The featured image illustrates a resistance-training class; it does not show participants in the studies discussed here.
Resistance-training studies have reported both positive and null cognitive findings. Keep the tested task and comparison beside each result.
This guide compares trial outcomes, review summaries, imaging findings and mortality research. Use the reading record below to keep those questions separate.
What the 155-woman trial actually tested
The 2010 Liu-Ambrose trial randomized 155 independently living women aged 65 to 75 to yearlong weekly lifting, twice-weekly lifting or twice-weekly balance and tone classes.
Its primary Stroop task assessed selective attention and conflict resolution using conflicting color words.
| Outcome | Trial finding | Interpretation limit |
|---|---|---|
| Stroop task | Both lifting groups better than balance and tone | Task-specific |
| Working memory and set shifting | No significant between-group difference | Keep null findings visible |
| Whole-brain volume | Reduced versus control | Not larger brains |
| Dementia or brain age | Not measured | No prevention estimate |
Stroop performance improved 12.6% weekly and 10.9% twice-weekly; controls deteriorated 0.5%. These are task-performance changes, not dementia-risk reductions.
Comparisons with balance and tone do not establish weekly versus twice-weekly lifting superiority.
Why newer reviews give different answers
In the review summaries below, list each named test or domain separately. Mark which comparison produced the result.
A 2025 resistance-exercise review reported 17 randomized trials and 739 older participants. It found benefits for overall cognition and working memory, but no significant pooled effect for processing speed, executive function or attention.
Its working-memory results showed possible publication bias in the full text, despite reassuring language in the abstract. Spatial memory span and verbal learning each relied on one study, so those findings should not be described as many independent replications.
A 2026 executive-function review identified 22 eligible trials and included 17 in its meta-analysis. It reported improvements in working memory and cognitive flexibility compared with control conditions, but no significant inhibitory-control benefit.
When that review compared resistance training with other exercise, it found no significant differences in those three domains. Its authors also reported low certainty, varied programs and substantial risk of bias.
These are different reviews with different eligibility rules, tests and comparison groups. Their results support a cautious domain-specific answer, rather than adding their trial counts together as if every study and participant were unique.
The comparison group can change the headline
A lifting program compared with no exercise answers a different question from lifting compared with another organized activity. The second comparison helps ask whether the resistance component adds something beyond exercise participation.
Likewise, improvement from a group’s own starting score is not sufficient to establish an advantage over a comparison group. Practice with a test, expectations and ordinary variation are reasons to inspect the between-group result.
| Comparison | Question it helps answer | What it cannot settle alone |
|---|---|---|
| Before versus after in lifters | Did that group’s measured score change? | Whether lifting caused a greater change than the control condition |
| Lifting versus inactive control | Did the lifting program outperform little or no added activity? | Whether it is better than walking, balance work or another exercise program |
| Lifting versus active exercise | Was there an additional advantage over the particular alternative? | Whether a nonsignificant difference proves the programs identical |
The 2025 bodyweight-training review makes this distinction especially relevant. It included nine trials, could not pool their diverse assessments and concluded that evidence was insufficient for definitive recommendations.
Its 2010 row labels balance and tone as bodyweight training and loaded lifting as comparison arms.
This compares different complete programs, not an isolated exercise or otherwise identical loaded and unloaded routines.
Brain scans, memory scores and dementia are different outcomes
A scan can measure a structural or tissue-related marker; a thinking test measures performance on a task. Neither automatically establishes how often people will develop a diagnosed illness years later.
A 2025 study of 44 adults with mild cognitive impairment illustrates the difference. It followed a training group and a control group for 24 weeks, assessing memory and MRI measures.
This was not a randomized trial. The primary paper says participants interested in training entered that group, while people without interest or availability formed the control group, leaving possible selection differences.
The findings included better verbal episodic-memory performance and differences in selected imaging measures. The training group had no reduction in the right hippocampus and precuneus, but did show reductions in those regions on the left.
These findings concern adults with mild cognitive impairment; they do not establish dementia prevention.
A 2026 study of 147,374 adults examined a different endpoint: deaths during up to 30 years of follow-up. Resistance-training habits were associated with lower neurological-disease mortality, after adjustment for aerobic activity and other measured factors.
Mortality from a category of diseases is not the same as developing dementia or improving a memory-test score. An observational association also cannot establish that assigning a particular lifting schedule would cause that reduction for an individual.
A five-question check for a brain-health claim
Before turning a headline into a training decision, complete the short record below. This is a way to read evidence, not a cognitive screening test or a treatment plan.
| Write down | What to record | Question to avoid skipping |
|---|---|---|
| Population | Eligibility, age and relevant diagnosis | Who does the claim concern? |
| Design | How groups were formed | Was allocation randomized or chosen? |
| Comparator | What the other group actually did | Better than what? |
| Endpoint and time | The named measurement and follow-up period | Is the headline substituting a different outcome? |
| Uncertainty | Null findings, limitations and missing information | What would change the conclusion? |
For example, “memory improved after training” leaves several blanks in the record. A useful follow-up is whether the improvement exceeded the control group’s change and whether memory was the outcome the study was designed to test.
Also distinguish a statistically significant result from a change that matters in daily life. A standardized effect size is not a percentage improvement in someone’s memory or a number of years removed from their age.
How often should older adults strength train?
Before copying a research schedule, ask whether your goal is general activity or a specific clinical concern. Keep the guidance below separate from the study comparisons.
CDC guidance for adults 65 and older includes muscle-strengthening activity on at least two days each week, aerobic activity and balance work. The strength component should involve all major muscle groups.
These are general health recommendations, not a prescription for a cognitive-test gain. CDC also advises people who cannot meet the recommendations to be as active as their abilities and conditions allow.
For a practical next step, discuss an accessible strength program with a qualified instructor, and involve your health care professional when medical conditions, symptoms or functional limitations affect training. Ask how exercises, support and progression will be adapted to you rather than copying a study’s laboratory load.
A useful training record can track attendance, exercises and tolerability. It cannot establish that a program is preventing dementia, and an online memory score should not be used to adjust a medical treatment.
When memory concerns need medical attention
The National Institute on Aging advises talking with a doctor about noticeable memory changes. Examples include repeatedly asking the same questions, getting lost in familiar places or having difficulty following directions.
Memory difficulties can have different causes, including medication effects, sleep problems and other health conditions. Do not wait for a training block to finish before seeking an assessment, and do not diagnose the cause from a workout response.
What to take away from strength training and brain health
Read a strength-training headline alongside its population, comparison, endpoint and uncertainty. The same record can organize questions for an instructor or health care professional.
For each claim, write the result and its limits in adjacent columns. Leave an unanswered field blank instead of converting it into a younger-brain or dementia-prevention promise.
Sources and further reading
- Liu-Ambrose and colleagues, 2010: Resistance training and executive functions: a 12-month randomized controlled trial.
- Wu and Huang, 2025: Resistance exercise and cognitive function in older adults.
- Jorquera-Cáceres and colleagues, 2026: Resistance training and executive functions in community-dwelling older adults.
- Levín Catrilao and colleagues, 2025: Body weight strength training, cognition and quality of life.
- Ribeiro and colleagues, 2025: Resistance training and imaging in adults with mild cognitive impairment.
- Zhang and colleagues, 2026: Long-term resistance training and cause-specific mortality.
- CDC: Older adult activity recommendations.
- National Institute on Aging: Memory problems, forgetfulness and aging.
Medical disclaimer: This article is general fitness and health information, not medical advice. Strength training is not a substitute for assessment or treatment of memory problems; if you have an injury, ongoing pain or a medical condition, talk to a doctor or physical therapist before changing how you train.
