The effect of grandparental care on social networking sites addiction in college: mediated by social anxiety and loneliness

Abstract

Objective

Grandparental care has become a common phenomenon, yet there is still limited research on the long-term psychological effects on children raised by their grandparents. This study aims to explore the impact of early grandparental care experiences on university students’ social networking sites addiction, social anxiety, and loneliness.

Methods

A random cluster sampling method was employed to select college students from a medical school in Jiangxi for a questionnaire survey. The study measured social networking sites addiction, social anxiety, and loneliness using the Bergen Social Media Addiction Scale (BSMAS), Interaction Anxiety Scale (IAS), and Loneliness Scale (ULS-6) respectively. Additionally, grandparental care was assessed through individual questions. Data analysis was conducted using mediated effects modeling.

Results

In total, 4318 participants were recruited. 45.2% of participants were male and 54.8% were female. The study found a significant positive association between grandparental care experience and social networking sites addiction, social anxiety, and loneliness. The study revealed that social anxiety mediated 20.0% of the effect between grandparental care experiences and social networking sites addiction, while loneliness mediated 16.0% of this effect. Moreover, social anxiety and loneliness together mediated 12.0% of the chained effects between grandparental care experiences and social networking sites addiction.

Conclusion

Grandparental care has no direct impact on social networking sites addiction in adulthood. Social anxiety and loneliness play a mediating role between grandparental care and social networking sites addiction. Therefore, schools and families should prioritize efforts to enhance the physical and mental well-being of individuals receiving grandparental care. This can be achieved through targeted health promotion initiatives.

Supplementary Information

The online version contains supplementary material available at 10.1186/s40359-024-02222-6.

Introduction

The rapid development of Internet technology in today’s world has significantly improved people’s quality of life, making the Internet an essential part of daily living. As of January 2023, there are over 5 billion Internet users globally, with 1.067 billion of them being Chinese [1]. Among this vast population, 32.2% are under 30 years old and highly active in online communication [2]. A growing number of young individuals have transitioned their social interactions from real life to the Internet, leading to a rise in social networking sites addiction, particularly among college students. Andreassen et al. [3] defines social networking sites addiction as an excessive focus on social media, an inability to control one’s usage, and spending excessive time and energy on these platforms, significantly impacting daily life. It was found that 60.31% exhibit a tendency towards addiction to social networking sites among Chinese college students [4]. The mechanisms of the addiction are complex and not yet clear, and various factors such as age, gender, family economic conditions, sleep quality, and psychological well-being are found to be related [5–9]. A recent report explores the neuropsychological mechanisms underlying internet addiction, identifying three main components: the ‘Must do’ pathway, the ‘Feels better’ pathway, and the ‘Stop now’ regulatory process [10]. The ‘Must do’ pathway involves both positive and negative reinforcement experiences, with the ventral striatum and dorsal striatum playing crucial roles. This pathway is primarily associated with compulsive behaviors, influenced by the dorsal striatum. The ‘Stop now’ process, controlled by the dorsolateral prefrontal cortex, regulates the first two pathways. The combined influence of these components ultimately contributes to the development of internet addiction.

Davis [11] developed the ‘cognitive-behavioral’ addiction theory model, which suggests that addiction formation is influenced by both proximal and distal factors. Proximal factors, occurring towards the end of development, directly impact the body through maladaptive cognitions. While distal factors including environmental influences, psychopathology, and other factors exist early and play indirect role in addiction occurrence. Examples of distal factors are grandparental care environment, social anxiety, and loneliness. Research has shown that adverse early childhood experiences are associated with adult mental illness [12]. Grandparental care is ofen regarded by most researchers as an adverse experiences, especially in developing countries like China, where young parents face greater economic pressures and household separations, leading to an increase in grandparental care and grandparents taking on the primary responsibility for child-rearing.

Grandparental care refers to a situation where grandparents take on the responsibility of raising their grandchildren either independently or in collaboration with the parents, with the grandparents assuming the role of primary caregiver for a duration usually over six months [13]. Research indicates that 73.29% of middle-aged and elderly individuals participate in caring for their grandchildren to different extents [14]. Additionally, the percentage of children being raised by their grandparents in intergenerational care settings has reached 34% [15]. The issue of grandparental care has garnered significant attention across various sectors. In contemporary China, grandparental care manifests in diverse forms. In urban settings, particularly in major cities, scenarios include parents and grandparents jointly raising children, as well as grandparents single-handedly caring for children. On the other hand, in remote and poor rural areas, children are largely raised by their grandparents. Research indicates that children raised under grandparental care often experience a disconnect from their parents, increased feelings of loneliness and vulnerability during their formative years, and exhibit more pronounced psychological challenges compared to those raised by their parents [16–18]. Grandparental care can negatively affect children’s social integration, leading to personality traits such as isolation, anxiety, depression, and reduced independence [19]. It is also associated with deficits in intellectual development and mental health, making individuals more susceptible to undesirable behaviors [20]. The impact of early grandparental care experiences also extends to the secondary school level. Zeng Diyang and Hong Yanbi found that these experiences negatively predicted middle school students’ mental health and self-rated health [21]. In a study conducted by Xing on the impact of grandparental care on the noncognitive abilities of middle school students, it was found that grandparental care had a notable detrimental effect on these abilities [22]. Grandparental care, in sum, has a significant adverse impact on the mental and social development of grandchildren. In China, intergenerational parenting is the predominant choice for families and spans the entire life cycle of grandchildren up to college. However, existing research on intergenerational parenting in China tends to focus on short-term effects during childhood and adolescence, with limited exploration of the long-term effects into adulthood, particularly within the context of the diverse intergenerational parenting practices influenced by complex social factors. Therefore, there is a gap in research regarding the long-term effects of intergenerational parenting on adulthood in China. In summary, Hypothesis H1 was proposed that early experiences of grandparental care positively correlate with social networking sites addiction, social anxiety, and loneliness among college students.

Social anxiety refers to the emotional experience of being nervous, anxious or panicky when interacting with people in real life [23]. The attention bias towards negative information is the key to the existence of social anxiety in an individual, and socially anxious people will deliberately pay attention to negative information in the environment, and then become extremely sensitive to interpersonal relationships [24]. Individuals with social anxiety tend to focus on negative information in their environment and become hypersensitive to relationships. To cope with these reactions, they often avoid face-to-face interactions in real life and rely on online communication through cell phones to alleviate social tension, fear, and pressure [25]. The high degree of use of social media to communicate avoiding the social anxiety that exists in the real interaction leads to their addiction on the Internet [26, 27].

Social anxiety is linked to early life experiences. Edwards compared children raised by grandparents to those raised by parents and found that the former group exhibited higher levels of anxiety (grandparental care group anxiety score: 56.11; parental care group anxiety score: 52.87; p < 0.01) [28]. A birth cohort study conducted in the UK indicated that childhood anxiety is associated with poor mental health in adulthood (anxiety: OR = 2.09, 95% CI: 1.63–2.69, P < 0.001) [29]. Children who experience social anxiety are likely to continue facing this psychological issue in adulthood [30]. Therefore, the hypothesis H2 that social anxiety mediates the relationship between grandparental care experience and social networking sites addiction was proposed.

Loneliness is a subjective psychological feeling or experience that arises when there is a discrepancy between an individual’s desire for social interaction and the actual level of interaction they receive. Research has indicated a strong correlation between loneliness and cell phone dependence [31–33]. According to Shapira’s substitutability model, individuals who experience higher levels of loneliness tend to spend more time online, which can lead to internet addiction [34]. Additionally, loneliness is linked to grandparental care. Research indicates that children receiving grandparental care exhibit lower emotional resilience compared to those under parental care, and experience higher levels of loneliness [35]. This decline in mental health is likely to persist into adulthood, affecting their overall well-being [36]. Hypothesis H3 was proposed that loneliness mediates the relationship between grandparental care experience and addiction to social network sites.

A substantial body of evidence indicates a correlation between loneliness and social anxiety [37, 38]. Lim’s study revealed a significant positive correlation between social anxiety and loneliness, with social anxiety being a predictor of loneliness [39]. Meanwhile, previous studies have established a link between grandparental care and anxiety levels in grandchildren [19, 28]. Like the study by Li et al. indicated that children who perceived greater grandparental care exhibited higher levels of anxiety, even after controlling for parental rearing practices and sociodemographic factors (β = 0.13, p < 0.05) [19]. According to the social needs theory, individuals possess an inherent need for social interaction, and when these interpersonal needs remain unmet, feelings of loneliness can emerge [40]. Children experiencing high levels of social anxiety often avoid interpersonal interactions, which exacerbates their unfulfilled social needs and subsequently triggers the feeling of loneliness. As this malaise intensifies, these children may turn to social media as a means of communication to escape their real-life challenges, potentially leading to social media addiction [41]. Consequently, hypothesis H4 was formulated: social anxiety and loneliness act as mediators between grandparental care and addiction to social networking sites.

This study aims to analyze the mechanisms through which grandparental care experiences before adulthood influence young people’s addiction to social networks, incorporating social anxiety and loneliness as mediating variables. Specifically, we will investigate the pathways through which grandparental care experiences affect social network addiction among college students, as well as the roles of social anxiety and loneliness in this relationship. The modeling assumptions are illustrated in Fig. 1. Our goal is to offer theoretical guidance for interventions addressing social network addiction within the youth with grandparental care experience.

Fig. 1.

Hypothetical diagram of the model of grandparental care and social network addiction

Method

Participants

A random cluster sampling method was utilized to select college students from a medical university in Jiangxi for a questionnaire survey. Before the survey started, we obtained the list of classes of the university, numbered them, and then randomly sampled the chosen classes using R software. The investigation was carried out during the evening when the respondents were studying in the classroom or rest in the dormitory. Prior to the survey, all respondents provided their consent and signature. Following the rule of thumb for determining sample size which suggests 5 to 10 times the number of variables, a sample size of 480 was estimated for the 48 variables in this questionnaire. In order to account for loss of follow-up, the sample size was expanded by 20%, resulting in a final required sample size of 576. Out of the 4390 questionnaires distributed, those with one-third or above of the questions missing, the same answers to all the questions, or other similar behaviors were considered invalid. After excluding 5 duplicate questionnaires, 56 questionnaires with many missing answers, and 10 questionnaires with the same answers, a total of 4,318 valid questionnaires were received, with the effective response rate of 98.4%. According to the study, the power of the test was 0.99.

Measures

Grandparental care measurement

The measure of grandparental care in this study was adapted from the China Education Tracking Survey (CEPS) questionnaire [15], with the duration of grandparental care experience extending from kindergarten to middle school, and the caregivers categorized into more types. The revised questionnaire included the following six questions: “Who primarily (secondarily) raised you in kindergarten?” “Who primarily (secondarily) raised you in elementary school?” and “Who primarily (secondarily) raised you in secondary school?” Furthermore, the response options included nine categories: grandfather, grandmother, father, mother, maternal grandfather, maternal grandmother, nanny, self, and others. Considering the common situation of co-parenting with 4 carers, i.e. mother, father, (maternal) grandmother, and (maternal) grandfather, the participants were allowed to select up to four options for their primary and secondary caregivers at each stage.

Based on this categorization, the participants who selected only one or more among grandfather, grandmother, maternal grandfather, or maternal grandmother, or those who identified grandparents as the primary caregivers with or without fathers or mothers as the secondary caregivers, were grouped under grandparental care. If the participants selected father and or mother as well as grandfather, grandmother, maternal grandfather, or maternal grandmother as their primary caregivers, they were categorized as grandparent and parent co-parenting. In the situation that only parental care was chosen or where parents were designated as the primary caregivers and grandparents as the secondary caregivers, we classified the respondents as parental care group. The participants who reported raised by other relatives, nannies, or other people were categorized as ‘others’. In the final analysis, we defined the students as “grandparental care” group if they had experienced intergenerational care in at least one of the three stages: in kindergarten, in elementary school and in secondary school. Direct parental care, joint parental care with grandparents, and other types of care were combined as “non-grandparental care”. For more details, please refer to Appendix S1.

The Bergen Social Media Addiction Scale (BSMAS)[42]

The BSMAS was adapted from the Bergen Facebook Addiction Scale (BFAS) [43] by replacing the word “Facebook” in the BFAS with social media in our study. Social media is defined as “content production and exchange platforms on the Internet based on user relationships, common social media in foreign countries include Facebook, Twitter, ins, etc., and common social media in China include WeChat, QQ, Zhihu, etc.“. There are 6 items that reflect the degree of addiction through six dimensions: salience, mood modification, tolerance, withdrawal, conflict and relapse. It is measured using a 5-point Likert scale: from 1 (almost none) to 5 (almost all). The Cronbach’s α coefficient for this scale in this study was 0.85.

The Interaction anxiety scale (IAS)

The IAS developed by Leary in 1983 [44] and translated into Chinese by Chunzi Peng in 2004 [45] was used in this study. The scale has a Cronbach’s α coefficient of 0.81 and a test-retest coefficient of 0.78, showing good reliability. The IAS is a 15-item scale with a Likert-5 point scale, with 1 being “not at all” and 5 being “completely”. The items “3”, “6”, “10”, and “15” are reverse scored, and the remaining items are positively scored. Higher scores indicate higher levels of social anxiety. In this study, the Cronbach’s α coefficient for the scale was 0.84.

The UCLA loneliness scale (ULS-6)

This study used the Chinese version of 6 item University of California, Los Angeles Loneliness Scale (the ULS-6) translated by Xiao Rong [46]. The ULS-6 is based on the ULS-8 [47], with the deletion of questions 3 and 6, resulting in a total of 6 questions. The questions were: (1) Do you often feel that you lack a partner, (2) Do you often feel that no one can be trusted, (3) Do you often feel left out, (4) Do you often feel distant from others, (5) Do you often feel unhappy because of loneliness, and (6) Do you often feel that no one cares about you even though you are surrounded by people. A 4-point Likert scale was used: 1 (never) to 4 (often). The total score of the scale ranges from 6 to 24, and higher scores mean greater loneliness level. The Cronbach’s α coefficient in this study is 0.91.

Statistical methods

Epidata 3.1 software was used for double entry of questionnaire data and consistency test was performed. Data were analyzed using R software. Quantitative data obeying normal distribution were described by mean ± standard deviation (±s), and those not obeying were described by median (P₇₅- P₂₅). Categorical data were described by constitutive ratio n (%). Continuous data obeying normal distribution were tested by t/F test, and those not obeying normal distribution were tested by Mann-Whitney / Kruskal-Wailis test. Correlation analysis was conducted for grandparental care, social networking sites addiction, social anxiety, and loneliness. The chain mediation model was constructed with social networking sites addiction as the dependent variable, social anxiety and loneliness as the mediator variables, and grandparental care as the independent variable. The mediation model was tested with the R software “BruceR” package. All data were analyzed based on R version 4.2.1.

Results

In this study, 45.2% of the participants were male and 54.8% were female. The age of the participants was 19.61 ± 2.0 years. Over 8 in ten (84.8%) of the study participants were not only child, and over 6 in ten (63.6%) were living in a family with general economic level. About one half (52.7%) of the college students lived in rural areas, and 47.3% lived in towns; 79.1% of the students used the Internet for > 3 h on weekdays, and 92.7% used the Internet on holidays > 3 h; 62.3% of students self-reported good health and above. Nearly 60% (58.0%) of the respondents had the livingcost in the range of 1000–1500 CNY. For details, see Table 1.

Table 1.

Description of demographic information (n = 4318)

Variable	Frequency	Percentage (%)
Gender
Male	1895	45.2
Female	2302	54.8
Missing	121
Only child
Yes	654	15.2
No	3659	84.8
Missing	5
Address
Urban	1895	47.3
Rural	2108	52.7
Missing	315
Family economic condition
Very poor	167	3.9
Poor	716	16.8
General	2711	63.6
Middle class families	641	15.0
Rich families	28	0.7
Missing	55
Weekday network usage time
< 1 h	85	2.0
1 ~ 3 h	811	18.9
3 ~ 5 h	1584	37.0
5 ~ 7 h	980	22.9
≥ 7 h	826	19.3
Missing	32
Network usage time on holiday days
< 1 h	59	1.5
1 ~ 3 h	190	4.8
3 ~ 5 h	783	19.6
5 ~ 7 h	1308	32.7
≥ 7 h	1656	41.4
Missing	322
Health status
Very good	446	10.3
Good	2246	52.0
General	1439	33.3
Poor	161	3.7
Very poor	24	0.6
Missing	2
Living costs
< 1000 CNY	483	11.2
1000 ~ 1500 CNY	2498	58.0
1500 ~ 2000 CNY	1067	25.2
≥ 2000 CNY	243	5.6
Missing	8

Table 2 was the description of the respondent’s primary care in kindergarten, primary and secondary school. At the preschool level, 50.9% of the children were raised by their parents, 34.4% by grandparents, and 12.7% jointly by parents and grandparents. At the elementary school level, 59.5% were raised by parents, 25.8% by grandparents in alternate care, and 12.1% in shared care. At the secondary school level, 70.5% of respondents were raised directly by their parents, 15.5% by their grandparents, and 9.7% by grandparents together with parents, with 4.3% raised by other caregivers.

Table 2.

Statistical description of the main rearing patterns of college students when they were in kindergartens, elementary schools or secondary schools (n = 4318)

Variable	Frequency	Percentage (%)
Kindergarten
Parent	2196	50.9
Grandparents with parent	547	12.7
Grandparent	1480	34.4
Others	78	1.8
Missing	17
Elementary schools
Parent	2560	59.5
Grandparents with parent	523	12.1
Grandparent	1112	25.8
Others	110	2.6
Missing	13
Secondary school
Parent	3036	70.5
Grandparents with parent	419	9.7
Grandparent	669	15.5
Others	182	4.2
Missing	12

Before conducting the correlation matrix and chain mediation analyses, we transformed the polytomous variable of primary caregivers, i.e. grandparent-raised, grandparent-parent co-raised, parent-raised, and others into a dichotomous variable: grandparental care versus non-grandparental care. Individuals who experienced grandparental care at any of the three stages before going to college were defined as engaging in intergenerational parenting. Ultimately, 39.8% (n = 1722) were included in the “grandparental care” group. In contrast, parental and paternal co-parenting, parental-only parenting, and parenting by others were grouped as “non-grandparental care” (n = 2587). After excluding cases with missing core variables (n = 107), a total of 4211 participants were included in the correlational matrix analysis and chain mediation analysis.

Table 3 presents the correlation matrix between grandparental care experience and social networking sites addiction, social anxiety, and loneliness. The results showed that grandparental care experience was positively correlated (p < 0.05) with social networking sites addiction, social anxiety, and loneliness. Social anxiety was positively correlated with loneliness (p < 0.05).

Table 3.

Correlation coefficients, means and standard deviations of grandparental care experiences with social networking sites addiction, social anxiety, and loneliness (n = 4211)

	1	2	3	4
1. Grandparental care	1
2. Social anxiety	0.078***	1
3. Loneliness	0.057***	0.310***	1
4.Social networking sites addiction	0.048**	0.210***	0.291***	1
Mean	-	43.60	11.09	13.47
SD	-	9.05	3.94	4.61

*P < 0.05; ** P < 0.01; ***P < 0.001

The chain-mediated regression model of grandparental care experience, social anxiety, loneliness and social networking sites addiction showed (Table 4; Fig. 2) that grandparental care experience positively predicted social anxiety (β = 1.436, p < 0.001) and loneliness (β = 0.257, p < 0.05); social anxiety positively predicted loneliness (β = 0.134, p < 0.001) and social networking sites addiction (β = 0.066, p < 0.001); loneliness positively predicted social networking sites addiction (β = 0.291, p < 0.001).

Table 4.

Chain-mediated regression analysis of grandparental care experiences, social anxiety, loneliness, and social networking sites addiction (n = 4211)

Predictor variable	Model 1	Model 2	Model 3
	Social anxiety	Loneliness	Social networking sites addiction
	β t	β t	β t
Grandparental care	1.436***	0.257*	0.223
Social anxiety		0.134***	0.066***
Loneliness			0.291***
R 2	0.006	0.098	0.101
F	26.43	228.2	157.7

*P < 0.05; ** P < 0.01; ***P < 0.001

Fig. 2.

Path diagram of the mediation effect tests. *P < 0.05; ** P < 0.01; ***P < 0.001

Table 5 presents the tests of mediating effects of grandparental care experiences, social anxiety, loneliness and social networking sites addiction. The direct effect of grandparental care experience on social networking sites addiction was not statistically significant (effect value = 0.223, Bootstrap 95% CI: -0.045-0.490). The total mediating effect was significant (effect value = 0.226, Bootstrap 95% CI: 0.135–0.322) accounting for 47.4% of the total effect. Social anxiety mediates 20.0% of the effect between grandparental care experience and social networking sites addiction (effect value = 0.095, Bootstrap 95% CI: 0.053–0.141). Loneliness mediates 16.0% of the effect between grandparental care experience and social networking sites addiction (effect value = 0.075, Bootstrap 95% CI: 0.008–0.144). Social anxiety and loneliness mediate 12.0% of the chain of effects between grandparental care experience and social networking addiction (effect value = 0.056, Bootstrap 95% CI: 0.033–0.082).

Table 5.

Tests of the mediating effects of grandparental care experiences, social anxiety, loneliness, and social networking sites addiction (n = 4211)

	Effect value	Standard error	P	Bootstrap 95%CI
				Lower	Upper
Direct effect	0.223	0.136	0.102	-0.045	0.490
Indirect effect	0.226	0.022	< 0.001	0.135	0.322
Grandparental careγ Social anxietyγ Social networking sites addiction	0.095	0.022	< 0.001	0.053	0.141
Grandparental careγ Lonelinessγ Social networking sites addiction	0.075	0.035	0.032	0.008	0.144
Grandparental careγSocial anxietyγ Lonelinessγ Social networking sites addiction	0.056	0.013	< 0.001	0.033	0.082
Total effect	0.476	0.145	< 0.001	0.189	0.765

Discussion

This study explores the lasting impact of intergenerational parenting experiences on the mental health of college students through the lens of grandparental care. It aims to investigate the connection between grandparental care experiences, social anxiety, loneliness, and addiction to social networking sites, in order to shed light on the underlying mechanisms of social networking sites addiction among young people.

Grandparental care experience positively predicted social anxiety and loneliness, as shown in our study. Children who are raised by their grandparents may exhibit excessively higher sensitivity in their lives. A community-based study conducted in the United States found that 21.9% of children raised by grandparents required psychiatric treatment for psychological problems [48], while 16.5% of children raised by their parents [49]. Compared to children raised by their parents, children raised by grandparents face significantly higher risks of psychological issues and adverse behaviors [50]. In the United States and many European countries, the majority of grandparents do not volunteer to raise their grandchildren. A great number of children are placed in the care of their grandparents due to various reasons, such as parental abuse, divorce, drug/alcohol abuse, incarceration, and death [51–53]. The situation in China is different, where grandparental care mainly exist when parents have to go out to work and could not care the children themselves. The parents in the middle generation often bear the responsibility of providing for the entire family, compelling them to seek employment in big cities and entrust their children’s care to their own parents. So Chinese grandparents raise their grandchildren in hometown while the parents work in big cities. But wherever it is, most grandparents raise their grandchildren not on their own accord really, but rather are forced by the situation with socially disadvantaged parents. Despite cultural differences, the impact of grandparenting on grandchildren is a universal, with grandparents raising their grandchildren by sacrifice their own health [54, 55], resulting in potential adverse effects on the grandchildren also [56]. At the same time, Chinese grandparents demand more quietness from their grandchildren, which results in children who may lack energy and curiosity and be shy and socially inept [57], which contribute to their social anxiety and loneliness lasting into adulthood.

According to our study, the direct effect of grandparental care experience with social network sites addiction is not significant. Griffiths identifies six main characteristics of addictive behavior, which are salience, mood change, tolerance, withdrawal, conflict, and relapse [58]. Social media addiction, as one of the many addictive behaviors, is not only psychologically traumatic but also physically damaging [59, 60]. This study supports the cognitive model of addiction, indicating that grandparental care experiences play a role in early upbringing but may do not directly cause social media addiction. Instead, they are linked to social media addiction through intricate pathways.

Social anxiety plays a significant mediating role in the relationship between experiences of grandparental care and addiction to social networking sites in the study, indicating a cognitive pathway from grandparental care to social networking sites addiction. Brand’s Interaction of Person-Affect-Cognition-Execution (I-PACE) [61, 62] proposes that an individual’s core factors, such as psychopathological and physiological influences, as well as affect, cognition, and executive functioning, all impact the individual’s behavior in using the Internet. Individuals with social anxiety tend to avoid in-person social interactions but feel more comfortable communicating through social software. This reliance on social networks can lead to decreased face-to-face socialization and potentially result in addiction to social networking sites. Research has shown that having a grandparental care experience significantly increases the occurrence of adverse psychological events such as depression and social anxiety [63]. Students raised by their grandparents may underestimate their self-image in social situations, feel fear about interacting with classmates, and worry negative evaluation or rejection, ultimately leading to significant social anxiety [24, 64]. In an effort to ease social tension, fear, and stress, individuals may seek solace in the virtual world, using their cell phones for online interpersonal interactions [25]. Various studies in evidence-based medicine indicate that behavioral inhibition is a significant predictor of Social Anxiety Disorder (SAD) development [65, 66]. Nearly 45% of children with behavioral inhibition eventually develop SAD [66]. This could be one of the reasons why social anxiety is commonly observed in intergenerational settings. It may represent a significant mechanism in behavioral psychology, illustrating how social anxiety influences the relationship between experiences of grandparental care and addiction to social networking sites.

On the other hand, from a neurobiological point of view, internet addiction has been found to have slow self-control, increased sensitivity to rewards, and compulsive usage and similar brain structure changes to drug abuse, showing parallels with the behavior patterns and neurobiological mechanisms to traditional addictions such as drug addiction [10]. Recent relevant studies have found that social anxiety is associated with over-activation of the amygdala and low blood oxytocin concentrations [67–70]. Prefrontal-amygdala circuits are involved in emotion regulation and behavioral control. Previous studies have consistently shown that individuals with addiction tend to exhibit more pronounced alterations in the amygdala and prefrontal lobes when compared to non-addicted control groups [71, 72], which suggests the linkage between addiction development and prefrontal-amygdala circuits. And Oxytocin is gradually coming into the limelight as a modulator, and researchers have found that oxytocin can reduce the amygdala and prefrontal lobe of individuals through the hypothalamic-pituitary-adrenal (HPA) axis [73]. In an animal model, Kovács found that oxytocin slowed addiction tolerance, and that this effect was correlated with the dose of oxytocin [74]. Low oxytocin concentrations in socially anxious individuals prevent effective regulation of the amygdala, which further amplifies the negative emotions of socially anxious individuals, and in an attempt to alleviate these negative emotions, they become addicted to the internet for gratification, resulting in internet addiction. The existing research provide preliminary support for the similarity between social networking sites and substance use disorders although the results are not always consistent. Identifying more specific vulnerability factors like childhood experience including grandparental care, and defining how the factors determine reward and relief experiences related to the use of social media applications may help to distinguish between rich and problematic use patterns, and finally explain the neurobiological mechanisms of social networking sites addiction.

Loneliness mediates the relationship between grandparental care experiences and social networking sites addiction. Self-determination theory posits that individuals possess an inherent need for intimacy and connection with important others [75]. The theory further suggests that personal or situational factors that support the satisfaction of this fundamental need contribute to overall well-being, whereas factors that hinder or obstruct its fulfillment can have negative effects on health. On the one hand, in today’s life, it is becoming common that parents are unable to raise their children in person due to various reasons, especially with the increasing divorce rate and the popularity of migrant work. This separation can prevent the fulfillment of the parent-child relationship, impacting the emotional well-being of the children [36], and care from grandparents cannot compensate, even leads to more psychological and behavioral negative issues. On the other hand, technological advancements and industrial development have facilitated the widespread and low-cost ownership of electronic devices among college students even since their childhood. Individuals experiencing loneliness, particularly teenagers and young adults, may turn to social networks which they are more familiar with than the real world as a means of coping with their feelings of depression and emotional distress. This behavior can be attributed to their underdeveloped psychological coping mechanisms, limited social support systems, and the accessibility of social networking platforms. The reliance on social media as a coping mechanism can ultimately result in social network addiction [76]. A recent meta-analysis on the neurobiological mechanisms of loneliness revealed associations with the ventral striatum, amygdala, and ventral striatum/vomeronasal nucleus in the brain [77]. This aligns with the brain activation patterns observed in internet addicts, which could potentially explain internet addiction in individuals experiencing loneliness [10].

Social anxiety and loneliness act as chain mediators between grandparental care and social networking sites addiction. Multiple studies have demonstrated that the experience of grandparental care can lead to adverse emotional outcomes in grandchildren, including depression, social anxiety, and loneliness. Furthermore, this type of care can also influence their cognitive development and physical well-being [19–21]. According to the I-PACE model [61, 62], when grandchildren are raised in an intergenerational environment, their fear of unfamiliar environments and attachment to their parents hinder their communication with the outside world in their daily lives. This fear of negative reactions from others leads to social anxiety and feelings of loneliness in grandchildren. Social anxiety and loneliness often occur together, and they share similar activation mechanisms [67–70, 77]. Individuals experiencing social anxiety and loneliness may seek out online social networks as a form of psychological compensation. Through virtual interactions, they can fulfill their social needs and even receive positive feedback that is difficult to obtain in their real lives. As this pattern continues, a strong connection forms between the individual’s positive (pleasure, reward) and/or negative (stress relief and negative emotions) reinforcement experiences and the gratification obtained from social networks, which could potentially lead to addiction. To gain web traffic and finally economic benefits, the algorithms constructed by the applications can further enhance the stickiness of young people on social networks, promote the compulsive use even if they know that excessive online use is harmful, they cannot easily get rid of it. While this study delves deeper into the mechanism of social network addiction among college students, it also broadens the understanding of this issue.

Life course theory posits that personal development is inextricably linked to social development [78]. Since the reform and opening up of society, dramatic social changes have disrupted traditional intergenerational relationships, gradually shifting the focus of family dynamics from grandparents to grandchildren. The prevalence of grandparental care has become a common phenomenon, resulting in both ideological and habitual clashes between grandparents and grandchildren, which can impact the physical and mental health of both parties [12]. The effects of these interactions extend into adulthood. Additionally, Generation Z has experienced significant events, such as the wave of educational expansion and the widespread availability of the Internet during their formative years. Compared to previous generations, this cohort exhibits a stronger intergenerational identity with their peers and notable generational differences from earlier groups. They are eager to integrate into the digital realm, adopting a networked lifestyle, while simultaneously being influenced by both the positive and negative consequences of the rise of Internet society and rapid technological advancements [79–81]. Social network addiction is a prevalent negative consequence of their digital integration. This study examines social network addiction among college students who have experienced grandparental care, utilizing a clear definition and precise measurement of intergenerational parenting within Chinese society. It explores in depth the pathways through which childhood grandparental care influences social network addiction in college students, analyzing the roles of social anxiety and loneliness in this context. The findings provide a scientific basis for assisting this demographic in reducing cell phone dependence and enhancing interpersonal skills and overall physical and mental well-being.

However, there are a few limitations to consider. First, the study did not account for confounding variables like demographic factors and time spent on mobile phones in the chain mediation analysis, which could affect the outcomes. Second, this retrospective study relied on the respondents’ recollection for their upbring experiences. Although childhood was divided into three rough stages and people were supposed to have a better understanding of their upbring experiences, but the recall bias was inevitable especially when measuring the experience in kindergarten. Third, this study examines the impact of grandparental caregiving during childhood on social media addiction in adulthood. However, the effects of grandparental caregiving at different developmental stages may have varying impacts on health. Future research could conduct more in-depth stratified analyses to explore how grandparental caregiving at different stages influences social media addiction in adulthood. Finally, while this study focused on the relationship between social anxiety and loneliness in the context of grandparental care and social network addiction, other psychological factors—such as depression and self-esteem—may also play a significant role. These factors warrant further exploration in subsequent studies.

Conclusion

This study examines the mechanism of social network addiction among college students, finding a positive relationship between intergenerational parenting experiences and social network addiction. Social anxiety and loneliness are identified as mediating factors in this relationship. The study suggests that schools, mental health departments, and families should prioritize support for college students with intergenerational upbringing to facilitate their integration into society and enhance their overall well-being.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

BSMAS

Bergen Social Media Addiction Scale

IAS

Interaction Anxiety Scale

ULS-6

University of California Los Angeles Loneliness Scale

I-PACE

Interaction of Person-Affect-Cognition-Execution

SAD

Social Anxiety Disorder

Author contributions

S. Z：Conceptualization, Methodology, Data curation, Writing- Original draft preparation. Y. L：Conceptualization, Methodology, Data curation.C. Z: Conceptualization, Methodology, Data curation. X. R: Conceptualization, Methodology, Data curation. C. L: Resources, Data curation. C. W: Writing- Original draft, Conceptualization, Methodology, Writing - Review & Editing, Supervision.

Funding

This study was supported by the Special Funds for Postgraduate Innovation in Jiangxi Province (Grant No. YC2023-S944).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

The Research Ethics Committee of Gannan Medical University approved the procedure. Written informed consent was obtained from each participant before starting the interview, and all principles of confidentiality, anonymity, and voluntary involvement, following the Declaration of Helsinki, were maintained.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.