You are at the kitchen counter with a softgel resting on your fingertip, the fish smell already on your thumb, and a label that prints EPA and DHA in milligrams you have nothing to compare them against. You bought it because someone said something about the brain. What you want to know is narrower than that: whether those two numbers mean different things for memory, and whether the bottle in your hand is the right one. So you turn the label over, read the fine print a second time, and get nothing. It says which fatty acids are in the capsule. It does not say which one carried the result people quote at you, how long anyone took it, or that most of the trials you have heard about were measuring something other than memory. That gap is the whole problem, so it is worth stating plainly before anything else. Omega-3 supplements have not been shown to improve memory in people whose memory is already declining. What the evidence does point at is narrower than the claim on the bottle: a set of reviews, a few populations, and one finding that keeps landing on the people who had not yet been diagnosed with anything.
Does this salmon do anything for memory?Which of the two numbers carried the resultWatch the video holding two labels side by side
DHA or EPA: which one does the memory work?
EPA (eicosapentaenoic acid) and DHA (docosahexaenoic acid) are the two omega-3s most of the research has focused on. Both occur in seafood, especially cold-water fish such as salmon, mackerel and tuna, and in shellfish and fish oil supplements. A third type, ALA (alpha-linolenic acid), sits in flaxseed, soybean and canola oils, chia seeds and walnuts, and it is not interchangeable with the other two for the questions people ask here. The supplement shelf divides on the same lines. Fish oil carries EPA and DHA. Krill oil carries omega-3s as phospholipids. Algal oil is a vegetarian source of DHA, and some algal products also carry EPA. Flaxseed oil carries ALA only, which is not what a memory shopper is usually looking for.
There is a memory-shaped answer here, and it belongs to a different condition. Reviews of omega-3s for depression have suggested that EPA may be more beneficial than DHA, while also warning that publication bias and differences among trials leave the question open and that the evidence quality is low to very low. Nothing of that kind has been established for memory. On memory, the split between the two fatty acids is not where the answer lives.
In the fifth month the bottle is half empty and you have moved it behind the vitamins, out of the line of sight. You still read the EPA and DHA numbers on any label you pick up, but you have stopped photographing them, and you quietly stopped asking your partner whether names come easier lately. The capsules go down with water because that is the routine, not because you are waiting on anything. What changes is the conversation: you answer fine, you ask about everyone else, and you file your own question somewhere it will not come up again.
What the trials actually measured
The table below is the honest version of the memory evidence, written as rows so the shape of it is visible at a glance. Healthy adults without a diagnosis of Alzheimer’s disease or dementia: a 2022 review of 33 studies, mixed dietary and supplemental intakes, a possible protective effect against cognitive decline was suggested. Adults already diagnosed with Alzheimer’s disease: the same review, no effect on cognition. Men and women aged 55 to 90 with mild to moderate Alzheimer’s disease: a 2018 review of 7 studies covering 795 participants, possibly helpful at the earliest stage of slight impairment, and not enough evidence to support supplementation as treatment. Carriers of the APOE4 gene with no signs of Alzheimer’s disease: a 2017 research review suggested DHA taken beforehand might matter. General population taking one gram a day of omega-3s: a 2019 review of three large cardiovascular trials, no memory outcome reported.
Read the rows together and the pattern is not about dose. It is about timing and about who was in the room. The finding that leans favorable sits with people who had not yet been diagnosed with anything, and the finding that leans neutral sits with people who already had a diagnosis and a clinician’s name for it. We keep coming back to that distinction, because it is the part the marketing sentence drops.
One more thing the rows show: the word you searched for, brain fog, does not appear in any of this. The material does not test that complaint, and good evidence on it is thin. The same goes for the language of repair. Nothing here shows omega-3 rebuilding or repairing brain tissue, and a supplement that made a claim like that would be going far past what any of these trials measured.
If you would rather see two labels held side by side than read another paragraph of it, we filmed the drawer, and you can watch the full video while your tea goes cold.
The column the table cannot fill
Every version of that table has a dose column, and this one stays empty. The reviews behind the memory outcomes do not report a single shared DHA or EPA dose that can be lifted out and repeated, so any number you see attached to a memory claim came from somewhere else. The one figure the material does give is one gram a day of omega-3 fatty acids, drawn from a 2019 review of the VITAL, ASCEND and REDUCE-IT trials, and those trials were about cardiovascular health, not memory. Likewise, a 2020 review of 23 studies with 43,998 participants found EPA and DHA reduce triglycerides by about 15 percent, which is a lipid result.
The point of that missing column is not that dose is irrelevant. It is that a dose quoted from a heart trial cannot be carried over to a memory question, and the moment you see a capsule advertised for memory next to a number from a cardiovascular study, you are looking at two different conversations.
Do you already eat fish?
This is the fork, and it decides which half of the evidence applies to you. On one side: you rarely eat seafood, your plate has no salmon, mackerel or tuna in a normal week, and the supplement is doing the work of a food you are not eating. On the other side: you already eat fish several times a week, so a capsule is an addition rather than a substitute. The Dietary Guidelines for Americans 2020-2025 recommend adults eat 8 ounces or more of a variety of seafood a week, including some choices higher in EPA and DHA.
Which side of the fork you are on
- If your plate rarely includes salmon, mackerel or tuna, the capsule works for a food you are not eating
- If you already eat fish several times a week, the capsule is an addition rather than a substitute
- If a memory claim sits beside a number from a heart trial, you are reading two different conversations
- If the change in memory or thinking persists or gets worse, that is worth a doctor’s time
For some health conditions, the evidence for benefits from seafood is stronger than the evidence for omega-3 supplements, and the reasons given are mundane. Other nutrients in seafood besides omega-3s may play a part. Some of the benefit may come from eating seafood in place of less healthful foods. And there is evidence that people who eat seafood have generally healthier lifestyles, which makes any comparison between fish eaters and non-fish eaters hard to read cleanly.
What the evidence does not do is split its memory results by how much fish a person was already eating. That split would settle the fork, and good evidence for it is thin. Until it exists, the question to bring to your doctor is not which brand, it is which side of the fork you are on and whether a capsule is answering a gap or duplicating a plate.
Why does the same word go missing at the same corner of the same errand, week after week, while other words arrive without effort? The pattern is what nags: it is repeatable, it is specific, and it seems to know where it is going before you do. What makes a particular name slippery at a particular moment, and what does that tell us about how a memory is held and how it comes loose?
Before the bottle
Side effects from omega-3 supplements are usually mild: unpleasant taste, bad breath, bad-smelling sweat, headache, and gastrointestinal symptoms such as heartburn, nausea and diarrhea. Fish allergy is a separate matter, and algal oils offer a vegetarian route to DHA, with some products also carrying EPA. If you take an anticoagulant such as warfarin, or any prescription medicine, take the bottle to your doctor or pharmacist before you start. The standing advice from NCCIH is to talk with your health care provider about any complementary approach you use, because some herbs and medicines interact in harmful ways. Dietary supplements are not approved by the FDA before they are sold.
Two more cautions worth keeping in the same paragraph. Causes of memory trouble are several, never one, so a capsule is not a diagnosis and a good week after starting one is not evidence. If a change in memory or thinking persists or gets worse, that is worth a doctor’s time, and a screening test is screening, never a diagnosis. The desk has no interest in telling you what you have or what you do not have.
The longer version sits where the label question started, so you can watch the full video before you commit to a bottle or to a number.
Everything on Supplements & food →
▶ Watch the memory video
