Behaviour and cognitionability
A sense of number
A bird, an insect and a plant, all responding to number — by three completely different mechanisms.
Many animals — and at least one plant — respond differently to different quantities. Almost none of them count in the human sense, and the difference between having a sense of number and counting is where most of the interesting detail lives.
This is the page that best shows why NatureHQ treats abilities as subjects. A carrion crow has neurons that fire for the number three. A honey bee places an empty card below a card with one shape. A Venus flytrap responds differently to two touches than to five — with no neurons at all. These are not variations on one ability; they are three different mechanisms that happen to be sensitive to quantity, and setting them beside each other makes the shape of the problem visible in a way no single subject page could.
Quick facts
- Counting is not required
- Most number-sensitive behaviour needs no enumeration at all
- The tell-tale sign
- A distance effect: larger differences are easier to tell apart
- No neurons needed
- The Venus flytrap counts action potentials with a decaying calcium signal
Where this appears
Every organism below has been linked to this page because the evidence links them. Each one carries its own evidence, and its own limits.
Ants
high confidenceintegrates stride counts to judge distance
Scope: Saharan desert ants
Crows
high confidencehas neurons tuned to specific quantities
Scope: carrion crows, in a trained task
Honey bee
moderate confidenceorders an empty set below one
Scope: western honey bees
Lion
high confidencecounts opponents by voice and compares against its own numbers before engaging
Scope: female lions, tested by playback in the Serengeti
One of the cleanest numerical-assessment experiments in any wild animal, because the intruders do not exist — nothing about them can vary except how many voices there are.
Venus flytrap
high confidencecounts electrical signals — with no nervous system
Scope: the Venus flytrap
The most important connection on this page, because it shows the ability does not require nerve cells at all.
Counting, as people do it, means assigning number words in order and treating the last one as the total. Almost nothing in the animal world does that. What is widespread is an approximate sense of quantity: an ability to treat "more" and "fewer" as properties of the world and act on them.
Researchers look for a specific signature. If an animal is treating quantities as points on a scale, it should find widely separated quantities easier to tell apart than close ones — the distance effect. Its presence is what turns "the animal chose the bigger pile" into evidence about number rather than about area or brightness.
- Carrion crow: individual neurons tuned to specific small quantities.
- Honey bee: an empty set ordered below one, with a clear distance effect.
- Saharan desert ant: distance judged by integrating stride counts.
- Venus flytrap: thresholds at two, three and five electrical signals, stored as a decaying calcium concentration.
Words used here
- Distance effect
- Quantities that are further apart are easier to distinguish. Its presence suggests the animal is treating number as a continuous scale.
- Subitising
- Recognising a small quantity — up to about four — instantly, without counting.
Everything below has been linked to numerical ability by the evidence, each with its own limits. The flytrap is the one worth pausing on: it demonstrates that number-based decisions do not require a nervous system, which reframes what the ability is for.
Do any of these mechanisms share a common origin?
Why it matters: Crows and bees last shared an ancestor before either had a brain worth mentioning. Whether their number sense is convergent or very deeply conserved is a real question.
Can honey bees really add and subtract?
Why it matters: The most widely repeated claim in this area rests on one study of fourteen bees, with substantive published criticism and no independent replication.
What would settle it: A replication with stimuli designed to eliminate correlated low-level cues.
The research behind this page
11 studies, newest first. Each one has a page explaining what it found and what it could not show.
Nature Plants
Calcium dynamics during trap closure visualized in transgenic Venus flytrap
Each touch produced a wave of calcium spreading from the trigger hair.
Science Advances
Numerical cognition in honeybees enables addition and subtraction
After extensive training, bees chose correctly above chance on both operations, including on some novel quantities.
Science
Numerical ordering of zero in honey bees
Bees chose the empty display above chance, and did so more reliably the larger the number it was compared against.
Current Biology
The Venus flytrap Dionaea muscipula counts prey-induced action potentials to induce sodium uptake
One electrical impulse does nothing.
Proceedings of the National Academy of Sciences
Neurons selective to the number of visual items in the corvid songbird endbrain
Individual neurons responded most strongly to a preferred quantity — some to one item, others to two, three or four — with responses tapering off for neighbouring numbers.
Science
Number-space mapping in the newborn chick resembles humans’ mental number line
Chicks preferentially approached the left panel when it showed a quantity smaller than the familiar number, and the right panel when it showed a larger one.
Animal Cognition
Grey parrot numerical competence: a review
The parrot labelled quantities up to six accurately, answered questions requiring a specific subset to be counted rather than the whole set, and showed some understanding of ordinality and of the absence of a quantity.
Science
The ant odometer: stepping on stilts and stumps
Ants on stilts overshot the nest; ants with shortened legs stopped short.
Animal Behaviour
Can honey bees count landmarks?
Search location shifted when the number of landmarks changed, indicating the bees used landmark count as part of their distance estimate — although flight distance dominated.
Animal Behaviour
Roaring and numerical assessment in contests between groups of female lions, Panthera leo
Lionesses approached the speaker when they outnumbered the simulated intruders and were much more cautious when they did not.
John Murray
Insectivorous Plants
Venus flytraps close only in response to mechanical stimulation of trigger hairs, and are insensitive to rain and to blown debris.
Where to go from here
Each of these follows from something on this page — a relationship in the evidence, a claim people ask about, or the next mechanism along.
They know your face. What that actually means is more specific — and more interesting — than the headline.
Crows · species group
A million neurons, a dance that gives directions, and an argument that took thirty years and a radar set to settle.
Honey bee · species
The colony solves problems no ant is solving. That is not a metaphor; it is the mechanism.
Ants · higher taxon
It counts to five. The counter is a chemical concentration that fades in half a minute.
Venus flytrap · plant
The obvious explanation for why lions live in groups was tested in the Serengeti, and it was wrong.
Lion · species
Can crows recognise human faces?
Crows · claim checked
How complete this page is, and what it is still missing
NatureHQ publishes its own gaps. This page is at 73% completeness against what we would call a finished subject, and was last reviewed on 2026-08-09. It carries 8 claims and answers 7 mapped search questions.
- 2 high-priority search intent(s) not yet covered
- no popular claim about this subject has been checked yet
- Primates, parrots and fish all have substantial numerical-cognition literatures not yet represented.
- Human infant work would be a useful comparison point and is absent.