Discoveries
New research does not get an article of its own here. It gets attached to the subject it belongs to, connected to the claims it bears on, and used to make that subject more complete than we found it. Every item below points back into the page it belongs to rather than standing in for it.
Showing 24 of 24 discoveries.
Aedes aegypti breeding detected in England: surveillance statement
Recorded as a dated official status rather than as a finding about the species, because that is what it is. Everything in it is time-sensitive: a status accurate on 3 September 2026 may not describe the situation later, in either direction. NatureHQ carries it with the date attached and with the four states — detected, breeding, established, transmitting — kept separate, because collapsing them is precisely how this kind of report becomes a headline the evidence does not support.
What it changes
What it does not show
A surveillance statement rather than a study; it reports a detection and an assessment, not an investigation.
Reference unverified. · institutional report
African elephants address one another with individually specific name-like calls
The reason this attracted so much attention is the contrast with dolphins and parrots, which address each other by copying the other's signature call. Here the label appears not to be an imitation, which would make it arbitrary — closer to how a human name works. It is recent, from two populations, and the acoustic analysis carries real uncertainty; it is a strong new finding rather than a settled one, and it is exactly the kind of result that should be revisited as replications appear.
What it changes
- supports Some animal calls point at specific things in the world, not just at how the caller feels Well supported
- supports Elephants may address each other with something like names Emerging evidence
- gives context to Elephants can learn to produce sounds they have heard, including ones no elephant makes Well supported
- supports African elephants appear to address particular individuals with calls specific to that animal — and unlike dolphins, they do not do it by imitating the animal they are calling. Emerging evidence
- qualifies Machine learning is very good at working out who made a sound and what kind of sound it is. It has no route to what the sound means. Established
What it does not show
Model accuracy was above chance but far from perfect.
Verified reference: doi.org/10.1038/s41559-024-02420-w · peer reviewed
Contextual and combinatorial structure in sperm whale vocalisations
This is the study behind a wave of "whales have an alphabet" coverage, and the gap between what it shows and what was reported is instructive. What was demonstrated is *structure*: the signal varies along more dimensions than anyone had catalogued, and the variation tracks conversational context rather than being noise. What was not demonstrated, and what the paper does not claim, is that any of it means anything. Finding combinatorial structure is a necessary step towards testing for a communication system with semantics; it is not evidence of one. Birdsong and whale song both have rich combinatorial structure without semantics. The honest position is that the search space has been mapped and the search has not been run.
What it changes
- supports The clicks have far more structure than anyone knew, and no demonstrated meaning Emerging evidence
- gives context to Several animals have one or two of the things that make human language what it is. Nothing has been shown to have the whole set. Well supported
- supports Machine analysis found more structure in sperm whale clicks than anyone had catalogued. What it means, nobody knows — and no whale has been tested. Preliminary
What it does not show
One clan in one region; whether the same structure appears elsewhere is untested.
Verified reference: doi.org/10.1038/s41467-024-47221-8 · peer reviewed
Wake-like skin patterning and neural activity during octopus sleep
This closes part of the gap the 2021 study left open: the active phase is not just twitching, the brain is doing something wake-like. It still cannot show that an octopus experiences a dream. The most defensible summary is that octopuses have a sleep state with the neural and behavioural signature of REM sleep, and that what it is like from the inside is unknown.
What it changes
- supports Octopuses have a sleep stage that looks like REM sleep Well supported
What it does not show
Small sample, as is normal for this technique.
Verified reference: doi.org/10.1038/s41586-023-06203-4 · peer reviewed
Targeted treatment of injured nestmates with antimicrobial compounds in an ant society
Reported almost everywhere as ants performing medicine, which imports intention the study does not test. What is demonstrated is precise, chemically specific and genuinely remarkable on its own terms: a secretion with antimicrobial activity applied to a wound, with a measurable survival benefit. Whether any ant is treating anything, in the sense a doctor treats a patient, is not a question this design asks.
What it changes
- supports Some ants carry injured nestmates home — but only the ones that call for help Well supported
What it does not show
Survival measured under laboratory conditions with controlled infection exposure.
Verified reference: doi.org/10.1038/s41467-023-43885-w · peer reviewed
Positive citation bias and overinterpreted results lead to misinformation on common mycorrhizal networks in forests
This is the most important record in this file, and it is a critique rather than a discovery. It does not say fungal networks are fictional; it says three specific popular claims are not well supported, and it documents how a modest 1997 result became a settled fact in the public mind without the intervening evidence. NatureHQ treats it as the current best summary of the state of the question, and keeps the underlying studies alongside it so a reader can see both.
What it changes
What it does not show
A review of the literature, not new field data.
Verified reference: doi.org/10.1038/s41559-023-01986-1 · peer reviewed
Re-examining the evidence for the mother tree hypothesis – resource sharing among trees via ectomycorrhizal networks
Recorded alongside the review above because two independent critiques reaching similar conclusions is more informative than one. The methodological point is the durable part: labelling experiments have real difficulty showing that carbon found near a seedling actually went into the seedling by way of the fungal network.
What it changes
What it does not show
A review, so it inherits the limitations of the studies it examines.
Verified reference: doi.org/10.1111/nph.18935 · peer reviewed
Associative learning in the box jellyfish Tripedalia cystophora
The importance here is not that a jellyfish is clever. It is that a nervous system of about a thousand neurons, arranged as four separate visual processing centres with no brain joining them, is enough for a form of learning that had been treated as requiring one. That is a claim about what brains are *for*, made by removing the brain and finding some of the function still there.
What it changes
- supports A jellyfish with no brain can learn from experience Emerging evidence
What it does not show
One species, small numbers, and a short observation window.
Verified reference: doi.org/10.1016/j.cub.2023.08.056 · peer reviewed
Nesting chinstrap penguins accrue large quantities of sleep through seconds-long microsleeps
This is a genuinely strange result and it should be reported as one. The obvious reading — that fragmented sleep works just as well — is not established, and the authors do not claim it. What was shown is that a bird under continuous predation and social pressure accumulates a normal daily quantity of sleep four seconds at a time, and successfully breeds. Whether those seconds do everything consolidated sleep does is exactly the open question, and it is more interesting than the headline. Note also that this is one species in one situation: it is a fact about incubating chinstraps, not about penguins.
What it changes
- supports A nesting penguin sleeps eleven hours a day, four seconds at a time Emerging evidence
What it does not show
One species during one demanding life stage; sleep architecture elsewhere in the year is unknown.
Verified reference: doi.org/10.1126/science.adh0771 · peer reviewed
Sounds emitted by plants under stress are airborne and informative
This is the paper behind every "plants scream" headline, and the gap between it and the headline is the point. What was measured is clicking, in a range no human can hear, at a rate of a few dozen an hour — closer to a dripping tap than to a scream, and produced by cavitation, water columns snapping under tension. There is no evidence it is a signal: nothing indicates the plant produces it for a receiver, and the paper does not say so. "Informative" is doing precise work in that title — it means the sound correlates with the plant's state well enough that a listener *could* learn something, which is the definition of a cue rather than a signal. Whether anything actually listens is the open question, and the honest answer today is that it is being investigated.
What it changes
- supports Plants do click under stress, in a range no human can hear, and nothing is known to listen Emerging evidence
What it does not show
Two crop species under controlled conditions; the greenhouse extension is a smaller dataset.
Verified reference: doi.org/10.1016/j.cell.2023.03.009 · peer reviewed
Electrical potentials in the ectomycorrhizal fungus Laccaria bicolor after a rainfall event
This is the study that answers the most obvious objection to the whole literature: that everything recorded so far comes from cultures in a laboratory, where an implanted electrode and an artificial substrate could plausibly account for the lot. Recording in the ground, in weather, is a real advance. It also demonstrates the plainest alternative reading of the activity — that it tracks what is happening to the fungus, water above all — which is a considerably duller explanation than language and considerably better supported.
What it changes
- supports Fungal mycelium produces electrical spikes, and the pattern changes when the fungus is disturbed. Well supported
- gives context to Fungi have not been shown to have words. The spikes are real; the vocabulary is an interpretation. Popular claim, unsupported
What it does not show
One species, one site, one rainfall event; a correlation across a single environmental change.
Verified reference: doi.org/10.1016/j.funeco.2023.101229 · peer reviewed
Social signal learning of the waggle dance in honey bees
This complicates the textbook version usefully. The dance was the classic example of a fully innate signal, and it turns out that part of it has to be learned by following other bees — which puts an insect into the same category as a songbird learning its song. The dance is still not a language, and this does not move it towards one. What it moves is where the boundary between instinct and tradition sits, which was drawn in the wrong place.
What it changes
- qualifies A great deal of animal signalling is produced automatically by the animal’s state, not chosen — and that is one of the things that makes it unlike speech. Well supported
- supports Bees raised with no experienced dancer to follow still dance — but they get the direction wrong more often, and never fully fix their distance signalling. Emerging evidence
What it does not show
A single species and a manipulated colony structure that does not occur naturally.
Verified reference: doi.org/10.1126/science.ade1702 · peer reviewed
Language of fungi derived from their electrical spiking activity
The measurement is the reliable part and it is not what travelled. Voltage fluctuations occur and cluster; that is data. Everything after it depends on a threshold chosen by the analyst for how long a gap ends a "word", and on an analogy to language that the analysis assumes rather than tests. Nothing in the design examines whether any fungus responds to a train, which is the minimum a communication claim needs, and no receiver was tested at all. Specialists said so publicly at the time, and the paper itself is more cautious than the coverage of it. NatureHQ therefore treats the spiking as demonstrated and the vocabulary as an interpretation with no supporting test.
What it changes
- partly supports Fungal mycelium produces electrical spikes, and the pattern changes when the fungus is disturbed. Well supported
- qualifies Fungi have not been shown to have words. The spikes are real; the vocabulary is an interpretation. Popular claim, unsupported
What it does not show
Four species, small numbers of electrode pairs, no receiver and no behavioural outcome measured.
Verified reference: doi.org/10.1098/rsos.211926 · peer reviewed
Termite sensitivity to temperature affects global wood decay rates
The study that stops "fungi decompose wood" from being the whole sentence. Across much of the warm world the animals get there first, and they respond to climate on a completely different curve from the fungi — which means a warming forest does not simply decay faster, it decays differently, by different organisms, in different places. It sits alongside the finding that insects do about 29% of deadwood decay globally and explains where that 29% is concentrated.
What it changes
- supports Warmth speeds up microbial decay of wood. It speeds up termite decay several times faster. Well supported
What it does not show
Standardised blocks of one wood type, which is what makes sites comparable and is not what a forest floor contains.
Verified reference: doi.org/10.1126/science.abo3856 · peer reviewed
Computational bioacoustics with deep learning: a review and roadmap
Carried because it is the honest description of what "AI is decoding animal communication" actually refers to. The work is real and useful: continuous recordings that would take a lifetime to listen to can be searched automatically, which is how population monitoring at scale became possible at all. Every task in the review is detection, classification or identification — questions of the form "what made this sound". None of them is a question about meaning, and no amount of improvement on them becomes translation, because the training data contains sounds and labels, never meanings.
What it changes
What it does not show
A review; it introduces no new data.
Verified reference: doi.org/10.7717/peerj.13152 · peer reviewed
Cyclic alternation of quiet and active sleep states in the octopus
This is where "do octopuses dream?" comes from. What was shown is a cyclic active sleep state — the same *structure* as REM sleep, arrived at independently. Whether anything is experienced during it is not something a camera can answer, and the researchers say so. Four animals of one species.
What it changes
- supports Octopuses have a sleep stage that looks like REM sleep Well supported
What it does not show
Four individuals.
Verified reference: doi.org/10.1016/j.isci.2021.102223 · peer reviewed
Male lyrebirds create a complex acoustic illusion of a mobbing flock during courtship and copulation
The single most interesting result in vocal mimicry, and a useful antidote to the idea that mimicry is an exuberance. This is context-specific, structured, deployed at a particular moment and imitates a *scene* rather than a sound — several species plus wingbeats, assembled into something that does not exist. The functional interpretation is a strong inference from timing rather than a demonstration; no playback experiment has confirmed what a female does when she hears it.
What it changes
What it does not show
Observational. The proposed function rests on when the display occurs, not on a manipulation.
Verified reference: doi.org/10.1016/j.cub.2021.02.003 · peer reviewed
Half a century of global decline in oceanic sharks and rays
Abundance indices assembled from fisheries data carry real uncertainty, and the headline percentage should be read as an estimate with a wide interval rather than a measurement. What is not in doubt is the direction and the mechanism. It also sets the honest proportions for the topic: the number of people killed by sharks each year is around ten, and the number of sharks killed by people is estimated in the tens of millions.
What it changes
- supports Oceanic sharks and rays are down by roughly three quarters since 1970 Well supported
- gives context to Greenland sharks live for centuries and do not reproduce until around 150 years old Well supported
What it does not show
Fishery-dependent data can reflect changes in fishing behaviour as well as in abundance.
Verified reference: doi.org/10.1038/s41586-020-03173-9 · peer reviewed
Wolf attacks on humans: an update for 2002–2020
This is the paper that lets a wolf page answer "are they dangerous" without either reassurance or sensationalism. The honest answer has a shape: the risk is genuinely very low, it is not zero, most fatal cases historically involve rabies rather than predation, and the variable that actually matters is whether wolves have learned to associate people with food. That last point makes it actionable — feeding wildlife is the mechanism, not proximity.
What it changes
What it does not show
Compiles reported incidents; under-reporting in some regions is likely and unquantified.
Reference unverified. · institutional report
Chest beats as an honest signal of body size in male mountain gorillas (Gorilla beringei beringei)
This is the useful replacement for the strength mythology, and it is a better story. The display is not bluster; it is a measurement other gorillas can take at a distance. Because frequency is set by the size of the resonating cavity, a small male cannot fake a large male's sound — the honesty is enforced by physics rather than by good faith. That is what "honest signal" means technically, and it is why the display works to settle contests without fighting, which is the thing the chest-beating gorilla of popular imagination is supposed to be doing and is not.
What it changes
- gives context to No measurement supports "ten times stronger than a human" Popular claim, unsupported
- supports The chest beat broadcasts body size, and cannot be faked Well supported
- supports A silverback is a group's protector and centre, not a permanently aggressive ruler Well supported
What it does not show
One subspecies at one site; other gorillas differ in habitat acoustics and social structure.
Verified reference: doi.org/10.1038/s41598-021-86261-8 · peer reviewed
Adaptation of sperm whales to open-boat whalers: rapid social learning on a large scale?
A remarkable use of an archive. Whaling logbooks were kept for commercial reasons, and they happen to record exactly what is needed: how many whales were seen and how many were caught, ship by ship, in a population encountering hunters for the first time. The collapse in strike rate over a few years is too fast for evolution and too widespread for individual learning, and the recorded change in behaviour — from a formation that works against orcas to fleeing upwind, which is the one thing a sailing ship cannot follow — is specific to the new threat. It is also, unavoidably, a reconstruction from records kept by people with other priorities, which is the honest limit on how hard to lean on it.
What it changes
- supports Sperm whales learn what to say from the whales they grow up with Well supported
- gives context to Around a third of the pre-whaling population, recovering very slowly Well supported
What it does not show
Logbook data were recorded for commercial purposes with inconsistent definitions of a sighting.
Verified reference: doi.org/10.1098/rsbl.2021.0030 · peer reviewed
Biofluorescence in the platypus (Ornithorhynchus anatinus)
Three museum specimens, and the finding is exactly that: platypus fur fluoresces. What has not been shown is that it matters. There is no evidence platypuses can see the effect, no evidence any other animal responds to it, and no evidence it does anything at all — and biofluorescence turns up in a great many biological materials for chemical reasons with no function attached. Subsequent work has found it in a long and growing list of mammals, which cuts both ways: it is more common than anyone thought, and being common makes a special explanation less likely rather than more. The honest framing is a real observation looking for a reason.
What it changes
- supports The fur does fluoresce under ultraviolet light, and nobody knows whether it means anything Emerging evidence
What it does not show
Three preserved specimens; preservation chemistry can itself alter fluorescence.
Verified reference: doi.org/10.1515/mammalia-2020-0027 · peer reviewed
Developmental genetics of color pattern establishment in cats
This settles the timing question the 2012 paper left open. The stripe exists before the animal has any hair to be striped, which is a genuinely surprising fact and a much better one than the version in circulation. It also explains the uniqueness that camera-trap work depends on: a developmental process of this kind produces a reliably tiger-shaped pattern that is never twice the same in detail.
What it changes
- supports The pattern is laid down in the skin before there is any hair — but the colour is in the hair Established
- gives context to No two tigers are marked the same, and this is how tigers are counted Established
What it does not show
Domestic cats; extended to tigers by homology, not by direct observation.
Verified reference: doi.org/10.1038/s41467-021-25348-2 · peer reviewed
On tree longevity
This supplies the distinction NatureHQ makes wherever an age appears: a crossdated ring count on a living stem, an estimate from growth rate, and the age of a clonal lineage are three different measurements with three different error bars, and they get compared as if they were one. It also contains a genuinely surprising pattern — the oldest trees are mostly in poor, cold, dry places, growing so slowly that they are hard to kill.
What it changes
What it does not show
A review of published ages rather than new dating.
Verified reference: doi.org/10.1111/nph.17148 · peer reviewed
Why this is not a news feed
A news item is complete in itself and disposable. Everything here is a way into something that lasts: the claim the finding bears on, the stance it takes towards that claim, what it does not show, and the earlier work it revises. If a claim’s evidence status changes later, this page changes with it, because nothing here is written separately from the page it describes.
The filters above deliberately do not change the address of this page. Every combination would otherwise become a thin, near-duplicate page competing with this one in search, which is exactly what NatureHQ’s routing model exists to prevent.