Social media use,mental fatigue,and sleep quality among undergraduate students: A cross-sectional study

Introduction

Social media has become deeply embedded in the daily routines of university students, shaping not only how they communicate but also how they study, relax, and manage their emotional lives. Importantly, social media engagement is not a unitary construct. Prior research distinguishes between functional domains of use, including academic, socialization, entertainment, and information-seeking purposes (Azizi et al., 2019; Gupta and Bashir, 2018; Ostic et al., 2021). These domains may differ in their cognitive demands and potential implications for fatigue and sleep, particularly in university settings where academic-related digital engagement often coexists with recreational use. While these platforms offer academic and social benefits including easier access to learning communities and timely information (Azizi et al., 2019), a growing body of research suggests that higher levels of social media use can carry significant cognitive and emotional costs. Among the most concerning are mental fatigue and sleep disturbances, two interconnected outcomes that can severely undermine students’ psychological well-being and academic functioning (Ostic et al., 2021; Alshoaibi et al., 2023; Qin et al., 2024; Yu et al., 2024).

Mental fatigue is a multidimensional state of cognitive and emotional exhaustion that may arise from sustained cognitive demands across multiple contexts, including but not limited to prolonged or intensive digital and social media engagement. In this study, mental fatigue is conceptualized as a general cognitive state rather than a platform-specific construct such as “social media fatigue,” which was not directly assessed. Mental fatigue has been associated with reduced attention, emotional dysregulation, and impaired academic performance, especially when social media use encroaches on study time or sleep (Karaduman et al., 2025; Malik et al., 2020). According to cognitive load theory (Sweller, 2011), the rapid task-switching and fragmented attention characteristic of social media engagement overloads working memory, draining cognitive resources and leaving users mentally depleted (Lepp et al., 2015; Vogel et al., 2014). This cognitive strain is further compounded by emotional demands and the compulsive nature of online interactions, which can mimic behavioral addiction (Aljawarneh et al., 2024; Griffiths et al., 2014).

Concurrently, sleep as an essential element for memory consolidation, emotional regulation, and executive function appears increasingly compromised among young adults engaged in nighttime social media use. Empirical studies have consistently shown that exposure to social media before bedtime interferes with melatonin secretion, prolongs sleep latency, and leads to fragmented, non-restorative sleep (Alonzo et al., 2021; Bányai et al., 2017). Poor sleep quality, in turn, has been linked to decreased cognitive performance, reduced academic success, and increased emotional reactivity (Almarzouki et al., 2022; Tomaso et al., 2021).

Importantly, recent literature highlights a distinction between adaptive and pathological patterns of social media engagement. While the former can support learning and social connection, the latter is marked by impaired self-regulation, addictive tendencies, and heightened psychological distress (Kennard et al., 2025). Emerging regional studies, including those from the United Arab Emirates, have revealed associations between problematic social media use and poor sleep quality, anxiety, and depressive symptoms in university students (Al Kazhali et al., 2023; Hasan et al., 2023), emphasizing the urgency of addressing these issues in higher education. However, much of the existing research treats social media use, mental fatigue, and sleep disruption as isolated phenomena. Few studies have examined how these factors interact—particularly whether cognitive exhaustion statistically account for the association between digital engagement and impaired sleep. This gap is particularly significant for cognitive psychology, as it limits our understanding of how modern digital habits shape attention, memory, and emotion regulation in young adults.

From a health psychology perspective, social media use can be conceptualized as a modifiable health-related behavior that interacts with stress processes, self-regulatory capacity, and recovery behaviors such as sleep (Baumeister et al., 2007; Cain and Gradisar, 2010). Frequent digital engagement may increase cognitive and emotional demands through attentional fragmentation, multitasking, and ongoing social evaluation, potentially taxing self-regulation and impairing recovery (Thomée, 2012; Twenge et al., 2018). Sleep quality represents a core health behavior and recovery indicator, while mental fatigue reflects cognitive strain that may co-occur with diminished recovery and sustained stress exposure (Ampofo et al., 2025; Brautsch et al., 2023). Framing these constructs within stress and self-regulation models provides a theoretical foundation for interpreting associations between digital behavior, mental fatigue, and sleep health in undergraduate populations.

Building on this framework, the present study adopts an exploratory, cross-sectional approach to examine the associations among social media use, mental fatigue, and sleep quality among undergraduate university students, using a health psychology perspective. Specifically, the study examined whether mental fatigue statistically accounts for the association between social media use and sleep quality, while acknowledging the cross-sectional and associative nature of the data. Rather than proposing a causal or mechanistic explanation, mediation is conceptualized as a regression-based indirect association intended to clarify patterns of association and inform hypothesis generation for future longitudinal and experimental research. Based on the literature, the study tested the following hypotheses: (H1) greater social media use is associated with higher levels of mental fatigue; (H2) greater social media use is associated with poorer sleep quality; (H3) higher mental fatigue is associated with poorer sleep quality; and (H4) mental fatigue statistically accounts for the association between social media use and sleep quality.

Methods

Results

Descriptive and subgroup analyses are presented to contextualize patterns of social media use, mental fatigue, and sleep quality prior to hypothesis-driven regression and mediation analyses.

Demographic characteristics of participants

Participant demographic characteristics and descriptive statistics for key study variables are presented in Table 1. The final sample consisted of 762 undergraduate students. The mean age of participants was 22.07 years (SD = 2.95; range = 17–40). The sample included students from multiple campuses, academic disciplines, and years of study, with representation of both genders. Overall, a high prevalence of poor sleep quality and moderate to high levels of mental fatigue was observed in the sample.

OPEN IN VIEWER

Table 1.

Demographic characteristics of the study participants (N = 762).

Variable	Category	n (%)
Age	Mean (SD)	22.07 (2.95)
Gender	Male	179 (23.5)
	Female	583 (76.5)
Marital Status	Single	602 (79.0)
	Married	137 (18.0)
	Divorced	13 (1.7)
	Widowed	10 (1.3)
Year of study	Year 1	91 (11.9)
	Year 2	171 (22.4)
	Year 3	246 (32.3)
	Year 4	254 (33.3)
Major	Health Science	257 (33.7)
	Business	181 (23.8)
	Computer Information Science	144 (18.9)
	Applied Media	97 (12.7)
	Engineering Technologies and Science	48 (6.3)
	Education	35 (4.6)
GPA	0.1–1.0	18 (2.4)
	1.1–2.0	75 (9.8)
	2.1–3.0	465 (61.0)
	3.1–4.0	204 (26.8)

Correlation between social media use, sleep quality, mental fatigue, and age

Pearson correlation analyses are presented in Table 2. Age demonstrated a weak but statistically significant positive association with sleep quality (r = 0.194, p < 0.001), indicating that older participants reported slightly better sleep quality. Academic-related social media use, defined as the use of social media platforms for coursework-related communication, academic information sharing, and peer collaboration, was positively associated with mental fatigue (r = 0.161, p < 0.001). Social media use for socialization was weakly but significantly associated with both mental fatigue (r = 0.192, p < 0.001) and sleep quality (r = 0.099, p = 0.006). Information-seeking social media use was also weakly associated with mental fatigue (r = 0.167, p < 0.001) and sleep quality (r = 0.081, p = 0.026).

OPEN IN VIEWER

Table 2.

Pearson’s correlations between age, social media use, sleep quality, and mental fatigue (N = 762).

Variable		1	2	3	4	5	6	7
Age	r	1	0.015	0.041	0.004	0.034	0.194**	0.252**
	p		0.686	0.260	0.920	0.346	<0.001	<0.001
Academic	r	0.015	1	0.841**	0.886**	0.861**	0.080*	0.161**
	p	0.686		<0.001	<0.001	<0.001	0.027	<0.001
Socialization	r	0.041	0.841**	1	0.806**	0.875**	0.099**	0.192**
	p	0.260	<0.001		<0.001	<0.001	0.006	<0.001
Entertainment	r	0.004	0.886**	0.806**	1	0.798**	0.060	0.143**
	p	0.920	<0.001	<0.001		<0.001	0.099	<0.001
Information	r	0.034	0.861**	0.875**	0.798**	1	0.081*	0.167**
	p	0.346	<0.001	<0.001	<0.001		0.026	<0.001
Sleep quality Total	r	0.194**	0.080*	0.099**	0.060	0.081*	1	0.564**
	p	<0.001	0.027	0.006	0.099	0.026		<0.001
Total Mental Fatigue	r	0.252**	0.161**	0.192**	0.143**	0.167**	0.564**	1
	p	<0.001	<0.001	<0.001	<0.001	<0.001	<0.001

**

Correlation is significant at the 0.01 level (two-tailed).

*

Correlation is significant at the 0.05 level (two-tailed).

Age group differences in social media use, mental fatigue, and sleep quality

Independent samples t-tests were conducted to examine differences in social media use, mental fatigue, and sleep quality between two age groups (18–23 years and 24–29 years). Students aged 24–29 years reported significantly higher total social media use (M = 93.23, SD = 14.34) compared with those aged 18–23 years (M = 88.17, SD = 15.76), t(762) = −3.11, p = 0.002. A significant difference was also observed for social media use related to socialization, with higher scores among the older group, t(762) = −3.22, p = 0.001. However, no significant difference was observed in academic-related social media use between the two groups, t(762) = 0.78, p = 0.361. In contrast, mental fatigue was significantly higher among students aged 24–29 years (M = 22.50, SD = 11.02) compared with those aged 18–23 years (M = 18.76, SD = 10.53), t(762) = −3.78, p < 0.001. A statistically significant difference in sleep quality was observed across age groups (t(762) = −2.43, p = 0.015)), with students aged 24–29 reporting poorer sleep quality than those aged 18–23. However, the magnitude of this difference was modest, suggesting limited practical significance.

Gender differences in sleep quality, mental fatigue, and social media use

Chi-square analyses indicated significant associations between gender and sleep quality, χ²(1, N = 762) = 187.23, p < 0.001, with a higher proportion of female students classified as having poor sleep quality. Gender was also significantly associated with mental fatigue, χ²(1, N = 762) = 193.52, p < 0.001, with higher fatigue prevalence among female students. A significant association was observed between gender and overall social media use, χ²(1, N = 762) = 97.82, p < 0.001, although the linear-by-linear association was not significant (p = 0.212), suggesting non-linear differences across usage categories.

Regression analysis

Multiple linear regression analyses were conducted to explore the predictors of sleep quality and mental fatigue (Table 3). The first model predicting sleep quality was statistically significant, F(3, 758) = 119.59, p < 0.001, and explained 32.1% of the variance (adjusted R² = 0.319). Mental fatigue emerged as the strongest predictor of sleep quality (β = 0.552, p < 0.001). Social media use was not a significant predictor (p = 0.674), while age showed a marginal effect (p = 0.074). A second regression model was conducted to explore reciprocal associations between mental fatigue and sleep quality, consistent with prior literature suggesting potential bidirectionality. The model was statistically significant, F(3, 758) = 139.36, p < 0.001, and explained 35.5% of the variance in mental fatigue (adjusted R² = 0.353). Sleep quality was the strongest predictor (β = 0.525, p < 0.001), followed by age (β = 0.147, p < 0.001) and social media use (β = 0.128, p < 0.001).

OPEN IN VIEWER

Table 3.

Regression analysis predicting mental fatigue and sleep quality (N = 762).

Model	Predictor	B	SE B	β	t	p	95% CI (Lower)	95% CI (Upper)
Model 1: Outcome: SQ	Social Media Use	−0.022	0.052	−0.013	−0.421	0.674	−0.124	0.08
	Mental Fatigue	1.111	0.063	0.552	17.585	<0.001	0.987	1.235
	Age	0.435	0.243	0.055	1.788	0.074	−0.043	0.912
Model 2: Outcome: MF	Social Media Use	0.108	0.025	0.128	4.358	<0.001	0.06	0.157
	Age	0.576	0.116	0.147	4.954	<0.001	0.347	0.804
	Sleep Quality	0.261	0.015	0.525	17.585	<0.001	0.232	0.29

Note. SMU: Social Media Use; MF: Mental Fatigue; SQ: Sleep Quality.

Mediation and sensitivity analyses

Regression-based mediation analysis indicated that social media use was positively associated with mental fatigue (B = 0.13, p = 0.041), and mental fatigue was strongly associated with poorer sleep quality (B = 1.04, p < 0.001). The indirect association between social media use and sleep quality through mental fatigue was statistically significant (indirect effect = 0.14, 95% bootstrap CI [0.08, 0.20]). After accounting for mental fatigue, the direct association between social media use and sleep quality was not statistically significant (B = −0.02, p = 0.674). A sensitivity analysis testing an alternative directional model (sleep quality → mental fatigue → social media use) yielded a smaller indirect association (indirect effect = 0.06, 95% CI [0.03, 0.09]), suggesting asymmetry in the observed associations while acknowledging the cross-sectional nature of the data.

Discussion

This study examined the associations among social media use, mental fatigue, and sleep quality among undergraduate students adopting an exploratory cross-sectional design. The findings indicate that higher levels of social media use were associated with greater mental fatigue, and that greater mental fatigue was associated with poorer sleep quality. These findings are interpreted as associative rather than causal, consistent with the cross-sectional nature of the data.

A prominent finding was the high prevalence of poor sleep quality among participants, with more than 60% reporting poor sleep. This aligns with international evidence documenting widespread sleep disturbances in university populations. Comparable prevalence rates have been reported among students in Nepal (Shrestha et al., 2021), Iran (Jalali et al., 2020), and the United Arab Emirates (Abedalqader et al., 2019), suggesting that sleep problems among undergraduates represent a global concern rather than a context-specific phenomenon. In parallel, a large proportion of students reported moderate to high levels of mental fatigue, reinforcing concerns about increasing cognitive burden in contemporary academic environments.

The observed association between social media use and mental fatigue is consistent with previous research indicating that intensive information and communication technology use is linked to mental strain and psychological distress among young adults (Thomée, 2012). Social media platforms frequently require sustained attention, rapid task switching, and continuous cognitive engagement, which may contribute to subjective experiences of exhaustion over time. Notably, the present findings suggest that academic-related social media use, in addition to recreational use, was associated with higher mental fatigue. This supports emerging evidence that digitally mediated academic engagement may blur boundaries between learning and leisure, thereby increasing overall cognitive load (Junco, 2015).

Mental fatigue demonstrated a strong association with poorer sleep quality, consistent with established literature linking cognitive and emotional exhaustion to sleep disturbance. Previous studies have shown that heightened mental strain is associated with difficulties initiating and maintaining sleep, as well as reduced restorative sleep (Akerstedt et al., 2002; Cain and Gradisar, 2010). From a psychophysiological perspective, mental fatigue may be accompanied by heightened cognitive arousal and rumination, which can interfere with the downregulation processes necessary for healthy sleep.

Regression-based mediation analysis provided additional insight into the pattern of associations, indicating that mental fatigue statistically accounted for the relationship between social media use and sleep quality. Although social media use was not directly associated with sleep quality after accounting for mental fatigue, the indirect association suggests that cognitive experiences may represent an important correlate linking digital engagement and sleep outcomes. This statistical mediation should not be interpreted as evidence of a causal mechanism.

Beyond individual behavior, the broader digital environment should also be considered when interpreting these findings. Many social media platforms are intentionally designed to maximize user engagement through persuasive design features such as infinite scrolling, algorithmic may encourage prolonged and repetitive use, making disengagement more difficult for young users, and increasing cumulative cognitive and emotional demands over time. From an addiction research perspective, such reinforcement mechanisms share characteristics with other habit-forming digital behaviors. More broadly, the social media industry has been described as a commercial determinant of health, whereby corporate incentives may conflict with user well-being when engagement is prioritized over psychological health (Gong and Liu 2023; Zenone et al., 2022). This wider context suggests that mental fatigue and poor sleep may reflect not only individual self-regulation challenges, but also exposure to digital environments structured to sustain attention and consumption.

Sensitivity analyses testing an alternative directional model yielded a weaker indirect association when sleep quality was specified as the predictor of social media use via mental fatigue. While this does not exclude the possibility of bidirectional relationships, it supports a cautious interpretation of asymmetry in the observed associations. Prior longitudinal studies have demonstrated reciprocal relationships between media use and sleep problems (Exelmans and Van Den Bulck, 2016; Tavernier and Willoughby, 2014), underscoring the limitations of cross-sectional data for establishing temporal ordering.

Demographic patterns further contextualized the findings. Female students reported higher levels of mental fatigue and poorer sleep quality than male students, consistent with prior literature documenting gender differences in sleep disturbance and stress-related outcomes (Fatima et al., 2016; Twenge et al., 2018). Older students reported greater mental fatigue and higher academic-related social media use, potentially reflecting cumulative academic demands and prolonged exposure to digitally intensive learning environments. These findings are interpreted as contextual rather than hypothesis-testing and align with the exploratory aims of the study.

Several limitations should be acknowledged. First, the cross-sectional design limits causal inference and prevents conclusions regarding temporal ordering. Second, the use of convenience sampling may limit the generalizability of findings to the wider undergraduate student population. Students who chose to participate may have differed systematically from non-participants, including the possibility that heavier social media users or students experiencing fatigue or sleep difficulties were more motivated to respond. Accordingly, the findings should be interpreted with caution. In addition, reliance on self-reported measures introduces the possibility of recall bias, social desirability bias, and common method variance. Academic-related social media use may overlap with broader academic workload, multitasking demands, or general screen exposure; therefore, the present measures cannot fully disentangle whether observed associations reflect social media behavior specifically or overall academic digital demands. Assessing predictors and outcomes within the same survey at a single time point may also inflate observed associations. Nevertheless, the study is strengthened by its large sample size, use of validated instruments, and inclusion of mediation and sensitivity analyses that enhance interpretive rigor.

Conclusion

This study found that higher levels of social media use were associated with greater mental fatigue, and that greater mental fatigue was associated with poorer sleep quality among undergraduate students. Regression-based mediation analysis indicated that mental fatigue statistically accounted for the association between social media use and sleep quality; however, this indirect association should be interpreted cautiously as exploratory given the cross-sectional design. Overall, the findings underscore the relevance of mental fatigue as a cognitive correlate in discussions of digital engagement and sleep health and provide a foundation for future longitudinal and intervention-based research aimed at supporting student wellbeing in increasingly digital learning contexts.