Insomnia is a prevalent and burdensome sleep disorder characterized by persistent difficulties initiating or maintaining sleep, accompanied by impaired daytime functioning. Beyond subjective sleep complaints, insomnia is associated with fatigue, reduc...
Insomnia is a prevalent and burdensome sleep disorder characterized by persistent difficulties initiating or maintaining sleep, accompanied by impaired daytime functioning. Beyond subjective sleep complaints, insomnia is associated with fatigue, reduced cognitive performance, emotional dysregulation, and elevated risks of psychiatric disorders and cardiometabolic conditions. The public health impact is substantial because insomnia is linked to decreased quality of life and productivity loss at the population level. Current evidence-based care emphasizes non-pharmacological management, particularly cognitive behavioral therapy for insomnia (CBT-I), which is widely recognized as an effective first-line approach. However, real-world implementation remains constrained by limited accessibility, the need for trained providers, and the sustained engagement required from patients. Pharmacological treatments can be effective in the short term, but concerns persist regarding tolerance, dependence, and adverse effects with prolonged use. These limitations have increased interest in complementary and non-invasive interventions that may support self-regulation of sleep-related neurophysiological processes.
The purpose of this study was to integrate a bibliometric analysis of global neurofeedback research with a meta-analysis of randomized controlled trials to quantify the clinical effects of neurofeedback for insomnia. Neurofeedback is a brain-based training technique in which individuals receive real-time feedback on neural activity and learn to modulate targeted patterns through repeated practice. In the context of insomnia, neurofeedback has been proposed as a potential intervention for addressing cortical hyperarousal and maladaptive neurophysiological regulation by reinforcing sleep-conducive rhythms (e.g., sensorimotor rhythm or alpha activity) and/or downregulating excessive high-frequency activation. Despite growing clinical and academic attention, the field has faced two persistent challenges: (1) an incomplete understanding of how insomnia-related research is positioned within the broader neurofeedback literature and (2) limited synthesis of randomized evidence regarding the magnitude and consistency of treatment effects, particularly across subjective and objective sleep outcomes. This study addresses these gaps through a mixed-methods design that integrates macro-level mapping of the research landscape with a quantitative evaluation of clinical trial outcomes.
A bibliometric analysis was conducted using records retrieved from PubMed and Web of Science Core Collection from January 2010 to October 2025. A total of 2,090 records were identified (PubMed n=450; Web of Science n=1,640). After de-duplication, 1,838 unique publications were included for analysis. To examine sleep- and insomnia-focused research in greater depth, a three-stage filtering procedure was applied to extract a sleep-focused subcorpus of 487 articles, representing 26.5% of the full corpus. Bibliometric indicators were used to evaluate publication growth, country and institutional productivity, and international collaboration patterns. Thematic structures were examined using keyword co-occurrence analysis and Latent Dirichlet Allocation topic modeling to identify major research clusters and to characterize the consolidation and differentiation of insomnia-related themes within the neurofeedback knowledge structure.
In parallel, a systematic review and meta-analysis of randomized controlled trials (RCTs) was performed in accordance with PRISMA 2020 guidelines. RCTs evaluating neurofeedback interventions among adults with insomnia were included. Eight trials met eligibility criteria (total N=320; intervention n=179; control n=141), encompassing diverse protocol types, including sensorimotor rhythm, alpha, infra-low frequency, and real-time functional magnetic resonance imaging-based neurofeedback. Risk of bias was assessed using the Cochrane Risk of Bias 2.0 tool. Outcomes were categorized into subjective sleep measures, assessed using the Insomnia Severity Index (ISI) and the Pittsburgh Sleep Quality Index (PSQI), and objective sleep duration, assessed using total sleep time (TST). Hedges’ g was calculated as the standardized effect size, and random-effects models were applied to account for between-study heterogeneity. Subgroup analyses examined moderation by control condition and protocol type. Exploratory meta-regressions evaluated dose-related indicators, including total number of sessions, session duration, weekly frequency, and cumulative training hours.
The bibliometric analysis demonstrated steady growth in neurofeedback research over the study period, with a compound annual growth rate of 14.21% and publication volume exceeding 200 per year by 2024. Country-level productivity indicated that the United States, Germany, and China were leading contributors, while South Korea ranked 12th globally and third in Asia, reflecting emerging yet expanding activity. Topic modeling identified eight major research clusters, with attention-deficit/hyperactivity disorder, sleep and insomnia, emotional regulation, and brain–computer interfaces representing prominent themes. Importantly, sleep and insomnia were characterized as an increasingly distinct research domain, showing both increasing publication volume and thematic consolidation in recent years.
Meta-analytic findings indicated significant improvements in subjective insomnia outcomes following neurofeedback. For the ISI, the pooled effect size was large (g = −1.18, 95% CI −1.70 to −0.65, p < .001). For the PSQI, a moderate effect was observed (g = −0.64, 95% CI −0.97 to −0.30, p < .001). In contrast, objective sleep duration did not significantly change (TST: g = .31, 95% CI −0.07 to 0.68, p = .107), suggesting a divergence between perceived symptom improvement and measurable changes in sleep duration. Subgroup analysis by control condition demonstrated larger effects relative to waitlist controls (g = −1.45) than to active controls (g = −0.85), with statistically significant between-group heterogeneity (Q = 4.89, p = .027). When compared specifically with sham neurofeedback controls, significant effects were maintained (g = −0.79, p = .001), suggesting that observed improvements may not be fully explained by nonspecific or placebo-related factors. Protocol-type differences were not statistically significant (Q = 3.12, p = .538), although interpretation is constrained by the limited number of trials within each protocol category. Meta-regressions did not identify significant dose–response relationships across total sessions (p = .327), session duration (p = .302), weekly frequency (p = .251), or cumulative training hours (p = .289). Sensitivity analyses supported robustness, and publication-bias assessments indicated limited evidence of systematic bias.
Several limitations warrant consideration. The number of eligible RCTs was modest, with relatively small sample sizes, which likely reduced statistical power for moderator analyses. Reliance on self-reported outcomes and challenges in blinding participants in neurofeedback trials may contribute to measurement and expectancy effects. Limited variability in training dosage parameters across studies may have constrained detection of dose–response patterns. Additionally, insufficient long-term follow-up data prevents firm conclusions regarding durability of effects.
In conclusion, this integrated synthesis indicates that insomnia has become a prominent and growing domain within neurofeedback research. RCT evidence suggests that neurofeedback is associated with meaningful improvements in subjective insomnia symptoms, including effects persisting against sham controls, while objective sleep duration remains unchanged. These findings support the view that neurofeedback may be positioned as a complementary, non-pharmacological intervention for perceived sleep disturbance. Future research should prioritize well-powered, multi-center trials with standardized protocols and reporting, refined sham designs, objective sleep assessments (e.g., polysomnography or high-resolution actigraphy), cost-effectiveness evaluations, and long-term follow-up across diverse clinical and cultural contexts, including Korea.