To the best of our knowledge, this is the first study investigating the relationship between affective temperaments and Hikikomori-like social withdrawal symptomatology in a cohort of young adults, by comparing clinically significant depressed versus not-depressed individuals and exploring differences between sexes. In particular, following more recent research directions which suggested to explore Hikikomori-like social withdrawal as a transdiagnostic specifier, particularly within the sample of individuals with depressive symptomatology [6, 34], our study specifically explored the association between Hikikomori as diagnostic specifier associated with depression and the identification of specific predominant associated affective temperaments. Overall, our findings found that, within the total recruited sample, HQ-25 mean total scores are overly higher compared to previous published studies carried out within an Italian sample [28, 35, 36], probably due to the younger age of our sample and the recruitment period which was after the COVID-19 pandemic. Comparable with previous studies [28, 35, 36], in our sample, males reported significantly higher HQ-25 scores, particularly in the subscales ‘isolation’ and ‘emotional support’, compared to females. In the total sample, Hikikomori-like social withdrawal symptomatology was found to be significantly predicted by higher levels of low arousal boredom (i.e., by manifesting dysphoria, feelings of emptiness and fatigue) and feelings of disengagement from meaningful and interesting life activities (as measured by MSBS). Moreover, Hikikomori-like social withdrawal has been significantly predicted by higher depressive levels, lower anxiety levels and by higher levels at irritable and depressive affective temperaments (as measured by TEMPS-M). Indeed, despite some studies having been carried out by exploring some child temperamental features and social isolation, there are no published studies specifically addressed to young adults [37, 38]. These studies found an association between child social isolation and the presence of the so-called behavioral inhibition temperament, i.e. the tendency to react following exposure to unfamiliar stimuli by developing anxiety and avoidance behavior [37, 38]. Akiskal already identified a possible association between social isolation and specific affective temperaments, such as cyclothymic (particularly in transient social isolation episodes), and depressive affective temperaments (more associated with the tendency to develop a social withdrawal) [39, 40]. Subjects with depressive affective temperaments tend to be sensitive to suffering, self-denying and devote themselves to others [40]. Their optimal balance is achieved when they are in harmony with others, adhering to social norms and roles [40]. According to this perspective, social isolation could be a maladaptive mechanism to escape suffering or the inability to find harmony with the world around them. In addition, such subjects are characterized by harm avoidance and low novelty seeking, which often leads the subject to a boring life and, hence, could result in a progressive isolation [40]. On the other hand, subjects with an irritable affective temperament, are characterized by the presence of irritable-lunatic mood with 'ill-humored joking' [30]. Such traits could result in disagreements with peers and social impairment leading to progressive social withdrawal both as distanced from others and as a maladaptive defense mechanism (such a person might feel not understood by others) [30]. This could be reinforced by the fact that such individuals have a tendency to brood and great impulsivity [30]. Disengagement in the process leading to boredom is characterized by a difficulty in the process of orienting and attributing attention to the environment resulting in a mismatch between fully experiencing an activity and paying attention to it [41, 42]. This process may explain why this dimension is related to social withdrawal. Indeed, the lack of attention to the environment results in a disinterest in what we find in the environment, with a progressive isolation. Finally, low arousal is connected with the proneness to boredom. In particular, subjects with a low arousal try to find some activities to enhance their arousal [43]. If this process is maladaptive, we can hypothesize that persistent low activation could lead to isolation through feelings of emptiness and fatigue, despite we should integrate these findings by using longitudinal cohort studies to demonstrate whether there is a causal relationship between boredom dimension and the onset of a HK-like social withdrawal within the depressed individuals.

Moreover, considering that our sample is mainly represented by females, we also carried out a sub-analysis assessing potentially sex-based differences in the clinical and predisposition to the development of a depression with or without a Hikikomori diagnostic specifier. In particular, according to our findings, within the male sample, Hikikomori-like social withdrawal conditions seemed to be significantly predicted by general higher boredom levels and the level of depressive severity. While only the presence of a predominant hyperthymic temperament seemed to be negatively associated with the presence of a Hikikomori-like social withdrawal, by suggesting a potential protective role, which should be further investigated and confirmed in larger longitudinal studies evaluating both depressed and not-depressed HK individuals and considering sex-based differences between both samples. Indeed, there are no published studies which allow us to confirm these findings, despite Akiskal previously suggested a possible association of transient social isolation episodes (not Hikikomori-like) and a predominant hyperthymic temperament, mainly occurring as a reaction to the social and seldom maladaptive consequences of their temperaments and the subsequent need to self-isolate in order to have not been exposed to a negative judgment from others [40]. However, social isolation among hyperthymic individuals, may indeed represent a transient reaction which rarely meets Hikikomori diagnostic criteria [6]. Furthermore, our findings reported that, among females, Hikikomori-like social withdrawal seemed to be significantly associated with low arousal and disengagement boredom levels, by depression severity and by irritable, anxious and depressive affective temperaments. Indeed, previous literature already documented that, among females, the most predominant affective temperaments are generally represented by anxious and depressive affective temperaments [20] by, hence, suggesting that probably the irritable affective temperament could indeed more likely be associated to the psychopathological trajectory leading to HK-like social withdrawal symptomatology [30]. However, also these preliminary findings should also be extensively confirmed and replicated in more larger sex-based cohort longitudinal studies, as there is no still published literature on the topic, either in depressed versus not depressed young adults.

Furthermore, after stratifying the entire sample according to the presence versus absence of a depressive symptomatology, according to our findings, depression associated with Hikikomori specifier seemed to be positively associated with higher general boredom levels, particularly low arousal boredom levels. Therefore, within the depressed sample, Hikikomori seemed to not be predicted by disengagement boredom levels as observed in the total sample. This could be explained by the fact that individuals in whom there is a greater component of the disengagement dimension, could display some difficulty in recognizing internal information (e.g., thoughts and emotions) [41]. On the other hand, this ability is often present in those suffering from depression, and this would result in the ability to direct attention to specific environmental elements, more likely responsible for the development of the depressive symptomatology. Moreover, the likelihood of developing Hikikomori-like social withdrawal symptomatology within the context of a depression seemed to be positively predicted by the presence of predominant depressive and cyclothymic affective temperaments, confirming data already observed by Akiskal [21, 30]. Generally, major depressive disorder (MDD) is associated with depressive and anxious temperament [20, 44]. Cyclothymic temperament is associated with the development of Bipolar Disorder (BD), in particular to type II [20, 44]. However, this temperament is also associated with forms of MDD lately evolving into BD, in those individuals who develop MDD but with a positive family history for BD and in atypical forms of MDD [20, 44]. Therefore, cyclothymic temperament could be associated with those clinical phenotypic depressive forms which could be different from the classic MDD clinical picture. Cyclothymic could evolve into depression associated with Hikikomori-like social withdrawal as a diagnostic specifier, manifested by mood swings, which could lead to increasingly frequent depressive episodes over time and, hence, resulting in the potential development of a progressive social and emotional isolation.

Therefore, based on our preliminary findings, one could argue that depression associated with Hikikomori-like social withdrawal symptomatology could represent a distinct type of depression which should be adequately investigated and clinically characterized from a diagnostic and therapeutic perspective, in order to build a personalized and tailored-based intervention. The association with specific affective temperamental profiles could also help clinicians in early identifying those depressed individuals at-risk to develop a clinical picture associated with Hikikomori specifier which, indeed, could potentially modify clinical course, outcomes and treatment strategies.

However, despite these preliminary and promising findings, our study has several limitations which should be adequately addressed and discussed before generalizing our results. Firstly, findings coming from the total sample could be influenced by sex unbalance, being our total sample mainly represented by females. Secondly, our sample could be influenced by the highly age-based selection, mainly represented by young adults (aged 18–35). Therefore, further studies should be carried out by including a more representative sample of individuals (independently by the presence of comorbid depressive symptomatology), in order to clearly confirm these findings only in primary Hikikomori and according to different age ranges. Conversely, when comparing both groups (depressed versus not-depressed), both samples are sex- and age-based homogeneously represented. Therefore, one could argue that findings coming from sub-analysis could be more easily generalizable to the sample of individuals affected by Hikikomori secondary to depression. Thirdly, the cross-sectional nature of our study does not allow us to draw up definitive conclusions regarding the relationship between depression and Hikikomori-like social withdrawal (i.e., identifying whether an individual is affected by a primary depression with Hikikomori or a depression secondary to Hikikomori). Fourthly, being our total sample mainly represented by females, our findings regarding the association between specific affective temperaments and Hikikomori-like social withdrawal symptomatology could be biased by the female-effect in temperamental profiles. Fifthly, our study is a nationwide population-based study and, hence, our findings could be influenced by selection bias, by the fact that our sample is a nonclinical one. Therefore, further larger longitudinal, multicentric and pan-European based studies should be conducted in order to replicate our preliminary findings as well as longitudinally identifying specific sex-based predictors influencing the clinical course, manifestation, treatment outcomes and prognosis of individuals affected by depression depending on the presence of Hikikomori diagnostic specifier. Furthermore, despite our preliminary findings also investigated as secondary exploratory outcomes, boredom dimensions and anxiety symptomatology, the cross-sectional nature of the study did not allow to draw uo definitive conclusions regarding the potential causal relationship between boredom dimension and subdimensions and the increased/decreased chance to develop Hikikomori-like social withdrawal both in depressed versus not depressed individuals, despite our findings could suggest also a potential association and role depending on the type of predominant affective temperament.

Overall, our preliminary findings could significantly help clinicians working with young adults manifesting depressive symptomatology by potentially shedding the light on the possible association between specific predominant affective temperamental profiles and the increased chance to develop a depression associated with the Hikikomori diagnostic specifier. However, our findings coming from a nationwide, Italy-based, nonclinical population-based study specifically recruiting young adults aged 18–35 which could indeed help in providing a current snapshot of the Italian situation regarding youth depression with/without Hikikomori, despite our findings should be extensively replicated in longitudinal clinical studies recruiting both primary and secondary Hikikomori subjects. An interesting result comes from our sex-based stratified sub-analysis, which suggested a potential different clinical phenotypization depending on the sex and also influenced by predominant affective temperament, which should also be confirmed and verified in larger longitudinal clinical studies. In particular, there is the need to confirm which affective temperaments could be protective (or risky) for the development of a Hikikomori-like social withdrawal symptomatology in depressed individuals. Early identification of affective temperament in patients with depression could help in predicting which will be the potential developing psychopathological trajectory leading to the onset of a Hikikomori-like social withdrawal associated with depressive symptomatology and which should be the tailored and personalized treatment to be adapted accordingly. Meanwhile, a comprehensive personological characterization of individuals who develop Hikikomori-like social withdrawal, considering both depressed versus not depressed individuals would be useful to better clinically characterize from a diagnostic and therapeutic perspectives these subjects, also investigating the (potential) mediatory role of the boredom dimensions as well as attachment style profiles and anxiety trait and state. Finally, following suggestions and research hypotheses of previous researchers [1, 2, 4, 10], it would be appropriate to clinically characterize depression associated with Hikikomori by identifying similarities and differences (if any) with the psychopathological construct of the Modern-Type Depression.