INTRODUCTION: Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) have been widely studied for cognitive enhancement. However, their clinical efficacy remains inconclusive in patients with Alzheimer's disease (AD), Parkinson's disease (PD), and type 2 diabetes mellitus (T2DM). We therefore conducted a domain-specific meta-analysis to evaluate their cognitive effects. METHODS: The PubMed, Web of Science, and Cochrane Library were searched for studies published up to January 2026. Eligible randomized controlled trials comparing GLP‑1 RAs to control groups, with baseline and post‑intervention cognitive scores, were included. Standardized mean differences (SMDs) with 95% confidence intervals (CIs) were pooled using a random‑effects model across six cognitive domains. RESULTS: Eleven trials met the inclusion criteria. GLP-1 RAs demonstrated no significant benefit across all cognitive domains: pooled SMD for overall cognition was 0.03 (95% CI, -0.11 to 0.18; P =  0.64); memory, 0.21 (95% CI, -0.19 to 0.60; P = 0.30); attention, 0.23 (95% CI, -0.19 to 0.65; P =  0.28); executive function, 0.14 (95% CI, -0.28 to 0.56; P =  0.51); visuospatial ability, 0.04 (95% CI, -0.43 to 0.50; P =  0.88); and motor function, 0.20 (95% CI, -0.19 to 0.59; P = 0.31). Subgroup analyses by disease type and drug class also revealed no statistically significant differences. DISCUSSION: Our findings suggest that the cognitive benefits of GLP‑1 RAs may be limited in patients with established disease. This may be attributable to irreversible neuropathological alterations in AD, PD, and T2DM patients, which hinder the translation of GLP‑1 RA-mediated neuroprotective actions into measurable symptomatic improvements. A key limitation of the present study is the relatively small sample size for several specific cognitive domains, which may affect the robustness of the results. CONCLUSIONS: The available evidence from this meta-analysis does not support significant benefits across six cognitive domains with GLP‑1 RAs in patients with AD, PD, and T2DM. Future largescale trials with domain-specific assessments are needed to evaluate their potential in earlier disease stages or specific subgroups.