The Hidden Variable That Decides Whether a Food Label Works: Your Working Memory

Why I care about this. I ran this study because of a question that sits at the heart of my research: do health policies actually work the same for everyone, or do we just assume they do? This trial was designed to test that assumption directly, and to make sure the answer wasn’t hiding somewhere we weren’t looking.

Try this: read a phone number once, then look away and repeat it back. Easy for some people, hard for others. Now imagine trying to do that while also deciding what to buy for dinner. That everyday mental juggling act — holding a few things in your head at once — is called working memory, and it turns out to be the hidden factor that decides whether a food label actually helps you or not.

A trial designed to catch something most trials miss

I recruited nearly 500 UK shoppers and had them choose between cereals under one of three conditions: no calorie label at all, a simple label (four broad categories), or a detailed label (eight precise categories). This is the same logic as a medical trial — people are randomly assigned to a group, so we can be confident any difference we see is caused by the label itself, not by who happened to be more health-conscious.

The twist is what I measured beforehand: each person’s working memory, using a quick memory game where they had to keep track of a sequence of letters. Most studies on food labels skip this step entirely. I included it because I suspected it was quietly deciding who a label actually works for.

The label that “wins” depends on who’s reading it

The results confirmed that suspicion. There wasn’t one label that was simply “better.” For people with more working memory to spare in the moment, the detailed label’s extra precision was genuinely useful. But for people with less spare mental capacity, that same detail didn’t help — and in some cases, it seemed to backfire, nudging them toward higher-calorie choices instead of lower ones.

In plain terms: the same label helped one group and quietly worked against another, and the difference had nothing to do with how much either group cared about eating healthily. It came down to something as simple as how much mental bandwidth they had left at that exact moment.

Why “more information is always better” is a myth

It’s tempting to think a label can only help, never hurt — that more detail is a safe, obvious improvement. This trial is direct evidence against that idea. Under carefully controlled conditions, where I could rule out every other explanation, the same piece of information helped some people and worked against others, purely depending on their mental capacity in that moment.

That’s a much bigger deal than it sounds, because working memory isn’t something you choose — it’s shaped by things like age, tiredness, stress, or simply how much else is competing for your attention that day. Which means label design isn’t just a communication challenge. It’s a fairness challenge.

A trial can tell you what happens to volunteers — not to the whole country

Because this was a controlled trial with people who volunteered to take part, it tells us something true and reliable about cause and effect — but it can’t, on its own, tell us what would happen if this label were rolled out to everyone in the UK, many of whom look nothing like the people who signed up for an online study. That’s a bigger, harder question, and it’s the one I tackled next by checking whether this trial’s finding still held up once scaled to the whole population. I’ve written a separate piece about what that follow-up revealed — the short version is: accounting for working memory changed the national-level answer by a lot.

For now, the finding from this trial alone is worth remembering: whether a food label helps or hurts isn’t just about what’s printed on the box. It’s about who’s reading it, and what else their mind is juggling at the time.

The full peer-reviewed study, along with my other published work, is available on my Google Scholar profile.