Introduction: Cognitive-motor training has been shown to prevent and treat cognitive decline; however, findings on its effects on postural control are not consistent. The objective of this study was to compare the effects of two multicomponent cognitive-motor training programs on postural control in independent and autonomous community-dwelling older adults. Methods: This was a randomized and controlled longitudinal single-blind clinical trial with two parallel groups of independent and autonomous community-dwelling adults, 60 years of age or older. The study was performed in the University of São Paulo. The blocks were distributed in two groups through a random process (4: 4) with blocks of 8 participants. The generated numbers were placed in opaque sealed envelopes that were opened by the physical therapist who applied the interventions. The multicomponent group (MT) underwent multicomponent training with task-oriented exercise. The multicomponent plus group (MT+) underwent the same training as the MT with the addition of a cognitive overload (memorization, arithmetic calculation and Stroop task). All participants completed fourteen 50-minute sessions, which were held twice weekly over seven weeks. The primary measurement was postural control, evaluated using the Mini-BESTest before and after the intervention and 30 days after the training program was completed. The study was registered in the Brazilian Clinical Trials Registry under study number RBR-99xnc6. Results: Between September 2017 and December 2018, 32 participants participated in the study. In total, three participants gave up. Twenty-nine participants were allocated in the MT (n = 14) or MT + (n = 15). There was a time effect and Total Mini-BESTest score (F (2; 54) = 20.8; p = 0.001; observed power = 1.0. Post-intervention improvement (Tukey post-hoc test; p = 0.0001), which remained in the follow-up assessment (Tukey's post-hoc test; p = 0.0002), regardless of the group. found time effect on the anticipatory domain (F (2.54) = 6.9, p = 0.002; observed power = 1.0); sensory orientation (F (2.54) = 8.2; p = 0.001; observed power = 1.0) and dynamic gait (F (2.54) = 9.5; p = 0.001; observed power = 1.0) There was an improvement in post-intervention evaluation (Tukey's post-hoc test, p <0.05) that remained in the follow-up evaluation (Tukey's post-hoc test; p <0.05), regardless of the group. The simple task timed up and go analysis of TUG showed a time effect (F (2.54) = 4.8 p = 0.011, observed power = 0.8) and group (F (1.27) = 5.2 p = 0.03, observed power = 0.6). Tukey's post-hoc test on the simple task TUG showed a difference between groups p = 0.03 and time effect p = 0.008 (baseline vs follow-up) regardless of the group. TUG analysis in cognitive task showed time effect (F (2.54) = 3.9; p = 0.026; observed power = 0.7). Tukey's post-hoc dual-task TUG test showed time effect p = 0.03 (baseline vs follow-up) regardless of group. Conclusion: Multicomponent training provided an improvement in postural control in independent elderly, and the addition of ...