Octopuses doing tricks on the internet and our search for non-human “intelligence”
People love to talk about octopus “intelligence.” The photographer specifies that the octopus wasn’t attacking him, but just wanted to steal his camera. Why? What use did he have for it?
If you look up “octopus” on youtube (which I hope you’ve done) you’ll come up with a billion videos of the animals solving puzzles. (Or carrying coconuts on their heads.) When I worked at Hatfield Marine Science Center, a story was told about the education center staff noticing that fish were disappearing from their tank, leaving no bodies but simply dwindling in number. One night, a security camera caught an octopus crawling across the floor into the fish tank, grabbing a snack and returning to its own tank. Whether it was intentionally trying to be sneaky or just didn’t like the conditions in the fish tank is unclear – nevertheless, stories like this are perfect fodder for support of cephalopod intelligence.
I’m clearly opening a huge can of worms here – or perhaps a jar of crabs – but can an octopus actually be described as “intelligent?” What is intelligence?
I’m on a cephalopod email list where there is an active discussion on the use of the word “intelligence” in reference to octopuses. One contributor, David Hill, wrote that one issue with the use of the word is its multiplicity of meanings depending on the circumstances. Here are some of his own examples of varying definitions:
1) Solves complicated problems
2) Integrates multiple sensory inputs into a behavioral sequence
3) Uses complex information to determine behavior
4) Displays ‘insight’ (?) when approaching a novel situation
5) Has a lot of neurons, or ‘higher’ interneurons
6) Demonstrates functional mastery of ‘difficult’ concepts
7) Behavior is highly visual, or otherwise a lot like human behavior
8) Uses a lot of longer-term memory, or can demonstrate this memory
9) Can communicate complex ideas to other members of the species
10) Has a complex communication system
11) Can communicate with people
Throughout this list, there are two main criteria: language and access to memory. However, the requirement for these two is consciousness. It’s hard to imagine storing and accessing information in the brain without having some sense of your own self. It’s consciousness that differentiates an automatic, mechanistic reaction by instinct from a thought-out (even if briefly) decision.
How can you qualitatively or quantitatively describe consciousness? We have no real language for this purpose because consciousness is purely experiential. For all I know, I am the only conscious being and you are all robots and this is some sick game. If we can’t even generalize across our own species, trying to imagine or articulate other levels of consciousness – a different mixture of thought and instinct – is nearly impossible. How can we know the self-awareness of a monkey or a dog?
In 2005, scientists from the Institute of Neurobiology in San Diego came up with a list of criteria for mammalian consciousness (article here). Now, I’m no neuroscientist, nor am I going to go through their entire list, but here are some of the most important physical brain criteria for determining consciousness:
- Variable EEG (electroencephalography) activity. An EEG is a test to measure the levels of electrical activity in the brain. An awake human has EEG levels of 20-70 Hz, while humans in unconscious states (sleep, under anesthesia, vegetative coma) have EEG levels of 4 Hz. If an animal has 2 different EEG states, it is presumed that they have some level of consciousness above zombie sleep. All mammals tested thus far exhibit this.
- Thalamocortical system. Many parts of the brain can be removed or damaged without loss of consciousness; not the thalamus or the cortex. These two parts of the brain (see colored portion of figure above) are well connected and significantly larger in humans than other mammals with similarly sized brains, leading scientists to believe that they are linked to consciousness.
- Widespread brain activity. When humans are actively thinking through a problem, there is activity all over the brain. However, when we’re performing tasks that have become routine or automatic, activity is restricted to smaller areas. Thus we could measure the brain activity of animals to see if they have different states of activity extension depending on comfort with a certain activity.
The authors suggest that these parameters, and 14 others that I’ve left out, can be tested in mammals to determine whether they are conscious, giving us information on the evolution of consciousness as well as supporting our constant anthropomorphization of cats, dogs, and walruses all over the internet.
All mammals are in the same class and show a great deal of homology between species, making it easy to compare them to our only known instance of consciousness, ourselves. Birds are more complicated because they have a different brain structure. However, as vertebrates, we are pretty closely related to birds. Studies have been able to map the bird brain onto the human brain, and the study of their consciousness is ongoing.
Octopuses, however, are a whole different ballgame. They don’t even have bones!! We clearly departed from the evolutionary pathway of octopuses a long time ago. Let’s quickly go over the 3 criteria above for mammalian consciousness to see if they apply for cephalopods.
- Variable EEG activity. In 1991, Bullock and Budelman measured EEG potentials in cuttlefish and found variation similar to humans, suggesting that cephalopods have multiple levels of consciousness. While this is the only study of its type and should be repeated, it is the strongest evidence thus far to support consciousness in cephalopods.
- Thalamocortical system. Clearly octopuses do not have a thalamocortical system. Humans and octopuses have very similar nerve cells, but their brains evolved separately so we will not find homology. The octopus does however have a central nerve cluster (“brain”) in its head, which is lateralized like vertebrates’ brains. It is about the size of the brain of a small vertebrate, and has huge optic nerves to support its fabulous eyes. No study has looked for a consciousness center, as we simply don’t know enough about the cephalopod nervous system yet to look for something that specific. Humans and octopuses do share some neurotransmitters, giving hope that we can do experiments using these as tools in the future.
- Widespread brain activity. Humans have a central brain with fewer nerve cells spread throughout the body. Surprise, surprise: octopuses have the opposite! Octopuses actually have more neurons in their tentacles than in their central brain. We humans visualize all our thinking as happening inside our heads – can we even imagine thinking in our arms? An octopus potentially having sensory nerves entangled with its “brain” nerve cells in its tentacles makes measuring electrical activity very complicated, don’t you think?
As you can see, it is very hard for us to draw conclusions about octopus consciousness based on anatomy. This is another example of how we humans only really know ourselves. It sometimes feels impossible to study organisms which have systems which don’t reflect our own; we don’t really know where to start.
But let’s get back to David Hill’s list of intelligent characteristics above: the two themes were memory and language. As the many videos on the internet show us, octopuses can learn. Scientific studies have demonstrated their ability to learn from experience and hold onto this knowledge as long as regularly presented with a similar challenge. (They do forget within just a few weeks if not required to utilize their learn skill again.)
Language is trickier. Octopuses and other cephalopods are solitary animals, interacting for territory and mating. They have an incredible ability to change the color and texture of their skin for camouflage and also use this for mating. Caribbean reef squid will flash different colors to signal enemies and mates. They will even halve themselves, each half painted a different color, to simultaneously signal to a female on one side and a competitive male on the other (see photo above and video here). But, once again – is this form of communication a conscious use of a physical language or instinct?
I have no answer here. I don’t know if octopuses are conscious, or are “intelligent.” They have many neurons; they can experiment, learn to solve problems and remember the solutions; they signal to one another through color; they use tools. What I think is more fascinating is our human need to find other organisms that are conscious. We constantly attribute empathy and emotion to animals that may not have the capacity. Maybe it makes us more kind to animals to imagine them as human-like, maybe it creates a space within which we can interact. Or maybe we just feel so alone in this big world… (Cue Bright Eyes now.)
In the end, I have a scientific brain: I was born to be skeptical of everything. So I do fall on the side of being skeptical about animal intelligence. We seem bred to seek intelligence in creatures and thus are predisposed to see it where it doesn’t exist. On the other hand, there wasn’t a switch flipped inside humans alone that turned on consciousness; there must be some continuum through the “lower animals.” (Oy, I hate that sort of language.) The truth is that we have no idea what a “lower level” of consciousness would feel like (or a higher one for that matter), or to have only a partial sense of self. I think it is a fascinating search and I commend it, but I don’t know if we will ever come up with an answer to any of these questions.
Edelman, D., Baars, B., & Seth, A. (2005). Identifying hallmarks of consciousness in non-mammalian species Consciousness and Cognition, 14 (1), 169-187 DOI: 10.1016/j.concog.2004.09.001
Mather, J. (2008). Cephalopod consciousness: Behavioural evidence Consciousness and Cognition, 17 (1), 37-48 DOI: 10.1016/j.concog.2006.11.006
Seth, A., Baars, B., & Edelman, D. (2005). Criteria for consciousness in humans and other mammals Consciousness and Cognition, 14 (1), 119-139 DOI: 10.1016/j.concog.2004.08.006