The Relationship between Big Five Personality Traits and Smartphone Addiction among University Students: Mindfulness as a Self-Regulatory Mediating Mechanism

1 Introduction

In today’s digitized society, smartphones are deeply embedded in university students’ daily lives, supporting learning, socialization, entertainment, and emotion regulation [1,2]. Alongside these benefits, escalating use has been accompanied by growing concerns about smartphone addiction, which has emerged as a widespread psychological and social issue [1]. Evidence from Taiwan further suggests that under demanding academic contexts, students may rely on smartphones to cope with stress and negative affect, potentially increasing the risk of excessive and dysregulated use [3,4].

Prior work highlights close links between problematic smartphone use and mental health-related outcomes, including associations with stress, depression, and loneliness [5]. In addition, evidence comparing self-reported and logged digital behavior suggests that behavioral indicators may complement traditional questionnaires and support early identification of risk patterns [6,7].

Problematic Smartphone Use (PSU) is often accompanied by deficits in emotional competence and fear of missing out (FoMO), and adolescents’ strong need for belonging, combined with limited inhibitory control, may further intensify overuse and its negative consequences [8]. These findings indicate that focusing solely on behavioral indicators or self-report symptom severity may be insufficient. A more integrative perspective is needed to explain why some individuals are more vulnerable to addictive patterns of smartphone use than others.

The Big Five personality traits provide a foundational framework for understanding stable individual differences in behavior. Accumulating evidence suggests that these traits are related to smartphone addiction, yet findings remain mixed across studies. Neuroticism is often associated with higher risk, whereas conscientiousness tends to be associated with a protective effect [4,9]. Extraversion and low neuroticism have also been linked to emotion-regulation tendencies, such as greater cognitive reappraisal and lower expressive suppression, which may support psychological adjustment [10]. However, the roles of openness and agreeableness appear less consistent across the literature [1,10]. Such variability suggests that personality may not operate as a simple direct predictor of addictive smartphone use; instead, intervening psychological mechanisms may be necessary to clarify how personality translates into risk or protection.

Mindfulness has been proposed as one such mechanism. As a relatively stable psychological resource, mindfulness emphasizes attending to present-moment experiences with nonjudgmental acceptance [11]. Empirical work indicates that mindfulness is associated with reduced automatic and compulsive tendencies and with enhanced emotion regulation and self-control [12,13]. At the same time, meta-analytic evidence has reported relatively small overall effect sizes, underscoring the need for large samples and rigorous designs to clarify when and how mindfulness exerts protective effects [14]. Accordingly, mindfulness may be relevant for explaining personality-related differences in smartphone addiction, but its role as a mediator requires more explicit testing.

Although prior studies have linked the Big Five traits to problematic smartphone use, evidence of mediation remains fragmented. For example, recent studies have examined psychosocial mechanisms (e.g., stress and self-regulation) that mediate associations between personality traits and problematic smartphone use in student populations [3,9].

To address these gaps, the present study tests mindfulness as a mediator linking each Big Five trait to smartphone addiction within a single unified model. Specifically, this study aims to: (i) examine associations between the Big Five personality traits and smartphone addiction; (ii) test whether mindfulness mediates the relationships between personality and smartphone addiction; and (iii) validate an integrated model with personality as antecedents, mindfulness as a mediator, and smartphone addiction as the outcome. The findings are expected to strengthen theoretical connections between personality psychology and technology-related addiction and to provide empirical implications for campus-based mindfulness education and preventive interventions [2,15].

2 Literature Review

2.1 Big Five Personality Traits

The Big Five Personality Traits constitute one of the most explanatory frameworks in contemporary personality psychology, widely used to predict behavioral tendencies and mental health [16]. The model includes Extraversion, Agreeableness, Conscientiousness, Neuroticism, and Openness, which correspond to relatively stable characteristics such as sociability, self-discipline, emotional stability, and cognitive flexibility [16].

Evidence shows that neuroticism is positively associated with anxiety and depression and demonstrates predictive utility for mental disorders [17,18]. In contrast, conscientiousness is positively associated with self-control, mindfulness, and behavioral regularity, which help mitigate digital addiction [19,20]. Further empirical findings indicate that extraversion and emotional stability are positively associated with cognitive reappraisal and negatively associated with expressive suppression, highlighting their central role in emotion regulation; by comparison, the associations of conscientiousness, agreeableness, and openness with emotion regulation are inconsistent [10]. Moreover, extraversion and emotional stability are core predictors of adolescent adaptive emotion regulation, facilitating healthy interpersonal interactions and reducing psychopathological symptoms [10].

Overall, neuroticism shows the strongest negative association with trait mindfulness, whereas conscientiousness shows the strongest positive association, underscoring the role of personality in cultivating mindfulness and regulating digital behaviors [12,21,22]. Accordingly, the Big Five provides a crucial theoretical framework for understanding the processes linking smartphone addiction and mindfulness [5,23].

2.2 Smartphone Addiction

Smartphone addiction (SA) is a form of behavioral addiction that involves excessive and compulsive phone use, leading to impairment in emotion regulation, interpersonal relationships, and academic or workplace performance [24]. Core symptoms include tolerance, withdrawal, loss of control, and functional impairment, which parallel the features observed in substance addictions [25]. From a motivational perspective, excessive smartphone use may arise from attempts to avoid stress, to obtain immediate gratification, or to regulate emotions [21,26].

Recent large-scale evidence also indicates that smartphone addiction remains prevalent among adolescents and is associated with identifiable patterns of smartphone use and risk factors in everyday settings [27]. Recent studies further describe Problematic Smartphone Use (PSU) as frequent overuse accompanied by regulatory difficulties, resulting in adverse consequences in daily life [8]. Adolescents and university students, given strong belonging needs and limited inhibitory control, are particularly vulnerable [8]. In addition, FoMO is positively associated with PSU, whereas emotional intelligence is negatively associated, suggesting that deficits in psychological regulation are important risk factors for addiction [8]. Beyond smartphone use alone, adjacent digital addictions may further compound daily impairment; for example, short-form video overuse has been linked to addictive tendencies and reduced attention control among adolescents [4], and social media addiction has been associated with poorer sleep quality [28]. Moreover, evidence comparing self-reports with logged behavior suggests that objective smartphone-use indicators can complement traditional questionnaires [6,7]. These findings highlight the tight interweaving between PSU and mental health and support the development of diverse approaches to early identification and intervention.

From a theoretical standpoint, self-regulation theory posits that addictive behaviors stem from difficulty delaying gratification and sustaining goal-directed action [29]; the compensatory internet use model proposes that smartphone use often serves as a compensatory strategy in response to negative emotions and real-world stress [30]. Consistent with these perspectives, recent evidence suggests that mental distress is linked to problematic smartphone and social media use through stress-related cognitive and affective processes [31]. Research further suggests that individuals high in neuroticism, due to emotional instability and anxiety proneness, are more likely to rely on smartphones to alleviate stress [9,32]. In contrast, mindfulness and self-control can effectively reduce the risk of overuse [13,33]. Intervention evidence also supports the broader psychological benefits of mindfulness in adolescents [34], and meta-analytic findings indicate that mindfulness is associated with lower learning burnout among university students [35], aligning with its proposed role in self-regulation. Understanding PSU, therefore, requires attention to both behavioral manifestations and the integration of personality with psychological processes, thereby enabling the construction of more explanatory models.

2.3 Mindfulness

Mindfulness is a psychological process of attending to present-moment internal and external experiences with an open and nonjudgmental attitude; it can also be conceptualized as a dispositional tendency to sustain such awareness [36,37]. In empirical research, trait mindfulness is assessed using self-report measures that emphasize mindful attention and awareness (e.g., the Mindful Attention Awareness Scale, MAAS) or broader multi-component conceptualizations (e.g., the Five Facet Mindfulness Questionnaire, FFMQ) [36,38]. In the present study, mindfulness was operationalized as mindful attention and awareness in daily life and measured using the five-item MAAS short form (MAAS-5) [36,39].

Mindfulness is considered beneficial for stress adaptation and behavioral control and shows protective effects against dependency-like digital behaviors such as smartphone addiction [12,21]. In theoretical terms, mindfulness is commonly defined as paying deliberate attention to present-moment experience rather than operating on “autopilot” [11,36]. Empirical work indicates that individuals with higher trait mindfulness are less likely to engage in avoidant or compulsive smartphone use when experiencing stress or emotional distress and display stronger self-control [14,40]. From an attention-regulation perspective, mindful awareness helps individuals notice habitual checking and mind-wandering, thereby reducing attentional impulsivity and supporting behavioral restraint in smartphone use [2,14]. Nevertheless, meta-analytic syntheses report small effect sizes, indicating a need for more rigorous research designs and larger sample sizes for validation [14,41]. It is also important to note that mindfulness is not a universal remedy; researchers should distinguish active ingredients from commercialized or popularized elements that lack substance [11]. This perspective aligns with the present study’s aim of examining how personality traits influence smartphone addiction through mindfulness, clarifying the underlying psychological processes, and offering theoretical and practical implications.

2.4 Hypotheses Development

2.4.1 Relationships between the Big Five Personality Traits and Smartphone Addiction

Given the observed prevalence of smartphone addiction and its links with distress-related processes in young populations [27,31], individual differences such as the Big Five personality traits have been widely examined as potential predictors of compulsive smartphone use.

Within individual-differences research on smartphone addiction, Openness has received growing attention. Openness reflects the degree to which individuals embrace new experiences, including curiosity, creativity, and exploratory tendencies [17]. Some studies report no significant association between openness and smartphone addiction [1,32]. In contrast, others argue that individuals high in openness, due to their enthusiasm for new knowledge and technologies, may use smartphones more frequently [42] and be more susceptible to problematic smartphone engagement [43]. Accordingly, this study proposes the following hypothesis:

H1a: Openness is positively associated with smartphone addiction.

Among predictors of smartphone addiction, Agreeableness often shows relatively weak predictive power, yet it remains theoretically meaningful. Agreeableness reflects tendencies toward cooperation, consideration, and interpersonal harmony and is typically negatively associated with impulsive control failures and addictive behaviors [19]. Although many studies report no significant association between agreeableness and smartphone addiction [1,32], some argue that individuals high in agreeableness, who seek to avoid conflict and please others, may increase smartphone use under interpersonal stress. However, evidence remains mixed [32,44]. Although findings are mixed, some studies that control for confounders have observed a small negative association between agreeableness and technology addiction [43,44]. Therefore, this study hypothesizes:

H1b: Agreeableness is negatively associated with smartphone addiction.

In the Big Five framework, Conscientiousness represents self-discipline, planning, goal orientation, and self-control [17]. Evidence shows that highly conscientious individuals manage time effectively, remain task-focused, and inhibit impulses, which reduces the likelihood of excessive smartphone use [19,45]. Marengo et al. [43] found that greater conscientiousness is associated with stronger emotional and behavioral regulation, which reduces addictive tendencies. Other work emphasizes the close links between conscientiousness and mindfulness or self-regulation, which enhance self-monitoring and inhibit compulsive smartphone use [15,21]. In contrast, those low in conscientiousness often lack self-discipline and goal focus, leading to uncontrolled use [43,44]. Ran [20] also indicates that via improved self-control, conscientiousness can indirectly alleviate smartphone addiction. Hence, this study proposes:

H1c: Conscientiousness is negatively associated with smartphone addiction.

Within the psychological mechanisms of smartphone addiction, Neuroticism is widely considered one of the strongest personality predictors. Individuals high in neuroticism often display emotional instability, anxiety, and impulsive behavior, and tend to use smartphones to escape reality and regulate negative emotions [9,32,46]. This trait can also weaken self-control, further increasing reliance on smartphones [4]. Doménech et al. [10] also report that neuroticism is associated with poorer emotion regulation, suggesting an indirect reinforcement of phone dependence. In line with compensatory accounts, stress and distress-related processes appear closely intertwined with problematic smartphone use [31]. Based on this evidence, it is hypothesized that:

H1d: Neuroticism is positively associated with smartphone addiction.

Many studies indicate that Extraversion may be unrelated or weakly protective regarding smartphone addiction. Bhayangkara et al. [1] found a significant negative association between extraversion and smartphone addiction, suggesting that extraverts, who prefer face-to-face interaction to satisfy social needs, rely less on virtual socializing [1,47]. Kayiş et al. [45] likewise reported that frequent engagement in social activities reduces the likelihood of smartphone overuse among extraverts. Recent evidence syntheses further suggest that extraversion is not a robust risk factor for smartphone addiction compared with traits such as neuroticism and conscientiousness [5]. Doménech et al. [10] add that extraversion correlates with stronger emotion regulation, such as cognitive reappraisal, which may indirectly lower addiction risk. In short, real-world social engagement may substitute for digital reliance among extraverts, thereby reducing addiction risk. Thus, it is hypothesized that:

H1e: Extraversion is negatively associated with smartphone addiction.

2.4.2 Relationship between Mindfulness and Smartphone Addiction

Mindfulness functions as a crucial protective factor against smartphone addiction, involving present-moment awareness and enhanced self-regulation. Individuals with high trait mindfulness can better detect internal cues and habitual impulses, thereby reducing engagement in compulsive behaviors such as excessive smartphone use [14,40]. Guided by conservation of resources theory, trait mindfulness can be conceptualized as a personal resource that supports emotion regulation and psychological well-being under stress [48,49]. Consistent with this view, evidence from randomized controlled trials indicates that mindfulness-based programs can improve psychological resources and stress-related outcomes [34,41]. Importantly, when mindfulness is operationalized as mindful attention and awareness (as in the MAAS-5), it is expected to reduce automatic or mindless smartphone checking by strengthening awareness of ongoing behavior [14,36]. Meta-analytic evidence indicates that the effects of mindfulness are generally small, underscoring the need for rigorous designs and large samples [14]. Mindfulness is not a panacea, and its active ingredients must be distinguished from popularized elements that lack substance [11].

Recent evidence nonetheless supports a significant risk-reducing role of mindfulness for smartphone addiction, achieved through enhanced self-control and emotion regulation that curb compulsive and avoidant use [14,50,51]. In addition, evidence synthesized across university samples indicates that mindfulness is associated with lower learning burnout [35], which is theoretically compatible with the view that mindfulness protects self-regulatory capacity under sustained demands. Therefore, this study hypothesizes:

H2: Trait mindfulness is negatively associated with smartphone addiction.

2.4.3 Relationships between the Big Five Traits and Mindfulness

Openness, characterized by receptivity to new experiences and cognitive flexibility, may relate to mindful awareness through curiosity and flexible attention. Prior research reports a small positive association between openness and trait mindfulness, especially for attention/awareness components [15]. Measurement also matters; attention/awareness measures (e.g., the MAAS) and multi-facet measures (e.g., the FFMQ) can yield different patterns of association with personality traits [19,52]. The introspection, self-exploration, and flexible thinking inherent in openness may support the cultivation of an attentive stance in daily life [15,19]. Although the overall correlation may be weaker than for other traits, openness remains meaningful for reflective engagement with experience and cognitive flexibility [53]. Errasti-Pérez et al. [11] further note that, although the effects of mindfulness are minor, mindful awareness remains theoretically relevant for self-regulation. Hence, this study proposes:

H3a: Openness is positively associated with trait mindfulness.

Agreeableness encompasses empathy, trust, and cooperation. These characteristics may facilitate mindful attention by reducing interpersonal conflict and promoting prosocial coping under stress, thereby supporting the maintenance of present-moment awareness [12,54]. Meta-analytic evidence indicates a small but reliable positive association between agreeableness and trait mindfulness [19]. In addition, mindfulness-related attitudes such as self-acceptance and an observational stance toward internal experience have been linked to psychological adjustment beyond the Big Five, suggesting possible pathways through which agreeableness may support mindful awareness [22]. Prior work also suggests that the supportive qualities of agreeableness may indirectly strengthen mindfulness-related self-regulation processes [53]. Doménech et al. [10] caution that associations between agreeableness and emotion regulation are not always consistent, which implies possible contextual moderation of their relationship with mindfulness. Thus, it is hypothesized that:

H3b: Agreeableness is positively associated with trait mindfulness.

Among the Big Five, Conscientiousness aligns closely with mindfulness, as reflected in self-discipline, responsibility, and planning. It is often the strongest predictor of personality tendencies in mindfulness. Individuals high in self-discipline and organization are better able to maintain steady mindfulness practice and sustained present-moment awareness [12], and meta-analytic work confirms a strong link [19]. Zhao [54] further notes that responsibility and personal standards bolster goal-directed action and awareness within mindfulness processes. Even when neuroticism is negatively associated with mindfulness, conscientiousness can compensate by helping individuals maintain mindfulness under stress [15]. Although mindfulness can be trained, conscientiousness provides stable behavioral persistence and motivational regulation that support long-term practice [53]. Therefore, this study hypothesizes:

H3c: Conscientiousness is positively associated with trait mindfulness.

Neuroticism shows a stable, significant negative association with mindfulness; its hallmarks of emotional instability, anxiety, and rumination conflict with present-moment awareness and nonjudgmental acceptance [12,15]. Meta-analytic evidence indicates that weaker self-regulation and emotional stability among individuals high in neuroticism hinder sustained mindfulness practice [19]. Avoidant and reactive modes in response to negative emotions impede entry into the nonreactive, accepting states required for mindfulness. Zhao [54] adds that intense affective lability under stress undermines the benefits of mindfulness for emotion regulation and mental health among those high in neuroticism. Doménech et al. [10] likewise link neuroticism to difficulties in emotion regulation, supporting its role as a negative predictor of mindfulness. Hence, this study proposes:

H3d: Neuroticism is negatively associated with trait mindfulness.

Extraversion reflects sociability, energetic engagement, and positive affectivity; however, its association with mindfulness is typically small and less consistent than that of neuroticism and conscientiousness [15,19]. Meta-analytic findings suggest that extraversion has limited predictive power for mindfulness outcomes, possibly because outward engagement can compete with sustained attentional monitoring in daily life [15]. Moreover, associations between extraversion and mindfulness can vary by measurement approach; attention/awareness measures, such as the MAAS, may yield different patterns than multi-facet measures, such as the FFMQ [52]. Doménech et al. [10] report a positive association between extraversion and cognitive reappraisal, suggesting a potential supportive role for extraversion in mindfulness-related regulation processes. Thus, it is hypothesized that:

H3e: Extraversion is positively associated with trait mindfulness, although the association is weaker than for other personality dimensions.

2.4.4 The Mediating Effect of Mindfulness between Big Five Personality Traits and Smartphone Addiction

Openness represents cognitive flexibility and a tendency to explore novel and complex experiences and is considered relevant to technology-use behaviors. Although some studies find no significant association with smartphone addiction [1,32], others argue that high-openness individuals, who prefer novel stimulation and technological media, may be more susceptible to problematic engagement with smartphones [42]. Mindfulness is a key protective factor that reduces avoidant and addictive tendencies via emotion regulation and self-control [12,36]. Studies also report a positive association between mindfulness and openness, especially for cognitive acceptance and experiential receptivity [15,53]. In addition, mindfulness can significantly lower smartphone addiction risk and may serve as a mediator between openness and addictive behaviors [3,55]. Among adolescents, mindfulness improves self-control and emotion regulation and reduces overuse driven by negative emotion or fear [8,21]. Errasti-Pérez et al. [11] add that, although mindfulness is not universally effective, it targets specific cognitive processes that may mediate the relationship between openness and smartphone addiction. Therefore, this study hypothesizes:

H4a: Mindfulness mediates the relationship between openness and smartphone addiction.

Individuals high in agreeableness typically display interpersonal harmony and cooperative coping under stress. These characteristics may support mindful attention by reducing conflict-related rumination and facilitating awareness of present-moment experience [12,54]. Mindfulness, operationalized as present-moment attention and awareness, can inhibit impulsive and avoidant behaviors, increase awareness and restraint in phone use, and reduce tendencies toward smartphone addiction [21,36]. Thus, mindfulness is expected to mediate the link between agreeableness and smartphone addiction, such that higher agreeableness is associated with higher mindfulness and, in turn, lower addiction risk [41,48,49]. Therefore, this study hypothesizes:

H4b: Mindfulness mediates the relationship between agreeableness and smartphone addiction.

Conscientiousness is a robust protective factor encompassing self-discipline, goal orientation, and behavioral control, and is closely associated with mindfulness tendencies [19]. Hanley and Garland [12] indicate that higher conscientiousness supports sustained present-moment awareness and nonjudgmental attitudes, thereby strengthening the self-monitoring and emotion-regulation components of mindfulness. Mindfulness is also a negative predictor of smartphone addiction, lowering impulsive and avoidant use [21]. Lan et al. [51] found that individuals high in mindfulness show greater awareness during phone use, thereby reducing compulsive and automatic behavior. Consistent with Doménech et al. [10], conscientiousness is associated with adaptive emotion regulation, which, in turn, further reduces addiction risk through mindfulness. Therefore, this study posits a mediating role for mindfulness in the relationship between conscientiousness and smartphone addiction, such that higher conscientiousness indirectly reduces addiction risk by increasing mindfulness, consistent with conservation of resources theory regarding the protection and conversion of self-regulatory resources [41,48,49,56]. Hence, this study proposes:

H4c: Mindfulness mediates the relationship between conscientiousness and smartphone addiction.

Neuroticism has long been viewed as a key risk factor for smartphone addiction and shows a stable negative relationship with mindfulness [12,15]. Due to emotional instability, anxiety, and rumination, individuals high in neuroticism are prone to avoidant strategies under stress, which can undermine sustained present-moment attention and increase mind-wandering. Such individuals may rely on smartphones to cope with distress and the risk of dependence [54]. Mindfulness, in contrast, strengthens awareness of thoughts and urges and supports self-control, thereby reducing compulsive smartphone use and technology-related addictions [14,21]. Individuals can better manage emotions and impulses by cultivating mindful attention, thereby mitigating neuroticism-driven smartphones [13]. Consistent with conservation of resources theory, mindfulness operates as a psychological resource that helps individuals high in neuroticism rely less on smartphones when facing anxiety and stress [41,48,49]. Prior work comparing self-reported and logged digital behavior indicates that self-reports may be insufficient for fully capturing relevant usage patterns, and that integrating behavioral indicators can improve measurement and inference [6,7].

H4d: Mindfulness mediates the relationship between neuroticism and smartphone addiction.

Although some evidence suggests that extraversion is a relatively weak and heterogeneous correlate of smartphone addiction [5,45], its association with trait mindfulness is also typically small and measure-dependent [15,19,52]. Frequent social engagement and positive affect among extraverts may support adaptive emotion regulation strategies (e.g., cognitive reappraisal), which could indirectly facilitate mindful awareness [10]. Since mindfulness promotes self-control and present-moment attention, reduces compulsive smartphone use [14,21], and can be conceptualized as a psychological resource within conservation of resources theory [41,48,49], a mediating role of mindfulness between extraversion and smartphone addiction is inferred [2,12]. Thus, it is hypothesized that:

H4e: Mindfulness mediates the relationship between extraversion and smartphone addiction.

3 Methods

3.1 Participants and Procedures

To reduce the risks of Type I and Type II errors, an a priori power analysis was conducted [57]. Using G*Power 3.1.9.7, we conducted a power analysis for multiple regression (fixed model, R2 deviation from zero), alpha set at 0.05, power set at 0.80, and an effect size of 0.15. The maximum number of predictors pointing to an endogenous construct in the proposed model was six. The minimum required sample size was estimated at 98 [58].

The survey was conducted in December 2023 using convenience sampling, which relies on voluntary participation and high accessibility and is commonly employed in studies focused on specific populations [59]. The questionnaire was distributed via Instagram, Facebook, LINE groups managed by university instructors, and digital learning platforms. Because the survey link was distributed via open online channels, the number of individuals who viewed the invitation could not be precisely determined; therefore, the response rate was unavailable. Before starting the questionnaire, participants read an information sheet and provided electronic informed consent.

All items were mandatory, and automatic validation checks were implemented to ensure accuracy and completeness. Invalid responses were identified and excluded based on duplicate submissions, extremely short completion times, and clear straightlining patterns (e.g., invariant responses across long item blocks). After screening and excluding 24 invalid responses, the final sample comprised 665 valid cases. Participants’ demographic characteristics are summarized in Table 1.

Sampling planning also referenced the minimum sample guideline of at least 100 participants proposed by Comrey and Lee [60] and the rule of five participants per scale item recommended by Gorsuch [61]. Regarding sample composition, 64.7% were female, and 35.3% were male. Most participants were under 30, with a mean age of 21.9 years, and the largest group was aged 21–30. By grade level, juniors accounted for 28.6% and seniors for 29%. By field of study, education-related majors accounted for the largest share at 61.5%.

3.2 Measures

All instruments used in this study were adapted from previously published questionnaires in international academic journals. Except for the Chinese Shortened Big Five Inventory, which used an existing Chinese version, the English versions of all other instruments were translated into Chinese by a professor proficient in English using the back-translation method [62]. After translation, five professors reviewed and refined the items to ensure consistency with the present research context and minimize potential bias in the scales.

3.2.1 Chinese Shortened Big Five Inventory (CBF-PI-15)

The Big Five personality traits were assessed using the Chinese Big Five Personality Inventory-15 (CBF-PI-15) [63]. The CBF-PI-15 is a 15-item shortened version derived from the Chinese Big Five Personality Inventory Brief Version (CBF-PI-B), consisting of five subscales: Neuroticism, Conscientiousness, Agreeableness, Openness, and Extraversion, with three items per trait.

In the scale development study, the CBF-PI-15 demonstrated acceptable internal consistency given its brevity, with Cronbach’s α ranging from 0.611 to 0.803 in Sample 1 (N = 10,738) and from 0.612 to 0.811 in Sample 2 (N = 256). Good model supported factorial validity fit in confirmatory factor analysis (MLRχ2 = 918.882, df = 80, CFI = 0.946, TLI = 0.929, RMSEA = 0.044, SRMR = 0.040), and measurement invariance was further supported across gender and age groups.

3.2.2 Mindful Attention Awareness Scale (MAAS)

The Mindful Attention Awareness Scale (MAAS) was originally developed as a 15-item measure of dispositional mindfulness (present-moment attention and awareness) [36]. In this study, we adopted a five-item short form (MAAS-5) adapted from the original MAAS to reduce respondent burden, drawing on prior cross-cultural validation of the MAAS-5 [39].

Items describe day-to-day lapses in attention and awareness (e.g., “I do jobs or tasks automatically, without being aware of what I’m doing”) and were rated on a 6-point frequency scale from 1 (almost always) to 6 (almost never). Scores were coded such that higher values indicate higher mindfulness. In the present sample, confirmatory factor analysis supported the adequacy of the five-item MAAS, with all item loadings exceeding 0.700 (CFI = 0.941, GFI = 0.948, NFI = 0.918, RMR = 0.039, RMSEA = 0.043). Composite reliability was 0.888, and Cronbach’s alpha was 0.844.

3.2.3 Smartphone Addiction Scale-Short Version (SAS-SV)

The Smartphone Addiction Scale-Short Version (SAS-SV), developed by Kwon et al. [64], was administered to assess smartphone addiction among university students. The original scale comprises 10 items, each rated on a 6-point Likert scale from 1 (strongly disagree) to 6 (strongly agree), with higher scores indicating greater addiction severity. The original instrument reported a Cronbach’s alpha of 0.966.

In this study, the SAS-SV items were evaluated as part of the measurement model assessment. Following standard criteria for indicator reliability and construct validity, items with inadequate performance were removed, and six items were retained in the final measurement model. For the retained items, standardized loadings exceeded 0.700, and internal consistency was satisfactory (composite reliability = 0.905; Cronbach’s alpha = 0.875), supporting the reliability of the retained SAS-SV indicators in the present sample.

Recent psychometric work in university samples further highlights the importance of evaluating reliability and measurement invariance when applying smartphone- and social-media-addiction scales across subgroups [65].

3.3 Statistical Analyses

IBM SPSS 27.0 (IBM Corp., Armonk, NY, USA) was used to analyze the observed variables, including descriptive statistics to summarize participant characteristics. Shapiro–Wilk tests were conducted to assess univariate normality of the observed variables. Given the departures from normality, bivariate associations were examined primarily using Spearman’s rank-order correlations, while Pearson correlations were also computed as a robustness check. In addition, independent-samples t-tests and one-way ANOVAs were conducted to examine group differences (e.g., gender and grade level) in the study variables. Harman’s single-factor test was performed to assess common method bias [66].

For the structural model, PLS-SEM analyses were conducted using SmartPLS 4.0 [67]. Given the present research’s emphasis on prediction, PLS-SEM was deemed appropriate. Hair et al. [68] recommend that prediction-oriented research adopt a confirmatory composite analysis (CCA) approach, and note that PLS-SEM facilitates the integration of theory testing and predictive goals [69]. Shmueli et al. [69] further proposed an evaluation framework tailored to the predictive nature of PLS-SEM, which has been widely applied in hospitality and tourism research [68].

Following standard PLS-SEM procedures, a two-step approach was employed [70,71]. First, the measurement model was evaluated with respect to indicator reliability, convergent validity, internal consistency, multicollinearity (variance inflation factor, VIF), and discriminant validity, assessed using the Heterotrait-Monotrait (HTMT) ratio of correlations [72]. Second, the structural model was assessed to test the hypotheses and predictive performance [71].

In addition to the PLS-SEM analyses, we conducted (a) a regularized Gaussian graphical model (GGM) network analysis using EBICglasso to examine conditional dependence relations among constructs, implemented in JASP (version 0.95.4.0) with an EBIC tuning parameter of 0.5 [73]; and (b) a prediction-oriented assessment using PLS-Predict with tenfold cross-validation to evaluate out-of-sample predictive performance at the construct and indicator levels [74].

4 Results

4.1 Preliminary Analyses

Before testing the measurement and structural models, distributional assumptions were examined. Shapiro-Wilk tests indicated that univariate normality was violated for the study variables (W = 0.938–0.985, all p-values < 0.001). Therefore, Spearman’s rank-order correlations are reported in Table 2. As shown in Table 2, smartphone addiction was negatively associated with mindfulness (ρ = −0.378, p < 0.001) and positively associated with neuroticism (ρ = 0.319, p < 0.001). Mindfulness was positively correlated with openness (ρ = 0.155, p < 0.001) and conscientiousness (ρ = 0.220, p < 0.001), and negatively correlated with neuroticism (ρ = −0.378, p < 0.001). The intercorrelations among the Big Five traits were generally in the expected directions. As a robustness check, Pearson correlations showed a highly similar pattern; the only notable difference was that the association between smartphone addiction and conscientiousness was marginal and non-significant under Spearman’s rho (ρ = −0.073, p = 0.059) but reached significance under Pearson’s r (r = −0.082, p = 0.034).

4.2 Assessment of the Extent of Common Method Variance

To evaluate the presence of common method variance (CMV), Harman’s single-factor test was applied to an unrotated exploratory factor analysis. The first factor accounted for 22.41% of the variance, below the 50% threshold, indicating no CMV concern in this study [66].

4.3 Measurement Model

Following Hair et al. [72], an outer loading of 0.70 was adopted as the indicator reliability benchmark, and all retained items met this criterion. Convergent validity was assessed using average variance extracted (AVE), with 0.50 as the reference value [75,76,77]. Internal consistency reliability was examined via composite reliability (CR), and all constructs exceeded 0.70 [72,75,76,77]. Based on these criteria, the final scales included three items for openness, agreeableness, conscientiousness, neuroticism, and extraversion, five items for mindfulness, and six items for smartphone addiction. As shown in Table 3, all variance inflation factor (VIF) values were below 5, indicating no multicollinearity concerns [78]. In addition, to facilitate interpretation for readers more familiar with covariance-based SEM, Table 3 also summarizes the classical CFA fit indices (e.g., CFI, GFI, NFI, RMR, and RMSEA) estimated for the current sample as supplementary information.

4.4 Discriminant Validity

Discriminant validity was tested using the Heterotrait-Monotrait ratio of correlations (HTMT) method [79], given recent critiques of the Fornell and Larcker [76] criterion. HTMT values above 0.85 [80] or 0.90 [81] may signal insufficient discriminant validity. As reported in Table 4, all HTMT values were below 0.85, supporting satisfactory discriminant validity among the study constructs [80].

4.5 Hypothesis Testing

Bootstrapping was used to examine the significance of hypothesized paths in the structural model [82]. This approach evaluates estimates’ statistical significance and sampling error [75] and allows inspection of standardized path coefficients (β), significance levels, and R2 for endogenous constructs. The structural model and standardized path coefficients are presented in Fig. 1, and the corresponding bootstrapping results are summarized in Table 5. The results in Fig. 1 show significant paths for conscientiousness → mindfulness (β = 0.224, p < 0.001), neuroticism → mindfulness (β = −0.372, p < 0.001), mindfulness → smartphone addiction (β = −0.305, p < 0.001), neuroticism → smartphone addiction (β = 0.278, p < 0.001), and openness → smartphone addiction (β = 0.163, p = 0.003). In contrast, agreeableness → smartphone addiction (β = −0.021, p = 0.687), conscientiousness → smartphone addiction (β = −0.018, p = 0.693), extraversion → smartphone addiction (β = −0.048, p = 0.379), openness → mindfulness (β = 0.008, p = 0.859), agreeableness → mindfulness (β = −0.049, p = 0.300), and extraversion → mindfulness (β = −0.040, p = 0.421) were not significant. In particular, the direct path from conscientiousness to smartphone addiction was not supported because its confidence interval included zero (95% CI [−0.100, 0.077]) [83].

Figure 1: Structural model. Note: SA, smartphone addiction; MI, mindfulness; OP, openness; AG, agreeableness; CO, conscientiousness; NE, neuroticism; EX, extraversion. Values on the structural paths represent standardized path coefficients (β). Values on the measurement model links (between indicators and constructs) represent standardized outer loadings. Values inside the endogenous construct circles represent explained variance (R2).

Following Hair et al. [72], mediation analysis was performed via bootstrapping to test the indirect effects of mindfulness on the relationship between personality traits and smartphone addiction (see Table 6). Mindfulness significantly mediated the effects of conscientiousness (β = −0.068, p < 0.001) and neuroticism (β = 0.113, p < 0.001) on smartphone addiction. Based on the Variance Accounted For (VAF), the mediation magnitude was substantial for conscientiousness (79.07%) and moderate for neuroticism (28.90%) [72]. Overall, the results highlight mindfulness as a key mechanism linking personality traits to smartphone addiction. Accordingly, H1a, H1d, and H2 were supported, whereas H1b, H1c, and H1e were not supported. Regarding antecedents of mindfulness, H3c and H3d were supported, whereas H3a, H3b, and H3e were not supported. In the mediation analysis, mindfulness significantly mediated the effects of conscientiousness (H4c) and neuroticism (H4d) on smartphone addiction, while H4a, H4b, and H4e were not supported.

4.6 Explanatory Power of the Model

Model explanatory power was evaluated using the coefficient of determination R2 obtained from the PLS algorithm in SmartPLS. All R2 values exceeded the recommended threshold of 0.10 [84]. The R2 for Mindfulness was 0.204, and for Smartphone addiction, 0.218.

4.7 Partial Correlation Network Analysis

To examine the conditional dependence structure among the study variables and to probe whether conscientiousness has a unique association with smartphone addiction beyond shared variance with mindfulness and the other Big Five traits, we estimated a regularized Gaussian graphical model (GGM) using EBICglasso [73,85]. Edges represent undirected partial correlations between nodes after controlling for all other nodes in the network. The analysis was performed in JASP (0.95.4.0) with weighted and signed edges and an EBIC tuning parameter of 0.5 (N = 665). Table 7 presents the resulting partial correlation weights, and Fig. 2 visualizes the network. The network comprised seven nodes with 14 non-zero edges (14/21; sparsity = 0.333). Mindfulness showed a negative partial correlation with smartphone addiction (weight = −0.262), whereas neuroticism showed a positive partial correlation with smartphone addiction (weight = 0.231) and a negative partial correlation with mindfulness (weight = −0.257). Conscientiousness was positively associated with mindfulness (weight = 0.148). Still, its edge with smartphone addiction was shrunk to zero (weight = 0.000), suggesting that conscientiousness is linked to smartphone addiction primarily through its conditional association with mindfulness rather than via a distinct direct connection.

Figure 2: Regularized partial correlation network (EBICglasso Gaussian graphical model). Note: Edges indicate undirected partial correlations controlling for all other nodes; edge thickness reflects absolute weight, and color indicates the sign of the association.

4.8 Predictive Power of the Model

In line with a prediction-oriented assessment procedure for PLS-SEM, PLS-Predict was conducted with tenfold cross-validation to generate predictions at the construct and indicator levels [86]. Predictive relevance was first assessed by verifying that the Q2 values for latent constructs were greater than zero. Next, predictive power was assessed by comparing the RMSE or MAE differences between the PLS and linear (LM) models. RMSE is used by default, while MAE is preferred when error distributions are highly skewed [86]. If all PLS-LM differences are negative, the model exhibits high predictive power; if most are negative, medium predictive power; if only a few are negative, low predictive power. Regarding predictive relevance, Q2 values were above zero for both endogenous constructs (mindfulness: Q2 = 0.066; smartphone addiction: Q2 = 0.135), indicating adequate predictive relevance. As summarized in Table 8, all Q2 values exceeded zero, and for most indicators, the PLS model showed a lower RMSE than the LM model, indicating medium predictive power [74,86].

4.9 Comparison of Differences in Gender, Grade, and Major on the Measures

Table 9 shows that mindfulness differed significantly by gender and major, whereas smartphone addiction differed significantly by grade. Females scored higher on mindfulness (Mean = 4.136, SD = 0.976) than males (Mean = 3.893, SD = 1.023), t = 2.977, p < 0.01. By grade, sophomores (Mean = 3.606, SD = 1.212) and seniors (Mean = 3.560, SD = 1.201) reported higher smartphone addiction than juniors (Mean = 3.236, SD = 1.431) and master’s students (Mean = 2.826, SD = 1.223). With respect to major, students in the “other majors” category showed higher mindfulness (Mean = 4.382, SD = 0.964) than those in education (Mean = 4.021, SD = 0.988), medical-agricultural-life sciences (Mean = 3.852, SD = 1.174), and science-engineering (Mean = 3.845, SD = 1.126). The latter three groups did not differ significantly from one another.

5 Discussion

5.1 Relationships between the Big Five Personality Traits and Smartphone Addiction

This study examined links between the Big Five personality traits and smartphone addiction. The results indicated that neuroticism and openness were the primary direct predictors of smartphone addiction, whereas conscientiousness primarily exerted an indirect protective effect through mindfulness.

Neuroticism showed a stable positive association with smartphone addiction, indicating that individuals high in neuroticism tend to use smartphones to regulate negative emotions, which elevates addiction risk [9,32,46]. In addition, neuroticism was negatively related to mindfulness, suggesting that affective lability and rumination undermine self-regulatory capacity and indirectly foster addictive behavior [12,15,22].

By contrast, conscientiousness functioned as a protective trait mainly via a mindfulness pathway. Individuals high in conscientiousness, characterized by self-discipline and regularity, were better able to sustain mindfulness processes and inhibit impulsive smartphone use [19]. Although the direct effect of conscientiousness on smartphone addiction was not significant, conscientiousness was positively associated with mindfulness, and mindfulness mediated the association between conscientiousness and smartphone addiction [20,53].

Openness was positively related to smartphone addiction, implying that preferences for novelty and sensory stimulation may increase usage frequency [42,87], even though some studies report unstable associations [1,32]. As for extraversion and agreeableness, the present study found no significant effects. This pattern aligns with evidence that their associations with smartphone addiction tend to be comparatively small and heterogeneous across studies relative to neuroticism and openness [5,44,45]. Notably, openness showed a negligible zero-order correlation with smartphone addiction (ρ = 0.019) but a significant positive direct effect in the structural model (β = 0.163). This discrepancy is consistent with a suppression effect in multivariable models, in which correlated predictors can mask a variable’s unique predictive component until other predictors are controlled [88].

Overall, neuroticism and openness emerged as key direct predictors of smartphone addiction, and the mediating role of mindfulness highlighted conscientiousness as an indirect protective factor, reinforcing the applied value of mindfulness in prevention and intervention strategies for smartphone addiction.

5.2 Relationship between Mindfulness and Smartphone Addiction

Mindfulness showed a significant negative association with smartphone addiction. Individuals with higher trait mindfulness were less likely to develop addictive behaviors, which supports mindfulness as a self-regulatory mechanism [14,21]. The facet of acting with awareness can effectively suppress dependence driven by attentional impulsivity [21].

Woodlief et al. reported that individuals low in mindfulness tend to experience diminished awareness during high-involvement use, which heightens addiction risk [13]. Mindfulness also strengthens executive control and reduces overuse; its protective effect is particularly evident among high-risk groups, such as university students [3]. Individuals with higher mindfulness can more effectively detect the emotions and impulses underlying phone use and respond deliberately [14,50,51].

Mindfulness may strengthen self-control and emotion regulation, thereby supporting psychological well-being and reducing the risk of addictive smartphone use [14,34,50].

In sum, mindfulness is a cultivable psychological disposition with empirical evidence supporting its use in preventing and regulating smartphone addiction. Future research should clarify its internal mechanisms and application strategies. Future work could also assess students’ digital health literacy using validated instruments to better tailor campus support and to interpret self-reported or behavioral indicators of use [89,90,91,92]. Family-level strategies may complement campus-based programs; for example, mindful parenting approaches and simple affect-labeling practices may help strengthen supportive environments and emotion regulation around digital habits [93,94,95,96].

5.3 Relationships between the Big Five Personality Traits and Mindfulness

The present study found significant associations between conscientiousness, neuroticism, and mindfulness, consistent with prior literature. Conscientiousness was positively associated with mindfulness, indicating that individuals with self-discipline and planning skills are more likely to demonstrate present-moment awareness and nonjudgmental attitudes [12,15,19]. The mindfulness facets of self-regulation and acting with awareness directly reflect conscientiousness and its benefits for emotion regulation [12].

In contrast, neuroticism showed a negative association with mindfulness. High-neuroticism individuals, marked by heightened emotional reactivity, rumination, and anxiety, find it difficult to maintain awareness and acceptance within mindfulness processes [15,22,54]. The emotional regulation and clarity emphasized by mindfulness are the psychological resources that individuals high in neuroticism tend to lack [12].

Although openness, agreeableness, and extraversion did not show significant associations with mindfulness here, prior work suggests that links between these traits and mindfulness are often small and can vary across operationalizations (e.g., attention/awareness-focused MAAS versus multi-facet measures) [15,19,52]. For example, openness may be more strongly related to measures of reflective awareness, whereas extraversion may show heterogeneous associations across samples and contexts [15,19]. These results support the status of conscientiousness and neuroticism as core personality predictors of mindfulness tendencies and strengthen the construct validity and applied implications of mindfulness within the context of personality psychology.

5.4 The Mediating Effect of Mindfulness between Big Five Personality Traits and Smartphone Addiction

This study confirmed a significant mediating role of mindfulness in the relationship between conscientiousness and neuroticism, on the one hand, and smartphone addiction, on the other, underscoring mindfulness as a key psychological mechanism in regulating digital behaviors. Individuals high in conscientiousness, who tend to have stronger self-discipline and self-standards, are likelier to cultivate stable mindfulness, thereby reducing impulsive and dependent phone use [12,20,21]. Mindfulness enhances awareness of motives and emotional reactions, promotes self-control, and protects against addictive behaviors [21].

Conversely, individuals high in neuroticism, characterized by anxiety, rumination, and affective instability, tend to lack the awareness and nonjudgmental stance emphasized by mindfulness [19,54]. The present findings indicate that mindfulness can attenuate the adverse link between neuroticism and smartphone addiction, reducing the tendency to use phones as an avoidance tool [9,22]. Through self-regulation and emotional acceptance, mindfulness mitigates the impact of neuroticism on digital addiction [14,97]. Related research also confirms that mindfulness interventions effectively reduce smartphone addiction and associated psychological risks [3,51].

Overall, this study verifies the mediating role of mindfulness in the relationship between personality and behavior and reinforces its practical potential for promoting digital well-being and for behavior regulation [12,13,41,48,49].

5.5 Comparison of Differences in Gender, Grade, and Major on the Measures

To contextualize the main model, mindfulness and smartphone addiction were further examined for differences across demographic subgroups. The study found that females scored significantly higher than males on mindfulness, indicating gender differences in awareness and self-regulation. This aligns with the gender-related perspective discussed by Hanley and Garland [12] and Banfi and Randall [15], and may also reflect cultural expectations regarding emotional expression and introspection [19].

By grade, sophomores and seniors exhibited higher tendencies toward smartphone addiction, possibly because career exploration and academic pressure promote avoidant use behaviors [9,54]. A possible interpretation is that smartphone reliance may fluctuate across academic stages: sophomores may increase their use after adapting to campus routines; juniors may reduce reliance under heavier coursework and capstone demands; seniors may show a rebound under career-planning pressure; and master’s students may report lower dependence due to clearer goals and stronger autonomy. This pattern is also consistent with the findings of Marengo et al. [43], who observed associations between neuroticism and smartphone use disorders.

Although differences by major were limited, higher than students in education, natural sciences and engineering, and medical, biological, and agricultural fields, but not significantly different from those in humanities, social sciences, and business management. As noted by Hanley and Garland [12], dispositional factors related to personality structure may be reflected in mindfulness tendencies, and humanities- and arts-related training may be associated with greater reflective attention to experience and self-acceptance [22].

Importantly, these demographic differences can be interpreted through the lens of the study’s self-regulation. Specifically, subgroup variation in mindfulness and smartphone addiction is consistent with the proposed mechanism that mindfulness-related regulatory capacity is inversely associated with addictive smartphone behaviors [14,21,50,98]. Because demographic factors were not modeled as moderators in the current structural model, future research may formally test whether the mediation pathways differ across demographic groups (e.g., moderated mediation), thereby clarifying the boundary conditions of the proposed mechanism.

6 Implications

6.1 Theoretical Implications

This study offers several theoretical contributions regarding the mediating role of mindfulness in the relationship between the Big Five personality traits and smartphone addiction. The findings indicate that neuroticism and conscientiousness were associated with mindfulness in directions consistent with the proposed model through negative and positive pathways, respectively, and that these pathways influence smartphone addiction, thereby extending existing frameworks that link personality to digital behavior [12,15].

The mediation model positions mindfulness as a key variable that bridges individual-differences models and self-regulation theory, addressing the limitations of prior research focused on single variables [21,54,56]. In addition, the positive association between openness and smartphone addiction suggests that a strong preference for novelty may heighten the risk of technology-related involvement, extending observations by Mahajan et al. [42] and Woodlief et al. [13]. However, the indirect effect of openness via mindfulness was not supported, indicating that mindful attention and awareness did not statistically explain the association between openness and smartphone addiction in this sample.

By testing both direct and indirect personality effects on mindfulness, the study addresses theoretical questions about inconsistent association strengths noted by Giluk [19] and Stein et al. [22], and strengthens the view of mindfulness as a “metatrait stability” construct [12]. The results also support considering neuroticism as a key target for psychological intervention, echoing recommendations by previous meta-analytic work [12,15,19] and Banfi and Randall [15] to integrate multilevel personality models into future mindfulness interventions.

Furthermore, the structural modeling evidence for mindfulness as a mediator is consistent with conservation of resources theory, which conceptualizes mindfulness as a psychological resource, underscoring its potential for digital health and addiction prevention [3,41,48,49,51]. In sum, the study advances an integrated model linking personality, mindfulness, and addiction, reinforces the theoretical position of mindfulness in technology-related behavior and personality psychology, and offers new perspectives for individualized digital intervention strategies [5,12,22].

6.2 Practical Implications

This study shows that mindfulness significantly mediates the relationships between the Big Five traits, especially neuroticism and conscientiousness, and smartphone addiction, yielding critical practical insights. At the policy level, governments should regard smartphone addiction as a priority issue in youth mental health [4,9]. This emphasis is further warranted given recent evidence of substantial smartphone addiction prevalence in adolescent populations [27].

Ministries of Education could launch a “Mindfulness Literacy Education Promotion Program”, incorporate it into higher education and mental health policies, and integrate it with personality screening to deliver mindfulness and digital regulation courses for students who are high in neuroticism and low in conscientiousness [15]. Systems can also integrate “Digital Health Literacy Indicators” to track trends in personality, mindfulness, and addiction risk among students.

In teaching practice, instructors can normalize mindfulness techniques such as mindful breathing, emotion journals, and guided audio to enhance students’ present-moment awareness and attentional stability. For students high in neuroticism, approaches that strengthen awareness of distress cues and reduce automatic smartphone checking in response to negative affect may be particularly relevant [54]. For highly conscientious students, task planning and reflective activities can consolidate self-discipline and goal alignment, thereby further supporting mindful self-regulation in daily routines [20]. Instructors should also attend to anxiety and attentional problems among students low in mindfulness and use mindfulness-based techniques to regulate emotion and academic stress [14].

Evidence from randomized controlled research also suggests that mindfulness interventions can strengthen adolescents’ psychological resources [34], supporting the feasibility of incorporating brief, structured practices into educational settings. To make these implications more actionable, universities may consider a stepped, low-burden intervention package aligned with students’ routine smartphone-use contexts.

For example, institutions could embed brief mindfulness micro-practices (3–5 min) at the start of large-enrollment lectures for 4–6 weeks, focusing on breath awareness and “pause-and-notice” attention resets. In parallel, counseling centers could provide weekly group sessions (30–45 min) led by trained counselors that emphasize breath awareness, emotion labeling, and coping with urges (e.g., noticing cravings to check the phone without immediately acting on them). These activities can be paired with app-based guided exercises and adherence reminders to support daily practice and reduce reliance on one-off workshops.

Importantly, a risk-informed delivery can be implemented in a non-stigmatizing manner: students with higher neuroticism and lower conscientiousness may benefit from more structured support and monitoring, whereas students high in openness may be offered alternative offline novelty and creative engagement options to reduce digital stimulation-seeking [12,15]. To improve intervention effectiveness, schools can provide mindfulness-teaching workshops and personality-oriented course-design training that enhance teachers’ sensitivity to stress and addiction risks [15]. Mindfulness can also support differentiated instruction by tailoring strategies to personality profiles, strengthening students’ self-discipline and psychological resilience.

In addition, synthesised evidence indicates that mindfulness is associated with lower learning burnout among university students [35], suggesting potential spillover benefits for academic functioning when mindfulness is cultivated consistently. For university students, proactive mindfulness practice is advisable to develop self-awareness for digital restraint and stress detection. Research indicates that mindfulness reduces automatic smartphone use and mitigates avoidant behaviors driven by academic burnout and self-regulatory fatigue [3,13]. Daily mindful breathing and brief awareness exercises for 5–10 min can strengthen present-moment attention and awareness of urges, supporting psychological clarity under pressure [22,36]. Mindfulness apps, behavior logs, and study tools can help students identify use contexts and emotional triggers [21].

For individuals high in openness, offering diverse offline activities and novelty-rich alternatives (e.g., creative projects or exploration-oriented hobbies) may help satisfy their need for stimulation while reducing reliance on smartphones [5]. For individuals high in neuroticism, establishing self-talk and reflective routines can address emotion-regulation difficulties and addiction risk [22]. Given the links between mental distress and problematic smartphone use observed in student populations [31], universities may also integrate routine screening and referral pathways that connect stress-management supports with digital-use counseling.

At the family level, parents’ understanding of mindfulness and personality differences can promote their children’s self-discipline and digital health. For children high in neuroticism, a noncritical stance can foster emotional safety and regulatory capacity [54]. Families can cultivate a culture of mindfulness through shared practices such as sitting meditation and emotion labeling [97]. Emphasizing routines and positive communication and sustaining support for highly conscientious children will also aid the development of long-term self-management skills [20].

7 Limitations and Scope for Future Research

This study has several limitations. First, convenience sampling was used, and the sample was concentrated in the field of education at a university in Taiwan with an imbalanced gender ratio, which may affect representativeness and external validity [99]. Accordingly, subgroup comparisons (e.g., gender and major) should be interpreted cautiously and replicated in more balanced samples. Future work should adopt stratified random or multi-campus sampling to enhance inferential power [100,101].

Second, data were collected via self-report questionnaires. Although Harman’s single-factor test was applied to address common method variance, response bias, and social desirability effects cannot be entirely ruled out [66,102,103]. In addition, mindfulness was assessed using the MAAS-5, which primarily captures mindful attention and awareness rather than the full range of facets; future studies may employ multi-facet measures (e.g., the FFMQ) to examine facet-specific mechanisms [38,39]. Subsequent studies can combine behavioral logs and third-party assessments to strengthen reliability and validity [6,7].

Third, the CBF-PI-15 short form was used. While it shows fundamental validity, each dimension contains only three items, which may not capture fine-grained personality differences [63,104]. Future research may employ the full version or integrate qualitative data to increase explanatory depth.

Fourth, because this study employed a cross-sectional design, the observed associations should not be interpreted causally, and the temporal ordering implied by the mediation model cannot be firmly established. Cross-sectional mediation may be subject to temporal bias and may not accurately represent longitudinal processes [105,106]. Future studies should employ longitudinal designs to validate the directional relationships among personality traits, mindfulness, and smartphone addiction. Fifth, PLS-SEM was used for prediction. Despite an explanatory and predictive performance, covariance-based structural equation modeling (CB-SEM) and multi-group analyses were not conducted for theory testing and comparison, thereby limiting theoretical inference [72]. Future research should combine CB-SEM and multi-group analysis (MGA) to examine model robustness.

Sixth, mindfulness and personality are culturally sensitive. Because this study was situated in a Chinese cultural context, direct generalization to other cultural groups is limited [107]. Cross-cultural comparisons and measurement invariance testing are encouraged to evaluate model generalizability. Overall, future studies should advance the field by employing broader sampling, multi-source data, optimized instruments, diverse designs, and tests of cultural generalizability to deepen theoretical understanding and practical applications.

8 Conclusions

Using an empirical survey of Taiwanese university students, this study confirmed positive associations between neuroticism and smartphone addiction and identified openness as an additional risk-related trait. Mindfulness was a robust protective factor and served as a key self-regulatory mechanism linking personality traits to addictive smartphone behavior.

Importantly, conscientiousness did not show a significant direct association with smartphone addiction; rather, it functioned primarily as an indirect protective factor through mindfulness. These findings strengthen the integration of personality psychology and digital behavior research and underscore the applied potential of mindfulness-based approaches for promoting healthier digital habits and student digital well-being.

In practice, schools are advised to implement mindfulness programs to enhance students’ self-awareness and self-regulatory capacity and to prevent smartphone addiction. Educational agencies and policymakers can develop targeted interventions for students with high-risk personality profiles. Because the sample was drawn from students in the education field and the design was cross-sectional, generalization and causal interpretation are limited. Future research can employ multi-group and longitudinal designs and incorporate multi-source data to verify findings.