Quick Read
EPA and DHA are two different omega-3 fatty acids found in fish oil, and they work differently in your brain. DHA is the structural building block of brain cells and membranes, while EPA is the maintenance specialist, reducing inflammation and supporting communication between brain cells. The research suggests both matter, but for different reasons.
For immediate cognitive sharpness in healthy people, EPA appears to have a slight edge based on one well-designed study. For long-term protection against Alzheimer’s plaques, DHA shows a stronger connection. However, most fish oil supplements contain both, and taking a meaningful dose (at least 1,000 to 2,000mg combined daily) matters more than getting the ratio perfect. The research also shows omega-3 works better before cognitive problems start, not after.
One important caveat: if you carry the APOE4 gene variant (a major Alzheimer’s risk factor), your brain may not absorb omega-3 as efficiently as others, potentially requiring higher doses. Fish oil is very safe at normal doses and worth considering, though eating oily fish two or three times weekly remains ideal.
Verdict: Take a combined EPA and DHA supplement at a meaningful dose now rather than waiting, but neither omega-3 alone is clearly superior to the other for brain health.
EPA vs DHA, Which Omega-3 Actually Matters More for Your Brain?
What if the fish oil capsule you’ve been taking every morning is doing something genuinely useful for your brain, but not quite in the way you think? Most people who take omega-3 assume it’s doing one thing, as a single force for good. In reality, the two main omega-3s in your supplement, EPA and DHA, are quite different molecules, with different jobs, different strengths, and different stories to tell. And the research, when you actually dig into it, suggests the answer to “which one matters more?” is far more interesting than a simple winner-takes-all verdict.
Vitacuity has analysed over 1.77 million research papers and selected the 15 most relevant studies on this topic. Here’s what they actually show, and what they honestly don’t.
The Science Behind EPA and DHA: Two Omega-3s, Two Different Roles
Both EPA (eicosapentaenoic acid) and DHA (docosahexaenoic acid) are long-chain omega-3 polyunsaturated fatty acids, the kind found in oily fish, algae and fish oil supplements. Your body can technically make them from the plant-based omega-3 ALA (found in flaxseed, walnuts and chia seeds), but the conversion is so inefficient it barely counts. For practical purposes, you need to get EPA and DHA directly from food or supplements [1].
They sound similar, but they operate quite differently inside the brain.
DHA is the structural heavyweight. It’s literally built into your brain cell membranes, making up a significant proportion of the grey matter, and it’s crucial for how neurons communicate, for maintaining membrane flexibility, and for normal brain development from before birth onwards [12]. Think of DHA as the architecture of your brain: it’s in the walls, the foundations, the structure itself.
EPA is less present in the brain tissue itself, but it’s far from passive. It’s the signalling specialist. EPA generates anti-inflammatory compounds called resolvins and protectins, it influences neurotransmitter function, and it appears to be particularly active in regulating mood and behaviour [12]. Think of EPA as the maintenance crew and the communications system, keeping things running smoothly and managing the inflammatory environment that, if left unchecked, can damage the structure DHA is working so hard to maintain.
Both work via cell membranes, and both generate what researchers call “neuroprotective metabolites”, essentially, molecules that help protect brain cells from damage [12]. The question is: for brain health specifically, which one pulls more weight?
What the Research Shows: EPA and Cognitive Function in Healthy Adults
One of the most striking pieces of evidence comes from a well-designed randomised controlled trial published in the *American Journal of Clinical Nutrition* in 2021 [6]. This trial recruited 310 healthy adults aged 25 to 49 and ran for 26 weeks, a decent duration for this kind of research. Participants were split into three groups: one received an EPA-rich oil (900mg EPA / 360mg DHA daily), one received a DHA-rich oil (900mg DHA / 270mg EPA daily), and one received olive oil as a placebo.
The results were genuinely surprising. EPA-rich oil, not DHA, significantly improved global cognitive accuracy and processing speed compared to both placebo and the DHA-rich group. Memory accuracy also improved with EPA compared to DHA. The researchers framed this through what they called a “neural efficiency” framework: both EPA and DHA showed trends toward reducing oxygenated haemoglobin in the prefrontal cortex, suggesting the brain was working more efficiently, achieving the same cognitive output with less effort.
Evidence grade: Promising. This is a well-run RCT with a good sample size and reasonable duration, but it was conducted in younger, healthy adults, so how well this translates to people over 50 with existing cognitive concerns remains an open question.
What the Research Shows: DHA and Brain Structure, Especially Amyloid Protection
While EPA may edge ahead on short-term cognitive performance in healthy people, DHA has the stronger structural case when it comes to Alzheimer’s disease pathology.
A 2025 observational study from Japan followed 97 older adults aged 75 to 89, measuring their serum DHA and EPA levels 6 to 9 years before conducting amyloid PET brain scans [3]. Amyloid plaques, the protein deposits associated with Alzheimer’s, were the key outcome. The findings showed that higher DHA levels years earlier were inversely associated with amyloid accumulation in the brain. In other words, people who had more DHA in their blood in their late 60s to early 70s had less Alzheimer’s-related plaque in their brains a decade later. EPA showed a weaker and less consistent relationship.
This is significant because amyloid accumulation starts decades before symptoms appear. If DHA really does slow that process, the implications for early supplementation are substantial.
Evidence grade: Promising. This is an observational study, not a controlled trial, it can show an association, not prove causation. And 97 participants is a small sample. But the methodology, measuring omega-3 levels years before imaging, is unusually rigorous, and the Japan population (known for high fish intake) makes the findings interesting to generalise from.
Supporting this, a 2023 review in *Current Opinion in Lipidology* noted that prospective studies and three meta-analyses suggest fish and omega-3 intake are associated with reduced development of mild cognitive decline and Alzheimer’s disease [10]. Crucially, DHA supplementation in RCTs for people with mild cognitive impairment showed cognitive benefit, while in those who already had diagnosed Alzheimer’s disease, the benefit was not demonstrated. Timing matters enormously here.
What the Research Shows: DHA Needs to Actually Reach Your Brain, and That’s Not Guaranteed
Here’s something the omega-3 conversation almost never addresses: just because you swallow DHA doesn’t mean your brain gets it. A small but important RCT published in *eBioMedicine* in 2020 tackled this directly [7].
Thirty-three participants were randomised to receive either 2,152mg of DHA per day or placebo for six months, a meaningfully high dose. The researchers actually measured DHA in cerebrospinal fluid (CSF), the fluid surrounding the brain and spinal cord, using lumbar punctures before and after the trial. This is about as direct a measure of brain delivery as you can get in a living human.
The results were encouraging on one level: DHA supplementation produced a 28% increase in CSF DHA and, interestingly, a 43% increase in CSF EPA. Yes, taking DHA raised EPA levels in the brain too. But here’s the critical finding: the increase in CSF EPA in people who carried the APOE4 gene variant (the most significant genetic risk factor for Alzheimer’s disease) was three times lower than in non-carriers. Brain delivery of omega-3 appears to be compromised in APOE4 carriers.
A complementary 2020 analysis of the Alzheimer’s Disease Cooperative Study’s DHA trial, involving 275 participants over 18 months, confirmed this pattern [8]. APOE4 carriers showed significantly lower increases in plasma DHA and EPA after supplementation compared to non-carriers. In APOE4 non-carriers, higher EPA levels were associated with less decline in right hippocampal volume, a brain region critical to memory. In APOE4 carriers, this association disappeared.
Evidence grade: Promising, with important caveats. These findings suggest that for people with APOE4 genetics, standard omega-3 supplementation may be less effective, possibly requiring higher doses or different formulations. The APOE4 finding is consistent across multiple studies, which strengthens confidence in it.
What the Research Shows: Animal Studies Separate the Two, With Important Nuance
To really understand what EPA and DHA do differently in the Alzheimer’s brain, animal research has been useful, even if it can’t be directly translated to humans yet.
A 2024 study using a well-established Alzheimer’s mouse model (3xTg-AD mice) compared three groups: a control group receiving no omega-3, a DHA group, and a group receiving both DHA and high-dose EPA [2]. After three months, DHA alone increased DHA in the frontal cortex by 19%. DHA plus high EPA increased it by only 8%, suggesting that very high EPA might actually compete with DHA for brain uptake.
Both omega-3 diets reduced arachidonic acid levels in the brain, arachidonic acid being a pro-inflammatory omega-6 fatty acid. Reducing this is considered beneficial for brain health. The study found that DHA and EPA had differential but complementary effects, and that the ratio between them appeared to matter as much as the absolute dose of either.
Evidence grade: Early stage. This is mouse research, and mouse models of Alzheimer’s are imperfect proxies for the human disease. But it raises the genuinely important point that EPA and DHA don’t just add together, they interact, and the ratio may matter.
What the Research Shows: Clinical Trials in MCI and Alzheimer’s, Honest Results
One of the most honest recent summaries of omega-3 clinical trials came from a 2023 scoping review covering 78 RCTs [13]. The headline finding: only 43.6% of all trials reported a positive cognitive outcome from omega-3 supplementation. That’s barely better than a coin flip across the whole literature.
But here’s where the nuance becomes important. When researchers looked specifically at trials in people with mild cognitive impairment (MCI), early-stage cognitive decline, 66.7% of those trials showed a positive outcome. The picture is considerably better for people who are already showing signs of cognitive slowing than for those who are completely healthy. This makes biological sense: if your brain is already under stress, the anti-inflammatory and structural support from omega-3 may make a more measurable difference.
The trials varied wildly in design: doses ranged from 79mg to 5,200mg per day; populations ranged from healthy middle-aged adults to people with diagnosed dementia; testing methods differed hugely. This heterogeneity is almost certainly why results are so inconsistent [4].
A 2025 dose-response meta-analysis covering 58 RCTs added more texture [5]. For every 2,000mg/day increment in omega-3 supplementation, the analysis found significant improvements in attention, perceptual speed, language, primary memory, visuospatial functions and global cognitive ability. The evidence grade from GRADE analysis was “moderate” for perceptual speed, primary memory and visuospatial functions, which is actually reasonable for nutritional research. Dose appears to matter: studies using lower doses have consistently shown weaker effects.
A separate 24-month Taiwanese RCT [15], which directly compared pure DHA (0.7g/day), pure EPA (1.6g/day), and their combination in 163 people with MCI or Alzheimer’s, found no significant difference between treatment and placebo groups on overall cognitive scores. EPA did reduce one specific inflammatory marker (CCL4), and both EPA and DHA reduced scores on spoken language subtests, though the clinical meaning of this is unclear. Overall, this trial found limited benefit, reinforcing that in people who already have Alzheimer’s, omega-3 supplementation may be too late to make a measurable difference.
Evidence grade: Conflicted. The reason for conflict is well understood, different doses, different populations, different timing of intervention, genetic variation in metabolism, and different cognitive tests used. The weight of evidence suggests omega-3 is most likely to show benefit when taken earlier (before or at the MCI stage), at meaningful doses (at least 1-2g combined EPA+DHA daily), and when considering individual factors like APOE genotype.
What We Don’t Know Yet
Let’s be honest about the gaps, because they matter.
We don’t know the optimal EPA:DHA ratio for the brain. Most trials use combined fish oil supplements with varying ratios, making it very hard to isolate which molecule is doing what. The 2021 RCT [6] comparing EPA-rich and DHA-rich oils head to head is rare precisely because this kind of design is uncommon.
We don’t know the ideal dose, reliably. The 2025 meta-analysis [5] found dose-dependent effects, but also found that the relationship wasn’t always linear, for episodic memory, the curve appeared to turn back down at higher doses. Whether there’s a sweet spot, and what it is, remains unclear.
We don’t fully understand the APOE4 problem. Multiple studies confirm that APOE4 carriers respond differently to omega-3 supplementation, with reduced brain delivery of both DHA and EPA [7,8]. But we don’t yet know whether higher doses overcome this, or whether a different form of omega-3 (like krill-derived phospholipid forms) might be more effective in these individuals.
Long-term RCTs are largely absent. Most trials run for six months to two years. Cognitive decline happens over decades. We don’t have the long-duration trial data to confidently confirm whether omega-3 supplementation prevents Alzheimer’s, only that it’s associated with slower decline in some contexts.
The placebo problem is real. Olive oil, used as a placebo in several trials, is itself anti-inflammatory. This may have blunted the apparent benefit of omega-3 in some studies, making results look less positive than they are [4].
Most trial data doesn’t separate EPA and DHA effects cleanly. Even when researchers use EPA-rich or DHA-rich preparations, there’s always some of both present. Truly isolating one from the other in humans remains difficult.
The Final Takeaway
So, EPA or DHA? The honest answer, based on all the evidence above, is: both, and probably in greater amounts than most people take.
Here’s how a sensible, well-informed person should think about this:
If you’re currently healthy and focused on maintaining sharpness: The 2021 RCT [6] suggests EPA may be the more active player for day-to-day cognitive performance, accuracy, speed, memory. An EPA-rich fish oil (where EPA exceeds DHA) is a reasonable choice for this goal.
If you’re thinking about long-term brain protection and Alzheimer’s risk: DHA’s role in brain structure and its inverse association with amyloid accumulation [3] makes it the priority for the long game. An omega-3 supplement with meaningful DHA, ideally above 500mg per day, makes sense here.
In practice: most people should simply take a good combined EPA+DHA supplement, at a meaningful dose. The research consistently shows that underdosing is a major reason trials fail. Aim for at least 1,000–2,000mg of combined EPA+DHA daily. The dose-response meta-analysis [5] specifically looked at 2,000mg/day increments, suggesting that many standard “one-a-day” capsules providing only 300–400mg of combined EPA+DHA may be too low to show measurable brain effects.
Don’t wait until cognitive symptoms appear. The clearest signal in the research is that omega-3 works better earlier, at the MCI stage or before, not once Alzheimer’s is established [10,13]. If you’re in your 40s or 50s, now is the time.
If you know you carry the APOE4 gene variant, the research suggests you may need higher doses to achieve the same brain delivery as non-carriers [7,8]. This is worth discussing with a GP, but it’s not a reason not to supplement; it may be a reason to supplement more.
Fish oil is very safe at normal supplementation doses. It’s not in the category of nutrients where accumulation causes toxicity. At 1,000–3,000mg of EPA+DHA daily, the safety profile is well established. If you’re on blood thinners, it’s worth mentioning to your doctor at your next appointment, but this shouldn’t stop most people from taking it.
Eating oily fish two to three times a week remains the gold standard, and adds nutrients beyond omega-3 that no capsule fully replicates. But for most people in the UK, where fish consumption is low and conversion of plant-based omega-3 is inefficient, supplementation is the practical, sensible default.
The research doesn’t let us say omega-3 prevents dementia. But it does tell us something genuinely useful: the people who eat the most oily fish and have the highest omega-3 levels consistently show slower cognitive ageing and less amyloid accumulation. EPA and DHA, together, are among the most evidence-backed nutrients for brain health we have. The honest answer isn’t “take EPA” or “take DHA”, it’s “take both, take enough, and start now.”
References
[1] An analysis of omega-3 clinical trials and a call for personalized supplementation for dementia prevention (2024). https://pubmed.ncbi.nlm.nih.gov/38379273/
[2] Differential impact of eicosapentaenoic acid and docosahexaenoic acid in an animal model of Alzheimer’s disease (2024). DOI: 10.1016/j.jlr.2024.100682 | https://pubmed.ncbi.nlm.nih.gov/39490923/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11650307/
[3] An inverse association of cerebral amyloid-β deposition and serum docosahexaenoic acid levels in cognitively normal older adults in Japan (2025). DOI: 10.1177/13872877251340688 | https://pubmed.ncbi.nlm.nih.gov/40336297/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12349894/
[4] Omega-3 polyunsaturated fatty acids in neurodegenerative disorders: Mixed designs = mixed results (2025). DOI: 10.1016/j.plipres.2025.101356 | https://pubmed.ncbi.nlm.nih.gov/40976313/
[5] A systematic review and dose response meta analysis of Omega 3 supplementation on cognitive function (2025). DOI: 10.1038/s41598-025-16129-8 | https://pubmed.ncbi.nlm.nih.gov/40836005/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC12368174/
[6] Supplementation with oil rich in eicosapentaenoic acid, but not in docosahexaenoic acid, improves global cognitive function in healthy, young adults: results from randomized controlled trials (2021). DOI: 10.1093/ajcn/nqab174 | https://pubmed.ncbi.nlm.nih.gov/34113957/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8408864/
[7] Brain delivery of supplemental docosahexaenoic acid (DHA): A randomized placebo-controlled clinical trial (2020). DOI: 10.1016/j.ebiom.2020.102883 | https://pubmed.ncbi.nlm.nih.gov/32690472/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7502665/
[8] Effect of APOE Genotype on Plasma Docosahexaenoic Acid (DHA), Eicosapentaenoic Acid, Arachidonic Acid, and Hippocampal Volume in the Alzheimer’s Disease Cooperative Study-Sponsored DHA Clinical Trial (2020). DOI: 10.3233/JAD-191017 | https://pubmed.ncbi.nlm.nih.gov/32116250/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7156328/
[10] Omega-3 fatty acids and cognitive function (2023). DOI: 10.1097/MOL.0000000000000862 | https://pubmed.ncbi.nlm.nih.gov/36637075/ | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11878108/
[12] Omega-3 DHA and EPA for cognition, behavior, and mood: clinical findings and structural-functional synergies with cell membrane phospholipids (2007). https://pubmed.ncbi.nlm.nih.gov/18072818/
[13] New perspectives on randomized controlled trials with omega-3 fatty acid supplements and cognition: A scoping review (2023). https://pubmed.ncbi.nlm.nih.gov/36603691/
[14] DHA Supplementation Alone or in Combination with Other Nutrients Does not Modulate Cerebral Hemodynamics or Cognitive Function in Healthy Older Adults (2016). https://pubmed.ncbi.nlm.nih.gov/26867200/
[15] Omega-3 fatty acids and blood-based biomarkers in Alzheimer’s disease and mild cognitive impairment: A randomized placebo-controlled trial (2022). DOI: 10.1016/j.bbi.2021.10.014 | https://pubmed.ncbi.nlm.nih.gov/34755655/
This article is for informational purposes only and does not constitute medical advice. Food supplements should not be used as a substitute for a varied and balanced diet and healthy lifestyle. If you are pregnant, breastfeeding, taking medication or have a medical condition, consult your doctor before taking any supplement. These statements have not been evaluated by the Food and Drug Administration (FDA) or the Medicines and Healthcare products Regulatory Agency (MHRA). This product is not intended to diagnose, treat, cure, or prevent any disease.