It is difficult to draw conclusions about the effect of resistance exercises on information processing speed and inhibitory control from previous studies due to possible underestimations of maximal strength and the lack of information on the intervention programs. To address this issue, a familiarization of resistance exercise was introduced before the strength test, and the repetition-to-fatigue method was used to calculate the 1RM (one repetition max). A two-arm RCT was conducted to evaluate the cognitive effect of resistance exercise. Male adults aged 50-65 years old performed a single bout of multiple joint, structural barbell resistance exercises (back squat, press, and deadlift) with 75% 1RM * 5 repetitions * 3 sets with 2-3 min rest between sets and exercises or a stretching exercise session (active-control intervention). This type of resistance exercise improved the information processing speed measured by Stroop task reaction time (t(23) = - 2.313, p = .030, M = - 16 ms, 95% CI [- 30, - 2]) and decreased the conflict-related neural activity measured by event-related potential N2b in both congruent (t(20) = 2.674, p = .015, M = 2.290 μv, 95% CI [0.504, 4.075]) and incongruent (t(20) = 2.851, p = .018, M = 2.291 μv, 95% CI [0.439, 4.142]) conditions. Resistance exercise significantly improved information processing speed and decrease conflict-related neural activity, but did not change inhibitory control in older adults compared to active control.Trial registration: NCT04534374 (01/09/2020).