Red light and your brain: a search of 81 studies

The short answer

Shining light at your head to help your memory sounds like something out of an advertisement. It is also one of the best studied uses of red and infrared light there is, and the results point consistently one way.

That is not the same as "it works". Where the evidence is strong and where it is thin differ sharply, and one important caveat that held a year ago has since been overtaken.

Does red light on your head improve memory?

Three independent, pre-registered meta-analyses land in the same place.

The first found that light therapy significantly improved global cognitive function in people with dementia, at SMD 0.63, with near-infrared LEDs and lasers outperforming ordinary visible light ^[4]. A second, on age-related decline, found a pooled effect across eleven randomised trials of SMD 0.51 ^[7]. A systematic review of 35 human studies on light to the head specifically found that 82.9 percent reported improvement, in healthy adults as well as in people with impairment ^[6].

Then a network meta-analysis set all the non-drug options for dementia side by side: 38 studies, 3,412 participants. Photobiomodulation came first, ahead of exercise therapy, cognitive training and enriched environments ^[5].

Does it work at home or only in a clinic?

Until recently this article carried an important qualification, and it is worth leaving on the page because it shows how fast this field moves.

The meta-analysis on age-related decline split its results by setting. Treatment in a clinic was significant. Home devices were not ^[7]. The direction was the same, the certainty was not, and that is exactly the distinction that matters if you are considering buying one.

That has now been overtaken.

That is a real result with a real placebo group, in precisely the setting where the evidence had been missing.

Is it better than brain stimulation?

2026 brought the first head-to-head comparison with the established forms of non-invasive brain stimulation, and it deserves a careful reading.

That distinction is worth understanding, because it recurs constantly in this kind of analysis. Ranking first is not the same as being demonstrably better. The first is a probability calculation, the second is a test. Read only the ranking and you read more certainty than is there.

How much red light, and how often?

One finding runs against intuition and is useful for anybody buying a device.

Tissue rich in mitochondria, such as brain, muscle, heart and nerve, responds to lower doses of light than tissue with fewer of them, such as skin, tendon or cartilage. And where mitochondria-rich tissue showed no benefit, that was more often because the dose was too high than too low ^[10].

More power is therefore not automatically better in this field, and for your head it is probably the opposite.

The same meta-analysis that separated clinic from home use also found that protocols using multiple wavelengths were significant while single-wavelength protocols were not ^[7]. That suggests the combination matters more than any single "best" number.

What we think

Of all the uses of red light we have reviewed, this is the one with the most coherent evidence. Several independent meta-analyses point the same way, the effect sizes are not trivial, and there is now a placebo-controlled trial in the setting where it was previously missing.

What is not there: replication by an independent group, follow-up longer than twelve weeks, and any evidence that it does something worthwhile in healthy people without complaints. The trials are about people with a diagnosis.

If you are considering this for yourself without memory problems, you are buying something whose evidence is about somebody else.