate Editor of Frontiers in Psychology, Section on Psychopathology, Frontiers in Psychiatry, and Internet Research.

Hamed Qahri-Saremi is an assistant professor of Information Systems at the College of Computing and Digital Media, DePaul University. He holds a Ph.D. in business administration with a concentartion on information systems. His research focuses on patterns of use of information systems at both individual and organizational levels. His work has appeared in Journal of MIS, Journal of Strategic Information Systems, Information & Management, Computers & Education, and Expert Systems with Applications.

References

Ajzen

(1991) The theory of planned behavior. Organizational Behavior and Human Decision Processes 50(2): 179–211.

Alonzo

Aiken

(2004) Flaming in electronic communication. Decision Support Systems 36(3): 205–213.

American Psychiatric Association (2013) Internet gaming disorder. In: Diagnostic and Statistical Manual of Mental Disorders. 5th ed. Arlington, VA: American Psychiatric Publishing, pp. 795–798.

Antoci

Delfino

Paglieri

et al . (2016) Civility vs. incivility in online social interactions: an evolutionary approach. PLoS ONE 11(11): e0164286.

Baumeister

(2002) Yielding to temptation: self-control failure, impulsive purchasing, and consumer behavior. Journal of Consumer Research 28(4): 670–676.

Benotsch

Snipes

Martin

et al . (2013) Sexting, substance use, and sexual risk behavior in young adults. Journal of Adolescent Health 52(3): 307–313.

Cao

Lin

W-Y

(2015) How do victims react to cyberbullying on social networking sites? The influence of previous cyberbullying victimization experiences. Computers in Human Behavior 52: 458–465.

Cattan

White

Bond

et al . (2005) Preventing social isolation and loneliness among older people: a systematic review of health promotion interventions. Ageing and Society 25(1): 41–67.

Chaiken

(1980) Heuristic versus systematic information processing and the use of source versus message cues in persuasion. Journal of Personality and Social Psychology 39(5): 752.

10.

Cohen

Kamarck

Mermelstein

(1983) A global measure of perceived stress. Journal of Health and Social Behavior 24(4): 385–396.

11.

Collins

Lapp

(1992) The Temptation and Restraint Inventory for measuring drinking restraint. British Journal of Addiction 87(4): 625–633.

12.

D’Arcy

Gupta

Tarafdar

et al . (2014) Reflecting on the "Dark Side" of information technology use. Communications of the Association for Information Systems 35(1): 109–118.

13.

Davis

(2011) F**k, It’s Hard to Keep Profanity off Your Facebook Wall. Available at: http://www.businessinsider.com/facebook-swearing-profanity-reppler-2011-5

14.

Devine

(1989) Stereotypes and prejudice: their automatic and controlled components. Journal of Personality and Social Psychology 56(1): 5–18.

15.

Evans

JSBT

(2006) Dual system theories of cognition: some issues. Proceedings of the Cognitive Science Society 28(28): 202–207.

16.

Evans

JSBT

(2008) Dual-processing accounts of reasoning, judgment, and social cognition. Annual Review of Psychology 59: 255–278.

17.

Evans

JSBT

(2009) How many dual-process theories do we need? One, two, or many? In: Evans

JSBT

Frankish

(eds) In Two Minds: Dual Processes and beyond. New York: Oxford University Press, pp. 33–54.

18.

Friese

Hofmann

(2009) Control me or I will control you: impulses, trait self-control, and the guidance of behavior. Journal of Research in Personality 43(5): 795–805.

19.

Gibbons

Gerrard

Blanton

et al . (1998) Reasoned action and social reaction: willingness and intention as independent predictors of health risk. Journal of Personality and Social Psychology 74(5): 1164–1180.

20.

Gitter

(2010) Shooting the Shit: Profanity, Self-Control, and Aggressive Behavior. Tallahassee, FL: Department of Psychology, Florida State University, 73 pp.

21.

Greitemeyer

Agthe

Turner

et al . (2012) Acting prosocially reduces retaliation: effects of prosocial video games on aggressive behavior. European Journal of Social Psychology 42(2): 235–242.

22.

Hanna

Rohm

Crittenden

(2011) We’re all connected: the power of the social media ecosystem. Business Horizons 54(3): 265–273.

23.

Hayes

Carr

Wohn

(2016) One click, many meanings: interpreting paralinguistic digital affordances in social media. Journal of Broadcasting & Electronic Media 60(1): 171–187.

24.

Turel

Bechara

(2017a) Brain anatomy alterations associated with Social Networking Site (SNS) addiction. Scientific Reports 7(paper 45064): 1–8.

25.

Turel

Brevers

et al . (2017b) Excess social media use in normal populations is associated with amygdala-striatal but not with prefrontal morphology. Psychiatry Research: Neuroimaging 269(1): 31–35.

26.

Hofmann

Baumeister

Foerster

et al . (2012) Everyday temptations: an experience sampling study of desire, conflict, and self-control. Journal of Personality and Social Psychology 102(6): 1318–1335.

27.

Hofmann

Friese

Strack

(2009) Impulse and self-control from a dual-systems perspective. Perspectives on Psychological Science 4(2): 162–176.

28.

Hofmann

Friese

Wiers

(2008) Impulsive versus reflective influences on health behavior: a theoretical framework and empirical review. Health Psychology Review 2(2): 111–137.

29.

Kahneman

(2011) Thinking, Fast and Slow. New York: Farrar, Straus and Giroux.

30.

Lindgren

Neighbors

Westgate

et al . (2014) Self-control and implicit drinking identity as predictors of alcohol consumption, problems, and cravings. Journal of Studies on Alcohol and Drugs 75(2): 290–298.

31.

Ljung

(2011) Swearing: A Cross-Cultural Linguistic Study. Basingstoke: Palgrave Macmillan.

32.

Meshi

Tamir

Heekeren

(2015) The emerging neuroscience of social media. Trends in Cognitive Sciences 19(12): 771–782.

33.

Morozov

(2011) The Net Delusion: The Dark Side of Internet Freedom. New York: Public Affairs.

34.

Noel

Brevers

Bechara

(2013) A neurocognitive approach to understanding the neurobiology of addiction. Current Opinion in Neurobiology 23(4): 632–638.

35.

O’Sullivan

Flanagin

(2003) Reconceptualizing “flaming” and other problematic messages. New Media & Society 5(1): 69–94.

36.

Pantic

(2014) Online social networking and mental health. Cyberpsychology, Behavior, and Social Networking 17(10): 652–657.

37.

Park

E-Y

Kim

E-M

(2014) The relationship between online activities, netiquette and cyberbullying. Children and Youth Services Review 42(Supplement C): 74–81.

38.

Patton

Hong

Ranney

et al . (2014) Social media as a vector for youth violence: a review of the literature. Computers in Human Behavior 35: 548–553.

39.

Pujol

Fenoll

Forns

et al . (2016) Video gaming in school children: how much is enough? Annals of Neurology 80(3): 424–433.

40.

Qahri-Saremi

Turel

(2016) School engagement, information technology use, and educational development: An empirical investigation of adolescents. Computers & Education 102: 65–78.

41.

Rassin

Muris

(2005) Why do women swear? An exploration of reasons for and perceived efficacy of swearing in Dutch female students. Personality and Individual Differences 38(7): 1669–1674.

42.

Rogers

(1975) A protection motivation theory of fear appeals and attitude change. The Journal of Psychology: Interdisciplinary and Applied 91(1): 93–114.

43.

Rosner

Winter

Kramer

(2016) Dangerous minds? Effects of uncivil online comments on aggressive cognitions, emotions, and behavior. Computers in Human Behavior 58: 461–470.

44.

Siegel

(2009) Against the Machine: How the Web Is Reshaping Culture and Commerce—And Why It Matters. New York: Spiegel & Grau.

45.

Smith

(2016) Facebook Statistics and Facts (2016). Available at: http://expandedramblings.com/index.php/by-the-numbers-17-amazing-facebook-stats/

46.

Soror

Hammer

Steelman

et al . (2015) Good habits gone bad: explaining negative consequences associated with the use of mobile phones from a dual-systems perspective. Information Systems Journal 25(4): 403–427.

47.

Stone

McMillan

Hazelton

et al . (2011) Wounding words: swearing and verbal aggression in an inpatient setting. Perspectives in Psychiatric Care 47(4): 194–203.

48.

Strack

Deutsch

(2004) Reflective and impulsive determinants of social behavior. Personality and Social Psychology Review 8(3): 220–247.

49.

Tarafdar

D’Arcy

Turel

et al . (2015) The dark side of information technology. MIT Sloan Management Review 56(2): 600–623.

50.

Tarafdar

Gupta

Turel

(2013) The dark side of information technology use. Information Systems Journal 23(3): 269–275.

51.

Thacker

Griffiths

(2012) An exploratory study of trolling in online video gaming. International Journal of Cyber Behavior, Psychology and Learning 2(4): 17–33.

52.

Thompsen

Foulger

(1996) Effects of pictographs and quoting on flaming in electronic mail. Computers in Human Behavior 12(2): 225–243.

53.

Turel

(2015) Quitting the use of a habituated hedonic information system: a theoretical model and empirical examination of Facebook users. European Journal of Information Systems 24(4): 431–446.

54.

Turel

(2016) Untangling the complex role of guilt in rational decisions to discontinue the use of a hedonic information system. European Journal of Information Systems 25(5): 432–447.

55.

Turel

Bechara

(2016a) A triadic reflective-impulsive-interoceptive awareness model of general and impulsive information system use: behavioral tests of neuro-cognitive theory. Frontiers in Psychology 7: 1–11.

56.

Turel

Bechara

(2016b) Social Networking Site use while driving: ADHD and the mediating roles of stress, self-esteem and craving. Frontiers in Psychology 7: 1–10.

57.

Turel

Bechara

(2017) Effects of motor impulsivity and sleep quality on swearing, interpersonally deviant and disadvantageous behaviors on online social networking sites. Personality and Individual Differences 108(1): 91–97.

58.

Turel

Brevers

et al (2017) Social networking sites and the morphology of a social-semantic brain network. Social Neuroscience: 1–9.

59.

Turel

Xue

et al . (2014) Examination of neural systems sub-serving Facebook "addiction". Psychological Reports 115(3): 675–695.

60.

Turel

Qahri-Saremi

(2016) Problematic use of social networking sites: antecedents and consequence from a dual-system theory perspective. Journal of Management Information Systems 33(4): 1087–1116.

61.

Turel

Serenko

(2012) The benefits and dangers of enjoyment with social networking websites. European Journal of Information Systems 21(5): 512–518.

62.

Turel

Serenko

Bontis

(2011a) Family and work-related consequences of addiction to organizational pervasive technologies. Information & Management 48(2–3): 88–95.

63.

Turel

Serenko

Giles

(2011b) Integrating technology addiction and use: An empirical investigation of online auction sites. MIS Quarterly 35(4): 1043–1061.

64.

Van Cleemput

Vandebosch

Pabian

(2014) Personal characteristics and contextual factors that determine “helping,” “joining in,” and “doing nothing” when witnessing cyberbullying. Aggressive Behavior 40(5): 383–396.

65.

Vandebosch

Van Cleemput

(2008) Defining cyberbullying: a qualitative research into the perceptions of youngsters. Cyberpsychology and Behavior 11(4): 499–503.

66.

Wood

SMW

Bechara

(2014) The neuroscience of dual (and triple) system in decision making. In: Reyna

Zayas

(eds) The Neuroscience of Risky Decision Making. Washington, DC: American Psychological Association, pp. 177–202.

67.

Turel

Yuan

(2012) Online game addiction among adolescents: motivation and prevention factors. European Journal of Information Systems 21(3): 321–340.

68.

Yen

et al . (2007) The comorbid psychiatric symptoms of Internet addiction: attention deficit and hyperactivity disorder (ADHD), depression, social phobia, and hostility. Journal of Adolescent Health 41(1): 93–98.

69.

Zhao

Qiao

Fan

et al . (2017) Modulation of Brain Activity with Noninvasive Transcranial Direct Current Stimulation (tDCS): Clinical Applications and Safety Concerns. Frontiers in Psychology 8 (685): 1–17