*2.2. Measure*

#### 2.2.1. Autonomy Need Dissatisfaction

The level of autonomy need dissatisfaction was measured in T1 by the Basic Need Satisfaction in General Scale, which consists of the domains of autonomy, competence, and relatedness needs satisfaction [47]. This scale has been tested and used in the Chinese context [48]. The autonomy subscale contains three negative items (e.g., There is not much opportunity for me to decide for myself how to do things in my daily life), which has been used to assess autonomy need dissatisfaction [49]. Participants rated the items on a 5-point Likert scale (1 = *not at all true*, 5 = *very true*), with higher scores indicating higher levels of autonomy need dissatisfaction. The internal consistency of this scale showed acceptable reliability (Cronbach α = 0.60).

#### 2.2.2. Boredom Proneness

The level of boredom proneness was measured in T2 by the short version of the Boredom Proneness Scale [27]. This scale contains eight items (e.g., I often find myself at "loose ends," not knowing what to do) with one-dimensional structure. Participants rated the items on a 7-point Likert scale (1 = *strongly disagree*, 7 = *strongly agree*), with higher scores indicating higher levels of boredom proneness. The internal consistency of this scale showed satisfactory reliability (Cronbach α = 0.94).

#### 2.2.3. Mobile Phone Gaming

The measure of mobile phone gaming in T2 was adapted from the Chinese Internet Usage Questionnaire [50] and the Mobile Phone Use Patterns Questionnaire [43]. In total, there were 10 items regarding Internet use and 17 items regarding mobile phone use. After instructing "according to your daily routine, ... ", only one item (i.e., I play mobile games on my phone) was used to assess the mobile phone gaming frequency on a daily basis. Participants rated the items on a 5-point Likert scale (1 = *never*, 5 = *always*), with higher scores indicating more frequent use for mobile games.

#### 2.2.4. Problematic Mobile Phone Use (PMPU)

Participants' severity of PMPU was assessed in T3 by the short version of the Mobile Phone Problem Use Scale [51], which has been validated in the Chinese context and showed good validity and reliability [17,52]. This scale contains 10 items (e.g., I find it difficult to switch off my mobile phone) with five aspects, including craving, withdrawal, peer dependence, loss of control, and negative life consequences. Participants rated the items on a 5-point Likert scale (1 = *strongly disagree*, 5 = *strongly agree*), with higher scores indicating more severe PMPU. The internal consistency of this scale showed satisfactory reliability (Cronbach's α = 0.87).

#### *2.3. Data Analyses*

Means, standard deviations, and Pearson correlations were calculated using SPSS 19.0. The hypothesized multiple mediation model was tested by structural equation modeling (SEM)

using Mplus 7.1 [53]. The model was evaluated by following model fit indices: the chi-square values (χ2), the comparative fit index (CFI), the Tucker–Lewis fit index (TLI), the root mean square error of approximation (RMSEA), and the standardized root mean square residual (SRMR). The CFI and TLI at 0.90 or above, and the RMSEA and SRMR at 0.08 or lower, indicate that the model is acceptable [54].

#### **3. Results**

#### *3.1. Descriptive Statistics and Correlations*

Means, standard deviations, and Pearson correlations are presented in Table 1. As shown, autonomy need dissatisfaction was significantly and positively correlated with boredom proneness and PMPU, but it was not correlated with mobile phone gaming. Furthermore, each two elements of boredom proneness, mobile phone gaming, and PMPU had a positive association.

**Table 1.** Means, standard deviations, and correlations among the main variables.


Note. T1 Autonomy = Autonomy need dissatisfaction in Time 1; T2 Boredom = Boredom proneness in Time 2; T2 Gaming = Mobile phone gaming in Time 2; T3 PMPU = Problematic mobile phone use in Time 3; \*\*\* *p* < 0.001.

#### *3.2. Examinations of the Measurement Model*

Before testing the hypothesized model by SEM, it was necessary to examine the measurement model. According to the recommendation from Wu and Wen [55], autonomy need dissatisfaction could be loaded by the three observed items; boredom proneness that has eight items with one-dimensional structure could be parceled into three indicators; mobile phone gaming with only one item could be loaded by the one item; PMPU with five aspects could be loaded by the five substructures. Altogether, the CFA results of the measurement model showed a good model fit: χ2/df = 3.17, CFI = 0.95, TLI = 0.93, RMSEA = 0.08, SRMR = 0.05, in that all the loadings on latent variables were significant (*p* < 0.001).

#### *3.3. Examinations of the Structural Model*

As hypothesized, a multiple model with T1 autonomy need dissatisfaction as the independent variable, T2 boredom proneness and mobile phone gaming as the mediators, and T3 PMPU as the dependent variable was established. The SEM results showed a good model fit: χ2/df = 3.15, CFI = 0.94, TLI = 0.92, RMSEA = 0.05, SRMR = 0.05. As shown in Figure 2, T1 autonomy need dissatisfaction significantly predicted T3 PMPU. Similarly, T1 autonomy need dissatisfaction positively predicted T2 boredom proneness, which in turn positively predicted T3 PMPU. However, T1 autonomy need dissatisfaction did not predict T2 mobile phone gaming, although T2 mobile phone gaming positively predicted T3 PMPU.

To further examine the significance of the indirect effects, bias-corrected bootstrap tests derived with 1000 samples were used. That the 95% confidence interval did not contain zero indicated statistical significance [56]. As shown in Table 2, T1 autonomy need dissatisfaction positively predicted T3 PMPU, supporting H1. Furthermore, T2 boredom proneness significantly mediated the association between T1 autonomy need dissatisfaction and T3 PMPU, supporting H2. Whereas, mobile phone gaming did not mediate the association between T1 autonomy need dissatisfaction and T3 PMPU, rejecting H3. Additionally, the chain of T2 boredom proneness and T2 mobile phone gaming significantly mediated the association between T1 autonomy need dissatisfaction and T3 PMPU, supporting H4.

**Figure 2.** The mediation model of the association between T1 autonomy need dissatisfaction and T3 PMPU. Note. All the loadings on latent variables were significant (*p* < 0.001). Autonomy = Autonomy need dissatisfaction, Boredom = Boredom proneness, Gaming = Mobile phone gaming, PMPU = Problematic mobile phone use. \* *p* < 0.05, \*\*\* *p* < 0.001.



Note. T1 Autonomy = Autonomy need dissatisfaction in Time 1; T3 PMPU = Problematic mobile phone use in Time 3; T2 Boredom = Boredom proneness in Time 2; T2 Gaming = Mobile phone gaming in Time 2.

#### **4. Discussion**

This study focused on autonomy need dissatisfaction and examined its potential effect on PMPU. Boredom proneness and mobile phone gaming were suggested to be incorporated into this association to elucidate the underlying mechanism. Based on three-wave data, the SEM model results showed that T1 autonomy need dissatisfaction not only directly predicted T3 PMPU, but also exerted effects on T3 PMPU via the mediating role of T2 boredom proneness and via the chain mediating role of T2 boredom proneness and T2 mobile phone gaming. Altogether, the findings provide empirical evidence to support the relation between specific psychological need and PMPU, which lends further insight into targeted prevention and interventions of problematic online behaviors.

#### *4.1. Autonomy Need Dissatisfaction, Boredom Proneness, and PMPU*

This study demonstrated that T1 autonomy need dissatisfaction directly predicted T3 PMPU, it also indirectly predicted T3 PMPU through the mediating role of T2 boredom proneness. According to self-determination theory [18], adolescents with autonomy need dissatisfaction have few opportunities to volitionally make choices and self-organize actions in daily life. Thus, they may have to participate in activities with little intrinsic motivation, which increases a tendency to experience boredom [29,45]. For instance, they may execute what others compel them to do, such as participating in extracurricular courses that are arranged by their parents. In this sense, they are more likely to experience boredom. This finding was consistent with the previous studies that the higher levels of autonomy need dissatisfaction that adolescents perceive, the more likely they would experience boredom [31].

Furthermore, bored adolescents are more likely to seek external stimulation to cope with boredom [27], and thus they may spend much time and resources on the internet (or via mobile phones), which further increases the risk of problematic behaviors, including PIU [32,33] and PMPU [24]. Altogether, adolescents with autonomy need dissatisfaction cannot freely make decisions and volitionally engage in what they are interested in, which chronically contributes to boredom proneness. These bored adolescents are more likely to frequently act on mobile phones, leading to problematic use. Thus, it seems that autonomy need dissatisfaction in daily life gives rise to boredom proneness, which in turn increases the risk of subsequent PMPU.

#### *4.2. Autonomy Need Dissatisfaction, Mobile Phone Gaming, and PMPU*

This study showed that T1 autonomy need dissatisfaction did not predict T2 mobile phone gaming although T2 gaming positively predicted T3 PMPU. This finding weakly supported the mediating role of mobile phone gaming in the mediation process because the first stage was not significant. One possibility may be that psychological need dissatisfaction plays a double-edged role in determining online gaming [37]. As mentioned earlier, games that provide adolescents with opportunities for actions can assist in compensating for unsatisfied autonomy need in the real world [21,57]. Therefore, adolescents with autonomy need dissatisfaction may resort to the internet (or via mobile phones) to compensate for this dissatisfaction [15]. For instance, when individuals feel psychologically pressured and constrained, they would use mobile phones for gaming to alleviate these undesirable feelings as a way to compensate because they are free to do whatever they want in the game world. This perspective implies that the higher levels of autonomy need dissatisfaction that adolescents perceive, the more frequent mobile phone gaming they would engage in [38].

Nevertheless, adolescents with autonomy need dissatisfaction have few opportunities to decide for themselves even though they may think that gaming is tempting [39,47]. Specifically, adolescents with autonomy need dissatisfaction may be under the restrictions of their parents, particularly when engaging in mobile phone gaming. This perspective suggests that the higher levels of autonomy need dissatisfaction that adolescents perceive, the fewer opportunities they might have to play mobile games. Taken these two perspectives together, the former compensatory effect (i.e., autonomy need dissatisfaction motivates mobile gaming as a compensator) may neutralize the later restriction effect (i.e., autonomy need dissatisfaction indicates few opportunities for mobile gaming). Thus, it is not surprising that autonomy need dissatisfaction in daily life was weakly associated with mobile phone gaming. Future studies are warranted to further examine the complicated association between autonomy need dissatisfaction and online gaming.

#### *4.3. A Multiple Mediation Model*

One intriguing finding was that the chain of T2 boredom proneness and T2 mobile phone gaming significantly mediated T1 autonomy need dissatisfaction and T3 PMPU. Consistent with the etiology of addictive symptoms of PIU [22], psychological needs dissatisfaction as a distal factor exerts effects on addiction tendencies through the mediating variables (i.e., boredom proneness and mobile phone gaming). Specifically, adolescents with autonomy need dissatisfaction in real life tend to have relatively fewer opportunities to make decisions; thus, they may have to engage what is not congruent with their authentic interests [47]. For instance, the parents of Chinese students may arrange for them to engage in repetitive and monotonous academic activities. In a long run, they may possess low levels of intrinsic motivation and exhibit high levels of boredom proneness. Concurring with the earlier findings [25,44], bored adolescents may engage in online gaming to alleviate boredom as they can obtain external stimulation and gain flow experiences when fully involving in gaming [39]. For instance, participants reported that they played a kind of multiplayer online battle arena game named *Arena of Valor* on mobile phones because they felt that doing so can swipe away boring time. Additionally, mobile

phone gaming has been identified as a high-risk factor for PMPU [26,40]. That is, adolescents who frequently engage in mobile phone gaming are at risk in developing problematic use and nurturing addiction tendencies. Taken together, it seems that autonomy need dissatisfaction in daily life positively predicts boredom proneness that contributes to frequent mobile phone gaming, which in turn leads to subsequent PMPU.

#### *4.4. Limitations, Future Directions, and Implications*

There are several limitations of this study. First, self-reported data may produce response bias although there was no serious common method bias using Harman's single factor test [58]. Future studies could record time of generalized use and gaming on mobile phones, which may provide objective data and enhance reliability and validity. Second, the reliability of the measure of autonomy need dissatisfaction appeared to be somewhat low although it has been used in several studies [47–49]. Thus, this scale should be further improved in future research. Third, this study recruited secondary school students only from a regular secondary school; therefore, generalization of the conclusions to other groups should be made with caution. Future studies could focus on adults and/or clinical groups, which may contribute to a broader application of these findings. Fourth, the mediating effects seemed to be relatively small, however small effects can assist in developing theories when the findings support the theoretical hypotheses [59]. In addition, small effects should not be disregarded because they might be accumulated to generate large effects with the changing conditions [60]. In this digital age in particular, the use of mobile phones has exponentially grown and corresponding problems (e.g., addiction tendencies) have been increasing and appear severe, thus possibly leading to large effects on PMPU in future studies.

Despite the limitations, notable implications are twofold. From a theoretical perspective, this study was the first of its kind to use a cross-temporal design and to exclusively examine the effect of autonomy need dissatisfaction on subsequent PMPU. On the one hand, this study focused on the role of specific need (i.e., autonomy need), instead of psychological needs as a single entity, in explaining maladaptive online behaviors. On the other hand, boredom proneness as an individual characteristic and gaming as possible coping strategies helped to elucidate the potential etiology of addiction-like symptoms associated with mobile phone use in the framework of self-determination. These findings based on the three-wave data revealed that autonomy need dissatisfaction not only directly predicted subsequent PMPU, but also exerted indirect effects via the mediating roles of boredom proneness and mobile phone gaming. These findings help to develop a better understanding of the formation process of PMPU, which provides support for prevention and intervention programs. For instance, excessive parental restrictions on children's online behaviors (e.g., limiting use time, monitoring online content) may backfire because these children perceive autonomy need dissatisfaction and may increase addiction-like tendencies [61,62]. In contrast, we recommend that families and schools provide adolescents with a certain degree of autonomy and encourage adolescents to self-organize their behaviors, which can reduce their tendencies to experience boredom. Accordingly, these less-bored adolescents are less likely to engage in high-frequency game play, thus decreasing the risk of engaging in PMPU. Additionally, families and school personnel could try to purposefully increase diverse activities to avoid boredom from adolescents, as well as to guide adolescents to increase appropriate mobile phone use and decrease excessive mobile gaming, which may be instrumental to prevent from addiction-like online behaviors.

#### **5. Conclusions**

This is one of very few studies to focus on the association between autonomy need dissatisfaction in daily life and PMPU. With boredom proneness and mobile phone gaming introduced, the mediation model may contribute to explaining the potential mechanism of this association. Based on three-wave data, the results showed that T1 autonomy need dissatisfaction not only directly predicted T3 PMPU, but also exerted effects on T3 PMPU via the mediating role of T2 boredom proneness and via the chain mediating role of T2 boredom proneness and T2 mobile phone gaming. Altogether, these findings reveal

*Int. J. Environ. Res. Public Health* **2020**, *17*, 5305

the unique role of specific psychological need satisfaction in PMPU, which suggests that promoting autonomy need satisfaction may prevent adolescents from mobile phone addiction.

**Author Contributions:** Conceptualization, W.H. and R.-D.L.; Formal analysis, W.H.; Funding acquisition, R.-D.L.; Investigation, R.Z., R.J. and X.F.; Project administration, R.-D.L.; Supervision, R.-D.L.; Validation, R.-D.L. and Y.D.; Writing—original draft, W.H.; Writing—review and editing, W.H. and Y.D. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by the National Social Science Fund of China, grant number 17BSH102.

**Acknowledgments:** We appreciated the support from the participating schools' students and teachers.

**Conflicts of Interest:** The authors declare no conflict of interest.
