# Are octopuses as intelligent as popular stories suggest?

*A one-time literature review of five claims. Sources were retrieved on 2026-09-25 (UTC). Structured data is in [`evidence.json`](evidence.json).*

## Short answer

The core of the popular picture holds up, but it is narrower than the stories suggest. Controlled experiments across several species and labs show that octopuses learn and remember. Captive common octopuses do solve new physical puzzles, but the evidence shows learned, trial-and-error solutions rather than reasoning. Tool use rests on field observations of two species, and whether it counts depends on how tool use is defined. Recognising individual humans rests on one small, non-blind study. Octopuses have a well-documented "active" sleep stage whose skin patterns and brain activity resemble waking, but no study has measured dreaming. Most results come from a few captive animals of one or two species, so they should not be read as describing all octopuses.

**How to read this report.** *Fact* means what a study measured. *Interpretation* means what authors or I infer from it. Where only an abstract was accessible, sample sizes and methods are marked **unavailable** rather than guessed.

## Judgement scale

| Judgement | Meaning |
|---|---|
| **Well supported** | Several controlled studies, from more than one lab or species, agree on the claim as worded. |
| **Supported in a narrower sense** | Good evidence for a narrower version of the claim, which is stated. |
| **Preliminary** | One or very few small studies with major limits; no independent replication found. |
| **Unsupported by the evidence found** | No supporting primary study located. This does not show the ability is absent. |
| **Unresolved** | Evidence conflicts, or current methods cannot test the claim as worded. |

## Evidence table

| Claim | Key primary studies (type · species · n · setting) | What was actually observed (fact) | Main limitations | Judgement |
|---|---|---|---|---|
| **1. Use tools** | [Finn et al. 2009](https://doi.org/10.1016/j.cub.2009.10.052) · field observation · *Amphioctopus marginatus* · n unavailable in primary text (press release: about 20 animals, 4 long-distance carries) · wild, Indonesia. [Godfrey-Smith et al. 2022](https://doi.org/10.1371/journal.pone.0276482) · field video · *Octopus tetricus* · about 10 animals, 102 throws · wild, Jervis Bay, Australia | Veined octopuses carried coconut-shell halves, walking awkwardly, and later assembled them as shelter. Gloomy octopuses jetted silt and shells; 17 throws hit another octopus, and the throws that hit differed in vigour, material and arm position. | No experiments. Few individuals, one site each. Whether shelter-carrying counts as tool use depends on the definition. The authors call targeted throwing only "provisional". | **Supported in a narrower sense.** Some wild octopuses of two species use objects in ways that meet published tool-use definitions. |
| **2. Solve unfamiliar physical problems** | [Fiorito et al. 1990](https://doi.org/10.1016/0163-1047(90)90441-8) · lab experiment · *O. vulgaris* · n unavailable. [Richter et al. 2016](https://doi.org/10.1371/journal.pone.0152048) · lab experiment · *O. vulgaris* · 7 · captive, Israel. [Fiorito et al. 1998](https://doi.org/10.1007/s100710050015) · lab · n unavailable | Octopuses learned to unplug jars to reach crabs, getting faster over trials. All 7 learned to pull an L-shaped container through a tight hole; performance dipped whenever the task changed, then recovered. Pre-exposure to parts of the jar task did not help. | Small samples, one species. No control groups in Richter et al. None of the designs separates insight from trial and error; Fiorito et al. (1990) themselves attribute success to association or trial and error. | **Supported in a narrower sense.** Captive *O. vulgaris* learn new manipulation tasks flexibly; reasoning is not shown. |
| **3. Learn and retain information** | [Boal et al. 2000](https://doi.org/10.1037/0735-7036.114.3.246) · *O. bimaculoides* · n unavailable. [Hvorecny et al. 2007](https://doi.org/10.1007/s10071-007-0085-4) · *O. bimaculoides* · 10. [Shomrat et al. 2008](https://doi.org/10.1016/j.cub.2008.01.056) · *O. vulgaris* · n unavailable. [Tricarico et al. 2011](https://doi.org/10.1371/journal.pone.0018710) · *O. vulgaris* · 24 pairs. [Fiorito & Scotto 1992](https://doi.org/10.1126/science.256.5056.545) · *O. vulgaris* · n unavailable. All lab experiments. | Octopuses remembered a burrow location for a week and relearned it after reversal. 6 of 10 solved a conditional maze discrimination. Cutting or tetanizing a learning centre (the vertical lobe) changed learning speed and impaired next-day recall. Octopuses remembered a familiar conspecific for at least a day. Observers copied demonstrators' colour choice. | Several studies read as abstract only. The 1992 observational-learning result drew [technical comments](https://doi.org/10.1126/science.259.5101.1627), which I did not access, and I found no independent replication. | **Well supported** for associative and spatial learning with retention. Observational learning: **unresolved**. |
| **4. Distinguish individual humans** | [Anderson et al. 2010](https://doi.org/10.1080/10888705.2010.483892) · captive experiment · *Enteroctopus dofleini* · 8 · Seattle Aquarium | Over 2 weeks, one person fed each octopus and another poked it with a bristly stick. In late trials, octopuses moved toward feeders (13/14) and away from irritators (5/11), showed a dark "Eyebar" more to irritators, and aimed their water jet at irritators (6/8) more than feeders (1/8). Respiration differed only in the 4 larger animals. | One study. Testers scored their own trials and were not blind. The testers looked different (hair, face, build) and other cues were not isolated. The same tester appeared in every pair. No replication found. | **Preliminary.** It shows discrimination between two people linked to different treatments, not face recognition. |
| **5. Sleep states that might involve dreaming** | [Pophale et al. 2023](https://www.nature.com/articles/s41586-023-06203-4) · lab experiment + electrophysiology · *O. laqueus* · 3–10 per analysis, 9 probe recordings · captive, Okinawa. [Medeiros et al. 2021](https://doi.org/10.1016/j.isci.2021.102223) · lab video · *O. insularis* · 4 · captive, Brazil. [Brown et al. 2006](https://doi.org/10.1016/j.bbr.2006.05.009) · *O. vulgaris* | Quiet sleep is interrupted about hourly by about 1-minute "active" bouts. During these bouts, arousal threshold is raised, bouts rebound after deprivation, and skin patterns and brain activity (LFP, strongest in learning-related lobes) resemble waking. Quiet sleep shows spindle-like 12–18 Hz oscillations. *O. insularis* shows the same alternation, with eye movements. | Few small captive animals of two species. Recording required head fixation. Dreaming (subjective experience) was not measured, and the Nature paper does not claim it. | **Supported in a narrower sense.** Two-stage sleep: **well supported** in these species. Dreaming: **unresolved**. |

### Why each judgement

1. **Tool use.** Two independent field studies of different species, with no experimental manipulation; the label depends on the definition ([Finn et al.](https://doi.org/10.1016/j.cub.2009.10.052); [Godfrey-Smith et al.](https://doi.org/10.1371/journal.pone.0276482)).
2. **Problem solving.** Controlled studies from two labs show learning of new physical tasks. *Interpretation:* calling this reasoning goes beyond the data, because trial-and-error learning explains the results ([Richter et al.](https://doi.org/10.1371/journal.pone.0152048)).
3. **Learning and memory.** The strongest claim: several species and labs, retention of a day to a week, and brain-lesion evidence ([Boal et al.](https://doi.org/10.1037/0735-7036.114.3.246); [Shomrat et al.](https://doi.org/10.1016/j.cub.2008.01.056)). "They learn by watching each other" rests largely on one 1992 experiment.
4. **Human recognition.** Plausible, since octopuses recognise other octopuses in controlled tests ([Tricarico et al.](https://doi.org/10.1371/journal.pone.0018710)), but the human evidence is one 8-animal, non-blind study. Responding differently to a handler is not human-like face recognition.
5. **Sleep and dreaming.** The sleep correlates are robust within the species studied. "Dreaming" implies subjective experience, which no current method can measure in octopuses; lack of evidence is not evidence of absence.

### Where the stories come from

- **Coconut octopus** stories trace to the Finn et al. observations ([press release](https://www.sciencedaily.com/releases/2009/12/091214121953.htm)).
- **"Octopuses recognise people"** articles cite the single Anderson et al. study.
- **Dreaming** stories trace to a 2019 viral clip of "Heidi", one day octopus (*O. cyanea*) in a scientist's home aquarium; the narrator hedged ("If she is dreaming…", [Newsweek](https://www.newsweek.com/octopus-color-sleeping-dreaming-scientist-1461314)). Reporting later attached the label to the 2021 and 2023 papers, which do not claim it.
- **Jar-opening** videos show practised captive animals. They are anecdotes; the controlled evidence is Fiorito et al. (1990).

## How an octopus nervous system differs from ours

*Facts:*
- A human brain has about 86 billion neurons ([Azevedo et al. 2009](https://doi.org/10.1002/cne.21974)).
- An octopus nervous system has about 500 million neurons, and about 330 million of them lie outside the central brain, mainly in the arms. This is J. Z. Young's 1963 count as cited by [Gendreau et al. 2025](https://doi.org/10.1242/bio.062011); I did not read Young's original.
- The central brain wraps around the oesophagus.
- Each arm contains a massive axial nerve cord that is segmented and carries a sucker-by-sucker map ([Olson et al. 2025](https://doi.org/10.1038/s41467-024-55475-5), *O. bimaculoides*).
- Separate nerve cords link arms to other arms without passing through the brain ([Kuuspalu et al. 2022](https://doi.org/10.1016/j.cub.2022.11.007)).
- Brain structure varies between octopus species with habitat and lifestyle ([Chung et al. 2022](https://doi.org/10.1016/j.cub.2021.10.070)).

*Interpretation:* octopus control is partly distributed or "embodied", with arms handling much local processing ([Hochner 2012](https://doi.org/10.1016/j.cub.2012.09.001)). This design evolved independently of ours, so octopus and human intelligence are not points on a single scale.

## Three common exaggerations, corrected

1. **"Octopuses have nine brains."** There is one central brain plus eight large arm nerve cords. The cords hold most of the neurons and do substantial local control, but anatomists do not describe them as brains. The accurate version: *much of an octopus's nervous system is in its arms, which can coordinate some actions without the brain.*
2. **"Scientists proved octopuses dream."** The measured finding is a sleep stage whose skin patterns and brain activity resemble waking. The authors suggest it *may* involve practising skin patterns or reactivating waking activity, "reminiscent of" memory consolidation in vertebrates, and they state that its function still needs testing ([Pophale et al. 2023](https://www.nature.com/articles/s41586-023-06203-4)). Whether an octopus experiences anything during it is unknown, but that is not the same as showing it does not.
3. **"Octopuses recognise your face and hold grudges."** One study of 8 giant Pacific octopuses found they treated a feeder and a stick-poker differently after two weeks ([Anderson et al. 2010](https://doi.org/10.1080/10888705.2010.483892)). Faces were never isolated as the cue, observers were not blind, and "grudge" adds a motive that no study measured. Aquarium keepers' anecdotes are consistent with the finding but are not tests.

## The most useful unanswered question

**Does active sleep replay or strengthen specific waking experiences?** If learning a new task or skin pattern changes later active-sleep activity, and selectively interrupting active sleep impairs next-day memory, that would link learning to wake-like sleep. It would also replace the untestable "dreaming" question with a testable one. Pophale et al. note that understanding active sleep "will require studying whether patterns can be manipulated". A blind, cue-controlled replication of the human-recognition study would be the next most valuable work.

## Limits of this review

- Only freely accessible texts were used.
- Six studies (Fiorito 1990, 1998; Boal 2000; Shomrat 2008; Hvorecny 2007; Fiorito & Scotto 1992) were read as abstracts only.
- Finn et al. 2009 numbers come from its press release.
- The search was not systematic, so replications or conflicting studies may have been missed.

## References

All retrieved 2026-09-25 (UTC).

- Anderson RC, Mather JA, Monette MQ, Zimsen SRM (2010). Octopuses (*Enteroctopus dofleini*) recognize individual humans. *J Appl Anim Welf Sci* 13:261–272. https://doi.org/10.1080/10888705.2010.483892
- Azevedo FA et al. (2009). Equal numbers of neuronal and nonneuronal cells make the human brain an isometrically scaled-up primate brain. *J Comp Neurol* 513:532–541. https://doi.org/10.1002/cne.21974
- Biederman GB, Davey VA (1993). Social learning in invertebrates. *Science* 259:1627–1628. https://doi.org/10.1126/science.259.5101.1627
- Boal JG, Dunham AW, Williams KT, Hanlon RT (2000). Experimental evidence for spatial learning in octopuses (*Octopus bimaculoides*). *J Comp Psychol* 114:246–252. https://doi.org/10.1037/0735-7036.114.3.246
- Brown ER, Piscopo S, De Stefano R, Giuditta A (2006). Brain and behavioural evidence for rest-activity cycles in *Octopus vulgaris*. *Behav Brain Res* 172:355–359. https://doi.org/10.1016/j.bbr.2006.05.009
- Chung WS, Kurniawan ND, Marshall NJ (2022). Comparative brain structure and visual processing in octopus from different habitats. *Curr Biol* 32:97–110. https://doi.org/10.1016/j.cub.2021.10.070
- Finn JK, Tregenza T, Norman MD (2009). Defensive tool use in a coconut-carrying octopus. *Curr Biol* 19:R1069–R1070. https://doi.org/10.1016/j.cub.2009.10.052
- Fiorito G, von Planta C, Scotto P (1990). Problem solving ability of *Octopus vulgaris* Lamarck. *Behav Neural Biol* 53:217–230. https://doi.org/10.1016/0163-1047(90)90441-8
- Fiorito G, Scotto P (1992). Observational learning in *Octopus vulgaris*. *Science* 256:545–547. https://doi.org/10.1126/science.256.5056.545
- Fiorito G, Biederman GB, Davey VA, Gherardi F (1998). The role of stimulus preexposure in problem solving by *Octopus vulgaris*. *Anim Cogn* 1:107–112. https://doi.org/10.1007/s100710050015
- Gendreau AJ et al. (2025). How octopuses use and recruit additional arms to find and manipulate visually hidden items. *Biol Open* 14:bio062011. https://doi.org/10.1242/bio.062011
- Godfrey-Smith P, Scheel D, Chancellor S, Linquist S, Lawrence M (2022). In the line of fire: debris throwing by wild octopuses. *PLOS ONE* 17:e0276482. https://doi.org/10.1371/journal.pone.0276482
- Hochner B (2012). An embodied view of octopus neurobiology. *Curr Biol* 22:R887–R892. https://doi.org/10.1016/j.cub.2012.09.001
- Hvorecny LM et al. (2007). Octopuses (*Octopus bimaculoides*) and cuttlefishes can conditionally discriminate. *Anim Cogn* 10:449–459. https://doi.org/10.1007/s10071-007-0085-4
- Kuuspalu A, Cody S, Hale ME (2022). Multiple nerve cords connect the arms of octopuses. *Curr Biol* 32:5415–5421. https://doi.org/10.1016/j.cub.2022.11.007
- Medeiros SLS et al. (2021). Cyclic alternation of quiet and active sleep states in the octopus. *iScience* 24:102223. https://doi.org/10.1016/j.isci.2021.102223
- Olson CS, Schulz NG, Ragsdale CW (2025). Neuronal segmentation in cephalopod arms. *Nat Commun* 16:443. https://doi.org/10.1038/s41467-024-55475-5
- Pophale A et al. (2023). Wake-like skin patterning and neural activity during octopus sleep. *Nature* 619:129–134. https://doi.org/10.1038/s41586-023-06203-4
- Richter JN, Hochner B, Kuba MJ (2016). Pull or push? Octopuses solve a puzzle problem. *PLOS ONE* 11:e0152048. https://doi.org/10.1371/journal.pone.0152048
- Shomrat T, Zarrella I, Fiorito G, Hochner B (2008). The octopus vertical lobe modulates short-term learning rate and uses LTP to acquire long-term memory. *Curr Biol* 18:337–342. https://doi.org/10.1016/j.cub.2008.01.056
- Tricarico E, Borrelli L, Gherardi F, Fiorito G (2011). I know my neighbour: individual recognition in *Octopus vulgaris*. *PLoS ONE* 6:e18710. https://doi.org/10.1371/journal.pone.0018710
- Popular sources: ScienceDaily (2009) press release, https://www.sciencedaily.com/releases/2009/12/091214121953.htm ; Newsweek (2019) on "Heidi", https://www.newsweek.com/octopus-color-sleeping-dreaming-scientist-1461314