Cardio sharpens specific mental skills in early cognitive decline, not overall scores

Mild cognitive impairment is the grey zone where memory and thinking have measurably slipped but daily life still functions, and it is the stage at which intervention feels most worthwhile. Exercise is repeatedly recommended for people in this position, yet exercise covers everything from brisk walking to stretching and strength work, and these are not interchangeable. Whether the aerobic kind specifically, the sort that raises heart rate and builds cardiorespiratory fitness, is what helps thinking has remained unclear. This review pooled trials that put aerobic and non-aerobic exercise head to head in the same population to find out.

The authors searched four databases, PubMed, Web of Science, MEDLINE and the Cochrane Library, for trials published between January 2000 and July 2025 in which people with mild cognitive impairment were randomly assigned to aerobic or non-aerobic exercise. Fifteen trials met the criteria, together covering 1,192 older adults. The pooled comparisons looked at peak oxygen uptake as a measure of cardiorespiratory fitness, the trail making test part B as a measure of cognitive flexibility, delayed recall as a measure of memory, and two broad screening tools, the Montreal cognitive assessment and the mini-mental state examination.

Aerobic exercise came out ahead on the specific measures. Peak oxygen uptake was 3.99 ml/kg/min higher than in the non-aerobic groups, the trail making test part B was completed 9.57 seconds faster, and delayed recall improved by 0.90 points. On the two broad cognition screens, the MoCA and the MMSE, there was no significant difference between groups. In other words, the aerobic advantage showed up in fitness, mental flexibility and memory retrieval, but not in overall cognitive scores.

Exercise in general has long been linked to better cognitive outcomes in older adults, but head-to-head evidence on which type works best in mild cognitive impairment has been thin. This pooled analysis suggests the aerobic component adds something specific, sharpening executive speed and delayed memory beyond what non-aerobic training achieves, while global screening tools stay flat. That split could mean the benefits are genuinely domain-specific, or that broad instruments like the MoCA and MMSE are too blunt to detect change over trial timescales. Which non-aerobic comparators work best, and whether combining both types beats either alone, remain open questions.

If you are exercising with your future brain in mind, or supporting a parent who is, this is a reason to make sure some of that movement genuinely raises the heart rate rather than being purely strength or flexibility work. The gains here were on specific abilities, planning speed and recalling information after a delay, rather than a wholesale lift in cognition, so expectations should be sized accordingly. It also cuts the other way: non-aerobic exercise was the comparison here, not a failure, and nothing in this analysis argues for abandoning it.