So, Does Exercise Make Healthy Adults Smarter?

The most honest answer to the title is a calibrated yes—and a clear no, depending on what “smarter” is meant to describe. Repeated exercise can improve some measurable cognitive abilities in adults without diagnosed major cognitive impairment. The evidence is most useful when it is read by domain: broad cognition, executive control, working memory, processing speed, and episodic memory do not show identical patterns. But the two reviews do not prove a universal rise in intelligence, a guaranteed memory improvement, or one exercise program that is best for everyone.

This final synthesis connects four ideas that have to stay together. Exercise is a repeated intervention, not a single feeling after a workout. Cognition is a collection of partly distinct abilities, not one score. Modality rankings are conditional comparisons, not a universal league table. And a group-level average becomes practically meaningful only when the intervention is feasible, safe, and repeatable for the person considering it.

flowchart TB E["Repeated exercise in non-impaired adults"] --> D["Cognitive goal"] D --> G["Global cognition"] D --> X["Executive function"] D --> W["Working memory or speed"] D --> M["Episodic memory"] G --> MC["Multicomponent: leading hypothesis"] X --> AE["Aerobic: leading hypothesis"] W --> RE["Resistance: domain evidence"] M --> U["No stable pooled winner"] MC --> F["Filter by certainty, safety, access, and adherence"] AE --> F RE --> F U --> F F --> J["Some domain-specific benefits are supported"] J --> B["Not a universal intelligence or dementia-prevention claim"]

A calibrated yes: exercise can improve measured cognition

The broad sedentary-adult review found a positive average effect for global cognition: a standardized mean difference of 0.57 across nine studies, with a confidence interval from 0.11 to 1.03. The resistance-focused review also found a positive global-cognition effect, Hedges’ g = 0.76, although its heterogeneity was substantial. These results support the statement that repeated exercise can improve broad cognitive performance relative to the comparison conditions used in the studies.

That statement is meaningful, but it is narrower than “exercise makes people smarter.” Global cognition is usually a composite or screening estimate. It can summarize performance across several abilities, but it can also conceal a split profile. A person might improve in executive control and processing speed while showing little change in delayed recall. A composite may register an average improvement without revealing that pattern. The result supports a change in measured performance, not a direct measurement of general intelligence.

Executive function provides another relatively consistent signal. The broad review found a small positive average effect, and the resistance-focused review found a similar positive effect for resistance exercise. The tasks represented inhibition, flexibility, and problem solving rather than simple recall. The broad review also found a stronger executive-function effect in studies with mean age below 68 than in studies with mean age at least 68, but that age split was based on the median age of the included studies and is not a biological cutoff. It is a clue about possible moderation, not a rule that exercise stops helping after a particular birthday.

The resistance review adds more specific support for working memory and processing speed. Its pooled effect was g = 0.37 for working memory and g = 0.27 for processing speed, with relatively low heterogeneity in both analyses. That makes resistance exercise a substantive part of the cognitive picture rather than a consolation prize for people who do not choose aerobic training. Yet the result remains domain-specific. Better performance on Digit Span, Letter–Number Sequencing, reaction-time, or psychomotor-speed measures does not automatically imply better judgment, broader knowledge, or higher intelligence in every setting.

So the positive answer has a clear shape.