One Workout Raised Brain Blood Flow—What It Does Not Prove

A 60-minute aerobic workout raised blood flow in brain regions involved in memory and movement the following morning in adults at metabolic risk. It also changed how those regions responded to intranasal insulin. That is an interesting human result, not evidence that one workout protects the brain from dementia.

The study was small, with 15 participants, and its most intriguing finding has two plausible explanations that the design cannot separate. Exercise may have increased the brain's insulin-related vascular responsiveness. Or it may simply have raised blood flow enough that intranasal insulin had less room to raise it further. A companion pilot adds another wrinkle: higher exercise lactate tracked with lower post-exercise cerebral blood flow, while a larger BDNF response tracked with higher flow.

The data describe an acute, region-specific physiological response. They do not yet tell us which workout prescription improves cognition over years.

What the crossover study found

Malin et al. (2025) tested 15 overweight, prediabetic, middle-aged adults in a counterbalanced crossover design. Participants averaged 56.7 years old, had a mean BMI of 31.8 kg/m² and HbA1c of 5.8%, and had low cardiorespiratory fitness for their age, with a mean VO₂max of 23.9 mL/kg/min.

On one condition, they completed 60 minutes of treadmill exercise at 70% of VO₂max. The next morning, while fasting, researchers measured cerebral blood flow with arterial spin labeling MRI. They then gave 40 IU intranasal insulin and measured the regional blood-flow response.

Compared with the control condition, the exercise condition produced statistically significant increases in fasting cerebral blood flow in the hippocampus, putamen, and pallidum (condition effects, P<0.05). The publication does not provide confidence intervals or effect sizes in the available report, so the size and precision of those increases remain unclear.

Those regions matter, but the endpoint is still a surrogate. The hippocampus contributes to memory, while the putamen and pallidum are parts of the basal ganglia with roles in movement and other functions. Higher perfusion the next morning does not establish improved memory, healthier neurons, or lower dementia risk.

Why the insulin finding is intriguing—and unresolved

After exercise, participants with higher baseline blood flow showed a smaller subsequent cerebral blood-flow response to intranasal insulin in the hippocampus, caudate, pallidum, and putamen. The association was negative. It did not show that exercise caused greater neuronal insulin sensitivity.

Intranasal insulin is a reasonable research probe because it can alter regional cerebral blood flow and cognitive performance in randomized placebo-controlled studies, as summarized in a 2023 systematic review by Nijssen et al. A 24-week phase 2 trial by Novak et al. also found measurable cognitive and gait effects from 40 IU intranasal insulin in older adults with and without type 2 diabetes.

Still, the MRI response used here is indirect. It may reflect insulin-related vascular dilation rather than insulin signaling inside neurons. And the crossover result supports two competing explanations. Exercise might have primed an insulin-responsive pathway, leaving less additional response to the insulin challenge. It might also have elevated blood flow toward a local ceiling, with no implication that insulin sensitivity changed. The experiment cannot choose between those explanations.

The finding is best read as evidence that one aerobic session altered a brain vascular response to insulin in this specific population—not as proof that exercise repaired brain insulin resistance.

Lactate and BDNF point in opposite directions

Morris et al. (2024), another small crossover study, examined why cerebral blood-flow responses differed among people. Exercise-induced lactate area under the curve predicted lower post-exercise cerebral blood flow measured by arterial spin labeling. In the same study, a larger BDNF response predicted higher blood flow.

Neither association establishes a cause. People with lower fitness can accumulate more lactate at a given absolute workload and may also have lower baseline cerebral perfusion. That fitness difference could produce the lactate-flow association without lactate directly reducing brain blood flow. Human studies have not established a causal lactate-to-cerebral-perfusion pathway during exercise.

The BDNF result needs equal restraint. Acute exercise does raise peripheral BDNF on average in older adults, according to a 2022 meta-analysis by Fleitas et al. and a systematic review and meta-analysis by Titus et al. Both also found substantial variation among studies. Morris et al. found that the BDNF rise tracked strongly with platelet activation, which points to platelet degranulation as the proximate source of much of the measured serum signal.

That distinction matters. Serum BDNF after exercise is not a direct measurement of BDNF production in the brain. No human mechanistic study has shown that platelet-derived peripheral BDNF crosses the blood-brain barrier at physiologically meaningful concentrations. A favorable blood biomarker can be real without acting as a delivery receipt from the brain.

Intensity may change the biomarker response. In late-middle-aged and older adults, Tsai et al. found that acute high-intensity interval training produced larger BDNF responses and better neurocognitive performance than moderate-intensity continuous exercise. That does not prove HIIT produces more cerebral blood flow, nor does it establish a universal best protocol. A review by Hortobágyi et al. concludes that intensity effects on neuroplasticity markers depend on the population and outcome measured.

The long-term cognitive test was less encouraging

The strongest check on acute-mechanism enthusiasm comes from EXERT, a phase 3 multicenter randomized trial and the largest aerobic-exercise trial in mild cognitive impairment. Baker et al. found that aerobic exercise did not slow cognitive decline relative to stretching controls.

EXERT does not refute Malin's next-morning blood-flow finding. The populations, intervention length, comparator, and endpoints differ. Malin studied a single exercise bout in overweight, prediabetic adults with low fitness; EXERT studied adults with mild cognitive impairment over a longer intervention and assessed cognitive decline against stretching. But EXERT breaks the easy story that repeated acute vascular improvements must add up to cognitive preservation.

Other longer-term studies also resist a simple biomarker prescription. A six-month cluster randomized trial of dual-task, power-based exercise in older adults found uncertain combined effects on cognition and related neurological and inflammatory markers, reported by Tait et al. A separate randomized trial in older women with mild cognitive impairment found improvements in cognition and BDNF with functional training, but its population and chronic training program differ substantially from the acute aerobic studies.

What to do with this evidence

For an overweight, prediabetic, low-fit middle-aged adult, one 60-minute session at about 70% VO₂max raised next-morning blood flow in specific brain regions in a statistically significant 15-person crossover study. That is the direct finding. Exercise also improves glycemic control and insulin resistance in prediabetes across modalities, as shown in a network meta-analysis by Huang et al.

Do not use post-workout lactate or serum BDNF as a personal brain-benefit score. The lactate result is correlational, serum BDNF largely reflects platelet release, and both pilot studies need replication. Evening exercise followed by next-morning imaging also leaves circadian effects difficult to separate from exercise effects.

Exercise remains worth doing for established cardiometabolic reasons. This particular evidence adds a credible acute cerebral-perfusion signal in a high-risk group. It does not yet supply a proven neuroprotective protocol, a lactate target, or a shortcut around the long and less glamorous work of improving fitness over time.