In August 2026, Real Simple ran a piece by Christina Manian, RDN, headlined "We Asked 3 Registered Dietitians to Name the Best Snack for Brain Health—They All Picked the Same One." The answer:
According to three registered dietitians, the best snack for brain health is tinned (or canned) fish—whether that be tuna, sardines, or salmon—as this increasingly trendy snack ticks a lot of the brain-boosting nutrient boxes.
The piece carries six numbered citations and a sources box. That apparatus does most of the persuading, because almost nobody clicks. We clicked.
Start with what the article gets right
Tinned fish is a genuinely good food. Canned sardines in tomato sauce carry about 1.19 g of EPA plus DHA per 3 oz serving, and canned pink salmon about 0.91 g, per Table 2 of the NIH Office of Dietary Supplements omega-3 fact sheet. It is cheap, it keeps for years, and the advice to pair it with whole grains or vegetables for fibre is sound. Higher omega-3 blood levels really do track with better cognitive outcomes in observational work. Nothing here is fabricated. What fails is the joinery — the step from those ingredients to "the best snack for brain health."
The six citations, one at a time
A mouse. For the proposition that selenium in sardines "offers its own unique brain health benefit," the article cites Liu et al. (J Trace Elem Med Biol, 2025) — a dietary selenium intervention in the 3×Tg-AD transgenic Alzheimer's mouse model. There are no human participants in it.
Three narrative reviews the authors themselves hedge. The magnesium review (Maier et al., Int J Mol Sci, 2022) has no primary data, rests largely on animal models, and concedes that for some associations it describes "no causative mechanisms have been demonstrated yet." The omega-3 review (von Schacky, Nutrients, 2021) is cited to support eating fish, but its argument runs the other way: dietary intake is a poor proxy for omega-3 status. Bioavailability varies roughly 13-fold between people, and food frequency questionnaires produce "some 50% implausible data."
The third, on fibre and the gut–brain axis (La Torre et al., Gut Microbiome, 2021), is the one citation used at about the right strength. Its authors flag that most human studies used fibre supplements rather than whole foods and that most evidence is associative — and Kayla Farrell's sentence duly says diets higher in fibre are "associated with better cognitive health." Credit where it is due.
A supplement dose you cannot eat. The claim that omega-3s and vitamin D are "two nutrients that can boost mood" is sourced to a Massachusetts General Hospital psychiatry blog post about a 12-week trial of omega-3 at up to 4 g/day in unmedicated depressed patients with BMI above 25 and hs-CRP of at least 3 mg/L. Vitamin D is not in the study. Fish is not in the study. The post is explicit that its findings apply "specifically to unmedicated, depressed patients with inflammation," and that over-the-counter omega-3 products supply "600–800mg" EPA against the study's 4 g.
A folate finding used to sell a B12 food. B vitamins, the article says, "can slow cognitive decline and potentially prevent dementia," citing Wang et al. (Nutrition Reviews, 2022), a meta-analysis of 95 studies and 46,175 participants. The supplementation benefit there is 0.14 points on the 30-point MMSE (95% CI 0.04–0.23): statistically real, clinically about half a percent of the instrument. And the dietary dementia result is folate-specific — folate intake, HR 0.61 (95% CI 0.47–0.78), while "higher intake of B12 or B6 was not associated with lower dementia risk." Tinned fish's headline B vitamin is B12.
The question the article never asks
The promise is an intervention: eat this, and your brain works better. None of the six citations tests that. Three bodies of evidence that do were not cited.
In the VITAL cognitive substudies (Kang et al., Alzheimer's & Dementia: TRCI, 2022), healthy adults aged 60+ took 1 g/day of marine n-3 including 840 mg EPA+DHA — roughly a tin of sardines' worth, every day — for two to three years. On the primary global cognitive composite there was no effect: pooled mean difference in annual rate of change −0.01 (95% CI −0.02 to 0.003; P = .15). "Marine n-3 supplementation (1 g/day) did not confer cognitive benefits over 2 to 3 years in community-dwelling older adults." Two to three years is short against a dementia timeline, and this is supplementation, not diet.
A Cochrane review (Sydenham et al., 3 RCTs, 4,080 participants) found "no benefit of omega-3 PUFA supplementation on cognitive function in cognitively healthy older people." That review is from 2012 — the state of the field then, not now.
The most recent and most favourable evidence is an umbrella review of nine systematic reviews, 14 RCTs and 26,881 participants aged 40+ without dementia or with mild cognitive impairment (Nutrients, 2025). It is positive: a pooled "statistically significant but modest improvement in MMSE scores (effect size: 0.16; 95% CI: 0.01–0.32)." Then put it in proportion. Two of the nine reviews rated critically low on AMSTAR-2, the analysis covered MMSE-only trials, and there was no dose–response relationship. A lower bound of 0.01 points on a 30-point scale is a signal worth chasing, not a basis for naming a best snack.
The NIH ODS fact sheet summarises the same literature: omega-3 supplementation "does not affect cognitive function in healthy older adults or in people with Alzheimer's disease compared to placebo," with possible unconfirmed benefit in mild cognitive impairment.
Two nutrient claims that do not survive their own tables
"Tuna is one of the best dietary sources of DHA and EPA," per Maggie Bell. Against the article's own three nominees it is the weakest by a wide margin: ODS Table 2 puts canned light tuna in water at 0.19 g EPA+DHA per 3 oz, roughly one-sixth of canned sardines and one-fifth of canned pink salmon.
On vitamin D, ODS Table 3 gives canned light tuna 1.0 mcg (40 IU, 5% DV) per 3 oz and two canned Atlantic sardines 1.2 mcg (46 IU, 6% DV). Cooked sockeye salmon is 14.2 mcg (570 IU, 71% DV) — but that row is cooked sockeye. The table has no canned-salmon row, so that figure cannot be moved into a tin.
The advisory that goes unmentioned
The article tells readers to eat canned tuna "whenever hunger strikes," straight from the tin, and to serve it to children as a vegetable dip. It never mentions mercury.
EPA and FDA jointly advise people who are pregnant, might become pregnant or are breastfeeding, and adults feeding children, to "eat 2 to 3 servings of fish a week from the 'Best Choices' list OR 1 serving from the 'Good Choices' list," with children's portions of 1 oz at ages 1–3 rising to 4 oz at 11+. We read the advisory page but not its downloadable per-species chart, so we make no claim about tier placements.
Consumer Reports tested 30 samples across five canned tuna brands in 2023: albacore averaged about three times the mercury of light or skipjack tuna, and one in five cans (six of 30 — a small sample, worth saying so) spiked high enough to change the FDA recommendation for that can. CR advised pregnant people to avoid canned tuna altogether, and other adults to cap intake at three 4-oz servings of light tuna weekly. None of that makes tinned fish unsafe. It makes "whenever hunger strikes" the wrong frequency advice to publish with no mention of servings per week.
Three experts, two commercial adjacencies
"We asked 3 registered dietitians... they all picked the same one" is a consensus frame built on n = 3 with no stated method: we are not told whether the question was open, or what else was named.
Two of the three affiliations also matter. Farrell is a dietitian at FRESH Communications, an agency for "better-for-you food, beverage, and lifestyle brands" whose services include Public Relations and Influencer Marketing. Bell's name links to wowmd.com, a direct-to-consumer supplement retailer, where she is the site's credentialed expert. There is no evidence any of the three was paid to name tinned fish, and we did not look for a client list. But Real Simple linked both affiliations without characterising either, and a reader weighing "three independent experts agreed" should have been told.
Why D
D — Misleading. Every component is real; the argument built from them is not. A mouse becomes a human benefit, a folate result justifies a B12 food, and a 0.14-point MMSE difference becomes "slow cognitive decline and potentially prevent dementia."
Not E: nothing here is contradicted outright. The nutrients are present, the food is good, and the omega-3 hypothesis remains live — the 2025 umbrella review is positive. Calling this false would earn a correct rebuttal.
Not C: C is for evidence genuinely split and honestly presented. This is sound material rearranged to reach a conclusion it does not carry, plus the omission of a federal advisory that speaks directly to the behaviour recommended.
Eat the sardines. Read the footnotes.