BACKGROUND: This study aims to explore the efficacy of Relaxation Response Meditation Training (RRMT) on elderly individuals with different levels of vividness of visual imagery. METHODS: In this randomized controlled, double-blind, multi-center clinical trial, we recruited a total of 136 elderly individuals who were over 60 years with nonorganic sleep disorders to participate in a 4-week RRMT intervention from October 2020 to October 2022. The intervention occurred twice a week, totaling eight times. These individuals were divided into high and low groups based on the vividness of visual imagery, and then randomly assigned to either the control or intervention groups, as follows: low-visualizers intervention group (LI group); low-visualizers control group (LC group); high-visualizers intervention group (HI group); high-visualizers control group (HC group). Their social and psychological parameters were assessed before and after the intervention by the Pittsburgh Sleep Quality Index (PSQI), the Revised Piper's fatigue scale (RPFS), General well-being scale (GWB), and Satisfaction rating. The alpha waves of patients were also collected through electroencephalogram to assess their level of relaxation. RESULTS: Compared to the LI group, the HI group had a greater reduction rate in the PSQI score [25.2 % (18.8 % to 31.7 %), P < 0∙001], shorter sleep latency (P = 0.001), lower frequency of sleep medication (P < 0.001), lower PSQI scores (P < 0.001), and higher GWB scores (P < 0.001). There were significant differences in all indicators in the HI group vs. HC group and in the LI group vs. LC group. In the first five relaxation training sessions, there was no statistically significant difference in the proportion of α waves between the LI group and the LC group; however, from the sixth session onward, we observed a statistically significant difference (t = 2.86, P = 0.019)，while The HI group and HC group showing significant differences in the first relaxation training session (t = 4.464, P < 0.001). There was a statistically significant difference in subjective satisfaction between the intervention group and the control group (x2 = 49.605, P < 0.001). CONCLUSION: In this study, we found that most elderly people benefitted from RRMT regardless of their vividness of visual imagery. However, low-visualizers experienced slower and less effective results, so these patients may benefit more from alternative approaches.