Neural correlates of non-clinical internet use in the motivation network and its modulation by subclinical autistic traits

Abstract

Background: Increasing evidence regarding the neural correlates of excessive or pathological internet use (IU) has accumulated in recent years, and comorbidity with depression and autism has been reported in multiple studies. However, psychological and neural correlates of non-clinical IU in healthy individuals remain unclear. Objectives: The aim of the current study was to investigate the relationships between non-clinical IU and functional connectivity (FC), focusing on the brain’s motivation network. We sought to clarify the influence of depression and autistic traits on these relationships in healthy individuals. Methods: Resting-state functional magnetic resonance imaging (fMRI) was performed in 119 healthy volunteers. IU, depression, and autistic traits were assessed using the Generalized Problematic Internet Use Scale 2 (GPIUS2), Beck Depression Inventory-II (BDI-II), and the autism spectrum quotient (AQ) scale, respectively. Correlational analyses were performed using CONN-software within the motivation-related network, which consisted of 22 brain regions defined by a previous response-conflict task-based fMRI study with a reward cue. We also performed mediation analyses via the bootstrap method. Results: Total GPIUS2 scores were positively correlated with FC between the (a) left middle frontal gyrus (MFG) and bilateral medial prefrontal cortex; (b) left MFG and right supplementary motor area (SMA); (c) left MFG and right anterior insula, and (d) right MFG and right insula. The “Mood Regulation” subscale of the GPIUS2 was positively correlated with FC between left MFG and right SMA. The “Deficient Self-Regulation” subscale was positively correlated with FC between right MFG and right anterior insula (statistical thresholds, FDR < 0.05). Among these significant correlations, those between GPIUS2 (total and “Mood Regulation” subscale) scores and FC became stronger after controlling for AQ scores (total and “Attention Switching” subscale), indicating significant mediation by AQ (95% CI < 0.05). In contrast, BDI-II had no mediating effect. Conclusion: Positive correlations between IU and FC in the motivation network may indicate health-promoting effects of non-clinical IU. However, this favorable association is attenuated in individuals with subclinical autistic traits, suggesting the importance of a personalized educational approach for these individuals in terms of adequate IU.