Decaf coffee also boosted memory in a small Irish study, and both kinds lowered perceived stress

A small trial from University College Cork found that people who resumed drinking decaffeinated coffee after two weeks without it improved on a learning and memory test, a benefit the caffeinated group did not show. Both groups also reported lower perceived stress, depression and impulsivity scores. The work was led by Professor John Cryan and Dr Serena Boscaini of APC Microbiome Ireland.

The study is not new. It was published in Nature Communications on April 21, 2026, and has since been re-run by news aggregators, including a ScienceDaily item dated September 27. What follows explains what 62 volunteers actually did, what changed in them, and what the authors say the results do not prove.

Two weeks off coffee, then a blinded restart

The paper, titled “Habitual coffee intake shapes the gut microbiome and modifies host physiology and cognition,” compared 31 habitual coffee drinkers with 31 people who did not drink coffee. Regular drinkers were defined as people consuming three to five cups a day, a range that matches European Food Safety Authority guidance that caffeine intakes up to 400 milligrams a day from all sources raise no safety concern for healthy adults.

Coffee drinkers first went 14 days without coffee while researchers ran psychological assessments and collected stool and urine samples. They then entered a 21-day reintroduction phase in which, according to the University College Cork release, half received caffeinated coffee and half received decaffeinated coffee without knowing which. That blinding is what lets the study separate caffeine from everything else in the cup, and the abstinence phase gave every participant a common starting point before the two coffees were compared.

What changed in both groups, and what only decaf did

The paper reports that “participants receiving caffeinated and decaffeinated coffee both showed lower perceived stress, depression, and impulsivity scores.” Those are self-reported scores, which is why the stress finding here is about perceived stress rather than a physiological marker. The gut microbiome also shifted in both groups, with the bacteria Eggerthella and Cryptobacterium curtum rising when coffee returned and falling during abstinence.

The memory result is the one that separates the decaf drinkers. In the paper’s words, “the decaffeinated group showed a significant decrease in PAL errors, unlike the caffeinated group, suggesting that non-caffeine components of coffee may improve learning and memory.” PAL refers to a paired-associate learning task, in which participants learn to match items with locations. Fewer errors on that task is the measured basis for the memory claim, and it is a narrower thing than everyday memory: a single lab test, taken before and after the reintroduction phase.

Caffeinated coffee did other things. The ScienceDaily summary reports that only the caffeinated group had reduced anxiety and improved vigilance and attention, along with a lower inflammation risk. Decaf, in other words, was not simply a weaker version of regular coffee: the two cups produced overlapping but different results, which is the reason the study is framed around components of coffee beyond caffeine.

Reading the result with the caveats attached

The sample is 62 people, the phases were short at 14 and 21 days, and the outcome measures were largely questionnaires and a single learning task. Only half of the 31 drinkers received decaf, which leaves roughly 15 or 16 people in the group behind the memory result, a size at which chance findings are a real possibility. The University College Cork team was direct about the limit. In the first ScienceDaily report on the study, which ran on May 2: the study’s framing was that coffee is more than caffeine, a dietary factor that interacts with gut microbes, metabolism and mood. The later item quotes Cryan on the limit of that idea, saying the findings “do not establish that coffee directly causes these changes in mood, cognition, or gut health.”

The funding is also relevant. The University College Cork release lists the Institute for Scientific Information on Coffee as the funder of the research. Nothing in the reports suggests the sponsor altered the results, but industry funding of a coffee trial is a standard reason to want an independent replication, particularly with a sample this small.

The stress finding also has a limit inside the paper itself. The authors reported lower perceived stress after coffee returned, but the coffee and non-coffee groups showed no significant difference in cortisol during a stress test, so the change appears in how people rated themselves, not in that hormonal measure.

Still unresolved is which components matter, and whether the effect lasts beyond the three weeks studied. The authors suggest compounds other than caffeine, such as polyphenols, may explain the decaf memory effect, but the trial did not isolate them. A follow-up that tested those compounds directly, in more people and over longer periods, would show whether the PAL result was chemistry or chance.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

More from Morning Overview