Quick Answer: Exercise changes the brain through several distinct mechanisms: increased blood flow, release of growth factors such as BDNF, reduced inflammation, and improved sleep quality. Aerobic exercise has the strongest evidence for memory and executive function, resistance training has growing support particularly for older adults, and a single 20 to 30 minute moderate session produces measurable cognitive benefits for roughly two hours afterward. The long-term effects require consistency across years.

The claim that exercise is good for your brain has become so common it’s stopped carrying information. What’s more useful is the specifics: which mechanisms, which types of exercise, what dose, how long the effects last, and what the evidence actually supports versus what gets overclaimed.

This is the long version.

This is general information, not medical advice. Persistent cognitive changes - memory problems that concern you or those around you, or a noticeable change in function - warrant medical assessment rather than a training program.

What actually happens in the brain during exercise?

Four mechanisms, operating on different timescales, and they explain most of the observed effects.

1. Blood flow, immediately. Exercise increases cerebral blood flow, and moderate aerobic activity in particular increases perfusion to regions including the hippocampus. More blood means more oxygen and glucose delivered to working tissue. This mechanism operates within minutes and is largely responsible for the acute post-exercise cognitive boost most people can feel directly.

2. BDNF and other growth factors, within hours. Brain-derived neurotrophic factor supports the survival of existing neurons and the growth of new connections, and it’s often described as fertiliser for the brain. Exercise reliably increases circulating BDNF, with larger increases from higher intensity work. Related factors including IGF-1 and VEGF also rise and support blood vessel growth in the brain.

3. Reduced neuroinflammation, over weeks to months. Chronic low-grade inflammation is associated with cognitive decline. Regular exercise reduces systemic inflammatory markers and appears to shift immune activity within the brain toward a less inflammatory profile.

4. Sleep, indirectly and substantially. Exercise improves sleep quality and duration, and sleep is when the brain consolidates memory and clears metabolic waste through the glymphatic system. A meaningful share of exercise’s cognitive benefit likely runs through sleep, which is easy to overlook because it’s indirect.

A fifth worth noting: structural change over years. Aerobic training has been associated with increased hippocampal volume in older adults in several trials - notably reversing part of the age-related shrinkage that would otherwise occur. This is among the more striking findings in the field, and it takes months to years rather than weeks.

How long does the effect of a single session last?

Longer than people expect, and it’s the most practically actionable finding in the whole area.

The acute window: roughly 1 to 2 hours. After a single moderate-intensity aerobic session of about 20 to 30 minutes, executive function, attention, and processing speed improve measurably, and the effect persists for approximately one to two hours before returning toward baseline.

What improves most:

• Executive function - planning, task switching, inhibiting distractions. The largest and most consistent acute effect.
• Attention and processing speed
• Mood and perceived energy, which affect cognitive performance indirectly
• Memory encoding - material learned shortly after exercise tends to be retained better

What this means practically: schedule your hardest cognitive work in the two hours after exercise. A 25-minute brisk walk before the writing you’ve been avoiding is not a nice-to-have; it’s a measurable performance intervention, and it’s free.

On intensity: moderate intensity produces the most reliable acute cognitive benefit. Very high intensity can temporarily impair cognitive performance immediately afterward due to fatigue, with the benefit appearing later once recovery begins. If you’re exercising to think better in the next hour, moderate is the target - a pace where you can talk in sentences but not comfortably sing.

On timing relative to learning: exercising shortly before or shortly after studying appears to aid retention, with some evidence that a delay of a few hours after learning may be particularly effective for long-term retention. The practical version: move at some point around the learning, rather than sitting still all day.

Which type of exercise does what?

The types are not interchangeable, and the evidence differs meaningfully between them.

Aerobic exercise - the strongest evidence overall.

Walking, running, cycling, swimming. The best-supported modality for memory and executive function, and the one behind the hippocampal volume findings. Trials in older adults using moderate aerobic exercise several times weekly over six to twelve months have shown improvements in memory and in brain structure.

• Dose: 150 minutes a week of moderate intensity, in sessions of 30 minutes or more
• Best for: memory, hippocampal health, general cognitive function

Resistance training - growing and increasingly convincing evidence.

Once considered irrelevant to cognition, now supported by a solid body of work, particularly in older adults. Effects on executive function appear comparable to aerobic exercise in several trials, and resistance training has the additional benefit of preserving muscle mass, which independently correlates with cognitive outcomes in ageing.

• Dose: twice weekly, progressive, hitting major muscle groups
• Best for: executive function, and everything else that muscle mass protects

Coordination and skill-based activity - underrated.

Dance, martial arts, racket sports, climbing. These combine aerobic demand with motor learning, spatial navigation, and decision-making under time pressure, and dance in particular has shown promising results in older adults. The novelty and complexity appear to matter independently of the physical demand.

• Dose: once or twice weekly as part of the mix

Yoga and mind-body practice. Evidence is more modest and more mixed for cognition specifically, though the stress-reduction effects are real and stress reduction supports cognition indirectly.

The practical conclusion: combine them. The strongest weekly structure includes aerobic work, resistance training, and something that requires learning a skill.

What Nobody Tells You About Exercise and Cognition

Six things that change how you’d interpret the whole literature.

The effect sizes are real but modest. Exercise produces meaningful improvements in cognitive measures - it does not produce transformation. Headlines describing exercise as making you dramatically smarter overstate what the trials show. What the evidence supports is a genuine, worthwhile, cumulative benefit, particularly for maintaining function over decades.

Consistency matters far more than intensity. Three moderate sessions a week for ten years beats an intense six-month block followed by nothing. The structural and vascular adaptations reverse with detraining, which makes this the rare area where ‘good enough forever’ clearly beats ‘excellent for a while’.

Fitness level predicts cognitive outcomes better than exercise minutes. Cardiorespiratory fitness - your actual measured capacity - correlates more strongly with cognitive outcomes than self-reported activity does. This matters because it means the goal is to genuinely improve fitness, not to accumulate minutes of easy movement.

The biggest effects appear in people who start from the lowest baseline. Going from sedentary to moderately active produces a larger cognitive benefit than going from moderately active to very active. The first 90 minutes a week is worth more than the second 90.

Sleep may be the largest single mediator, and it’s rarely mentioned. If exercise improves your sleep, a substantial portion of the cognitive benefit likely flows through that. Which also means exercising in a way that damages your sleep - very late high-intensity sessions, for some people - can work against the goal.

Midlife is the highest-leverage period for long-term brain health. The associations between midlife cardiovascular fitness and later-life cognitive outcomes are among the most consistent in the literature. What you do at 45 appears to matter for what happens at 75, which is a difficult thing to act on and an important one.

What does an evidence-informed week look like?

A structure that covers all three modalities in about four and a half hours.

Monday - Aerobic, moderate, 40 minutes. Brisk walk, easy run, or cycle. Conversational pace but working. Schedule demanding cognitive work in the two hours after.

Tuesday - Resistance training, 40 minutes. Squat, hinge, push, pull, carry. Progressive - the weight goes up over months.

Wednesday - Skill-based, 45-60 minutes. Dance class, tennis, climbing, martial arts. The requirement is that you’re learning something, not just moving.

Thursday - Aerobic, including some intervals, 30 minutes. Warm up, then 6 x 1 minute hard with 90 seconds easy, then cool down. Higher intensity drives larger BDNF increases.

Friday - Resistance training, 40 minutes.

Weekend - One longer easy aerobic session, 60-90 minutes. A hike, a long walk, an easy ride. Outdoors if possible - there’s a reasonable literature on nature exposure and attention restoration that’s independent of the exercise itself.

Daily - walking. Not as training, just as movement. The dose-response for cognition appears to continue well past the minimum.

The supporting factors that matter as much as the training:

• Sleep: 7-9 hours. The single most important variable for cognitive function, and exercise helps you get it.
• Social interaction, which is independently associated with cognitive outcomes - part of why group exercise and team sports may outperform solo work
• Learning something genuinely new on an ongoing basis
• Managing cardiovascular risk factors - blood pressure, blood glucose, and lipids all affect brain vasculature directly

If you can only do one thing: 30 minutes of brisk walking, five days a week. It covers the aerobic base, it’s the dose most of the evidence is built on, and it’s the one people actually sustain.

FAQ: Exercise and the Brain, Answered

Does exercise prevent dementia?

The evidence supports reduced risk rather than prevention. Physical activity is consistently identified as one of the modifiable risk factors for dementia in large reviews, alongside hearing loss, hypertension, and others. Reduced risk across a population is not a guarantee for an individual, and it’s still one of the most substantial levers available.

Is walking enough or do I need to run?

Brisk walking is enough to hit the moderate-intensity threshold most of the evidence is built on, and it’s what a large share of the successful trials used. Higher intensity produces larger acute BDNF responses, but adherence over years matters more than intensity, and walking wins decisively on adherence.

Does exercising in the morning help more than the evening?

For the acute cognitive boost, exercise before the work that needs it - the timing that matters is relative to your cognitive demands, not the clock. For sleep, very intense late-evening exercise disrupts sleep for some people and not others; if you sleep fine after evening training, there’s no reason to change.

What about brain training apps compared with exercise?

Brain training apps largely improve performance on the trained tasks, with limited transfer to general cognitive function. The evidence for aerobic exercise affecting general cognition and brain structure is considerably stronger. If you have thirty minutes, walking is the better investment.

Does it help with ADHD, anxiety, or low mood?

Exercise has reasonable evidence for improvements in attention in ADHD, and a stronger body of evidence for depression and anxiety symptoms - in some trials comparable to other first-line interventions for mild to moderate depression. It complements rather than replaces treatment, and it’s worth discussing with whoever manages your care.

TL;DR:

• Four mechanisms: blood flow, growth factors like BDNF, reduced inflammation, and improved sleep
• A single moderate 20-30 minute session improves executive function for roughly one to two hours
• Schedule demanding cognitive work in that window - it’s the most actionable finding in the field
• Aerobic has the strongest evidence; resistance training is close behind; skill-based activity adds something distinct
• Effect sizes are real but modest, and consistency across years matters far more than intensity
• Midlife fitness predicts later-life cognition, and the first 90 minutes a week is worth the most
• Combine aerobic work, resistance training, and one activity that requires learning a genuinely new skill
• Sleep is probably the largest single mediator, so protect it rather than training at its expense

Exercise doesn’t make you smarter this afternoon so much as it protects what you have across decades - with a genuine, usable two-hour boost thrown in each time you do it.