Answers are grounded in this paper and cite their sources.
Internet gaming disorder (IGD) is a behavioral addiction whose resting-state brain signature may be expressed in transient connectivity configurations rather than in a scan-wide average. Resting-state fMRI data from a predominantly adolescent cohort of 58 participants with IGD and 56 healthy controls were analysed with Dosenbach-160 regional time series, 60-TR sliding windows, and a four-state k-means solution evaluated for stability and robustness. State-specific AUC graph metrics and temporal metrics were compared with covariate-adjusted models and residual-permutation inference. Linear SVR with leave-one-out cross-validation tested prediction of Internet Addiction Test (IAT) scores within the IGD group, with repeated nested cross-validation as a robustness analysis. Under within-state FDR correction, group differences in graph topology were detected only in State 4 under the present state-wise analyses: IGD showed lower clustering, global efficiency, local efficiency, normalised clustering, and small-worldness, together with longer characteristic path length. Local efficiency, normalised clustering, and small-worldness remained significant after correction across all 28 state-by-metric tests. IGD showed lower fractional occupancy and shorter mean dwell time in States 1 and 4 and higher fractional occupancy and longer mean dwell time in State 3 (nine-outcome FDR-adjusted q ≤ 0.024); State 2 temporal measures and total transitions did not differ. State-specific graph features predicted IAT scores in State 2 ( r = 0.56, FDR-adjusted q = 0.010, n = 58) and State 4 ( r = 0.57, FDR-adjusted q = 0.014, n = 45); repeated nested cross-validation yielded median r values of 0.54 and 0.58, respectively. IGD was characterised by a state-dependent redistribution of temporal occupancy, together with topological group differences detected only in State 4 (a moderately strongly connected configuration accounting for 18.9% of recording time, in which functional integration and segregation are both reduced) under the present state-wise analyses. State-resolved graph features also tracked addiction severity in States 2 and 4. These findings support complementary temporal and topological characterisation of dynamic brain-network alterations in IGD, while requiring replication in larger independent samples.
No takes yet. Share an insight, caveat, or question.
Guo et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: