Problematic Internet use (PIU) refers to a preoccupation with the Internet accompanied by various symptoms, which can cause functional impairments and psychosocial distress (Lozano-Blasco et al., 2022). Despite controversies over its diagnostic criteria and assessment instruments (Casale and Fioravanti, 2023, Laconi et al., 2014), concerns about PIU’s rising prevalence and potential harms have led to its recognition as a national threat in China (Kuss et al., 2021). A meta-analysis revealed this increasing trend, with eastern societies (8.9 %) showing a higher prevalence than western societies (4.6 %) (Pan et al., 2020).

Adolescents’ psychological immaturity and neurobiological plasticity make them particularly vulnerable to PIU’s health impact (Cerniglia et al., 2017). While PIU has been associated with anxiety, depression, and sleep problems (Alimoradi et al., 2019, Cai et al., 2023), its relationship with psychotic-like experiences (PLEs) – a common phenomenon in adolescence – remains understudied. PLEs are subclinical experiences that resemble positive symptoms of psychosis, such as hallucinations and delusions (Kelleher et al., 2012). Neurodevelopmentally, adolescence represents a critical period of psychosis proneness, marked by typically transitory PLEs (van Os et al., 2009). Along the psychosis continuum, environmental exposures during this time can trigger progression from transitory to persistent PLEs and eventually psychotic disorders (van Os et al., 2009).

Evidence supports PIU as a stressor for PLEs, though most studies are cross-sectional or focus on adults. For instance, PIU has been linked to concurrent PLEs among Korean (Lee et al., 2019) and Irish (McMahon et al., 2021) adolescents, and to increased PLEs over two months in 170 adults (Mittal et al., 2013). In addition, PIU was found to predict PLEs over one year in 636 Chinese college students (Peng & Zou, 2025). Notably, only one study has empirically identified PIU as a risk factor for future PLEs during adolescence (Narita et al., 2024). However, its minimum prediction interval of four years limits its applicability to near-future interventions. Given the transitory nature of developmental PLEs in adolescence (van Os et al., 2009), shorter-interval studies are essential to better capture their progression stemming from PIU.

Sleep disturbance, such as insomnia symptoms and poor sleep quality, is associated with PIU (Alimoradi et al., 2019). Internet use can displace sleep duration (Kokka et al., 2021) and disrupt sleep quality through limited physical activity and nature exposure (Shin et al., 2020, Stepanski and Wyatt, 2003). For example, prolonged digital media use at various daily intervals could all worsen future sleep outcomes (Chen et al., 2024). Moreover, preference for pre-bedtime Internet use may heighten psychological arousal (Brautsch et al., 2023, Hysing et al., 2015) and cause insomnia (Harvey, 2002, Riemann et al., 2010). Empirically, pre-sleep arousal was found a mediator between various online behaviors and sleep outcomes (Harbard et al., 2016, Tu et al., 2023, Wang and Scherr, 2022).

Sleep disturbance, in turn, is a well-documented risk factor for PLEs in general populations (Barton et al., 2018, Koyanagi and Stickley, 2015), with prospective studies showing its role in the presence, onset, and persistence of PLEs in adolescence (Reeve and Bell, 2023, Thompson et al., 2015, Wang et al., 2022). The diathesis-stress hypothesis proposes that general sleep problems during adolescence act as stressors, interfering with neuromaturation and contributing to psychosis development (Lunsford-Avery & Mittal, 2013). However, the mediating role of sleep disturbance in the longitudinal PIU-PLEs link remains unexplored.

Psychological resilience, the ability to positively adapt to stressful experiences, is critical for protecting mental health (Hu et al., 2015). According to an integrated model, resilience can mitigate stressor effects by hampering the transformation of adaptive into maladaptive processes or by hindering the development of psychopathologies (Davydov et al., 2010). In the current mediation model, where PIU acts as a stressor for sleep disturbance (a maladaptive process (Drake et al., 2014)) and PLEs (a psychopathology (Narita et al., 2024)), and where sleep disturbance serves as a stressor for PLEs (Reeve et al., 2015), resilience may buffer these pathways. While resilience is a known protective factor for PIU (Hidalgo-Fuentes et al., 2023), sleep disturbance (Arora et al., 2022), and PLEs (DeLuca et al., 2022), its moderating role in the PIU-PLEs link remains untested.

Accordingly, this study aims to examine: (1) the longitudinal association between PIU and PLEs over six months among adolescents; (2) the mediating role of sleep disturbance; and (3) the moderating role of resilience. The proposed moderated mediation model is illustrated in Fig. 1.