How Much of Your Will Is Yours? Zombie Ants, Hairworms, and Gut Bacteria
A parasite does not kill the host. It changes the heading. The body stays alive exactly as it was; only what it wants belongs to someone else.
Killing the host is the clumsy option.
A dead body does not move, and food that does not move rots.
A more efficient arrangement has evolved independently many times. Leave the body as it is and change the heading.
Only the drive to run is switched off
The jewel wasp (Ampulex compressa) stings a cockroach twice.
The first sting goes into a thoracic ganglion. The front legs are briefly paralysed. The point is to stop resistance, and it wears off in minutes.
The second is the problem. The wasp finds the subesophageal ganglion and the cerebral ganglia inside the head and delivers venom there.
Afterwards the cockroach is in this state:
| Condition | |
|---|---|
| Legs | intact |
| Muscle | alive |
| Response to stimulus | present |
| Starting to walk on its own | does not happen |
This is not paralysis. The ability to walk remains and only the initiation is off.
There is proof. Inject an octopamine agonist into a roach in this state and locomotion returns. What was switched off was not the motor circuit but the signal that starts it.
The wasp clips an antenna, drinks the haemolymph, and leads the roach by the stump. The cockroach walks into the burrow on its own legs.
One egg is fixed to a leg. The larva bores in and eats the tissues the host can survive without first. Heart and major nerves are left for later.
Not out of cruelty. Using a living meat store for a little longer simply paid better.
Guarding the ones that ate you
Wasps of the genus Glyptapanteles lay eggs inside a moth caterpillar. Dozens grow inside, then chew out through the body and spin cocoons beside it.
It does not end there.
The host does not die. It stops feeding, stays beside the cocoons, and swings its head at approaching predators. Then it dies.
In a 2008 study, predation losses on the pupae fell by more than half when this guarding occurred.
One or two larvae are thought to remain inside the host to sustain the behavior. Those that stay never become adults. They take the steering role and nothing else.
Someone else’s eggs, tended as its own
Sacculina carcini is a relative of the barnacle. It attaches to a crab, sends roots through the body, and forms an egg sac under the abdomen.
The host loses its ability to reproduce. And its behavior changes.
The crab tends that sac as though it were its own brood — grooming it, fanning water over it, and at the right moment flexing its abdomen to release the larvae into the sea.
The same happens in male crabs. The abdomen broadens as a female’s would, and the animal performs brooding behavior it has never performed.
Its own line is closed off, and it spends its labour propagating another’s.
Into the water
Hairworms (Nematomorpha) grow inside crickets and mantises. When mature they need water. The host does not.
Infected hosts jump into water. The worm emerges and mates there.
A 2005 proteomic analysis found altered expression of proteins related to neurotransmission and nervous-system development in the central nervous system of infected hosts.
One correction is needed here.
Hairworms do not parasitise humans.
Humans are not in the life cycle. The very rare human findings are accidental passage after swallowing contaminated water or an insect; nothing develops inside a person and no behavior changes.
Waiting all night on a blade of grass
Dicrocoelium dendriticum lives in the liver of sheep and cattle. To get there it must enter a grazing mouth.
The route: snail → ant → grazing animal.
Dozens of larvae enter the ant. But only one goes somewhere else: the subesophageal ganglion.
Once it settles, the ant’s behavior changes. At dusk it climbs to the tip of a grass blade, clamps its mandibles and hangs there all night — exactly where a grazing mouth passes.
In the morning, as the temperature rises, it releases and climbs down. Otherwise the ant would die of heat, and everything inside it with it.
And the one in the ganglion never becomes an adult. Even if the ant is eaten, it does not reproduce.
There is a separate post for steering, and it is a dead end.
Zombie ants — the fungus does not eat the brain
An ant infected with Ophiocordyceps unilateralis leaves the colony and descends toward the forest floor. It bites into a vein on the underside of a leaf and locks there. The position and timing are remarkably consistent.
It dies in that posture, and a stalk erupts from behind the head to release spores.
The most widely repeated explanation of this is wrong.
In 2017 researchers reconstructed an infected ant in three dimensions. The result:
| Location | Fungal cells |
|---|---|
| Around muscle fibres throughout the body | dense — wrapping fibres, forming an interconnected network |
| Inside the brain | none found |
The fungus does not enter the brain. It takes the muscle directly and acts on the nervous system through secreted compounds.
The mandible muscles atrophy so the bite cannot be released. The grip is not a decision; it is a lock that has broken shut.
“It eats the brain to control the ant” sounds stronger than it is, and it does not match the observation. What actually happens is stranger. The organ that decides is left alone and the organ that executes is taken.
The shared grammar
Lineages far apart from one another arrived at the same structure.
| Parasite | Switched off | Switched on | Payoff |
|---|---|---|---|
| Jewel wasp | walking initiation | being led | fresh larder |
| Glyptapanteles | feeding | guarding | pupal survival |
| Sacculina | reproduction | brooding | larval dispersal |
| Hairworm | water avoidance | moving to water | mating site |
| Dicrocoelium | returning to nest | clamping on grass | next host’s mouth |
| Ophiocordyceps | returning to colony | biting a leaf vein | spore release point |
One thing they share. None of them destroys a capacity. Every host is alive, moving, and behaving with precision.
What changed is not the capacity but what it is pointed at.
Are humans exempt?
Care is required here. This is where exaggeration enters.
Toxoplasma gondii reproduces sexually only in cats. In its rodent intermediate host, it needs the rodent to be eaten by a cat.
Infected rats lose their innate aversion to cat odour. That was confirmed in 2000 and the finding is solid.
Humans are another matter.
- Studies have reported associations with personality traits, traffic accident risk and schizophrenia risk
- But infection status travels with diet, hygiene and socioeconomic conditions. Confounding is heavy
- A large birth-cohort analysis in 2016 found no association with most psychiatric and behavioural measures
The honest summary: correlation has been reported; causation has not been established.
Holding that line is the only way to be believed on this subject. Push the rat result onto humans and the rest of the facts get doubted along with it.
Who decides what you want to eat?
One human-side area does have a concrete mechanism: appetite.
ClpB. A heat-shock protein made by E. coli, structurally similar to α-MSH, the peptide that signals satiety. A 2014 study reported that it can interfere with appetite-regulating circuits.
A molecule made by bacteria imitating a human satiety signal.
The channel has been identified too. Mice fed a particular lactobacillus showed changes in behavior and in brain receptor expression — and cutting the vagus nerve abolished the effect. Not a vague influence through the blood, but a specific nerve.
One common misunderstanding has to be cut off here.
Most of the body’s serotonin is in the gut, and gut bacteria regulate its production. True. But gut serotonin does not cross the blood-brain barrier. “Gut serotonin makes your mood” skips that step.
The real routes are more indirect: short-chain fatty acids, the vagus nerve, immune signalling, precursor supply.
Where the evidence stops
The boundary, drawn plainly.
| Status | |
|---|---|
| Behavioural manipulation in insects, crustaceans, rodents | established, with much of the mechanism known |
| Gut bacteria involved in human appetite | mechanistic evidence; human causation in progress |
| Gut bacteria and human mood or anxiety | strong in animals, early in humans |
| Microbes determining human personality or sexual behavior | no support |
That last line matters. Appetite has molecules attached to it. Carry the claim through to personality and the piece leaves science behind.
And what is established is already cold enough.
This article’s definition
People fear having the body taken — money, home, position.
The order nature shows is different. The purpose goes before the body does.
The jewel wasp’s cockroach did not lose its legs. Sacculina’s crab did not lose the ability to care for young. The zombie ant did not lose its bite.
Every capacity remained. It was simply spent on something else.
The most completely swallowed host is not the one bleeding. It is the one that has begun to feel another’s purpose as its own will.
Fixed here as this article’s definition: manipulation is not the removal of a capacity but the substitution of its direction.
And the reason the definition is uncomfortable is simple. From the inside it is indistinguishable. The cockroach presumably felt that it walked in by itself.
In short
- The jewel wasp switches off the signal that starts walking, not the ability; an octopamine agonist restores it
- The host caterpillar guards the pupae that ate it, halving their predation losses
- Male crabs infected by Sacculina brood and release another organism’s larvae
- Hairworms send hosts into water. They do not parasitise humans
- Of dozens of Dicrocoelium larvae, one goes to the ganglion — and never reproduces
- The zombie-ant fungus does not enter the brain; it wraps the muscle and locks it
- The shared grammar is substitution of direction, not destruction of capacity
- Toxoplasma and human personality: correlation reported, causation not established
- E. coli’s ClpB mimics α-MSH, and bacterial behavioural effects vanish when the vagus is cut
- Gut serotonin does not cross the blood-brain barrier; the routes are more indirect
This piece is rebuilt for search readers from “The One Who Swallowed the Host” of the serial essay The Great Corona Swindle. The original’s point of view is kept; primary sources and mechanisms have been attached to each case.
Frequently asked
- Do parasites control hosts by eating their brains?
- Usually not. A 2017 three-dimensional reconstruction of ants infected by Ophiocordyceps found fungal cells wrapped around muscle fibres throughout the body but none inside the brain. Manipulation appears to work through direct control of peripheral muscle and secreted compounds. Brain invasion is a popular misconception.
- Do hairworms parasitise humans?
- No. Humans are not in the nematomorph life cycle; they infect insects and mate in water. The very rare human findings are accidental passage after swallowing contaminated water or an insect. They do not develop inside a person or alter behavior.
- Does Toxoplasma change human personality?
- Not established. That infected rodents lose their innate aversion to cat odour is a well-replicated result. In humans, studies have reported associations with personality traits, traffic accident risk and schizophrenia, but confounding is heavy, and a large birth-cohort analysis found no association with most psychiatric and behavioural measures. Correlation reported; causation not established.
- Do gut bacteria really control appetite?
- There is mechanistic evidence. The ClpB protein produced by E. coli structurally resembles the satiety peptide α-MSH and has been reported to interfere with appetite-regulating circuits. And the behavioural effects of certain lactobacilli disappear when the vagus nerve is cut. Most of this is animal work; there is no basis for saying bacteria determine a person's eating habits or personality.
References
- Gal, R. & Libersat, F. (2010). A wasp manipulates neuronal activity in the sub-esophageal ganglion to decrease the drive for walking in its cockroach prey. PLoS ONE, 5(4), e10019. — The venom acts on the subesophageal ganglion, lowering only the drive to walk; an octopamine agonist restores it.
- Libersat, F. & Gal, R. (2013). What can parasitoid wasps teach us about decision-making in insects? Journal of Experimental Biology, 216(1), 47–55. — Separating motor capacity from behavioural initiation.
- Grosman, A. H. et al. (2008). Parasitoid increases survival of its pupae by inducing hosts to fight predators. PLoS ONE, 3(6), e2276. — The host caterpillar guards the pupae that ate it, halving predation losses.
- Fredericksen, M. A. et al. (2017). Three-dimensional visualization and a deep-learning model reveal complex fungal parasite networks in behaviorally manipulated ants. PNAS, 114(47), 12590–12595. — Fungal cells surround muscle but are absent from the brain.
- Hughes, D. P. et al. (2011). Behavioral mechanisms and morphological symptoms of zombie ants dying from fungal infection. BMC Ecology, 11, 13. — The death grip on a leaf vein, and the consistency of its position and timing.
- Biron, D. G. et al. (2005). Behavioural manipulation in a grasshopper harbouring hairworm: a proteomics approach. Proceedings of the Royal Society B, 272(1577), 2117–2126. — Altered central-nervous-system protein expression in hairworm-infected hosts.
- Berdoy, M., Webster, J. P., Macdonald, D. W. (2000). Fatal attraction in rats infected with Toxoplasma gondii. Proceedings of the Royal Society B, 267(1452), 1591–1594. — Infected rats lose their innate aversion to cat odour.
- Sugden, K. et al. (2016). Is Toxoplasma gondii infection related to brain and behavior impairments in humans? PLoS ONE, 11(2), e0148435. — No association with most psychiatric and behavioural measures in a population-representative cohort.
- Tennoune, N. et al. (2014). Bacterial ClpB heat-shock protein, an antigen-mimetic of the anorexigenic peptide α-MSH, at the origin of eating disorders. Translational Psychiatry, 4, e458. — E. coli ClpB mimics the satiety signal α-MSH.
- Bravo, J. A. et al. (2011). Ingestion of Lactobacillus strain regulates emotional behavior and central GABA receptor expression in a mouse via the vagus nerve. PNAS, 108(38), 16050–16055. — Vagotomy abolishes the bacterial effect — the channel is identified.
- Yano, J. M. et al. (2015). Indigenous bacteria from the gut microbiota regulate host serotonin biosynthesis. Cell, 161(2), 264–276. — Gut bacteria regulate gut serotonin synthesis — which does not cross the blood-brain barrier.
Published September 27, 2026