A young reef octopus probes coral rubble while fish gather around it. One fish trails behind, ready to seize prey flushed from a crevice. Another darts in to defend a patch of reef. To a casual observer, both are simply fish getting close. To the octopus, a new field study suggests, they are very different encounters.
Researchers recorded 38 juvenile Octopus insularis in shallow water around oceanic islands off northeastern Brazil. Across 101 interactions involving 10 fish species, the octopuses did not strike indiscriminately. Arm slaps were part of fast defensive exchanges with territorial fish, alongside flinching and a distinctive body pattern. Nonterritorial followers remained with a hunting octopus for far longer.
The contrast is sharp: territorial encounters averaged about 14 seconds, while nonterritorial interactions averaged about 46 seconds, more than three times as long.
The researchers watched a reef neighborhood
The study was published online in Behavioural Processes in July 2026 by Michaella Pereira Andrade, Jennifer Mather and Charles Morphy D. Santos. Its title, “Hunting in a complex neighborhood,” describes the central problem. A foraging octopus is not moving through empty space. It is entering a shifting community of fishes with different diets, territories and reasons to approach.
The team snorkeled near juvenile octopuses and analyzed video of naturally occurring encounters. This was not a staged experiment in which researchers placed a fish beside an octopus or provoked attacks. The field design captured what happened during ordinary hunts, including the animals’ sequence of approaches, body-pattern changes and physical actions.
That naturalism is a strength, but it sets a boundary. Video can show which behavior followed another and how long an encounter lasted. It cannot directly reveal an animal’s intention, prove that a follower helped the octopus, or establish that an arm slap was punishment rather than defense.
The 101 recorded interactions included seven nonterritorial fish species and three territorial species. Those labels describe the ecological role fish played around the hunting octopus. They are not simple personality types, and the study does not imply that every member of a species behaves identically in every setting.
Territorial fish compressed the encounter
When territorial fish approached, the tempo changed. Their approaches were faster, and both participants were more likely to alter what they were doing. The fish used behaviors the researchers classified as jabs and swipes. The octopuses flinched, struck with an arm and sometimes displayed a contrasting “half blotch” body pattern.
Those exchanges ended quickly, after an average of roughly 14 seconds. Encounters with nonterritorial fish lasted about 46 seconds on average. That 3.3-fold difference is descriptive, not a stopwatch test of friendship. Still, it captures two recognizable situations: a short disruption by a fish defending space and a follower remaining beside an octopus while the hunt continues.
This is where the “underwater fight club” image needs restraint. The scientists did not discover a permanent combat ring or octopuses eager to hit anything within reach. They documented brief defensive sequences concentrated in interactions with territorial fish. The slap was contingent on the kind of encounter unfolding around the animal.
Territorial damselfish are small, but size does not remove conflict. A fish defending a feeding or shelter area can jab at an intruder much larger than itself. A juvenile octopus searching through that defended patch has to keep hunting, retreat or push the defender back.
An arm slap is not a generic punch
An octopus arm has no bones and no permanent joints, yet it can temporarily stiffen sections, send a bend along its length and sweep toward a moving fish. A 2023 field analysis of the octopus arm “slap” found that the animal combines translation and rotation into a controlled action. The movement can repel a fish while other arms stabilize the body against the reef.
The 2026 observations add ecological context. In these juveniles, arm strikes clustered in encounters that already contained territorial approaches, fish jabs or swipes and rapid behavioral changes. That pattern supports a defensive interpretation more strongly than an indiscriminate-aggression story.
Octopus arms are capable of many actions that can look similar in a short clip. A previous ScienceBlog report on wild octopus arms described combinations of bending, shortening, elongating and twisting used in exploration, locomotion, camouflage and hunting. “Slap” is therefore a behavioral category with context, not a synonym for every brisk sweep of an arm.
The half-blotch display also resists a simple translation. It occurred only in territorial encounters, making it a candidate component of warning or defense. But the study did not test whether fish read the pattern as a signal. Calling it an octopus threat display remains a reasonable hypothesis, not a decoded sentence.
Followers entered a different behavioral script
A nonterritorial fish was more likely to follow or circle. The octopus, rather than breaking off to repel it, generally continued searching for food. A blue tang was among the followers described around the young octopuses.
Why follow an octopus? Eight flexible arms can reach into cracks and disturb animals hidden in reef structure. A nearby fish may intercept prey that escapes. That makes following potentially profitable even if the octopus receives nothing in return. The relationship could be commensal, opportunistic or kleptoparasitic in some moments, and more mutually useful in others.
The present study did not measure prey capture rates with and without followers, so “cooperation” would outrun the evidence. Longer proximity shows tolerance and shared hunting space, not a negotiated partnership.
There is strong evidence that some other octopus-fish associations are organized. In 2024, ScienceBlog reported on multispecies hunting groups in which fish influenced where a group moved while an adult day octopus influenced when it moved. The underlying three-dimensional tracking research involved Octopus cyanea in another region and a different social system. It is valuable context, but it cannot be pasted wholesale onto juvenile O. insularis in Brazil.
The order of actions carried the result
The team did not rely only on totals. It used lag sequential analysis, a method that asks whether one coded behavior tends to follow another more often than expected. Territorial and nonterritorial encounters produced distinct, predictable chains.
In the territorial chain, rapid fish approaches and jabs or swipes appeared with octopus flinches, arm slaps and abrupt behavioral changes. In the nonterritorial chain, following and circling accompanied continued foraging. The difference was therefore not merely that one group stayed longer. The interactions were organized around different actions.
Sequence analysis is useful because a highlight reel can be misleading. Ten dramatic slaps placed back to back might make a young octopus look habitually violent. Restoring what happened immediately before and after each movement shows that the strike belonged to a narrow setting.
It does not read minds. It finds statistical structure in observations.
The same caution applies to color and texture. Octopus skin is a fast, active part of behavior, and an earlier ScienceBlog article described light sensing in octopus skin outside the central visual system. That physiological sophistication makes the half blotch interesting, but it does not turn a body pattern into proof of anger, fear or conscious messaging.
One species, one life stage, many open questions
The observations concern juvenile O. insularis at particular Brazilian reef sites. They do not establish that adults, other octopus species or juveniles in other habitats sort nearby fish in exactly the same way. Reef visibility, fish density, prey type and the structure of a defended territory could all change an encounter.
The averages need similar care. Fourteen seconds versus 46 seconds does not prove why the difference arose. A territorial fish may force a rapid reset, while a follower simply has repeated opportunities to remain. Fish identity and the stage of the hunt could influence both duration and behavior. Individual octopuses also appeared to vary in reactivity, but repeated tracking would be needed to show a stable personality rather than a temporary state.
Future experiments could compare hunting success with followers present or absent, test whether the half blotch changes fish behavior, and follow known individuals as they mature. More video across sites could also reveal whether the territorial sequence generalizes beyond the 38 juveniles in this sample.
For now, the careful conclusion is more interesting than a tale of random underwater violence. These young octopuses adjusted their responses to the ecological role of the fish beside them. They tolerated some neighbors through extended hunts and rapidly repelled others when territorial defense disrupted the search.
The reef was crowded, but the slaps were selective.
