Science1 distinct publisher2 min readPublished
A KU Leuven and Valladolid team gave 18 commercial colonies a choice of robotic flowers, and only the ones exposed to a larva-killing fungus learned to prefer the antimicrobial nectar, though the study did not measure whether that preference actually helps the colony.
The Scientist · Science desk
science
What you expect from your own old age shows up a decade later in who you still see1 distinct publisher
science
Researchers trace a third of Central Valley groundwater loss to 122 named producers1 distinct publisher
science
The sharp crystal seed was never there: 8,000 nuclei imaged atom by atom, all gradients1 distinct publisher
science
Eastern US extreme rain is pooling into fewer, wider storms, and station records hide it1 distinct publisher
Compiled by The ScientistSomething wrong?How this is made
There's an asymmetry worth noting: *Ascosphaera apis* kills larvae rather than adults [7], so the bees whose foraging changed were not themselves the ones at risk. *Serratia marcescens*, which infects adult bees [8], is the pathogen a forager might plausibly be expected to dose itself against, and that arm showed no significant drift toward quercetin [11]. If the effect is real it looks more like brood care than like self-treatment, which is presumably why the authors declined to call it either: their reading is that "pathogen identity shapes behavioral plasticity in nectar foraging" [13], and that the data do not amount to "a definitive demonstration of formal self- or social medication" [15]. The paper, from Kamiel Debeuckelaere and colleagues, carries its own question mark in the title [16].
A null in the bacterium arm also has duller readings available. An adult gut infection could depress foraging activity or slow learning without redirecting choice, and this design does not separate that from indifference. The denominator is worth stating plainly: 18 colonies across three exposure groups [3][6] is six per arm if the arms were equal [18]. Sensors logged every visit [9], which yields a lot of rows, but the thing that varies with treatment is the colony, and there were not many colonies.
Two features of the rig do real work. The flowers refilled automatically after each visit [9], so a growing preference cannot be an artefact of the plain flowers running dry. And quercetin occurs in nectar and pollen already [5], so the choice on offer is one a bee could meet outside a tent. What the account does not settle is colour. Four blue and four yellow flowers dispensed the two nectars, and the report does not say how quercetin was assigned across them [4]. In a bumblebee, a nectar preference is very easily a colour preference that happened to pay off.
For anyone shipping managed pollinators, the operational content sits in the design rather than the result. These were commercial colonies [3], and they had a behaviour to express only because two nectars were on the menu. A single fixed supplemental feed gives a colony nothing to titrate against; a pathogen-conditioned shift in foraging has precedent in kind, since honey bees with some infections collect more protective plant resin [17].
Whether any of this helps is still unknown. Fungal load and larval survival went unmeasured [14], and the write-up gives the direction and significance of the preference change without a magnitude [19]. Until a follow-up connects the extra visits to fewer dead larvae, quercetin preference is best read as a well-instrumented marker of exposure.
Ranked by verification strength, evidence, and original report placement.
A study published in Royal Society Open Science suggests buff-tailed bumblebees (Bombus terrestris) may self-medicate when exposed to a fungal brood disease.
A team at KU Leuven in Belgium and the University of Valladolid in Spain investigated whether buff-tailed bumblebees change their foraging habits when exposed to different pathogens.
The scientists studied 18 commercial bumblebee colonies placed inside flight tents.
In each tent, eight robotic flowers (four blue and four yellow) produced microdoses of artificial nectar containing either plain sugar water or sugar water with quercetin; the report does not state how quercetin nectar was assigned across the two flower colours.
Quercetin is a natural compound found in plant nectar and pollen and is known to have antimicrobial properties.
Before the experiment the colonies were split into three groups for two weeks: some exposed to Ascosphaera apis, others to Serratia marcescens, and a control group fed pollen with a sterile sugar solution.
Distinct publishers with included, body-backed reporting in this cluster.
phys.org
1 article · August 28, 2026
Follow any of these and your For You feed starts watching them — no settings page required.
Evidence-backed comparisons of source perspectives and observed adoption signals. Read the methodology
Which Builder, Operator, and Investor concerns the observed source mix emphasized—not a truth score.
Evidence, demonstrated adoption, hype gap, incentives, and confidence are assessed independently, each on its own current evidence. How these are measured.
One peer-reviewed paper, one retelling
The spine of this story is a Royal Society Open Science paper with a named lead author and a live DOI, which is a real anchor — and it is the only anchor. Every fact, including the caveats, reaches readers through phys.org's summary of it. The design is described in enough detail to be judged, the results are not: direction and significance without a single visit count, drinking time or per-arm colony number.
Nothing deployed to observe
This is a five-day tent experiment reported the week it published. Nobody in this reporting is using the finding for anything — no breeder programme, no hive-management change, no follow-up trial — and inventing such a use would be worse than leaving the box empty.
The verb outruns the measurement
"Self-medicate" appears in the headline; what was measured is a preference shift. The gap is modest rather than egregious because the paper itself hedges in its own title with a question mark, and phys.org quotes the authors saying their results are not a definitive demonstration of self- or social medication. Readers who stop at the headline get a stronger claim than readers who finish the piece — the correction is present, just downstream of the framing.
Mild, and stated out loud
The pressures visible here are small and disclosed. Phys.org closes with a donation appeal and names its author, editor and fact-checker, which is a reader-funding interest declared rather than hidden. The researchers' interest is the ordinary one of publishing an eye-catching result, and their own title's question mark suggests it was held in check. What we cannot see — who funded the work, who supplied the commercial colonies — the reporting never mentions, so we do not score it.
Firm on what the bees did, thin on why it matters
We are reasonably sure the experiment happened as described and that only the fungus-exposed colonies drifted toward quercetin — that is a clean, specific, published claim reported without embellishment. Confidence drops on interpretation: with no numbers, no arm sizes, no second reading of the paper and no measured benefit to the colony, the distance between "changed its foraging" and "treated itself" stays unmeasured.