Effects of Neuroticism on Differences in Symptom Structure of Life Satisfaction and Depression-Anxiety among College Students: A Network Analysis

Object: Numerous studies show that depression and anxiety have an adverse effect on life satisfaction among college students. Moreover, neuroticism affects depression, anxiety, and life satisfaction. Comparing the low-neuroticism and high-neuroticism groups, the current study used network analysis to examine the relationship between depression, anxiety, and life satisfaction among college students. Methods: A sample consisted of 1233 college students from China who completed the Big Five Inventory-2 (BFI-2), Self-rating Anxiety Scale (SAS), Self-rating Depression Scale (SDS), and Satisfaction With Life Scale (SWLS).All students were divided into two groups according to levels of neuroticism. Depression-anxiety symptom networks and flow networks were formed. Results: “Insomnia” (SAS19) and “Sleep disturbance” (SDS4) are bridge symptoms of groups with varying neuroticism. In addition, compared to the group with low levels of neuroticism, the group with high levels of neuroticism showed more depression symptoms in bridge symptoms and greater global strength. Many depression-anxiety symptoms are negatively associated with life satisfaction, and “Emptiness” (SDS18) is an important symptom in the high-neuroticism group’s flow network. Conclusion: This study contributes to our understanding of the connection between depression, anxiety, neuroticism, and life satisfaction. In addition, the current study identified the essential symptoms to target in depression and anxiety intervention and life satisfaction enhancement among college students.


Introduction
With the development of positive psychology there is burgeoning literature focused on life satisfaction. Life satisfaction, a significant indicator of mental health [1], is an individual's cognitive and global evaluation of their happiness with life [2]. Numerous studies have documented the relationship between life satisfaction and positive social, interpersonal, and intrapersonal outcomes [1,3]. Life satisfaction is also important for college students because college is a pivotal time when they face personal development challenges and must make numerous life-altering decisions [4,5]. For college students, literature shows that life satisfaction is positively associated with academic performance [6], self-esteem [7], physical activity [8], and mental health [9]. Thus, improvement of the level of life satisfaction of college students is conducive to their healthy growth.
To improve college students' life satisfaction, we cannot ignore the effect of psychopathological variables on life satisfaction [1], particularly depression and anxiety. According to a study of 200 college students, those who experienced depression, anxiety, and

Participants
Participants (N = 1238) in the current study were recruited from a university in Harbin, China. In 2021, participants completed the questionnaire through the Wenjuanxing online questionnaire platform (https://www.wjx.cn/, accessed on 9 June 2023). As Sawyer et al. [42] suggested defining the age range of 10-24 years as adolescence, in this study, five participants who did not meet the above age criteria were excluded and 1233 subjects (female = 618, Mean age =18.29, SD age = 0.78) were included in later analysis. Using the mean score of the neuroticism subscale (Mean neuroticism = 32.38, SD neuroticism = 8.26), we divided the subjects into two groups. A total of 685 subjects (female = 361, Mean age = 18.30, SD age = 0.75) with scores above the mean score on the neuroticism subscale divided into the high-neuroticism group, and 548 subjects (female = 257, Mean age =18.30, SD age = 0. 80) were divided into the low-neuroticism group. The research was examined and approved by the ethics committee of Beijing Normal University (Reference number: 202112220084).

Big Five Inventory-2 (BFI-2)
We used the sub-dimension of the neuroticism of BFI-2 in Chinese to measure the neuroticism of participants [43]. The sub-dimension of neuroticism contains 12 items, each scored on a 5-point Likert scale. Higher scores on the sub-dimension of neuroticism mean that the individuals show higher levels of neuroticism. The BFI-2 has good reliability and validity in China [43]. In the current study, the Cronbach's α score of neuroticism sub-dimension was 0.872.

Self-Rating Anxiety Scale (SAS)
The Self-rating Anxiety Scale (SAS), which was designed by Zung [44], was used in this research to measure the anxiety level of participants. The Chinese version is widely used and has solid reliability and validity [45]. This inventory has 20 items and each is scored on a 4-point Likert scale ranging from "1" (no or little time) to "4" (most or all of the time). After reversing the score of 5 items, the anxiety level of participants can be measured by the total score of all items, and a higher score means that the individual has a higher level of anxiety. This inventory shows a high level of internal consistency in the current study with a Cronbach's α score of 0.890.

Self-Rating Depression Scale (SDS)
The Self-rating Depression Scale (SDS) in Chinese was used to measure the level of depression [46,47]. This inventory has 20 items with 10 items that are reverse scored; each item is graded from "1" (no or little time) to "4" (most or all of the time). Participants with high total scores on this inventory may show higher levels of depression. The SDS has a Cronbach's α score of 0.863, showing great internal consistency in the current study.

Satisfaction with Life Scale (SWLS)
The SWLS was developed by Diener et al. [48] and was used to measure the level of life satisfaction. This scale contains five items and is graded from "1" to "7". Each item score is added up to a total score for the SWLS with a higher total score in this inventory corresponding with higher levels of life satisfaction. The reliability and validity of the SWLS in China have been established by a previous study [49]. In this study, the Cronbach's α score of SWLS was 0.878.

Data Analysis
In this study, all data were analyzed using R (Version 4.3.0) [50]. As a first step, we calculated the Means, standard deviations (SDs), skewness, and kurtosis of all SAS, SDS, and SWLS items. Additionally, we used the function descrTable of the R package compareGroups to compare the scores of items between the high-neuroticism group and low-neuroticism group [51].
Secondly, we used the R package qgraph [52] to estimate and visualize the anxietydepression symptom network of all people-the low-neuroticism group and the highneuroticism group, respectively. In this process, a Gaussian Graphical Model (GGM) was used to estimate associations between different symptoms and to construct the network between different symptoms [53]. Consequently, the Extended Bayesian Information Criteria (EBIC) and graphical least absolute shrinkage and selection operator (LASSO) models were used to regularize the GGM [54,55]. In the network, each node represents a symptom and each edge represents the regularized partial correlation between two nodes. Additionally, as to the edge, green lines mean that two nodes have positive correlations, red lines denote negative correlations, and thicker edges represent stronger relationships.
To explore the significance of each symptom within the network, previous studies usually used the centrality index strength (i.e., the sum of the absolute edge weights connected to a specific node) which purports to be the most reliable centrality index [56,57]. However, Robinaugh et al. [58] suggested that strength may not be a reliable indicator when the network contains both positive and negative edges and that Expected Influence (EI) may prove a more reliable index. Thus, we used EI in the current study to measure the importance of each symptom within the network. We also used the R package mgm [57] to calculate the predictability (i.e., R 2 ) of each node which represents the variance with which a node can be explained by neighboring nodes [59]. To identify the bridge symptoms in the networks-the overlapping symptoms of two disorders [60,61]-we calculated the bridge-expected influence (1-step) [58]. As in previous research, we set the standardized values of the bridge EI ≥ 1 as the criterion of bridge symptoms [19].
To compare the difference between the high-neuroticism group and low-neuroticism group, we used the R package NetworkComparisonTest [62] to conduct the Network Comparison Test (NCT). Additionally, we also conducted a bootstrap procedure to ensure that the networks constructed were robust via the R package bootnet [54]. Firstly, we conducted a nonparametric bootstrap, calculating the bootstrapped confidence intervals (95% CIs) to test the accuracy of edges in networks [54]. Secondly, we evaluated the differences between edge weights and centrality indices by utilizing bootstrapped difference tests. Lastly, we computed the correlation stability coefficient (CS-C) through case-dropping bootstrapping to examine the stability of the network model. The value of CS-C should be at least higher than 0.25 and preferably higher than 0.5 [54].
Finally, we used the function flow to plot a flow diagram and further explore the relationship between life satisfaction and other symptoms. The flow diagram places the node life satisfaction on the left side, then revealing direct and indirect relationships between life satisfaction and other symptoms.

Results
The Means and standard deviations (SD) of all items, and the results of t-tests between the high-neuroticism group and the low-neuroticism group are reported in Table 1. Table 1 shows that except for "Dyspnea" (SAS13) and Age, the scores of other symptoms show significant differences between the low-neuroticism group and the high-neuroticism group.

Network Structure
The anxiety-depression symptom networks of different groups are shown in Figure 1. Part A of Figure 1 shows the network of all participants. This network has 40 nodes and 780 edges with 330 non-zero (42.31%) edges. Among all edges in this network, "Confusion" (SDS11) and "Psychomotor retardation" (SDS12) show the strongest correlation, followed by the correlation between "Personal devaluation" (SDS17) and "Emptiness" (SDS18), and the correlation between "Panic" (SAS3) and "Fear" (SAS2) (refer to Table S1).
As shown in Part B of Figure 1, in the anxiety-depression network of the low-neuroticism group, there are 40 nodes and 276 non-zero edges (35.38%). The correlation between "Confusion" (SDS11) and "Psychomotor retardation" (SDS12) is the strongest one. We also found that the correlation between "Dizziness" (SAS11) and "Faintness" (SAS12) as well as the correlation between "Fear" (SAS2) and "Panic" (SAS3) displays strongly in this network (refer to Table S2).
The anxiety-depression symptom network of the high-neuroticism group is shown in Part C of Figure 1. The network of the high-neuroticism group has 40 edges and 313 non-zero edges (40.13%). Among all edges in this network, the edge between "Personal devaluation" (SDS17) and "Emptiness" (SDS18), the edge between "Fear" (SAS2) and "Panic" (SAS3), and the edge between "Dizziness" (SAS11) and "Faintness" (SAS12) are the three strongest edges (refer to Table S3). Figure 2 shows the standardized EI of all symptoms for all participants (Part A)-the low-neuroticism group (Part B), and the high-neuroticism group (Part C). The symptoms with a high standardized EI may be considered central symptoms. In the network of all participants, the standardized EI of "Faintness" (SAS12), "Paresthesias" (SAS14), and "Depressed affect" (SDS1) ranked in the top three. Among these three nodes, "Faintness" (SAS12) was also the node with the highest standardized EI in the high-neuroticism group, and "Paresthesias" (SAS14) also had a high standardized EI in the low-neuroticism group. Additionally, in the low-neuroticism group network, "Dissatisfaction" (SDS20) and "Dizziness" (SAS11) also showed high standardized EI. In the high-neuroticism group, "Hopelessness" (SDS14) and "Mental disintegration" (SAS4) revealed high standardized EI. The results of standardized strength were shown in Figure S1.
The flow networks of all participants (Part A), the low-neuroticism group (Part B), and high-neuroticism group (Part C) are shown in Figure 4. In Figure 4 we find that, among all symptoms directly related to life satisfaction, "Emptiness" (SDS18) was the symptom most highly associated with life satisfaction for all three groups and a higher level of emptiness corresponded with lower life satisfaction.. Additionally, in the low-neuroticism group, the relationship between "Easy fatiguability & weakness" (SAS8) and life satisfaction was as strong as the relationship between "Emptiness" (SDS18) and life satisfaction. In the highneuroticism group, "Weight loss" (SDS7) was the second strongest symptom associated with life satisfaction, and greater weight loss corresponded with greater life satisfaction. Behav. Sci. 2023, 13, x FOR PEER REVIEW 7 of 18  (SAS12) was also the node with the highest standardized EI in the high-neuroticism group, and "Paresthesias" (SAS14) also had a high standardized EI in the low-neuroticism group. Additionally, in the low-neuroticism group network, "Dissatisfaction" (SDS20) and "Dizziness" (SAS11) also showed high standardized EI. In the high-neuroticism group, "Hopelessness" (SDS14) and "Mental disintegration" (SAS4) revealed high standardized EI. The results of standardized strength were shown in Figure S1. The bridge symptoms of different networks are shown in Figure 3. Across three networks, "Insomnia" (SAS19) was the symptom with the highest standardized bridge EI. The standardized bridge EI "Sleep disturbance" (SDS4) was also ranked in the top three across three networks. Additionally, in the network of all participants, "Easy fatiguability & weakness" (SAS8), "Restlessness" (SAS9), "Nightmares" (SAS20), "Depressed affect" (SDS1), and "Tachycardia" (SDS9) were also bridge symptoms. In the network of lowneuroticism, other symptoms revealing high standardized bridge EI were "Mental disintegration" (SAS4), "Easy fatiguability & weakness" (SAS8), "Restlessness" (SAS9), "Nightmares" (SAS20), "Crying spells" (SDS3), and "Fatigue" (SDS10). In the network of high-neuroticism, "Mental disintegration" (SAS4), "Apprehension" (SAS5), "Depressed affect" (SDS1), "Decreased appetite" (SDS5), and "Tachycardia" (SDS9) were symptoms for which standardized bridge EI was higher than 1.  The flow networks of all participants (Part A), the low-neuroticism group (Part B), and high-neuroticism group (Part C) are shown in Figure 4. In Figure 4 we find that, among all symptoms directly related to life satisfaction, "Emptiness" (SDS18) was the symptom most highly associated with life satisfaction for all three groups and a higher level of emptiness corresponded with lower life satisfaction.. Additionally, in the lowneuroticism group, the relationship between "Easy fatiguability & weakness" (SAS8) and life satisfaction was as strong as the relationship between "Emptiness" (SDS18) and life satisfaction. In the high-neuroticism group, "Weight loss" (SDS7) was the second strongest symptom associated with life satisfaction, and greater weight loss corresponded with greater life satisfaction.

Network Comparison
The results of the network comparison are shown in Figure 5. Part A in Figure 5 shows the difference in the distribution of edge weights between the low-neuroticism group and the high-neuroticism group as non-significant (M = 0.192, p = 0.193). Part B in Figure 5 shows a significant difference between the global strength of the two groups (S =

Network Comparison
The results of the network comparison are shown in Figure 5. Part A in Figure 5 shows the difference in the distribution of edge weights between the low-neuroticism group and the high-neuroticism group as non-significant (M = 0.192, p = 0.193). Part B in Figure 5 shows a significant difference between the global strength of the two groups (S = 1.068, p = 0.015).

Network Accuracy and Stability
To ensure the networks in the current study were accurate and stable, we conducted a bootstrap procedure. As Figure S2, the 95% CIs are narrow across the groups of all participants (Part A), the low-neuroticism group (Part B), and the high-neuroticism group (Part C). As seen in Figures S3 and S4, results show that most comparisons between edge weights and centrality indices were significant. Figure S5 reports the results of the casedropping bootstrap, the CS-C values of all participants, the low-neuroticism group, and the high-neuroticism group are 0.7, 0.7, and 0.7, respectively.

Discussion
The current research conducted network analysis to investigate the relationship between depression, anxiety, and life satisfaction, factoring in level of neuroticism. Our study identified anxiety or depression symptoms associated with life satisfaction. Our study also identified the difference in global strength and bridge symptoms between the low-neuroticism group and the high-neuroticism group. Several significant findings require further discussion.

Bridge Symptoms of Depression-Anxiety Networks between the Low and High Neuroticism Groups
The bridge symptoms of the two groups reveal both similarities and differences. Specifically, the commonality between the two groups is that "Insomnia" (SAS19) and "Sleep disturbance" (SDS4) are bridge symptoms within both groups. These two symptoms are associated with sleep, an indispensable process in our daily life. Studies focused on sleep in adolescence document that sleep plays a crucial role in learning, attention, and cognitive processes [63]. Additionally, previous research also finds that sleep has a significant effect on mental health [64]. One study measured and followed up on the sleep quality and the mental health of college students, finding that poor sleep quality was associated with mental health problems (i.e., depression, anxiety, and stress) [65]. Consistent with this, research also finds that sleep loss is associated with heightened emotional reactivity to visual stimuli and changes in emotional memory processing [63,66,67], further support-

Network Accuracy and Stability
To ensure the networks in the current study were accurate and stable, we conducted a bootstrap procedure. As Figure S2, the 95% CIs are narrow across the groups of all participants (Part A), the low-neuroticism group (Part B), and the high-neuroticism group (Part C). As seen in Figures S3 and S4, results show that most comparisons between edge weights and centrality indices were significant. Figure S5 reports the results of the casedropping bootstrap, the CS-C values of all participants, the low-neuroticism group, and the high-neuroticism group are 0.7, 0.7, and 0.7, respectively.

Discussion
The current research conducted network analysis to investigate the relationship between depression, anxiety, and life satisfaction, factoring in level of neuroticism. Our study identified anxiety or depression symptoms associated with life satisfaction. Our study also identified the difference in global strength and bridge symptoms between the low-neuroticism group and the high-neuroticism group. Several significant findings require further discussion.

Bridge Symptoms of Depression-Anxiety Networks between the Low and High Neuroticism Groups
The bridge symptoms of the two groups reveal both similarities and differences. Specifically, the commonality between the two groups is that "Insomnia" (SAS19) and "Sleep disturbance" (SDS4) are bridge symptoms within both groups. These two symptoms are associated with sleep, an indispensable process in our daily life. Studies focused on sleep in adolescence document that sleep plays a crucial role in learning, attention, and cognitive processes [63]. Additionally, previous research also finds that sleep has a significant effect on mental health [64]. One study measured and followed up on the sleep quality and the mental health of college students, finding that poor sleep quality was associated with mental health problems (i.e., depression, anxiety, and stress) [65]. Consistent with this, research also finds that sleep loss is associated with heightened emotional reactivity to visual stimuli and changes in emotional memory processing [63,66,67], further supporting the notion that sleep is a significant indicator of mental health difficulties [68][69][70][71]. As for the relationship between sleep, depression, and anxiety, sleep problems are identified as common symptoms of both depression [72,73] and anxiety [74,75]. Furthermore, the relationship between depression, anxiety, and sleep may be bidirectional, which means that not only are people with anxiety and depression prone to having sleep problems, but people with sleep problems are also susceptible to developing other symptoms of anxiety and depression [76,77]. Presenting in the networks, sleep-related symptoms may serve as bridge symptoms in the anxiety-depression networks of the low-neuroticism group and the high-neuroticism group, consistent with the results of the current study.
The difference between the bridge symptoms of the two groups is that the highneuroticism group reveals more symptoms of depression in bridge symptoms. Specifically, in the high-neuroticism group, there are four bridge symptoms related to depression and three bridge symptoms related to anxiety. In the low-neuroticism group, there are three bridge symptoms related to depression and five symptoms related to anxiety. Although anxiety and depression are both related to negative affect and stressful life events-their overlapping features-several studies also suggest that they have distinctive features [25,78]. As to their relationship with affect, compared to anxiety, depression is more associated with the absence of positive affect (i.e., happiness, interest, energy, and confidence) [79,80]. Supporting the difference between anxiety and depression, Khazanov and Ruscio [81] conducted a meta-analysis of longitudinal studies to examine the relationship between positive affect, anxiety, and depression and results showed that higher level of depression and anxiety is associated with less positive affect, but the relationship between positive affect and depression was greater. Additionally, neuroticism is also a factor that is related to positive affect, supported by the findings of Hisler et al. [82], who found that neuroticism has a negative influence on the average level of positive affect. Similarly, results of other research also documented the relationship between high levels of neuroticism and a decrease in positive affect [83]. Thus, we may infer that people with high levels of neuroticism, experiencing less positive affect, will be more susceptible to developing depression symptoms, and that the symptoms of depression may further activate other symptoms of depression or symptoms of anxiety [84]. However, these results must be interpreted with caution because, in the data, discrepancies still existed in the relationship between neuroticism and positive affect [82,85]. Future studies may further examine the effect of neuroticism on positive affect as well as the mechanisms by which neuroticism influences depression and anxiety.

Difference in Global Strength between the High-Neuroticism Group and Low-Neuroticism Group
Our study found that the global strength of the high-neuroticism group was higher than that of the low-neuroticism group. In other words, tighter connectivity existed between the symptoms in the high-neuroticism group, which may contribute to the vulnerability of individuals with high neuroticism to develop or maintain anxiety and depression symptoms [86]. Research by Smith et al. [87] found that neuroticism significantly predicted depression and anxiety and suggested that cognitive biases may explain the detrimental effect of neuroticism. Other studies likewise provide supportive evidence for the negative effects of neuroticism on anxiety and depression and are consistent with results of the current study [38,39]. Additionally, a review of neuroticism shows that neuroticism has its biological (i.e., autonomic nervous system) and psychological basis (i.e., cognitive processing of emotional information) [88] helping us understand the effect of neuroticism on depression and anxiety as well as the possible mechanisms behind this effect. According to these findings, at least, more studies are needed to investigate and to verify through which mechanisms neuroticism affects anxiety and depression, and to determine methods for the delivery of timely and effective interventions for highly neurotic college students with depression or anxiety.

Flow Network between the Low and High Neuroticism Groups
As seen in the results of the flow network, in the high-neuroticism group and lowneuroticism group, most symptoms of depression and anxiety were directly or indirectly negatively correlated with life satisfaction. Significantly, our study identified the significant relationship between the symptoms of "Emptiness" (SDS18) and life satisfaction in the high-neuroticism group. "Emptiness" (SDS18) reflects whether the individual thinks "My life is meaningless" [89]. Similar to the findings of the current study, Ritchie et al. [90] investigated the relationship between neuroticism, meaninglessness, subjective well-being, and self-concept clarity and results showed that meaninglessness was positively associated with neuroticism while it was negatively related to life satisfaction. Furthermore, the implication of this finding is that among neurotic people with depression or anxiety, meaninglessness may be the key symptom to target to improve life satisfaction. Julom and de Guzmán [91] conducted a Logotherapy program, an existential therapy founded by Frankl [92], to intervene in the meaninglessness felt by paralyzed in-patients; the intervention ultimately decreased the meaninglessness felt by the experimental group. Consistent with prior research, several studies have also demonstrated the effectiveness of logotherapy in college students [93,94]. Notably, group logotherapy has been found to effectively alleviate depression and enhance life satisfaction among college students [93]. However, it is necessary to further investigate whether logotherapy can effectively address feelings of meaninglessness and enhance life satisfaction specifically in neurotic college students experiencing depression or anxiety.
These findings can be understood from the perspective of different facets of happiness, namely hedonic well-being and eudaimonic well-being [95,96]. Hedonic well-being principles encompass notions such as pleasure, enjoyment, and satisfaction, emphasizing the significance of life satisfaction and affective components [95,96]. In contrast, eudaimonic well-being focuses on optimal psychological functioning, encompassing concepts like personal growth, purpose in life, and a sense of autonomy [95]. Our study revealed a negative association between most depression and anxiety symptoms and scores on the Satisfaction with Life Scale (SWLS), which reflects hedonic well-being [97]. This result can potentially be explained by the physiological mechanisms underlying anxiety and depression, as changes in these mechanisms may influence an individual's emotional perception [98]. Furthermore, the prominent role of "Emptiness" (SDS18) observed in the high-neuroticism group may indicate the impairment of eudaimonic well-being in neurotic individuals, as feelings of meaninglessness and depression are often associated with lower levels of eudaimonic well-being [99]. However, it is important to note that our study did not directly measure eudaimonic well-being, thus necessitating future research to further explore this hypothesis.

Limitations and Conclusions
Despite the significant findings, the present study has several weaknesses. Firstly, the current results are based on cross-sectional data. Therefore, the directed and causal relationship between symptoms of life satisfaction, neuroticism, depression, and anxiety is still largely unknown. Future studies may collect longitudinal data and use statistical methods such as cross-lagged panel network analysis (CLPN) to explore the directed or predictive relationship between symptoms and these variables. Secondly, the current study used self-report questionnaire to measure the level of life satisfaction, neuroticism, depression, and anxiety, which was mainly subjective. Future research should use behavioral experiments and use other objective neurophysiological indicators (i.e., fMRI, EEG) to test the results of the current study. Thirdly, the participants of this study were all from a university in China and results may not be generalizable to all cultures and all populations. Cross-cultural and cross-group research is needed to further examine the results of our study. The present research conducted network analysis and analyzed the relationship between life satisfaction and the comorbidity of depression and anxiety in groups with different levels of neuroticism. In this study, we found that the depression-anxiety networks between the high-neuroticism group and the low-neuroticism group have similarities; "Insomnia" (SAS19) and "Sleep disturbance" (SDS4) are bridge symptoms of both groups. As for the difference between the two groups, our study found that the high-neuroticism group has higher global strength and shows more symptoms of depression in bridge symptoms. Furthermore, in addition to finding that most symptoms of depression and anxiety have direct or indirect negative effects on life satisfaction, this study also identified "Emptiness" (SDS18) as an important symptom that is negatively associated with life satisfaction in the high-neuroticism group. Despite its limitations, the present study adds to the understanding of the relationship between depression, anxiety, neuroticism, and life satisfaction, generating fresh insights from the perspective of network analysis into the interventions for depression and anxiety as well as the improvement of life satisfaction among people with different levels of neuroticism. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

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