Leisure satisfaction refers to positive views or feelings obtained in leisure activities, which reflects the overall satisfaction degree of leisure experience (Beard and Ragheb, 1980). According to the compensation theory (Wilson, 1980), individuals can compensate for negative aspects in other areas of life through well-experienced leisure activities. Some hold that leisure satisfaction may even be more important than other sources of life satisfaction (Tokay Argan and Mersin, 2021). From an ecological perspective, both positive leisure activities and consequent flow experiences (wonderful experiences of being fully involved in the present moment) not only can help prevent problematic behaviors such as aggression (Park et al., 2017) and substance use (Santini et al., 2020) among young people but also promote their development and adjustment (Mahoney et al., 2009; Freire and Teixeira, 2018). Specifically, excellent leisure experience is associated with a clear sense of self-identity (Mansfield et al., 2020), the formation of social relationships (Kim et al., 2016a), and relief from stress (Chun et al., 2021). All these factors are conducive to the development of psychosexual adjustment (Bigras et al., 2015; Kim and Jang, 2020). Considering leisure is a kind of social construction, Bandura’s social learning theory (Bandura and Walters, 1977) can provide us insights into the understanding of how people promote their psychosexual adjustment through leisure activities. Positive participation and meaningful experience in leisure activities allow the youth to learn how to socialize properly (Granzin and Haggard, 2000) and gain interpersonal support and confidence (Kim et al., 2016b), which have an important impact on their psychosexual adjustment (Parchomiuk, 2019; Barrada and Castro, 2020).

In contrast, a lack of pleasure in leisure seems to be associated with an unhealthy status. For example, leisure boredom could help predict substance use and Internet use in adolescents (Weybright et al., 2015; Wang, 2019). In the field of sexual psychology, studies have shown a correlation between leisure boredom and adolescent risky sexual behaviors (Miller et al., 2014). Based on these facts, we proposed the following first hypothesis:

Personality is a mode of thinking, feeling and behaving that help to differentiate people from each other. It affects a person’s response to all kinds of situations (Caspi, 1998). Personality is relatively stable throughout life even during adolescence, fluctuating only within a certain range (Brown and Prinstein, 2011). The five-factor model of personality (Van Leeuwen et al., 2004) is one of the most widely used trait models with high reliability and validity at different developmental stages of the life cycle (De Clercq et al., 2009). It measures personality from five dimensions: extraversion, openness, agreeableness, conscientiousness, and neuroticism (also known as emotional stability). However, individuals are usually affected by different trait dimensions simultaneously (Atkins and Hart, 2008). According to the five-factor model, some studies have confirmed that the personality structure can be divided into four classes: the moderate, protected, vulnerable, and undercontrolled groups (De Clercq et al., 2012) or into three classes: the resilient, undercontrolled, and overcontrolled groups (Zhang et al., 2015). In a longitudinal study of Finnish adults, the researchers identified five personality types: the resilient, overcontrolled, undercontrolled, reserved, and ordinary groups (Kinnunen et al., 2012). Overall, resilient/protected individuals typically have low neuroticism and high levels of the other four dimensions. Overcontrolled/vulnerable individuals are exactly the opposite, with high neuroticism and low levels of the other dimensions. Undercontrolled individuals are characterized by low conscientiousness, slightly high neuroticism, and high openness and extraversion (Zhang et al., 2015). Reserved people usually have low openness. Moreover, all personality traits of moderate/ordinary individuals are at the middle levels (Kinnunen et al., 2012).

As a typical individual difference variable, the role of personality on sexual psychology and behavior has also been widely studied. Eysenck and Wilson (1976) suggested that extroverts are thrill seekers, while neurotic people do not enjoy sexual behavior as much because they are extremely tense. Moreover, extraversion is positively correlated with sexting, while agreeableness and emotional stability are negatively correlated (Delevi and Weisskirch, 2013). Openness is negatively correlated with sexual prejudice, for people who are higher in openness are more likely to accept social changes (Miller et al., 2012). Furthermore, personality types and personality disorders caused by the mixture of multiple personality traits are also related to psychosexual adjustment. Children of the undercontrolled type are more likely to have sex before the age of 16 (Atkins and Hart, 2008). Women with borderline personality disorder have been found to have chaotic relationships and more sexual partners in a short period of time (Thompson et al., 2019). However, previous studies on the influence of personality on psychosexual health have mostly focused on unhealthy sexuality and morbid personality and have not explored the normal groups. Therefore, we proposed the second hypothesis:

There is an interaction between personality traits and leisure satisfaction. Scholars generally hold that leisure has a profound and unique reflection on our personality (Munusturlar and Argan, 2016). Furthermore, the relationship between personality and leisure satisfaction has physiological basis. Eysenck and Wilson (1976) suggested that extroverts have high cortical inhibition and low baseline arousal, which causes them to seek external stimulation. Introverts, on the other hand, have a low arousal threshold and need less stimulation to maintain optimal levels. People with lower emotional stability are less playful and get less enjoyment from social activities (Lepp et al., 2015). Additionally, people of neuroticism may feel unfulfilled in leisure and be less cheerful than other personality types (Codish and Ravid, 2015). People of the traditional type of personality can obtain higher educational satisfaction in leisure than the investigative, enterprising, realistic, and artistic types. They score higher on social satisfaction than the enterprising type as well (Park, 2009).

We have found the relationship between leisure satisfaction and psychosexual adjustment as well as the relationship between personality and psychosexual adjustment. However, as far as we know, no one has investigated the relationship among these three variables. Under the theoretical framework of personal-environment fit models (Iso-Ahola and Weissinger, 1990), studies conducted have proved that the overcontrolled adolescent maladaptive behavior (such as criminal behavior) is a strategy to buffer leisure boredom (Spaeth et al., 2015). Moreover, it has been shown that there is an interaction between leisure satisfaction and personality on subjective well-being (Liu, 2014). Based on the research mentioned above, we proposed the third hypothesis:

In general, this study aims to explore the relationship between leisure satisfaction and psychosexual adjustment based on the compensation theory and the social learning theory. We also explored the differences in the psychosexual adjustment across different personality types. Furthermore, we examined the moderating effect of personality types between leisure satisfaction and psychosexual adjustment. We propose the following three hypotheses: H1: Leisure satisfaction is positively correlated with psychosexual adjustment. H2: Psychosexual adjustment differs across different personality types. H3: Leisure satisfaction and personality have an interactive effect on psychosexual adjustment. The theoretical model is shown in Figure 1.

Participants in the study were undergraduate students from a university in China. The research team recruited participants through posters and online notifications. Our team informed them of the principles of confidentiality and voluntary. Participants would take part in the survey in designated classrooms at specified time, and would not get rewarded. In addition, the survey would be done anonymously and the results would be kept private. Eventually, a total of 1,161 students were recruited to answer the questionnaires. Data collecting was conducted from November 2021 to January 2022. One research group member and one school teacher were designated and trained as the experimenters. To start with, the research group member read the instructions. The participants were asked to complete the questionnaires in 25 min.

Invalid questionnaires (answering regularly or missing more than five questions) were removed after all the questionnaires were collected. Following the removal, 1,075 valid questionnaires were obtained. The rate of effective recovery was 92.6%. The average age of the participants was 18.7 (SD = 1.07). There were 511 males (47.5%) and 564 females (52.5%) participants. A total of 652 (60.7%) participants were rural residents, and 423 (39.3%) were urban residents.

Mplus 8.0 was used for the confirmatory factor analysis and classification of personality types using latent profile analysis. SPSS 23.0 was used for the descriptive statistics analysis and Pearson correlation analysis. The difference in psychosexual adjustment across different personality types was analyzed based on the hierarchical linear models in SPSS. The moderating effects of personality on leisure satisfaction (independent variable) and psychosexual adjustment (dependent variable) were tested based on the hierarchical multiple regression models in SPSS. And p < 0.05 was set as of statistical significance.

The Harman single-factor test was used to test the method bias. The variance explained by the first factor (“I have passion in leisure activities” from leisure satisfaction questionnaire) was 20.14%, below the 40% threshold. The result of the principal component factor analysis without rotation showed that 19 factors had eigenvalues greater than 1. Therefore, common method bias did not affect the results of this study.

To test hypothesis 1, we used Pearson correlations. The means (M) and standard deviations (SD) of personality, leisure satisfaction and psychosexual adjustment are shown in Table 1. Table 1 also shows the correlation matrix of these variables. Among all five personality traits, extraversion, openness, agreeableness, and conscientiousness were positively correlated with leisure satisfaction and psychosexual adjustment (p < 0.01). Neuroticism was negatively correlated with leisure satisfaction and psychosexual adjustment (p < 0.01). Leisure satisfaction was positively correlated with psychosexual adjustment (p < 0.01).

To explore the potential patterns of personality traits, this study used latent profile analysis (LPA) with five personality traits as observation variables and one profile as the starting point. The results are shown in Table 2.

Table 2 contains the AIC, BIC, ABIC, entropy, and LMR statistics for each of the tested models. An optimal model fit is defined by lower AIC, BIC, and ABIC values (Finch, 2015). For all models, these three values decreased as the number of profiles increased from 1 to 5, providing support for the 5-profile solution. If the LMR value of model K reached the significance level (p < 0.05), it indicated that model K had a higher variance explanation rate than model K-1. According to the model fitting results in Table 2, the LMR values of the 2-profile model, 3-profile model and 4-profile model all reached the level of significance, showing that these three models were superior to the 5-profile solution. Furthermore, there was little difference in the index between the 4-profile model and the 5-profile model. Therefore, we preferred the 4-profile solution. Entropy represents classification accuracy, which generally has a high standard of 0.80 (Clark, 2010). Ultimately, the 4-profile model was chosen as the optimal latent profile analysis model.

Table 3 shows the Z scores of five personality traits for each personality profile. The overcontrolled group has higher neuroticism but lower scores in the other four dimensions, and Profile 1 matched these features. Profile 4 conformed to the definition of the resilient group, showing the lowest level of neuroticism and higher levels of the other four dimensions. All personality dimensions of Profile 2 and Profile 3 were almost at the average level. The only difference was that the scores of Profile 2 were slightly lower than those of Profile 3. Therefore, Profile 2 and Profile 3 were identified as typical personality types, with Profile 2 named the low-moderate group and Profile 3 named the high-moderate group.

For the test of hypothesis 2, bar charts were used to show the differences in the Z scores of college students with different personality types in psychosexual adjustment (see Figure 2). Furthermore, four types of personality were included in the hierarchical linear models. Profiles 1, 2, and 3 were compared as reference groups to test their differences in self-adjustment, social adjustment, sexual control, and psychosexual adjustment. The results are shown in Table 4. There was no significant difference between Profile 1 and Profile 3 in the sexual control dimension (p = 0.326). Except for that dimension, all personality types showed differences in psychosexual adjustment.

For the test of hypothesis 3, the hierarchical multiple regression models were used to investigate the moderating effects of personality on leisure satisfaction and psychosexual adjustment. Among the four personality types, we selected the overcontrolled group as the reference group. Three dummy codes were used for the high-moderate group, the low-moderate group, and the resilient group. To reduce the collinearity between variables, we standardized the independent variable (leisure satisfaction) first. After that, the interaction term was calculated between the standardized independent variable and the moderating variable (personality types). Leisure satisfaction was put into the first layer. Then, dummy codes of the moderating variable were added to the second layer. Finally, the interaction terms were added into the regression equation as the third level. The results show that the interaction of the leisure satisfaction and the dummy variable for the high-moderate group, the low-moderate group, and the resilient group was all statistically significant, with p < 0.001, p < 0.001, and p = 0.011, respectively, as shown in Table 5.

To further explore the moderating effects of personality, we divided participants into two groups: a low leisure satisfaction group (scored one standard deviation below the average) and a high leisure satisfaction group (scored one standard deviation above the average). Figure 3 illustrates the moderating effects of personality. For the accurate demonstration of the moderating effects of each personality type, a linear regression model test was conducted for each group and the path coefficients were 0.24 for the overcontrolled group (t = 1.74, p = 0.089), 0.22 for the high-moderate group (t = 4.28, p < 0.001), 0.30 for the low-moderate group (t = 7.75, p < 0.001), and 0.47 for the resilient group (t = 3.49, p = 0.001).

As expected, the results confirmed a link between leisure satisfaction and psychosexual adjustment. Consistent with previous research (Zerengok et al., 2018), when college students actively participated in leisure activities, they showed progress in social adjustment and reported psychological and physical benefits. This result can be interpreted based on social learning theory. People learn communication skills and norms, and learn to understand themselves and others appropriately through entertainment on informal occasions. Individuals could derive pleasure and social benefits from interpersonal interactions in leisure activities (Chao, 2013). In other words, positive leisure can enable people to achieve psychosexual socialization (Shin and You, 2013). Apart from this, leisure is regarded as a potential resource for people in dealing with some forms of stigma such as sexual orientation (Mock et al., 2013). Leisure could provide sexual minorities with a sense of belonging which brings benefits to their psychosexual selfhood, but this requires favorable social and relaxing conditions (Outley and McKenzie, 2007; Jones and McCarthy, 2010). As previous research has found, engaging in meaningful social leisure activities can positively impact current and future mental health (Timonen et al., 2021).

Based on the compensation theory, positive experience in leisure seems to be a powerful source of self-control efficacy (Stebbins, 2006). Due to the lack of leisure resources and opportunities, bored young people may turn to risky activities (Sharp et al., 2011). Leisure interventions that can change the leisure cognition of teenagers and help improve their leisure satisfaction, such as HealthWise (Motamedi et al., 2020), have been shown having the potential to reduce the likelihood of early sexual activity in boys.

Among all the relationships between leisure satisfaction and psychosexual adjustment, the relationship between contact satisfaction and self-adjustment was the strongest. One explanation is that self-esteem can be gained from social support (Erango and Ayka, 2015). When people get along well with others and receive positive feedback about themselves from others, they can improve their self-acceptance concerning physiology and psychology. This finding was also expressed by Cooley’s conception of the “looking-glass self” (Cooley, 1902), in which the actor constantly builds his self-image based on the reactions of others to his performances. Similarly, social interaction has explained more than half of the association between green space use and self-satisfaction (Dadvand et al., 2019).

We further explored the difference in psychosexual adjustment across different personality types. In the first place, we used the same five-factor model to identify different personality types as many human-centered studies do. Our results were consistent with the findings of other researchers but somewhat different. We found that the vast majority of Chinese college students in our research could be classified as the moderate group (there were differences between the high- and low-moderate groups), a small number of students (approximately 4.7%) could be classified as the resilient group and approximately 4.3% could be classified as the overcontrolled group. Unlike a previous study (Alessandri et al., 2014), we did not actually identify an undercontrolled group. This result may, in part, be explained by the stricter control of Chinese parents over teenagers and the corresponding more reserved qualities teenagers have (Ho, 1994; Wang et al., 2018).

The hierarchical linear models seem to indicate that the overcontrolled group was at a disadvantage in all dimensions of psychosexual adjustment compared with other groups. Only in the sex control dimension was there no significant difference between the overcontrolled group and the low-moderate group. The overcontrolled group scored relatively low in the dimension of extraversion and relatively high in neuroticism, showing inflexibility in cognition and behavior and high attention to detail processing (Gilbert et al., 2020). Besides, the overcontrolled group is usually associated with a higher likelihood of internalizing problems (Asendorpf et al., 2001). This situation is mainly due to the fact that overcontrolled individuals may have more difficulty participating in social activities or engaging fully in intimate relationships (Denissen et al., 2008). Nevertheless, this situation also entails that such individuals have a higher degree of control than undercontrolled people (Block and Block, 1980). This fact may explain why the overcontrolled group did not differ from the low-moderate group in sexual control.

The resilient group, the opposite of the overcontrolled group, showed an advantage across all dimensions. Adolescence is a time when young people can sort out their sexual identities and sexual orientation (Katz-Wise, 2015). Meanwhile, in this phase the youth might worry about their sexual attractiveness. Individuals in the resilient group excel at this stage because they are able to adapt to a variety of environmental needs (Asendorpf, 2006). Specifically, higher levels of resilience are associated with higher levels of self-acceptance and self-control (Baeva et al., 2016), both of which contribute to the development of psychosexual adjustment (Woodford et al., 2014; Magnusson et al., 2019).

Comparing Profiles 2 and 3, both of which were considered moderate, we found that the former profile performed better in psychosexual adjustment. But overall, they scored in the middle in psychosexual adjustment. Consistent with previous findings (Rettew et al., 2008), individuals with moderate characteristics belong to a stable class. They do not pursue stimulating experience too excessively and have a certain degree of self-protection. In addition, more agreeable and more open individuals may have greater psychological and physiological adjustment (Boyce and Wood, 2011; Ó Súilleabháin et al., 2018).

Considering people with different personality types process information in different ways, we analyzed the interaction between leisure satisfaction and personality to understand how individuals apply leisure experience. The moderating effects of all personality types were positive, and the moderating effect of the resilient group was the most prominent. These results confirmed the applicability of personal-environment fit models to adolescent psychosexual adjustment. As a risk-protective factor, resilience can promote positive development during times of adversity (Luthar et al., 2000). In high-risk family settings, resilient youth are 3.5 times more likely to be free of health problems throughout their lifetime than their peers with low resilience (Hopkins et al., 2015). The moderating effects of the remaining personality types were also positive, but not as good as that of the resilience group. Moreover, it is worth emphasizing that leisure satisfaction seems to make a major contribution to the existence of a moderating effect. This further suggests that adolescents of different types could benefit from well-experienced leisure (Bradley and Inglis, 2012).