When it comes to nutrition, were all too eager to ignore the evidence. Heres why.

July 2024 · 7 minute read

Let’s do a little thought experiment. Imagine a new study about Ginkgo biloba and memory. For the study, a researcher gave one guy — let’s call him Walter — Ginkgo biloba every day for a year. When the year was up, she asked Walter if his memory improved, and Walter reported that, yes indeed, he doesn’t forget as many names and he almost never misplaces the car keys. New England Journal of Medicine, here we come!

You’d roll your eyes, right? One guy, no control group, subjective outcome. Our researcher gets laughed out of peer review.

And yet, studies just like that are probably the biggest single determinant of what people believe about how diet and nutrition affect health. Of course we don’t believe it when it’s Walter. But we believe it with every fiber of our being when it’s us.

Most dietary supplements don’t do anything. Why do we spend $35 billion a year on them?

Did you cut out sugar and feel more energetic, take vitamin C and stop getting colds, switch to eating only meat and watch anxiety melt away? It’s going to be very hard for anyone to talk you out of that. (Full disclosure: I’m not going to try.)

Advertisement

Those of you following along at home may remember that I wrote about dietary supplements last month. I found that, with a few exceptions, they don’t do much of anything, and there are some potential risks. And boy, did I hear about it! While about half the mail was from people who agreed, the other half was the Tamar-you-ignorant-slut kind (and, yes, all my SNL references are last-century). “I take [fill in supplement here] religiously, and I never [fill in bad health outcome here]!”

I really liked one of those letters, from a guy named Walter (get it?):

“This 93-year-old, who has only been in a hospital only once in 77 years, hasn’t had a cold in 20, who is younger in appearance than his chronological age, who volunteers for two organizations ... believes that the many years of taking nutritional supplements in quantities some would consider excessive, have contributed to his good health. Your anti-supplement bias piece cherry-picked examples of minor issues without addressing the overall issue of a balanced nutritional program. In fairness, this response should be published in response to your biased piece.”

Advertisement

Okay, Walter. Not only will I publish your letter, I will name the ability personal experience has to override research findings: It’s the Walter Effect.

Here’s what the government’s dietary guidelines should really say

Why is the Walter Effect so much more powerful than, say, PubMed, the repository of journal articles where I spend about half my life? I asked Caltech professor Christopher Hitchcock, who studies causal reasoning and is the co-editor of “The Oxford Handbook of Causation” (causation has a handbook!).

For starters, making decisions based on experience is our evolutionary endowment. “For the vast majority of our evolutionary history, the only way we had to learn about a causal relationship was through our own observations, or through firsthand testimony of someone in our immediate social group,” Hitchcock wrote me in an email. In evolutionary terms, we’ve had PubMed for seven nanoseconds, so it’s not surprising it can’t override a decision-making apparatus that has been evolving since the primordial ooze.

Advertisement

We’re hard-wired to connect dots. When Thing 1 happens, and then Thing 2 happens, we humans are very likely to conclude that Thing 1 caused Thing 2, even if they’re completely unrelated; it’s a phenomenon psychologists call the “illusion of causality.”

We’re not the only species susceptible to the Walter Effect. If it’s any consolation, pigeons are, too. Way back in 1948, behaviorist B.F. Skinner did an experiment in which he put hungry pigeons in cages and sent food pellets down at random intervals. The poor birds ended up believing that what they happened to be doing just before the feeding made the pellet appear, and they kept repeating that behavior. One turned around in the cage (only counterclockwise). One repeatedly stuck its head in the corner. One made pecking motions toward the floor.

How processed food makes us fat

That’s us, people. We are Skinner’s pigeons. We are, all of us, Walter.

Advertisement

But Hitchcock goes on to point out that causation isn’t always an illusion. Sometimes, Walter is just plain right. No matter what happens in double-blind, placebo-controlled studies, the supplement/food/diet may actually be helping you. It could be a placebo effect (which Hitchcock notes is particularly powerful for effects that are hard to measure, such as energy level or pain), but it also could be that you are particularly responsive to the supplement/food/diet and experience an effect that isn’t common or powerful enough to change research results, which average out results over populations.

Share this articleShare

Take dieting to lose weight. If you go keto and lose 20 pounds, there’s no question that the effect is real. The keto diet is helping you even though it doesn’t outperform other diets in trials.

So, it could be that the effect is all in your head. It could be that the effect is real, whether it’s placebo pain relief or measurable weight loss. But either way, if your experience flies in the face of research results, you’re probably going to go with your experience. And Hitchcock says that could be a completely rational decision. If the cost of continuing (say, paying for a supplement) is small compared to the risk of discontinuing (and potentially giving up the perceived benefit), it makes sense to keep on keeping on.

Advertisement

Add to this the pesky cognitive trick that is confirmation bias — seizing on information that supports our beliefs and finding ways to reject information that conflicts — and you have a recipe for not changing your mind. “Once we form beliefs, it is very hard to give them up,” Hitchcock wrote. “This is especially true if the beliefs relate to our own self-conception. If I think of myself as being very health-conscious, paying special attention to my diet, it is hard for me to accept that what I have been doing isn’t actually doing me any good.”

If something feels like it’s doing you good, and it’s definitely not causing harm (an important caveat that could be a column in itself), why would anyone want to talk you out of it? If you’re 93 years old, and still healthy enough to do volunteer work and write cranky letters to journalists, I’m sure not going to argue. You go, Walter.

But there is something I’m going to try to talk everyone out of: saying out loud that whatever your experience leads you to believe is The Truth. A goodly portion of food Twitter is dedicated to people proclaiming that the thing that works for them is — duh! — the thing that will work for everyone.

All processed food isn’t bad. We took a nutritionist shopping to find reasonable options.

All of this scares me, since it’s my job to figure out what will work for everyone, and I’m as susceptible as the next guy to the Walter Effect (unless the next guy is actually Walter, who seems pretty dug in). I asked Hitchcock if he had any advice for those of us trying to figure out if we’re just spinning in our cage, counterclockwise. The first step, he says, is to simply understand our tendency to leap to causal conclusions. If we do that, we get better at thinking of our own experiences “as hypotheses to be tested, rather than beliefs to be clung to.” We may be less susceptible to the illusion of causality if we understand how it works, he says.

Advertisement

I’m fortunate in that I also have a fail-safe: readers who will let me know, early and often, if I get derailed from the best evidence. I’m counting on you.

Just be more polite than Dan Aykroyd was, okay?

More from Food:

Most dietary supplements don’t do anything. Why do we spend $35 billion a year on them?

Here’s what the government’s dietary guidelines should really say

How processed food makes us fat

All processed food isn’t bad. We took a nutritionist shopping to find reasonable options.

Unearthed column archive

ncG1vNJzZmivp6x7uK3SoaCnn6Sku7G70q1lnKedZLmqssSsq7KklWSzsLvDaK6hnZ5itrV5wqiknqtdqbxuutStqaKsmaS7bsPEq5xmmZyherW7zmacmp%2BVp3q1u4yinqenopp6tbTEZpyvoZSau6SxjKGcq52jYsSpxY5rZ2toX2V%2FcH6SaJttnJRtgnSAjG5rmnBdZn6mrYxynG1vXWqGeXyTm5xqnJObr6C%2F06ipsmaYqbqt