Hostname: page-component-74656f9884-rngvg Total loading time: 0 Render date: 2026-11-04T05:39:26.537Z Has data issue: false hasContentIssue false
  • English
  • Français

Social withdrawal among university students in the UK: the emerging hikikomori phenomena

Published online by Cambridge University Press:  20 July 2026

Ismail Hassan ,
Kieran K. Chumun ,
Hussien Elkholy Open the ORCID record for Hussien Elkholy [Opens in a new window]  and
Yoko Nagai Open the ORCID record for Yoko Nagai [Opens in a new window]
Ismail Hassan
Affiliation:
Department of Clinical Neuroscience, Brighton and Sussex Medical School, University of Sussex, UK
Kieran K. Chumun
Affiliation:
Department of Clinical Neuroscience, Brighton and Sussex Medical School, University of Sussex, UK
Hussien Elkholy
Affiliation:
Department of Clinical Neuroscience, Brighton and Sussex Medical School, University of Sussex, UK Department of Neurology and Psychiatry, Faculty of Medicine, Ain Shams University, Egypt Sussex Partnership NHS Foundation Trust, Brighton, UK
Yoko Nagai*
Affiliation:
Department of Clinical Neuroscience, Brighton and Sussex Medical School, University of Sussex, UK
*
Correspondence: Yoko Nagai. Email: y.nagai@bsms.ac.uk
Rights & Permissions [Opens in a new window]

Abstract

Background

Hikikomori, first recognised and defined in Japan as a culture-bound syndrome of prolonged social withdrawal, is increasingly reported worldwide. There is now a broad consensus that hikikomori is not confined to Japan. Its relevance to the UK context is raised by the presence of approximately one million young adults not in education, employment or training (NEET).

Aims

To examine the occurrence of hikikomori and hikikomori-like behaviours among UK university students and to explore the association between mental health difficulties, neurodevelopmental traits and gaming disorder in its development.

Method

A cross-sectional online survey was conducted, with a final sample of 105 students completing validated measures of social withdrawal, psychological distress, functional impairment and associated psychiatric traits.

Results

One in five students (21.0%) reported some form of hikikomori behaviour; 14.3% met criteria for pathological hikikomori, although no cases of severe hikikomori (home confinement >6 months) were identified. Hikikomori tendencies correlated strongly with psychological distress (r = 0.73, p < 0.001), autism-related traits (r = 0.72 p < 0.001) and internet gaming disorder (r = 0.70, p < 0.001). Moderate positive associations were found with depression and anxiety, while weaker negative associations emerged with psychotic-like experiences, agreeableness and conscientiousness. Gaming disorder, underlying mental health difficulties and autistic traits emerged as strong predictors of hikikomori behaviours.

Conclusions

Hikikomori-like behaviour may be more prevalent in the UK student population than previously recognised. The findings underline the importance of screening for social withdrawal in young adults, with particular attention to gaming behaviours, mental health problems and autistic traits. Furthermore, identifying and addressing pre-hikikomori cases (social withdrawal lasting less than three months) may be crucial for early intervention and prevention.

Keywords

Hikikomoriuniversity studentsmental healthautistic traitsgaming disorder

Information

Type
Original Article
Information
The British Journal of Psychiatry , First View , pp. 1 - 9
Check for updates
Copyright
© The Author(s), 2026. Published by Cambridge University Press on behalf of Royal College of Psychiatrists

Hikikomori, a term that originated in Japan, refers to a condition of profound social withdrawal, typically affecting adolescents and young adults who isolate themselves at home for six months or longer. Reference Borovoy1 While it was initially considered a culture-bound syndrome unique to Japanese society, growing evidence of similar cases worldwide has reframed hikikomori as a global phenomenon. This shift reflects a broader understanding of it as a complex societal issue rather than one confined to a single cultural context.

Hikikomori, or severe social withdrawal, is influenced by societal pressures, family dynamics and cultural expectations. Reference Nagai, Kartar, Pfaff and Elkholy2 In Japan, intense pressure to succeed and conform, along with a highly competitive education system, often overwhelms young individuals, leading to withdrawal. Family-related factors such as poor communication and increasingly fragmented household structures further contribute to isolation. Culturally, the tension between personal identity and societal roles, along with a hierarchical society, deepens the psychological roots of the condition.

While originally seen as unique to Japan, hikikomori is now observed globally. This observation is partly driven by universal stressors, such as the impact of the COVID-19 lockdowns, which normalised prolonged isolation. Reference Rooksby, Furuhashi and McLeod3 Additionally, the widespread availability of the internet and digital entertainment offers a convenient escape from societal expectations and real-world responsibilities. Reference Kuss and Griffiths4 These factors, present in many societies, have created fertile ground for hikikomori-like behaviours to emerge beyond Japan, highlighting its relevance as a global mental health concern.

In the UK, the growing number of youth, not in education, employment or training (NEET) presents a striking parallel to hikikomori. Between October and December 2024, 987 000 individuals aged 16–24 (over 13% of this age group) were classified as NEET; a 1.3% rise from the previous year. 5 NEET status is associated with negative outcomes, including increased risks of depression, anxiety, substance abuse, loneliness and long-term social isolation. Reference Bambra6,Reference Silva, Resurrección, Antunes, Frasquilho and Cardoso7 These individuals often struggle with low social integration, financial dependence and unstructured routines, contributing to chronic loneliness and lower life satisfaction. Reference Lindblad, Ravn, Gaardsted, Hansen, Lauritzen and Melgaard8 The trend of young adults remaining in the family home longer and the widespread accessibility of digital entertainment have further contributed to what some scholars describe as ‘digital retreatism’ or ‘digital escapism’, the gradual withdrawal from real-world obligations in favour of virtual spaces. Reference Sales-Filho, Bandeira, Argollo and Lucena9

Some researchers argue that hikikomori and NEET should not be considered separate conditions but rather points along a continuum spectrum of social withdrawal. Reference Uchida and Norasakkunkit10 Individuals at risk often share vulnerabilities such as social anxiety, low self-efficacy and difficulty meeting societal expectations. Hikikomori may represent the severe end of this spectrum, while NEET status could be an earlier or less severe stage of the same withdrawal process. Reference Uchida and Norasakkunkit10 The NEET/Hikikomori Risk (NHR) scale highlights shared predictors across both groups, including a preference for a freer lifestyle, lack of personal ambition and diminished self-competence. Reference Bambra6,Reference Uchida and Norasakkunkit10,Reference Kato, Kanba and Teo11

There is a pressing need to assess hikikomori within Western contexts like the UK, which is currently experiencing what has been described as a ‘loneliness epidemic’. Reference Holt-Lunstad12 Although comorbid psychiatric conditions such as depression, anxiety and autism traits are common among hikikomori individuals, Reference Gorman, Bacon, May and Minton13 many cases do not align neatly with conventional psychiatric diagnoses and occupy a diagnostic grey area – neither fully attributable to mental illness nor entirely reducible to cultural or social dynamics. Reference Kato, Kanba and Teo11

University students may be particularly vulnerable to social withdrawal, given their transitional life stage, academic pressures and increasing digital dependency. Reference McVarnock and Closson14 Emerging adulthood is a critical period for the onset of mental health issues, with individuals navigating academic, financial and social challenges that contribute to psychological distress. Reference Schulenberg, Sameroff and Cicchetti15 Rising rates of anxiety, depression and loneliness in this demographic Reference Wu, Chou, Catmur and Lau16 underscore the importance of early intervention. As students remain at least partially engaged in academic life, universities offer a unique opportunity to identify early signs of hikikomori before individuals become fully withdrawn from society.

Aims

This study investigates the occurrence and correlates of hikikomori among UK university students – a group situated at the intersection of academic stress, digital reliance and the psychosocial challenges of emerging adulthood. Reference Gonçalves and King17 Using a cross-sectional survey design, we apply the Hikikomori Diagnostic Evaluation Screening Tool (HiDE-S) alongside validated mental health and personality assessments. Reference Teo, Horie, Kurahara and Kato18 The goal is to identify key demographic, psychological, behavioural and neurodevelopmental predictors of social withdrawal. Particular attention is also given to how gender and personality influence these experiences and whether distinct subtypes of hikikomori may be present. This research aims to improve understanding of how hikikomori manifests in the UK and inform tailored intervention strategies.

Method

Participants

This study recruited 203 students from universities across the UK, targeting young adults during a pivotal stage in their academic and social development. Eligibility criteria required participants to be currently enrolled at a UK university and aged 18 or older. Recruitment was carried out via university mailing lists, social media platforms and digital posters. Ethical approval was obtained from the Brighton and Sussex Medical School (BSMS) Ethics Committee (ref. ER/BSMS9N6B/1).

After applying the exclusion criteria (participants aged under 18, incomplete responses and non-UK residents), the final sample consisted of 105 participants (mean age = 24.06 years, s.d. = 3.39). Participants who did not complete the survey, those who finished it in under five minutes and individuals who were not university students were also excluded. The gender distribution was 54.29% male, 43.81% female and 2.86% preferring not to disclose. These demographics allowed for the analysis of gender differences in hikikomori-like behaviour. Missing data was minimal (<5%) and handled using listwise deletion (Fig. 1).

Fig. 1

Recruitment flow chart. Flow chart of filtering process for included and excluded participants.

Flowchart illustrating the filtering process for included and excluded participants in a survey.
Fig. 1 Long description

Flowchart illustrating the filtering process for included and excluded participants in a survey. The process begins with the screening phase, where the number of people who filled out the survey is 203. In the filtered phase, participants are excluded based on several criteria: incomplete surveys (30), under 5 minutes completion time (25), non-university students (24), geolocation filtration for the UK (16), and under 18 years old (3), totaling 98 excluded participants. The final included phase shows that 105 respondents are kept in the survey.

Procedure

Data was collected through an anonymous, self-administered Qualtrics survey. Participants were recruited opportunistically from four UK universities via online advertisements distributed through university mailing lists and student social media platforms; no stratification or demographic weighting procedures were applied as this was a cross-sectional study aiming for descriptive analysis. Prior to participation, individuals were provided with an informed consent form and a participant information sheet detailing the study’s purpose and procedures. Consent was obtained electronically. Confidentiality and anonymity were strictly maintained in accordance with approved ethical standards. To encourage participation, respondents were entered into a randomised draw to win one of 40 £25 Amazon vouchers.

The survey remained open for two months, ensuring sufficient response time and reducing potential response biases. Participants retained the right to withdraw at any time before survey closure.

Questionnaires

To comprehensively assess social withdrawal and associated psychological constructs, a range of validated instruments were employed.

  1. (a) Hikikomori Diagnostic Evaluation – Screening Tool (HiDE-S): Designed to identify individuals exhibiting hikikomori-like behaviour, the HiDE-S evaluates social withdrawal based on diagnostic criteria including frequency, duration, functional impairment and subjective distress. It accounts for both self-identified distress and external concern from others. Originally developed in Japan, the HiDE-S has been adapted for international use. Reference Teo, Horie, Kurahara and Kato18

  2. (b) General Health Questionnaire (GHQ-28): A widely used screening tool for general mental health, the GHQ-28 assesses somatic symptoms, anxiety, depression and social dysfunction. Reference Goldberg and Hillier19 Its 28-item format is well-suited for large-scale studies and cross-cultural comparisons.

  3. (c) Hospital Anxiety and Depression Scale (HADS): The HADS consists of two subscales, HADS-A (Anxiety) and HADS-D (Depression), each with seven items. It minimises the confounding effects of physical health and has shown strong validity in identifying anxiety and depression across diverse populations. Reference Zigmond and Snaith20

  4. (d) Ritvo Autism Asperger Diagnostic Scale – 14 Item Version (RAADS-14): A brief screening tool for identifying autism traits in adults without intellectual disabilities. RAADS-14 assesses mentalising deficits, social anxiety and sensory sensitivity. It was included to explore possible neurodevelopmental links to social withdrawal. Reference Ritvo, Ritvo, Guthrie, Yuwiler, Ritvo and Weisbender21

  5. (e) Psychosis Screening Questionnaire (PSQ): This tool screens for psychotic-like experiences, such as paranoia, thought insertion, perceptual disturbances and unusual experiences. It distinguishes between past-year and lifetime prevalence, helping to identify potential psychotic contributions to social withdrawal. Reference Bebbington and Nayani22

  6. (f) Internet Gaming Disorder Test (IGDT-10): Based on DSM-5 criteria for internet gaming disorder, the IGDT-10 assesses the impact of gaming on social, academic and occupational functioning. Reference Király, Sleczka, Pontes, Urbán, Griffiths and Demetrovics23 It was included to examine the potential role of gaming in hikikomori-like behaviour.

  7. (g) Big Five Inventory – 10 Item Version (BFI-10): This concise personality assessment measures five traits: openness, conscientiousness, extraversion, agreeableness and neuroticism. Reference Goldberg24 Despite its brevity, it has demonstrated adequate psychometric reliability and is effective for large-scale research.

  8. (h) Self-Developed Informant Survey: Participants who reported knowing someone with hikikomori-like behaviour completed an additional questionnaire about the individual. This included questions on demographics, technology use, substance use and diagnoses (e.g. attention-deficit hyperactivity disorder, autism). These responses provided qualitative insights into contextual and environmental factors related to social withdrawal.

All measures were selected for their validated use in cross-cultural studies of social withdrawal and associated psychopathology.

Statistical analysis

We used descriptive statistics to explore the psychological, demographic and behavioural factors associated with hikikomori-like behaviour. Normality was assessed using the Shapiro–Wilk test. As the primary questionnaire (HiDE-S) was not normally distributed, Spearman’s correlation was applied. Statistical analyses were performed using SPSS (IBM SPSS Statistics, Version 27 (2020) for macOS, IBM Corp., Armonk, New York, USA; https://www.ibm.com/products/spss-statistics. Descriptive Statistics summarised sample characteristics and psychological scale scores. Percentage and distributions were reported for hikikomori-like behaviour and other relevant constructs. HiDE-S is fundamentally composed of two distinct sections: section 1, which assesses the frequency and duration of physical isolation, and section 2, which evaluates distress and functional impairment. While the main outcome of HiDE-S is a diagnostic classification, the continuous variable of the HiDE-S score was also used as it indicates overall severity. The continuous score was conceptualised as an index of overall withdrawal severity rather than representing physical isolation or psychological distress independently. The composite score was calculated combining section 1 and 2 into a single continuous value reflecting the global withdrawal experience. This approach is consistent with the clinical understanding that physical and psychological aspects of social withdrawal are interdependent and jointly contribute to overall burden. In group comparisons, independent sample t-tests were conducted to assess differences across gender and other demographic variables.

Spearman correlation coefficients were calculated to assess relationships between social withdrawal, psychological symptoms, personality traits and digital engagement. Additionally, a regression model was constructed with HiDE-S scores as the dependent variable. Predictors demonstrating significant correlation with HiDE-S scores were entered into the regression model. These included psychological distress, depression, anxiety, psychotic symptoms, autism traits and IGD scores. Assumptions of normality, linearity, homoscedasticity and multicollinearity were tested. Variance inflation factors (VIF < 4) were used to confirm acceptable collinearity levels. Residual plots were reviewed for homoscedasticity. Finally, mediation analysis was conducted to investigate the relationship with the different predisposing psychological factors. A combination of the magnitude of the univariate B coefficients and R 2 values in regression analysis was utilised to inform selection of variables for inclusion in the subsequent mediation analysis model.

Results

Occurrence of hikikomori in the UK university population

Among the 105 UK university students surveyed, 41 respondents (39.0%) met the preliminary criteria for physical hikikomori, reporting that they left home fewer than 4 days per week for short errands or similar outings (Table 1). Of those who passed the initial screening, further assessment of outing frequency revealed that 20 students (19.0%) were classified as experiencing mild hikikomori (going out 2 to 3 days per week), while 14 (13.3%) exhibited moderate to severe withdrawal, leaving home only once a week or less. Regarding the duration of withdrawal, 15 respondents (14.3%) indicated periods lasting 3 to 6 months, but notably, no participants reported social withdrawal exceeding 6 months – a criterion often associated with chronic hikikomori.

Table 1

Prevalence of hikikomori categories in a UK university student sample (N = 105)

Table showing prevalence of hikikomori categories among UK university students, detailing initial screening, severity, duration, and classification.
Table 1 Long description

The table presents data on the prevalence of hikikomori among 105 UK university students. It includes categories such as initial screening, severity of withdrawal, duration of withdrawal, and integrated classification. The table has 10 rows and 4 columns, with headers labeled Category, Definition, n, and percentage. Key data points include 41 students (39.0%) meeting the criteria for physical hikikomori, 20 students (19.0%) experiencing mild hikikomori, and 14 students (13.3%) showing moderate to severe withdrawal. Additionally, 15 students (14.3%) reported withdrawal durations of 3 to 6 months, with no cases of chronic hikikomori exceeding 6 months.

When psychological distress and functional impairment were considered, 15 students (14.3%) were classified as experiencing pathological hikikomori, meaning their withdrawal was accompanied by significant distress or disruption to daily life. An additional 8 respondents (7.6%) met criteria for non-pathological hikikomori, indicating withdrawal without notable distress. Taken together, 22 participants (21.0%) in this sample displayed some form of hikikomori behaviour. These findings suggest that social withdrawal tendencies are present within this university population and may warrant further attention in student mental health initiatives.

Correlational link

HiDE-S score was not normally distributed. Thus, Spearman’s correlation was used for the analysis. Spearman correlation analysis was used to examine the relationship between hikikomori tendencies, psychiatric conditions, personality traits and behavioural factors. A strong positive correlation was revealed between HiDE-S scores and general psychological distress with GHQ-28 (r = 0.729, p < 0.001). Similarly, significant correlations were observed between HiDE-S and depression (r = 0.517, p < 0.001) and anxiety (r = 0.504, p < 0.001), reinforcing the association between hikikomori-like behaviour and mood-related distress. Hikikomori symptoms are linked to behavioural tendencies, with internet gaming disorder showing a strong correlation with higher HiDE-S scores (r = 0.696, p < 0.001). This suggests that excessive internet and gaming use could play a role in social withdrawal patterns. Autism-related traits demonstrated a significant positive correlation with HiDE-S (r = 0.715, p < 0.001). Interestingly, psychotic-like experiences, measured by PSQ, showed a statistically significant negative correlation with HiDE-S (r = −0.272, p = 0.005). Personality traits also showed associations with higher HiDE-S scores, with inverse correlations being shown in agreeableness (r = −0.231, p = 0.018) and conscientiousness (r = −0.234, p = 0.016).

Beyond HiDE-S, general psychological stress was linked with a total score of anxiety and depression (r = 0.767, p < 0.001), internet gaming disorder (r = 0.521, p < 0.001) and autism trait (r = 0.767, p < 0.001), nevertheless fewer psychotic symptoms (r = −0.360, p = 0.026), less agreeableness (r = −0.324, p < 0.001) and less conscientiousness (r = −0.301, p < 0.001). Internet gaming disorder was significantly associated with autism-related traits (r = 0.489, p < 0.001).

Finally, mediation analysis was conducted to examine whether the association between the variables was explained by an indirect pathway. The model (PROCESS Model 4: parallel mediation model) Reference Hays25 was tested using three predictors of GHQ-28, IGDT-10 and RAADS-14 in the previous regression analysis (see Supplementary material available at https://doi.org/10.1192/bjp.2026.10688), which presented as strong predictors. Among the tested models, IGDT-10 (internet gaming disorder) emerged as the best direct predictor of HiDE-S scores. Mediation analysis indicated a significant total effect of IGDT-10 on HiDE-S (B = 0.742, 95% CI 0.59–0.89, p < 0.001). The overall model predicting HiDE-S was significant, F(3, 101) = 94.15, p < 0.0001, R 2 = 0.74. Both mediators contributed significantly: the indirect effect via RAADS (B = 0.187, bootstrapped 95% CI 0.085–0.33, p < 0.001), F(1, 103) = 32.84, p < 0.0001, R 2 = 0.49, accounting for 25.48% of the total effect, and the indirect effect via GHQ-28 (B = 0.141, bootstrapped 95% CI 0.07–0.22, p < 0.001), F(1, 103) = 25.73, p < 0.0001, R 2 = 0.20, accounting for 18.99% of the total effect. These results suggest that autistic traits and psychological distress partially mediate the association between gaming behaviour and hikikomori behaviour. The findings highlight the importance of considering psychological and neurodevelopmental factors when addressing the social withdrawal associated with problematic gaming, with implications for targeted interventions and prevention strategies in at-risk populations (Figs. 2 and 3).

Fig. 2

(a) HiDE-S versus GHQ-28 – a positive correlation between hikikomori tendencies and general psychological distress (r = 0.695, p < 0.001). (b) HiDE-S versus RAADS-14 – a strong association between hikikomori tendencies and autism-related traits (r = 0.736, p < 0.001). (c) HiDE-S versus IGDT-10 – hikikomori tendencies show a significant link to problematic gaming behaviours (r = 0.694, p < 0.001). GHQ-28, General Health Questionnaire; HIDE-S, Hikikomori Diagnostic Evaluation Screening Tool; RAADS-14, Ritvo Autism Asperger Diagnostic Scale – 14 Item Version; IGDT-10, Internet Gaming Disorder Test.

A scatter plot showing the relationship between hikikomori tendencies and various psychological factors.
Fig. 2 Long description

A scatter plot showing the relationship between hikikomori tendencies and various psychological factors. The plot consists of three subplots. Subplot (a) shows a positive correlation between hikikomori tendencies, measured by the HiDE-S score, and general psychological distress, measured by the GHQ-28 score. The x-axis represents the HiDE-S score, ranging from 0 to 20, and the y-axis represents the GHQ-28 score, ranging from 0 to 80. The data points show a positive correlation with a correlation coefficient of 0.695 and a p-value of less than 0.001. Subplot (b) shows a strong association between hikikomori tendencies and autism-related traits, measured by the RAADS-14 score. The x-axis represents the HiDE-S score, ranging from 0 to 20, and the y-axis represents the RAADS-14 score, ranging from 0 to 40. The data points show a strong association with a correlation coefficient of 0.736 and a p-value of less than 0.001. Subplot (c) shows a significant link between hikikomori tendencies and problematic gaming behaviors, measured by the IGDT-10 score. The x-axis represents the HiDE-S score, ranging from 0 to 20, and the y-axis represents the IGDT-10 score, ranging from 0 to 20. The data points show a significant link with a correlation coefficient of 0.694 and a p-value of less than 0.001. Each subplot includes a regression line indicating the trend. All values are approximated.

Fig. 3

Mediation analysis. Mediation analysis showed IGDT-10 as the strongest direct predictor of HiDE-S scores. Significant indirect effects were found through RAADS (25.48%) and GHQ-28 (18.99%), indicating partial mediation. These findings suggest autistic traits and psychological distress mediate gaming–hikikomori links. RAADS, Ritvo Autism Asperger Diagnostic Scale; IGDT-10, Internet Gaming Disorder Test; HIDE-S, Hikikomori Diagnostic Evaluation Screening Tool; GHQ-28, General Health Questionnaire.

A mediation analysis diagram showing relationships between IGDT, RAADS, GHQ, and HiDE-S scores.
Fig. 3 Long description

The mediation analysis diagram illustrates the relationships between IGDT, RAADS, GHQ, and HiDE-S scores. IGDT is connected to RAADS with a B value of 1.01 and a P value of less than 0.000. IGDT is also connected to HiDE-S with a B value of 0.411 and a P value of less than 0.000. RAADS is connected to HiDE-S with a B value of 1.1871 and a P value of less than 0.000. Additionally, IGDT is connected to GHQ with a B value of 1.301 and a P value of less than 0.000. GHQ is connected to HiDE-S with a B value of 0.1083 and a P value of less than 0.000. The diagram indicates that IGDT is the strongest direct predictor of HiDE-S scores, with significant indirect effects through RAADS and GHQ, suggesting partial mediation.

Gender differences

Independent t-test revealed that there was no significant difference between males (mean = 4.32, s.d. = 5.45) and females (mean = 4.43, s.d. = 5.05) in HIDE-S scores, t(101) = −0.114, p = 0.910, Cohen’s d = −0.022. However, significant gender differences were found in general psychological distress, anxiety and neuroticism. Females had significantly higher GHQ-28 (p = 0.02), HADS anxiety (p = 0.002) and Big 5 neuroticism (p < 0.001) scores than males. No significant differences were found in the other questionnaires or in age.

Discussion

The present study provides the first systematic examination of hikikomori-like behaviours among UK university students, revealing a notable occurrence of social withdrawal patterns that raises serious consideration. Our findings indicate that approximately 1 in 5 students (21.0%) exhibited some form of hikikomori behaviour, with 14.3% meeting criteria for pathological hikikomori, characterised by significant psychological distress or functional impairment, although the study did not identify any students who met the criteria for severe hikikomori, defined as remaining at home for more than 6 months. The findings further highlighted that computer game disorder, along with underlying mental health issues and autistic traits, were strong predictors of hikikomori, suggesting that these factors may play a central role in its development and persistence.

Pre-hikikomori data implication

Hikikomori, defined as prolonged social withdrawal lasting 6 months or longer, has gained international attention as a significant mental health phenomenon. Reference Kanai, Kitamura, Zha, Tanaka, Ikeda and Sobue26 While the standard definition requires a minimum duration of 6 months, there is growing interest in understanding shorter-duration withdrawal patterns, particularly ‘pre-hikikomori’ cases lasting 3–6 months. However, the available research literature provides limited specific data on prevalence rates for this intermediate duration category across different countries, making it difficult to synthesise a global comparison. This also represents a significant gap in the literature regarding pre-hikikomori (duration of 3 to 6 months) prevalence. Nevertheless, the systematic review noted that studies conducted outside Japan showed shorter hikikomori duration compared with Japanese studies, Reference Nonaka, Takeda and Sakai27 suggesting that international cases may include more shorter-duration episodes. Such data is important because early intervention (within 3–6 months) may be more effective, shorter withdrawal periods may be more common outside Japan and understanding occurrence could guide prevention strategies and resource allocation (Fig. 4). Without duration-specific data, targeted early interventions are harder to develop. In the UK university context, the concentration of cases within a 3-month period likely reflects the protected nature of the university setting, where students have access to peers, structured routines and mental-health services. Our findings suggest that some students showing hikikomori-like features also report symptoms consistent with other conditions, such as depression and autistic traits, making early support or treatment more accessible and potentially preventing progression beyond 6 months. This pattern may be characteristic of UK university environments rather than a broader national phenomenon. Methodology also plays a role: individuals who remain socially withdrawn for longer than 6 months may disengage from studies or drop out, making them unreachable through our student-based recruitment approach. Importantly, the occurrence of pre-hikikomori presentations in this UK university sample deserves careful attention.

Fig. 4

Comorbid condition of hikikomori. Conceptual framework illustrating comorbid conditions associated with hikikomori and the significance of early intervention. The diagram highlights the progression from pre-hikikomori states to full hikikomori, emphasising how early identification and targeted support strategies may interrupt this trajectory. Early intervention is proposed as a key factor in reducing symptom escalation, mitigating comorbid psychological and social difficulties and preventing the transition into prolonged social withdrawal.

A diagram illustrating comorbid conditions associated with hikikomori and the importance of early intervention.
Fig. 4 Long description

The diagram shows various comorbid conditions such as schizophrenia, autism spectrum disorder, psychosomatic symptoms, gaming disorder, personality disorder, and anxiety depression, all leading to pre-hikikomori, which is social withdrawal lasting 3 to 6 months. Pre-hikikomori can progress to hikikomori, defined as social withdrawal lasting more than 6 months. Early intervention is depicted as a crucial factor in preventing this progression.

Predictive factors of hikikomori

The analysis of our study presents evidence for a multifactorial model of hikikomori tendencies, explaining 80% of variance, consistent with Choi et al, Reference Choi, Yoon and Park28 who reported similar explanatory power in Korean adolescents using loneliness, education level and internet addiction. Gaming disorder was the strongest predictor, aligning with the other who documented high rates of internet gaming disorder among socially withdrawn individuals. Reference Fong, Cheng, Pai, Kwan, Wong, Cheung and Yip29 Furthermore, loneliness mediates the gaming–withdrawal link, Reference Shah, Helmy, Vally and Akaishi30 suggesting gaming serves as escape, social substitution and reinforcement. Psychological distress was another key predictor, supporting theories that hikikomori is a maladaptive coping response to mental health difficulties. Reference Carpita, Bonelli, Giovannoni, Parri, Gambini and Nardi31 Similarly, distinct distress patterns in socially withdrawn depressive patients were identified, reinforcing its role as a foundational vulnerability factor. Reference Orsolini, Longo, Bellagamba, Kato and Volpe32 Autism traits were significantly associated with withdrawal, consistent with the previous finding, Reference Dell’Osso, Nardi, Muti, De Felice, Tognini and Parri33,Reference Carpita, Nardi, Giovannoni, Parri, Cerofolini and Bonelli34 indicating that intrinsic social communication and sensory challenges contribute to risk. It is also indicated that individuals with autism face an increased risk of problematic internet use, in part because online platforms appear to compensate for difficulties with real-world social interaction. Reference DellOsso, Amatori, Muti, Giovannoni, Parri and Violi35

Unexpectedly, conscientiousness was positively associated with withdrawal, possibly reflecting perfectionism or rigidity, while neuroticism was inversely related, potentially due to emotional reactivity fostering social contact, findings that challenge traditional personality withdrawal assumptions.

Hikikomori and NEET

The most striking finding in this study is the observed association between computer game disorder and hikikomori. Excessive gaming, particularly when pathological, has been shown to contribute to social withdrawal by disrupting daily functioning. Reference Kuss and Griffiths4 Our results suggest that problematic gaming may not only maintain but also drive the emergence of prolonged withdrawal states characteristic of hikikomori.

This link also has implications for the NEET (not in education, employment or training) population. In Japan, hikikomori has been associated with mental health issues such as depression and social anxiety. Reference Teo, Fetters, Stufflebam, Tateno, Balhara and Choi36,Reference Kato, Kanba and Teo37 Similar mechanisms may underlie withdrawal among some NEET individuals, raising the possibility of overlap between the two groups. Importantly, it is necessary to distinguish between pathological withdrawal, where mental health conditions are central, and non-pathological withdrawal, where disengagement is voluntary. Reference Nonaka and Sakai38 This distinction is crucial, as intervention strategies must be tailored accordingly.

In the UK, NEET has been described as a hidden social crisis. Reference Yates and Payne39 If some NEET individuals also meet criteria for hikikomori, the public health implications could be substantial, given the long-term economic and social costs of persistent disengagement. Future research should therefore examine the overlap between hikikomori and NEET, the role of gaming in their development and the underlying mental health factors. Such work is essential for developing effective, targeted interventions and addressing what may represent an emerging social and public health challenge.

Gender impact

The present findings indicate that there was no significant difference in hikikomori-like behaviour between male and female participants in the UK survey. This pattern is consistent with earlier observations reported a decade ago, Reference Tajan40 that hikikomori is not predominantly male. This suggests that such behaviours are increasingly recognised, challenging earlier assumptions that hikikomori is a predominantly male Japanese sociocultural phenomenon. Reference Teo and Gaw41,Reference Kato, Shinfuku, Sartorius and Kanba42 However, it remains plausible that the original formation of hikikomori in Japan is deeply rooted in cultural elements such as social expectations, educational pressure and family dynamics Reference Furlong43,Reference Li and Wong44 where being male carries specific implications.

The cultural dimension may therefore continue to play a significant role in shaping the manifestation and experience of hikikomori within Japan, even as the behaviour becomes observable across diverse contexts.

In contrast, the overall vulnerability of female students to mental health problems during university years has been consistently documented. Reference Ibrahim, Kelly, Adams and Glazebrook45,Reference Macaskill46 Although female students reported slightly higher mental health vulnerability, our results show no significant gender differences in internet gaming disorder (IGD-10). This finding implies that excessive gaming itself is less likely to be the primary driver of mental health difficulties among female students exhibiting hikikomori-like tendencies. Instead, the mental health challenges experienced by female students may arise from alternative psychosocial stressors such as academic pressure, social comparison or gender-specific expectations. Reference Bayram and Bilgel47,Reference Auerbach, Alonso, Axinn, Cuijpers, Ebert and Green48 This highlights an important point: mental health problems and gaming disorder may represent distinct and independent causes of social withdrawal, requiring different forms of intervention.

Taken together, these findings suggest that while hikikomori-like behaviour is transcending cultural boundaries, its aetiology may still differ between regions. In Western contexts such as the UK, the behaviour may align more with general social withdrawal and underlying psychological vulnerabilities, whereas in Japan, cultural factors may amplify or reinforce its development. Future research should therefore examine not only cross-cultural prevalence but also the contextual determinants of hikikomori, with attention to gendered mental health risks.

Limitations

A significant number of the initial participants were removed through stringent inclusion and exclusion criteria. This substantial reduction reflects efforts to preserve model robustness by ensuring reliable and coherent responses to inform the analysis. While these exclusions were necessary for maintaining data quality, they resulted in a smaller analytic sample, which may limit the breadth of perspectives captured and requires cautious interpretation of the findings. The tested mediation model reflects one theoretically plausible pathway among several possible alternatives; however, given the cross-sectional design, alternative directional models cannot be excluded, and causal inference remains limited. Additionally, this study used a small, opportunistic sample drawn from four UK universities without stratification or demographic weighting, limiting population-level inference and introducing potential selection bias. Nevertheless, the exploratory design was appropriate for identifying under-recognised hikikomori-like patterns and their associations with psychological distress, autistic traits and gaming disorder, providing an important foundation for future larger, stratified and weighted studies to improve generalisability.

Implications

The substantial occurrence of hikikomori behaviours identified in this study has important implications for both clinical practice and institutional policy. Mental health professionals working with university populations should be trained to recognise and assess social withdrawal patterns, particularly given that withdrawn individuals may be less likely to present for services voluntarily. Universities may need to consider policy changes to better support students experiencing social withdrawal, including flexible attendance requirements, alternative assessment methods and enhanced digital support services. The development of specialised intervention programmes targeting social withdrawal could represent an important addition to existing student mental health services.

Supplementary material

The supplementary material is available online at https://doi.org/10.1192/bjp.2026.10688

Data availability

The data can be obtained from the corresponding author upon reasonable request.

Author contributions

Y.N. and H.E. conceptualised the study. I.H., K.K.C., H.E. and Y.N. designed the study, and I.H. and K.K.C. collected the data. All authors contributed to the analysis and interpretation. Y.N. is the main contributor to the writing with the support of I.H., K.K.C. and H.E.

Funding

This work was supported by internal funding from the University of Sussex, Brighton and Sussex Medical School.

Declaration of interest

None.

Footnotes

*

Joint first authors.

References

Borovoy, A. Japan’s hidden youths: mainstreaming the emotionally distressed in Japan. Cult Med Psychiatry 2008; 32: 552–76.10.1007/s11013-008-9106-2CrossRefGoogle ScholarPubMed
Nagai, Y, Kartar, A, Pfaff, M, Elkholy, H. The paradox of hikikomori through a transcultural lens. BJPsych Int 2025; 22: 22–4.10.1192/bji.2024.38CrossRefGoogle ScholarPubMed
Rooksby, M, Furuhashi, T, McLeod, HJ. Hikikomori: a hidden mental health need following the COVID-19 pandemic. World Psychiatry 2020; 19: 399–400.10.1002/wps.20804CrossRefGoogle ScholarPubMed
Kuss, DJ, Griffiths, MD. Online gaming addiction in children and adolescents: a review of empirical research. J Behav Addict 2012; 1: 3–22.10.1556/JBA.1.2012.1.1CrossRefGoogle ScholarPubMed
Office for National Statistics. Young People Not in Education, Employment or Training (NEET). ONS, 2025.Google Scholar
Bambra, C. Yesterday once more? Unemployment and health in the 21st century. J Epidemiol Community Health 2010; 64: 213–5.10.1136/jech.2009.090621CrossRefGoogle ScholarPubMed
Silva, M, Resurrección, DM, Antunes, A, Frasquilho, D, Cardoso, G. Impact of economic crises on mental health care: a systematic review. Epidemiol Psychiatr Sci 2018; 29: e7.10.1017/S2045796018000641CrossRefGoogle ScholarPubMed
Lindblad, V, Ravn, RL, Gaardsted, PS, Hansen, LEM, Lauritzen, FF, Melgaard, D. Beyond the mind: understanding the influence of mental health on youth NEET status in Europe: a systematic literature review. J Adolesc 2024; 96: 1428–44.10.1002/jad.12374CrossRefGoogle ScholarPubMed
Sales-Filho, GS, Bandeira, ID, Argollo, N, Lucena, R. Hikikomori syndrome and digital technologies: a systematic review. Harv Rev Psychiatry 2023; 31: 50–9.10.1097/HRP.0000000000000362CrossRefGoogle ScholarPubMed
Uchida, Y, Norasakkunkit, V. The NEET and Hikikomori spectrum: assessing the risks and consequences of becoming culturally marginalized. Front Psychol 2015; 6: 1117.10.3389/fpsyg.2015.01117CrossRefGoogle ScholarPubMed
Kato, TA, Kanba, S, Teo, AR. Multidimensional understanding, assessment, and future international perspectives. Psychiatry Clin Neurosci 2019; 73: 427–40.10.1111/pcn.12895CrossRefGoogle ScholarPubMed
Holt-Lunstad, J. The potential public health relevance of social isolation and loneliness: prevalence, epidemiology, and risk factors. Public Policy Aging Rep 2017; 27: 127–30.10.1093/ppar/prx030CrossRefGoogle Scholar
Gorman, G, Bacon, A, May, J, Minton, S. Hikikomori risk in the UK. Int J Soc Psychiatry 2025; 71: 1621–32.10.1177/00207640251348058CrossRefGoogle ScholarPubMed
McVarnock, AM, Closson, LM. Motivations for social withdrawal and academic adjustment in emerging adulthood. Br J Dev Psychol 2022; 40: 352–67.10.1111/bjdp.12411CrossRefGoogle ScholarPubMed
Schulenberg, JE, Sameroff, AJ, Cicchetti, D. The transition to adulthood as a critical juncture in the course of psychopathology and mental health. Dev Psychopathol 2004; 16: 799–806.10.1017/S0954579404040015CrossRefGoogle ScholarPubMed
Wu, AF-W, Chou, T-L, Catmur, C, Lau, JYF. Understanding the links between self-concept, sociocultural deviance and mental health problems in pathological social withdrawal. Curr Psychol 2023; 42: 5290–6.10.1007/s12144-021-01865-7CrossRefGoogle Scholar
Gonçalves, LNA, King, A. Digital dependence in the past decade: a systematic review. J Addict Res Adolesc Behav 2023; 6: 1–18.Google Scholar
Teo, AR, Horie, K, Kurahara, K, Kato, TA. The Hikikomori Diagnostic Evaluation (HiDE): a proposal for a structured assessment of pathological social withdrawal. World Psychiatry 2023; 22: 478–9.10.1002/wps.21123CrossRefGoogle ScholarPubMed
Goldberg, DP, Hillier, VF. A scaled version of the General Health Questionnaire. Psychol Med 1979; 9: 139–45.10.1017/S0033291700021644CrossRefGoogle ScholarPubMed
Zigmond, AS, Snaith, RP. The hospital anxiety and depression scale. Acta Psychiatr Scand 1983; 67: 361–70.10.1111/j.1600-0447.1983.tb09716.xCrossRefGoogle ScholarPubMed
Ritvo, RA, Ritvo, ER, Guthrie, D, Yuwiler, A, Ritvo, MJ, Weisbender, L. A scale to assist the diagnosis of autism and Asperger’s disorder in adults (RAADS): a pilot study. J Autism Dev Disord 2008; 38: 213–23.10.1007/s10803-007-0380-6CrossRefGoogle ScholarPubMed
Bebbington, PE, Nayani, T. The Psychosis Screening Questionnaire. Int J Methods Psychiatr Res 1995; 5, 11–9.Google Scholar
Király, O, Sleczka, P, Pontes, HM, Urbán, R, Griffiths, MD, Demetrovics, Z. Validation of the Ten-Item Internet Gaming Disorder Test (IGDT-10) and evaluation of the nine DSM-5 Internet Gaming Disorder criteria. Addict Behav 2017; 64: 253–60.10.1016/j.addbeh.2015.11.005CrossRefGoogle ScholarPubMed
Goldberg, LR. An alternative ‘description of personality’: the Big-Five factor structure. J Person Soc Psychol 1990; 59: 1216–29.10.1037/0022-3514.59.6.1216CrossRefGoogle ScholarPubMed
Hays, AF. Introduction to Mediation, Moderation, and Conditional Process Analysis: A Regression-Based Approach. Guilford Press, 2022.Google Scholar
Kanai, K, Kitamura, Y, Zha, L, Tanaka, K, Ikeda, M, Sobue, T. Prevalence of and factors influencing Hikikomori in Osaka City, Japan: a population-based cross-sectional study. Int J Soc Psychiatry 2024; 70: 967–80.10.1177/00207640241245926CrossRefGoogle ScholarPubMed
Nonaka, S, Takeda, T, Sakai, M. Who are hikikomori? Demographic and clinical features of hikikomori (prolonged social withdrawal): a systematic review. Aust N Z J Psychiatry 2022; 56: 1542–54.10.1177/00048674221085917CrossRefGoogle ScholarPubMed
Choi, E, Yoon, E-H, Park, M-H. Game-based digital therapeutics for children and adolescents: their therapeutic effects on mental health problems, the sustainability of the therapeutic effects and the transfer of cognitive functions. Front Psychiatry 2022; 13: 986687.10.3389/fpsyt.2022.986687CrossRefGoogle ScholarPubMed
Fong, TCT, Cheng, Q, Pai, CY, Kwan, I, Wong, C, Cheung, S-H, Yip, PSF. Uncovering sample heterogeneity in gaming and social withdrawal behaviors in adolescent and young adult gamers in Hong Kong. Soc Sci Med 2023; 321: 115774.10.1016/j.socscimed.2023.115774CrossRefGoogle Scholar
Shah, H, Helmy, M, Vally, Z, Akaishi, R. Hikikomori in the Middle East: the role of problematic gaming, social media use, and loneliness. PLOS One 2024; 19: e0312818.10.1371/journal.pone.0312818CrossRefGoogle ScholarPubMed
Carpita, B, Bonelli, C, Giovannoni, F, Parri, F, Gambini, M, Nardi, B, et al. Case report: hikikomori syndrome in Italy and its link with autistic traits and internet gaming disorder. Front Psychiatry 2024; 15: 1378572.10.3389/fpsyt.2024.1378572CrossRefGoogle ScholarPubMed
Orsolini, L, Longo, G, Bellagamba, S, Kato, TA, Volpe, U. Predominant affective temperaments in depressive patients with severe social withdrawal. Ann Gen Psychiatry 2024; 23: 12.10.1186/s12991-024-00496-zCrossRefGoogle ScholarPubMed
Dell’Osso, L, Nardi, B, Muti, D, De Felice, C, Tognini, V, Parri, F, et al. Comorbidity between hikikomori and autistic traits may be identified as a phenotypical presentation characterized by greater severity. Brain Sci 2025; 15: 496.10.3390/brainsci15050496CrossRefGoogle ScholarPubMed
Carpita, B, Nardi, B, Giovannoni, F, Parri, F, Cerofolini, G, Bonelli, C, et al. Exploring the relationship among hikikomori tendencies, autistic traits, computer game use and eating disorder symptoms. CNS Spectr 2024; 29: 670–81.10.1017/S1092852924002335CrossRefGoogle ScholarPubMed
DellOsso, L, Amatori, G, Muti, D, Giovannoni, F, Parri, F, Violi, M, et al. Autism spectrum, hikikomori syndrome and internet gaming disorder: is there a link? Brain Sci 2023; 13: 1116.Google ScholarPubMed
Teo, AR, Fetters, MD, Stufflebam, K, Tateno, M, Balhara, Y, Choi, TY, et al. Identification of the hikikomori syndrome of social withdrawal: psychosocial features and treatment preferences in four countries. Int J Soc Psychiatry 2015; 61: 64–72.10.1177/0020764014535758CrossRefGoogle ScholarPubMed
Kato, TA, Kanba, S, Teo, AR. Defining pathological social withdrawal: proposed diagnostic criteria for hikikomori. World Psychiatry 2020; 19: 116–7.10.1002/wps.20705CrossRefGoogle ScholarPubMed
Nonaka, S, Sakai, M. Psychological factors associated with social withdrawal (hikikomori). Psychiatry Investig 2021; 18: 463–70.10.30773/pi.2021.0050CrossRefGoogle ScholarPubMed
Yates, S, Payne, M. Not so NEET? A critique of the use of ‘NEET’ in setting targets for interventions with young people. J Youth Stud 2006; 9, 329–44.10.1080/13676260600805671CrossRefGoogle Scholar
Tajan, N. Mental Health and Social Withdrawal in Contemporary Japan: Beyond the Hikikomori Spectrum. Routledge, 2021.10.4324/9781351260800CrossRefGoogle Scholar
Teo, AR, Gaw, AC. Hikikomori, a Japanese culture-bound syndrome of social withdrawal?: A proposal for DSM-5. J Nerv Ment Dis 2010; 198: 444–9.10.1097/NMD.0b013e3181e086b1CrossRefGoogle ScholarPubMed
Kato, TA, Shinfuku, N, Sartorius, N, Kanba, S. Are Japan’s hikikomori and depression in young people spreading abroad? Lancet 2011; 378, 1070.10.1016/S0140-6736(11)61475-XCrossRefGoogle Scholar
Furlong, A. The Japanese hikikomori phenomenon: acute social withdrawal among young people. Sociol Rev 2008; 56: 309–25.10.1111/j.1467-954X.2008.00790.xCrossRefGoogle Scholar
Li, TMH, Wong, PWC. Youth social withdrawal behavior (hikikomori): a systematic review of qualitative and quantitative studies. Aust N Z J Psychiatry 2015; 49: 595–609.10.1177/0004867415581179CrossRefGoogle ScholarPubMed
Ibrahim, AK, Kelly, SJ, Adams, CE, Glazebrook, C. A systematic review of studies of depression prevalence in university students. J Psychiatr Res 2013; 47: 391–400.10.1016/j.jpsychires.2012.11.015CrossRefGoogle ScholarPubMed
Macaskill, A. The mental health of university students in the United Kingdom. Br J Guidance Counsell 2013; 41: 178–85.10.1080/03069885.2012.743110CrossRefGoogle Scholar
Bayram, N, Bilgel, N. The prevalence and socio-demographic correlations of depression, anxiety and stress among a group of university students. Soc Psychiatry Psychiatr Epidemiol 2008; 43: 667–72.10.1007/s00127-008-0345-xCrossRefGoogle ScholarPubMed
Auerbach, RP, Alonso, J, Axinn, WG, Cuijpers, P, Ebert, DD, Green, JG, et al. Mental disorders among college students in the World Health Organization World Mental Health Surveys. Psychol Med 2016; 46: 2955–70.10.1017/S0033291716001665CrossRefGoogle ScholarPubMed
Figure 0

Fig. 1 Fig. 1 long description.Recruitment flow chart. Flow chart of filtering process for included and excluded participants.

Figure 1

Table 1 Prevalence of hikikomori categories in a UK university student sample (N = 105)Table 1 long description.

Figure 2

Fig. 2 Fig. 2 long description.(a) HiDE-S versus GHQ-28 – a positive correlation between hikikomori tendencies and general psychological distress (r = 0.695, p < 0.001). (b) HiDE-S versus RAADS-14 – a strong association between hikikomori tendencies and autism-related traits (r = 0.736, p < 0.001). (c) HiDE-S versus IGDT-10 – hikikomori tendencies show a significant link to problematic gaming behaviours (r = 0.694, p < 0.001). GHQ-28, General Health Questionnaire; HIDE-S, Hikikomori Diagnostic Evaluation Screening Tool; RAADS-14, Ritvo Autism Asperger Diagnostic Scale – 14 Item Version; IGDT-10, Internet Gaming Disorder Test.

Figure 3

Fig. 3 Fig. 3 long description.Mediation analysis. Mediation analysis showed IGDT-10 as the strongest direct predictor of HiDE-S scores. Significant indirect effects were found through RAADS (25.48%) and GHQ-28 (18.99%), indicating partial mediation. These findings suggest autistic traits and psychological distress mediate gaming–hikikomori links. RAADS, Ritvo Autism Asperger Diagnostic Scale; IGDT-10, Internet Gaming Disorder Test; HIDE-S, Hikikomori Diagnostic Evaluation Screening Tool; GHQ-28, General Health Questionnaire.

Figure 4

Fig. 4 Fig. 4 long description.Comorbid condition of hikikomori. Conceptual framework illustrating comorbid conditions associated with hikikomori and the significance of early intervention. The diagram highlights the progression from pre-hikikomori states to full hikikomori, emphasising how early identification and targeted support strategies may interrupt this trajectory. Early intervention is proposed as a key factor in reducing symptom escalation, mitigating comorbid psychological and social difficulties and preventing the transition into prolonged social withdrawal.

Supplementary material: File

Hassan et al. supplementary material

Hassan et al. supplementary material
Download Hassan et al. supplementary material(File)
File 23 KB

This journal is not currently accepting new eletters.

eLetters

No eLetters have been published for this article.