Anterior temporal lobectomy (ATL) is the standard surgical treatment for drug-resistant mesial temporal lobe epilepsy (MTLE); however, it is associated with cognitive decline and visual field deficits. In the current era of personalized medicine and quality-of-life-focused care, minimally invasive surgical strategies are increasingly needed. Optimized stereoelectroencephalography (SEEG)-guided radiofrequency thermocoagulation (RF-TC) is a promising option for preserving cognitive and visual function while providing effective seizure control.Methods: This assessor-blinded randomized, controlled pilot trial enrolled patients with drug-resistant MTLE at Xuanwu Hospital between August 2019 and October 2022. Participants were randomly allocated (1:1) to the ATL or SEEG-guided RF-TC groups. The primary outcome was the change in cognitive function at the 1-year follow-up, assessed using the Chinese version of the Wechsler Adult Intelligence Scale-IV (WAIS-IV-C). Secondary outcomes included the Wechsler Memory Scale-IV (WMS-IV-C), seizure control (Engel classification), visual fields (Humphrey perimetry), quality of life (QOLIE-89), and adverse events.Thirty-eight patients (19 male [50.0 %]; mean [standard deviation] age, 28.3 [7.1] years) were enrolled, with 20 and 18 patients assigned to the ATL and RF-TC groups, respectively. At 1 year, the overall WAIS-IV-C scores did not differ between the ATL and RF-TC groups (P > 0.050). However, patients who underwent dominant-side RF-TC demonstrated superior preservation in the full-scale intelligence quotient (96.91 ± 15.83 vs. 113.44 ± 17.50, P = 0.040), cognitive productivity index (97.36 ± 15.64 vs. 116.89 ± 16.54, P = 0.014), working memory index (97.73 ± 12.89 vs. 115.56 ± 18.80, P = 0.022) and processing speed index (97.91 ± 15.44 vs. 115.00 ± 18.41, P = 0.036) compared to the ATL group. Visual field defects occurred more frequently in the ATL group (15/20, 75.0 %) than in the RF-TC group (3/18, 16.7 %; P < 0.010). The 1-year seizure freedom (Engel I) was comparable (ATL vs. RF-TC, 85.0 % vs. 72.2 %, P = 0.440; Engel IA: 65.0 % vs. 33.3 %, P = 0.103). The RF-TC group experienced fewer adverse events than the ATL group.Optimized SEEG-guided RF-TC preserved cognitive and visual function better in dominant-side MTLE than in ATL, with comparable seizure control and quality of life outcomes. Further multicenter studies with larger sample sizes and extended follow-up are needed to validate its efficacy and long-term safety.Clinicaltrials.gov Identifiers: NCT03941613.Copyright © 2025 The Author(s). Published by Elsevier B.V. All rights reserved.