What is ‘intelligence’?
What drives the Evolution of Intelligence?
How do we measure Intelligence?
Evidence on Coleoid Intelligence from Behavioural studies
Specific Coleoid Intelligence Evolutionary routes – Why Coleoids?
Implications of Advanced Intelligence
How are Neuroanatomy and Intelligence linked?
Evidence on Coleoid Intelligence from Physiological/ Anatomical/ Neurological Studies
How Diverged is Coleoid Anatomy? Comparisons of Neurobiology across taxa
Discussion & Conclusions
How intelligent are coleoid cephalopods? Why has their intelligence evolved, and how do neurological or anatomical features enable it? This paper reviews behavioural and neurological studies of to help answer these questions.
Firstly, for background I will summarise scientific definitions of intelligence, how it is measured in animals, and the factors typically driving its evolution. I will then move onto studies focussing on coleoid cephalopods (octopus, cuttlefish, squid) using behavioural experiments and observations. I will summarise current hypotheses of evolutionary pathways – why coleoids are intelligent, and briefly consider potential moral and legal implications.
I will then move onto the neurological (how) section of the review, which aims to reinforce behavioural findings with physiological results, demonstrating connections between advanced intelligence and complex neurobiology.
The paper concludes that Coleoid Cephalopods have remarkably advanced cognition, being not only the most intelligent invertebrates, but in many respects comparable to vertebrate species such as primates, cetaceans, and corvids. Furthermore, coleoid neuroanatomy suggests some convergence of brain structures between coleoids and intelligent vertebrates, despite deeply diverged taxa.
Cephalopods are an invertebrate animal family within the phylum Mollusca, which contains two extant subclasses: Nautiluses and Coleoids (Catalani, 2008). Nautiluses are more basal in terms of intelligence and neurological complexity (Mather, 2007), influencing the focus of this paper to the Coleoid group. The subclass Coleoidea contains hundreds of species of squid, cuttlefish, and octopus. Coleoids are referred to as the ‘soft-bodied’ cephalopods, due to their internalised or absent shell, which makes them distinct from Nautiluses, utilising a complex aragonite (calcium carbonate) shell for buoyancy and protection (Buchardt, 1981).
To investigate intelligence in a subclass of organisms, we must firstly define what we mean by this term. Intelligence is a concept with many definitions, often specific to the single organism being discussed.
However, biologists must identify a common characteristic or set of skills, which can be implemented across all taxa, with the aim of setting an overall definition of ‘intelligence’. This comes with many challenges and has been highly debated within the scientific community; not only do all animals have varying neurological restraints, but also differing physical specialisations. This issue is summarised by the saying ‘if you judge a fish by its ability to climb a tree, it will live its whole life believing that it is stupid’, which states that each species is highly skilled in its own way, depending upon it’s ecological niche. Consequently, biologists are wary of investigating species-specific traits when defining intelligence, but instead focussing on mental and behavioural flexibility.
In the first section of this paper, I will review the behavioural studies of coleoid intelligence, and then consider why they might have evolved this high-level ability.
What evolutionary niches have shaped coleoid biology?
Do their abilities open new niches?
Why are they so different from other molluscs and invertebrates?
What factors led to the evolution of large, complex brains in other organisms, and are there any parallels suggesting that elevated cognition is convergent under certain conditions?
The second part of this report will focus on the actual neurobiology of coleoids. The neuroscience segment is going to ask how these creatures are so intelligent, by analysing any physiological, anatomical, neurological, and chemical evidence currently available.
Therefore, the structure of this report is essentially two-fold, first going into the why questions of coleoid intelligence, then following with the how investigation into their neurobiology.
Then, collating these findings to present a cohesive argument as to why and how coleoids are so intelligent. I also want to discuss whether coleoid intelligence could have any further implications, such as legislation changes for their treatment, but more fundamentally, whether we need to continue updating what we understand as sentient life. In summary, as stated by Catalini: “intelligence cannot be measured directly, thus, it is typically estimated through morphological and behavioural proxies; namely brain features, and behavioural flexibility” (Catalani, 2008, p. 47), both of which will be examined in this report.
What is ‘intelligence’?
The concept of intelligence is complex in biology, as definitions are open to interpretation across taxa. There is no quantitative definition of what intelligence means (Troy, 1991); no nominal scale to measure behaviours or neurobiology for their level of intelligence.
When studying animal cognition, we must either define a set of species-specific traits (when investigating an individual against the average intelligence of its species), or accept a more general collection of qualities, which can be used to compare even the most diverged of species (Kamil, 1994).
A 2007 paper collated hundreds of definitions of the word ‘intelligence’ from many sources: psychologists, AI developers, biologists, and dictionaries, aiming to pull commonly occurring features. This study concluded that “intelligence measures an agent’s ability to achieve goals in a wide range of environments” (Legg, 2007, p. 22). This is reinforced by Sarah Farris in 2015, who states that animal intelligence should be investigated using the idea of ‘general intelligence’: this concept includes all behaviours which demonstrate the ability to carry out innovative, flexible actions, in response to novel problems (Farris, 2015).
These interpretations of intelligence theoretically protect us from the dilemma described in the book ‘Evolution of the Brain and Intelligence’: viewing traits which evolved under species-specific selection pressures to a niche, as high-level intelligence (Jerison, 2012) (Figure 1).
A secondary pitfall in intelligence studies is the assimilation of human-like behaviour and advanced intellect. The human condition seeks anthropomorphisms, but few studies aim to answer why we are fixated on discovering humanoid behaviours in other species, other than the reasoning that we find it difficult to relate to beings dissimilar to us (Horowitz, 2007). Viewing animal behaviour through the lens of human experience often leads to misunderstanding of innate behaviours. For instance, the interpretation of animal signals as analogous to humanlike-language is common (Wynne, 2004): examples include the ‘dance’ communication of honey bees (Wynne, 2004), and memorising of words by dolphins, dogs, and chimpanzees (Herman, 1986) (Gardner, 1989). These experiments mistook memory and recall for true understanding and utilisation of language.
Similarly, this same outlook leads to certain animal behaviours being overlooked, despite their complexity, because of how ‘alien’ they appear (Wynne, 2004).
In aiming to avoid these errors, when addressing the question of intelligence, I will be referring to an organism’s ability to form novel behaviours, and apply intelligence in a flexible manner, enabling the skills of: problem-solving, decision making, learning, planning and foresight, communication, tool use, self-recognition, play, ‘theory of mind’, and the development of ‘personality’.
What drives the Evolution of Intelligence?
Intelligence, as with all other animal characteristics, arose by process of evolution (Jerison, 2012). Here I will examine the evolutionary conditions which generally lead to advanced cognition. Although speciation occurs under diverse circumstances, due to varying biotic (energy-producers, competitors, predators, parasites) and abiotic factors (light, water, temperature), there are common themes which appear conducive to the development of higher intelligence (Farris, 2015).
A 2002 study aimed to identify reasons for cetaceans and primates developing complex neurobiology, despite undergoing separate evolutionary histories in drastically different environments (Marino, 2002). This study concluded that there must be general principles that underlie the development of advanced cognition. Despite these families being deeply diverged, and adapting to physically different environments, they appear to converge on many complex behaviours, such as elaborate social relationships, detailed communication, and the ability of self-recognition (Marino, 2002). Due to this, they commonly represent alternate pathways by which species can arrive at similar cognitive flexibility.
Additionally, the African Grey Parrot was investigated for cognitive capacities (Pepperberg, 1990) (Figure 2). It’s concluded that their results are evidence for a similar intelligence levels to basal primate species, and in some tasks, comparable to chimpanzees (Pepperberg, 1990). This study produced quantifiable (statistically significant) results for many complex tasks (such as long-term memory, object permanence, and self-recognition) over a period of 12 years.
When referencing avian intelligence, an additional parrot species heavily researched is the New Zealand Kea. When a 2006 study focussed on Kea, the aforementioned principle is applied: where ability to act in response to stimuli flexibly is key to defining intelligent behaviour (Huber, 2006). Using this definition, these studies concluded that the same mechanism which often explains the primate leap in intelligence can also be applied to certain birds: namely, the Technical Intelligence hypothesis. This theory claims that the principle driving force behind advanced cognition is the selection pressure of an increased efficiency and flexibility in foraging techniques (Huber, 2006, p. 3).
This theory has been cited in intelligence studies directly referencing coleoids, which will be discussed later.
Corvids (crows, jackdaws, ravens, jays) are also significant when discussing bird intelligence. They have been shown to exhibit complex tool use, social relationships, and the ability to travel mentally in time and space (Emery, 2004). Corvids are argued to have ape-like intelligence in multiple studies (Emery, 2004) (Clayton, 2005) (Roth, 2005). Along with parrots, corvids have forebrains that are extremely large in relation to their body size; particularly the areas thought to be analogous to the mammalian prefrontal cortex. This enlargement (encephalisation) of the “avian prefrontal cortex” – nidopallium and mesopallium- may reflect an increase of intelligence in birds (Taylor, 2014).
These studies, and many others, show recurring themes relevant to the evolution of intelligence. In order to evolve a new trait such as intelligence, species must enter an ‘adaptive zone’, which is a set of ecological/environmental factors that allow the diversification of homogenous organisms, into novel forms in different niches, leading to radiation and proliferation of these new traits (Roth, 2005).
What we can conclude is that, advanced intelligence – being defined by behavioural flexibility and abstract cognition – has evolved across many lineages (Roth, 2005). The general scientific consensus on the cause of the evolution of this trait, is the requirement to make complex decisions and utilise variable foraging techniques, as part of a generalist diet.
These factors can be seen when looking at the species named above, as well as other relatively smart species, such as rats and dogs (Bitterman, 1965), which are all omnivores and implement a wide variety of foraging methods. Thus, again suggesting that flexibility rather than specific skills is the key to intelligence.
How do we measure Intelligence?
Before analysing Coleoid intelligence, we must define the term and assess its causes, but also examine how scientists have investigated it previously. Which methodologies are common and successful in this field, are there standards by which other species have been assessed?
To overcome the communication barrier between human researchers and animals, methodology must be engineered to reduce of control variables. This usually leads to experiments being carried out in vitro conditions, as observations and anecdotes from the field (in vivo) contain too many variable factors (Sopher, 2015).
Some landmark experiments in animal intelligence include:
- Edward Thorndike’s 1898 consolidation of theoretical and empirical contributions. Thorndike’s approach is systematic and comprehensive experimentation using a variety of animals and tasks within a laboratory setting, which was ground-breaking in the field of animal cognition at the time, and paved the way to making the study of animal behaviour a respected field in experimental laboratory science (Thorndike, 1898).
- The ‘mirror test’, in which self-recognition and sense-of-self is the focus. A mark is placed on an animal’s face using a non-odorous and non-irritant dye, and when shown a mirror, some species will attempt to remove the mark from the mirror, while others will recognise the reflection, and try to remove the mark from their own face (Gallup, 2002).
- The ability to understand and act upon principles of delayed-gratification was exhibited in Chimpanzees in 2018, which is often equated with higher intelligence, due abilities of foresight and future planning (Beran, 2018).
- Comparative studies on Monkeys, Dogs, and Cats, which investigated adaptive intelligence, through learning and tool-use in 1915 were greatly significant. These studies showed to what extent species could solve a task in order to receive a food reward (Sheperd, 1915). Similar methodology was replicated in 1998, with experiments which put hungry animals in enclosures from which they could escape by a problem-solving action, such as pulling at a loop of string, pressing a lever, or stepping on a platform.
It was recorded how long each species on average took to create an association between the action and reward (Thorndike, 1998).
- Collective intelligence in species of social insects, fish schools, and bird flocks has shown that each of these utilises collective decision making, through positive feedback loops, as well as complex communication between sometimes hundreds of thousands of individuals, all acting towards a common goal (Pratt, 2010).
- In 1980, Roger Thomas et al. aimed to firstly quantify and list the features that constitute our understanding of ‘intelligence’ (Figure 3). Then from here, assess a variety of species against this hierarchy, with the aim of collating a selection of animals in order of intelligence (Thomas, 1980).
Many methods of quantifying a species’ intelligence exist in experimental science currently, and each of the studies described provides an example of methodology that has led to successful results in this field of research.
This area of research comprises of many thousands of experiments, from which a few impactful examples have been selected, in order to illustrate firstly the common themes in intelligence studies, but also demonstrate the level of knowledge we currently possess, acquired from behavioural studies.
Evidence on Coleoid Intelligence from Behavioural studies:
An important distinction must be made between types of studies: behavioural experiments use living subjects and observe their actions, while neurological methodologies use dissection, implanted electrodes, MRI’s, and chemical/hormonal measurements in the nervous system.
First, I will collate notable behavioural studies in the area of coleoid cognition.
John Catalini’s 2008 review of Cephalopod Intelligence states: experiments on coleoids have been carried out for ~60 years, the majority have used Octopus species as subjects, specifically Octopus vulgaris, which is as close as we get to a model-organism in this family (Catalani, 2008).
Amodio’s review named problem solving and tool-usage as hallmarks of intelligent behaviour, along with complex anti-predatory methods, and high sociality (Amodio, 2018); similarly, I will be structuring my evaluation thematically rather than study-by-study.
Learning is defined as the acquisition and storing of information, for future application (Amodio, 2018). Many behavioural studies investigate coleoids’ ability in this skill.
A 1992 experiment demonstrated that octopus can be taught cause-and-effect conditioning (Fiorito, 1992); specifically, individuals were taught that selecting a red button, opposed to the white button that was also available, resulted in a food reward, and an incorrect selection would cause a punishment. The learning was considered successful when the individual selected the correct button 5 times consecutively. The conditioning was ultimately successful in every individual (Fiorito, 1992).
To add further complexity to this experiment, researchers also wanted to ascertain whether coleoids had the ability to learn from each other, despite their limited social repertoire. To assess this, they placed a control octopus in an adjacent tank, to observe the conditioned individuals carry out the button selection and receive rewards, then placed them in the assessment tank. The unconditioned octopus, then consistently selected the button which they had observed produced a reward. The results of this study suggested that not only are coleoids successful at conditioned-learning tasks, but also learn from one another through observation alone (Fiorito, 1992). Methodologies analogous to this have been repeated since 1992, each producing similar outcomes, reinforcing the conclusion that coleoids, specifically octopus, have a high propensity to learn and retain positively-reinforced information (Boycott, 1995) (Mehrkam, 2019) (Mather, 2007).
Another assessment of coleoid learning from 2000, aimed to solve the mystery of how Octopus stray far away from their den to forage, but are able to instantly relocate their home after many days have passed (Boal, 2000). This study showed that Octopus bimaculoides learn the spatial location of their burrow within 24 hours, by memorising landmarks and distance between them, and subsequently retained this information for 7 days (Boal, 2000).
A similar study in 2008 examined coleoid spatial learning: Christelle Alves observes the relationship between cephalopod sensory organs and their learning abilities, claiming that their vertical lobe plays a similar role to the mammalian hippocampus, storing information as memories. Alves proved both octopus and cuttlefish species utilise landmarks, substrate texture, and spatial measurements to avoid artificial detours researchers placed as diversions (Alves, 2008), and still relocate their den after multiple days.
Such studies provide us with evidence suggesting coleoids have the ability to learn information – visual, spatial, navigational, as well as causal relationships – for many days, despite the learning process in these experiments only lasting hours.
A second form of intelligence, once we have confirmed an organism’s ability to learn, is to assess their capacity to then process information, and utilise it flexibly to achieve goals (D’Zurilla, 1971).
Coleoid foraging strategies often require multi-level problem-solving in order to reach the food reward. They are highly diverse and skilled predators, but some octopus specialised to feeding on bivalves (clams, oyster, mussels, etc.), which are heavily armoured prey items (Fiorito, 1990). Coleoids have developed complex methods to open their shells, from using their arms to pry them, to drilling holes in the shells, and injecting toxins. Additionally, the method chosen will differ depending on the size and species of the prey (Fiorito, 1999). This suggests that coleoids have the ability to assess a situation and apply the most appropriate technique to solve it.
A 2016 study concluded that coleoids possess high flexibility when presented with complex problems; specifically an L-shaped container which contained a food-reward, was placed behind an opaque plastic partition, and must be both pushed and pulled at specific orientations in order to retrieve it through the complex hole. Every Octopus vulgaris individual was successful, after a series of trial-and-error attempts (Richter, 2016).
Additionally, octopus have been shown to solve complicated mazes, and open containers that require multi-level understanding of obstacles (Figure 4) (Fiorito, 1990).
Another mechanism for problem-solving is the use of tools, specifically the potential for ‘composite tool-use’, which indicates a benchmark of higher-level thinking across species, due to the multiple elements used in combination to solve a problem (Amodio, 2018).
Specific cases with coleoids exist in the field and lab conditions; a notable example is the observation of Veined Octopus using two halves of coconut shells that have been found on the ocean floor, holding them close to its body using its arm suction, as a means of protection, as a ‘suit of armour’ (Finn, 2009). This observation shows composite thinking: firstly, the shells were identified as a useful tool, amongst other innocuous seafloor debris (displaying behavioural innovation), then each shell is examined for suitability (size/shape), before being transported for use in the future, showing planning, and an ability to understand delayed-gratification, as the transportation of the shells poses a threat to the octopus (due to conspicuousness therefore alerting predators), but the benefits outweigh this risk in the long run, due to the protection the coconut shells provide (Finn, 2009) (Amodio, 2018).
Another instance of tool-use is the coleoid use of their water ‘jets’ (Amodio, 2018). Many coleoids squirt water jets from their funnels (using water as a tool) for a variety of purposes: to push away fish that scavenge on their prey, burrowing techniques, or remove food remains from their dens (Mann, 2013).
Communication and Social Behaviour
The limited social behaviours in coleoids may appear an outlier, as most large brained animals all exbibit complicated social interactions and relationships, and this is thought to be a key factor to the evolution of intelligence (Amodio, 2018). However, despite coleoids’ often solitary existence, there are still instances of complex communication.
Coleoids have the ability to alter the colour and texture of their skin, which is primarily used as an anti-predatory tactic (Amodio, 2018), but also serves communicative functions (Hanlon, 2018). It was previously thought that the ability to change colours was simply a hard-wired reflex to a sensory input, such as proximity of a predator or potential mate, rather than a conscious skill (Catalani, 2008). However, recent evidence shows octopus can choose, or be trained, to break camouflage when offered a food reward, demonstrating there is decision-making involved (Hanlon, 2018).
With this knowledge, we can examine a phenomenon seen in Mourning Cuttlefish (Sepia plangon), where both honest and deceptive signals are displayed simultaneously (Umbers, 2016) (Schnell, 2019). When in the presence of a fertile female and a rival male, the male of this species can split its body longitudinally, and “express courtship displays towards a receptive female on one side of their body and deceptive female colourations towards the rival male on the other” (Amodio, 2018) (Brown, 2012) (Figure 5). However, this tactic is abandoned in the presence of multiple females, possibly because the masquerade is broken (Brown, 2012), further reinforcing the notion that some level of cognition is involved in the skin displays of coleoids.
A 2017 study aimed to classify coleoid communication, and concluded that Oval Squid utilise 27 categories of communicative body-patterning, which were observed both in the wild and captivity (Lin, 2017). Each of these categories conveyed different information using displays of the skin (Lin, 2017).
Further proof of coleoid communication is the apparent co-operative relationship between octopus’ species (Octopus cyanea, Octopus macropus) and Groupers. True inter-specific cooperation is unusual in nature (Vail, 2013), but these species have found a beneficial niche by which the high manoeuvrability of the octopus is used to move through small crevices in coral reefs, while groupers protect the octopus from predators, then eating the fish that are flushed by the moving octopus (Unsworth, 2012). The behaviour is initiated by an octopus, who circles the grouper, signalling the process to begin (Hanlon, 2018). It appears that this is a learned behaviour, which benefits both actors, and increases food diversity and feeding efficiency for both species (Diamant, 1985). Similar behaviours have been observed across the world, with instances in the Australian barrier reef, Egyptian red sea, and the Mediterranean (Unsworth, 2012) (Diamant, 1985) (Quetglas, 2015). True cross-species mutualism is uncommon, and more research is needed on such relationships to determine the cause, but regardless of causation, there is clearly a complex level of planning and communication that goes into this foraging strategy (Hanlon, 2018).
So, although coleoids don’t live in congregations or practice regular sociality, they have the capability to consciously display complex communicative information in natural environments and through conditioning (Hanlon, 2018).
Additional Higher Abilities
This section evaluates findings of behavioural studies focussed on more nuanced characteristics of high intellect, which go beyond functionality, and into innovative or creative behaviours. Examples are playfulness, self-identification, and ‘consciousness’.
Play in animals is defined by Burghardt, who devised 5 criteria to determine whether a behaviour was truly ‘play’, which can be condensed into spontaneous and pleasurable acts ‘done for its own sake’, which differ from functional behaviours being exaggerated or modified (Kuba, 2003).
Several species of octopus have been observed carrying out play-like behaviours: a 2003 study gave 7 Octopus vulgaris a series of floating objects for one hour daily (Kuba, 2003), and classified their subsequent behaviours into categories, from exploratory interaction, to entirely fulfilling Burghardt’s criteria. Each individual showed multiple actions of play, from pushing and pulling the buoyant object, to bouncing, and dragging it underground (Kuba, 2003).
Mather’s 1999 experiment used floating bottles as stimulus to octopus (Mather, 1999). Two individuals showed play through utilising their water jets to push the floating bottle into the current of the tank filter, chasing it, then repeating the process; a seemingly futile exercise, but one that appears enjoyable nonetheless (Mather, 1999).
Many studies replicate these findings, recording actions that appear spontaneous, pleasurable, unaffected by factors such as hunger and age, and ultimately, functionally pointless (Burghardt, 2010) (Kuba, 2006) (Kuba, 2014). One hypothesis on why coleoids play is to enable learning, similar to other species, where play-fighting teaches hunting etc (Kuba, 2014). Nevertheless, the process of play displays curiosity and a break-away from an existence of pure function.
Other higher-intellect abilities observed in coleoids include apparent individualised personalities (distinct temperamental differences & unique responses to identical stimulus across populations) (Sinn, 2001), theory of mind (Mather, 2008), and most importantly, a potential level of ‘primary consciousness’ (Mather, 2007).
Jennifer Mather has developed many pioneering experiments: her 2007 paper ‘Cephalopod consciousness’ is core literature on the topic (Mather, 2007). She claims that despite cuttlefish failing ‘the mirror test’, they may still possess a simpler sense of self, as this failure may be due to coleoids not possessing the relevant neural anatomy to allow detailed information of the individual’s position in space (Mather, 2007). However, in other experiments, Squid have been shown to pass ‘the mirror test’ to the same level as chimpanzees (Ikeda, 2009). Mather’s paper concludes that if we define intelligence by flexibility and broad learning abilities, then we can consider adding ‘cephalopods to the groups of animals that might have primary consciousness’ (Mather, 2007).
Taking all the data discussed in the behavioural section, the weight of evidence appears to suggest, that although the intelligence of coleoids is not equal with that of primates in terms of consciousness and self-awareness, lower-level cognitive tasks such as problem-solving, learning, and forms of communication, can be performed at an advanced degree; at some points out-competing other highly intelligent organisms (cetaceans, monkeys).
However, more research is needed to determine the quantified intelligence level of coleoids through behavioural studies, as the field of animal intelligence is relatively young to science, and coleoids have only been studied in the last 50-60 years (Catalani, 2008). Furthermore, majority of these studies focus on Octopus vulgaris, which inevitably limits our knowledge of the family as a whole. These issues will likely resolve over time, as our research widens and lengthens.
Even so, I believe that from the evidence examined above, we can conclude that coleoid intelligence is abnormally high and shows potential to be a form of primary consciousness.
Specific Coleoid Intelligence Evolutionary routes – Why Coleoids?
After examining alternative evolutionary routes to advanced cognition, and defining and measuring coleoid behavioural intelligence, I will combine this with the aim of identifying the specific coleoid evolutionary pathway.
As previously stated, one of the leading theories behind intelligence is the Technical Intelligence Hypothesis (Byrne, 1997), which has parallels with the Ecological Intelligence Hypothesis (Amodio, 2018). Both of these pose that advanced cognition developed in response to the challenges of complex foraging and feeding strategies, such as being a generalist predator, with dispersed prey, that requires flexible extraction techniques (Byrne, 1997).
In contrast to this, the Social Intelligence Hypothesis (Schaik, 2011), which names the demands of group living (“such as maintaining social bonds, deception, cooperation, or social learning from conspecifics” (Amodio, 2018, p. 46)), as the cause of high intelligence development.
There is evidence which supports both hypotheses in the case of coleoids, and contributing factors from these theories are applicable here. Furthermore, coleoid intelligence appears to be both convergent and divergent, with their high behavioural flexibility and learning abilities, contrasting with their lack of social bonding, fast-life history, and low reproductive care (Amodio, 2018).
Most coleoid species live 12-18 months, only surviving a singular breeding cycle (semelparity), additionally they provide little-to-no maintenance to their offspring after laying eggs (Catalani, 2008). Contrasting with the majority of other intelligent species, which often have extensive gestational periods, multiple reproductive windows (iteroparity), and comprehensive protection with teaching of skills to offspring (Amodio, 2018). Despite this, coleoids converge with other intelligent species, which leads the question, why coleoids?
The leading hypothesis to answer this question links back to their cephalopod relatives, the nautiluses. They differ greatly, with their protective shell, slow mobility/reaction time, limited sensory organs, and extended lifespan (Catalani, 2008). Although research on Nautilus is difficult, due to their deep-water habitats, scientists have attempted to recreate Pavlovian style methodologies which were successful with coleoids, on nautiluses in captivity (Basil, 2008). But they showed little or marginal responses to novel stimulus (Basil, 2008). Although nautiluses were proven to have some capacity for long-term memory, it is understood that their other cognitive abilities are much lower than their close relatives (Basil, 2008).
The leading theory behind this jump in cognitive complexity is directly tied to their distinct morphological differences(Figure 6); the coleoids all have an internalised (squid, cuttlefish) or absent shell (octopus) (Amodio, 2018), which has created new selection pressures upon them, that the basal nautiluses do not face. Coleoids must be faster, smarter, and more flexible, due to the loss in shell causing a dramatic increase in predatory pressure (which also may have influenced the short life-history (Catalani, 2008)).
Implications of Advanced Intelligence:
Other species considered to be of high intelligence are provided protections, including legislation condemning abuse and exploitation (Rogers, 2004) (Sunstein, 2003). Opinion has shifted from animals being considered objects of human property, to the understanding of nonhuman suffering and sentience (Shyam, 2015).
The recognised baseline criteria for animal welfare were written by Roger Brambell in 1965, which proposed the ‘5 Freedoms’ by which all intelligent nonhumans should be treated:
- Freedom from Hunger and Thirst
- Freedom from Discomfort
- Freedom from Pain, Injury, or Disease
- Freedom to Express Normal Behaviour
- Freedom from Fear and Distress (Francione, 2010).
In most of the western world, these principles are applied to our treatment of intelligent animals, whether that be in agriculture, entertainment, as pets, or in scientific experiments (Jones, 2013).
In 2010, a landmark decision made by the EU, added living cephalopods to their list of animals protected within scientific procedures, entitling coleoids the 5 Freedoms in captivity (Fiorito, 2015). This was the first instance of an invertebrate family being included in such a list.
In 2015 a paper published in the Journal ‘Laboratory Animals’ defined guidelines on cephalopod welfare, concluding that, avoidance of excessive pain through anaesthesia and analgesia, development of less invasive procedures (ultrasounds, fMRI), and stricter standards of care (enrichment and naturalistic enclosures), should be standard in captive cephalopods (Fiorito, 2015).
The potential sentience of coleoids would determine whether their legal status needs to be altered. If so, there would be implications for their care, as ‘welfare’ would go beyond basic health care, but also include the animals psychological wellbeing (Fiorito, 2015).
How are Neuroanatomy and Intelligence linked?
In this review it is necessary to consider both neuroanatomy and behaviour, as there is extensive evidence that neurological structure has direct effects on intellectual ability (Bitterman, 1965) (Farris, 2015); parallel consideration of these fields enables understanding of connections between observed behaviours and internal workings of neurobiology, thereby answering the questions of why and how.
Despite the Giant Squid axon being one of the most ground-breaking and extensively studied pieces of neural tissue, most other aspects of coleoid neuroanatomy are little known or studied (Williamson, 2004).
Evidence on Coleoid Intelligence from Physiological/ Anatomical/ Neurological Studies:
Before examining the functions of coleoid brain sections, a brief overview of its structure is necessary. Figure 7 illustrates basic octopus and squid anatomy (Hochner, 2017), with the octopus diagram displaying the central nervous system (CNS) in blue, and peripheral nervous system (PNS) in yellow; in octopus, only 1/3 of neurons are in the CNS (what we understand as the brain), while the remaining 2/3 of neurological tissue is spread through the rest of the body as the PNS, travelling down each of the arms (Huffard, 2013). The anatomy of squid and cuttlefish is less extreme, with proportionally more neurons in the CNS, as their arms are less tactile (Huffard, 2013). In coleoids, their brain is partially wrapped around their oesophagus, which leads to extended feeding time, to avoid large objects passing through the brain centre, and potentially causing damage (Williamson, 2004).
Selected significant studies and findinds on Coleoid Neuroanatomy:
An introductory paper (Huffard, 2013) on neurobiology states that coleoids possess the largest brains relative to body mass, of any invertebrates, and have as many neurons as a domestic dog (~500 million in Octopus vulgaris) (Hochner, 2008). Experiments across hundreds of taxa have concluded that proportional brain size is more relevant to intelligence level than total brain size, for instance: even though whales have physically much larger brains than humans, our brain-mass ratio is the highest in the animal kingdom (Huffard, 2013). From studies across species, scientists developed an ‘encephalisation quotient’ equation. This is used to calculate the difference between expected and observed brain size: the higher the brain-mass ratio is, the more likely an animal is to have advanced intelligence (Williamson, 2004).
Chrachri and Williamson studied neural networks associated with learning and memory in coleoids, and found that there are two distinct memory centres: one for visual the other for tactile (Williamson, 2004). This enables information to be stored in groupings, allowing faster retrieving of knowledge in problem-solving and decision-making situations. Furthermore, other coleoid systems, such as arm control, the olfactory system, and oculomotor system, represent large proportions of neural tissue, and contribute to skills such as complex object-manipulation, and detailed vision (Williamson, 2004).
An octopus brain has ~40 distinct lobes (Yan Wang, 2019), each with a differentiated functions; experiments using fMRI and inserted electrodes can show us which brain regions are active when carrying out certain tasks (Figure 8) (Young, 1971).
The Vertical Lobe has been identified as the region responsible for coleoids’ intelligence (Shomrat, 2015), and is thought to be analogous with the mammalian prefrontal cortex and avian nidopallium, both of which are used for executive function and higher cognitive tasks (Emery, 2005). This is one of many systems comparative with vertebrate neurobiology, suggesting deep homology or convergence on intelligence between profoundly diverged species (Shigeno, 2018).
The mechanisms enabling coleoid appearance changes are complex and display intelligence. The chromatophores are composite sacculus that contain pigment (either chromatophore-colours, leucophore-white, iridophore-reflective), with a crown of radial muscle fibres that open/close the cell sac (Anadon, 2019) (Figure 9). Each chromatophore is controlled by a single motor neuron (Boycott, 1953), which allows exceptional accuracy, and a huge array of combination-patterns can be created; this in conjunction with octopus/cuttlefish tubercles and protuberances, which also have specific nerves associated gives the ability to change colour and texture virtually instantaneously, due to the direct neuronal access (Dubas, 1986). Additionally, in many species of pelagic cephalopods, they also possess photophores – specialized skin organs that emit light (Anadon, 2019).
Another physiological characteristic believed to contribute to coleoid cognitive potential, is their ability to re-write the ‘central-dogma’ of biology (the flow of genetic information from DNA>RNA>Proteins) (Guarino, 2017). This is a new area of research, therefore, studies are limited, but there is evidence that species of octopus are able to edit their RNA to adapt to their environment (Liscovitch-Brauer, 2017). Usually, when a multicellular organism adapts, a random genetic mutation takes place (change in DNA), which is followed by the code being transcribed by RNA, and translated into corresponding proteins; this change to certain proteins may radiate if successful through evolution, and therefore lead to a species being better suited to its environment. However, it appears from initial evidence, that coleoids may be one of the few species that can skip this step, and their RNA edits directly, in response to external pressures, meaning the individuals are directly benefitting from these genetic changes (Guarino, 2017), due to greater transcription plasticity, enabling rapid adaptation (Figure 10).
Therefore, unlike in other species, where genetic mutations are mostly deleterious, in cephalopods recoding appears to be evolutionarily conserved (Liscovitch-Brauer, 2017), enabling higher transcriptomic plasticity (ability to individually adapt to your niche).
Core literature on this topic includes the 2012 discovery that polar octopuses can edit their RNA to speed up the function of K+ channels, so that these ion channels can continue to function in exceptionally cold conditions. (Garrett, 2012).
Followed by the 2015 research that showed ~60% of squid RNA in neurological tissue had been edited over its lifetime (Alon, 2015). This led to huge changes in brain physiology, and the process creates huge protein diversity, which may in part explain the advanced behaviours of coleoids (Alon, 2015).
All current evidence suggests that coleoids have the ability to edit their RNA to adapt to selection pressures that may lead to evolution of other species, but coleoids overcome this by ‘evolving themselves’ (Garrett, 2012). These findings have been linked to intelligence because firstly, their closest relatives – nautilus – do not possess the same abilities (Liscovitch-Brauer, 2017), and secondly because of the high proportion of editing which occurs in neurological tissue (Alon, 2015).
However, study of more diverse species is needed to discover the true extent that cold-blooded organisms might utilize RNA-editing to respond to environmental variables, such as temperature changes.
How Diverged is Coleoid Anatomy? Comparisons of Neurobiology across taxa:
Aiming to pinpoint the evolutionary jump in neurological complexity, I will examine anatomy and neurochemistry in other families, to assess whether coleoid composition is truly diverged, or whether higher intelligence inherently must converge on features.
Some argue that overall brain size is the pivotal factor (Deaner, 2007) (Radinsky, 1982), but the consensus within the scientific community is that, the higher a species’ encephalisation quotient, the more likely advanced intelligence is, with Homo sapiens having the highest encephalisation quotient of all (Harvey & Krebs , 1990). Coleoids have the highest brain: body ratio of all invertebrates studied, meaning they fit to this trend, suggesting that intelligence converges on this trait.
As well as brain size, certain brain regions also appear relevant to the development of a species’ intelligence. In the case of coleoids, this region has been identified as the Vertical Lobe (and frontal lobe in some papers) (Moroz, 2009), which is associated with learning memories, and problem-solving. Studies have used evolutionary context to question the possible ancestral form of the cerebral cortex in both invertebrates and vertebrates (Shigeno, 2018); the behavioural evidence on coleoids suggest that they must have an equivalent, which is where the vertical lobe is often referenced. Ground-breaking research of vertebrate-coleoid brain comparisons is provided by Young and Hobbs (Hobbs & Young , 1973) (Young, 1995), which conclude that although thought-centres in avian and mammalian brains are not directly connected, they are certainly analogous, and serve similar function in similar ways (Hobbs & Young , 1973). Common features of auxiliary centres across phyla include “synaptic long-term potentiation, neurotransmitter function, and heterogeneity of neurochemical identity” (Shigeno, 2018, p. 952).
The centres believed to be comparable are the vertical and frontal lobes (coleoids) (Jung, 2018), the cerebral cortex (mammals) (Clayton, 2015), and the nidopallium (birds) (Shigeno, 2018) (Emery, 2005), with the relevant brain regions being displayed in Figures 11 & 12. These conclusions suggest convergence (homoplasy) and deep homology, that these regions conserved across species that benefit from advanced intelligence (Farris, 2015).
Furthermore, these regions and traits are not found in the close relatives of coleoids, which possess lesser intelligence, such as others in the mollusc families and nautiluses (Purchon, 1977).
Homoplasy in higher brain centres is more difficult to track than that of primary sensory centres etc, because cognitive regions receive input from multiple sensory centres, are active in a huge variety of behaviours, and were influenced by different selection pressures in different animal groups (Farris, 2015), making connections between the neurobiology across taxa difficult to analyse.
It is clear that many common attributes exist in coleoid anatomy and other species of advanced intelligence. Encephalisation, number of neurons, and possession of a core, enlarged processing region appear necessary for higher intelligence in all taxa, suggesting deep homology in bilaterian/triploblastic organism nervous systems, despite varying evolutionary pressures (Farris, 2008) (Emery, 2005) (Farris, 2015).
Discussion & Conclusions:
When evaluating the weight of evidence, the vast majority suggests that coleoid cephalopods do possess advanced intelligence, when analysed in terms of the principle of ‘general intelligence’, which places significance on flexibility of behaviour rather than species-specific skills. These claims are supported by extensive behavioural studies, showing innovative cognition in learning, problem-solving, communication, as well as additional higher abilities such as play, personality, and potential consciousness.
Furthermore, this behavioural flexibility is reinforced by the majority of physiological and neuroanatomical studies, which display complex neural systems, with analogous regions and functionality to complex vertebrate systems, such as comparable size and regional homoplasy, which leads us to conclude that certain neural systems are necessary and convergent for advanced intelligence. In addition, I have identified general rules by which intelligence can be defined and the probable evolutionary pathways, namely the requirement of a generalist, predatory, and complex foraging strategies, as well as evolutionary pressures of a flexible and fast lifestyle, as well as the specific evolutionary route of coleoids.
Overall, this paper has summarised the introductory questions of coleoid intelligence and neurobiology – why it exists, to what extent, and how it functions.
Alon, S., 2015. The majority of transcripts in the squid nervous system are extensively recoded by A-to-I RNA editing. Elife – Genetics and Genomics , 4(1), p. e05198.
Alves, C., 2008. Short-distance navigation in cephalopods: a review and synthesis. Cognitive Process, 9(1), pp. 239-247.
Amodio, P., 2018. Grow Smart and Die Young: Why Did Cephalopods Evolve Intelligence?. Trends in Ecology and Evolution, 34(1), pp. 45-56.
Anadon, R., 2019. Functional Histology: The Tissues of Common Coleoid Cephalopods. In: S. Pascual, C. Gestal, Guerra, A & G. Fiorito, eds. Handbook of Pathogens and Diseases in Cephalopods. s.l.:Springer, pp. 39-85.
Basil, C., 2008. A role for nautilus in studies of the evolution of brain and behavior. Communicative & Integrative Biology, 1(1), pp. 18-19.
Beran, M., 2018. Self-Control in Chimpanzees Relates to General Intelligence. Current Biology, 28(4), pp. 160-162.
Bitterman, M., 1965. The Evolution of Intelligence. Scientific American , 212(1), pp. 92-101.
Boal, J., 2000. Experimental evidence for spatial learning in octopuses (Octopus bimaculoides).. Journal of Comparative Psychology, 114(3), pp. 246-252.
Boycott, B., 1953. The Chromatophore System of Cephalopods. Proceedings Linnean Society London, 164(2), pp. 235-240.
Boycott, B., 1995. Learning in the Octopus. Scientific American, 212(3), pp. 42-51.
Brown, C., 2012. “It pays to cheat: tactical deception in a cephalopod social signalling system.. Biology Letters, 8(5), pp. 729-732.
Buchardt, B., 1981. Diagenesis of aragonite from Upper Cretaceous ammonites: a geochemical case-study. Sedimentology, 28(3), pp. 423-438.
Burghardt, G. &., 2010. Current Perspectives on the Biological Study of Play: Signs of Progress. The Quarterly Review of Biology, 85(4), pp. 393-418.
Byrne, R., 1997. The technical intelligence hypothesis: an additional evolutionary stimulus to intelligence.. In: A. W. &. R. Byrne, ed. Machiavellian intelligence II. s.l.:Cambridge University Press, pp. 289-311.
Catalani, J., 2008. Cephalopod Intelligence. American Paleontologist, 16(3), pp. 35-39.
Clayton, N., 2005. Corvid Cognition. Current Biology, 15(3), pp. 1-3.
Clayton, N., 2015. Avian Models for Human Cognitive Neuroscience: A Proposal. Neuron, 86(6), pp. 1330-1342.
Deaner, R., 2007. Overall brain size, and not encephalization quotient, best predicts cognitive ability across non-human primates.. Brain, behavior and evolution , 70(2), pp. 115-124.
Diamant, A., 1985. Interspecific feeding associations of groupers (Teleostei serranidae) with octopuses and moray eels in the Gulf of Eilat (Agaba). Environmental Biology of Fishes , 13(2), pp. 153-159.
Dubas, F., 1986. Chromatophore motoneurons in the brain of the squid, Lolliguncula brevis: a HRP study. Brain Research, 374(1), pp. 21-29.
D’Zurilla, T., 1971. Problem solving and behaviour modifcation. Journal of Abnormal Psychology, 78(1), pp. 107-126.
Emery, N., 2004. The Mentality of Crows: Convergent Evolution of Intelligence in Corvids and Apes. Science, 306(5703), pp. 1903-1907.
Emery, N., 2005. Evolution of the avian brain and intelligence. Current Biology , 15(23), pp. 946-950.
Farris, S., 2008. Evolutionary Convergence of Higher Brain Centers Spanning the Protostome-Deuterostome Boundary. Brain, Behaviour and Evolution, 72(1), pp. 106-122.
Farris, S., 2015. Evolution of Brain Elaboration. Philosophical Transactions, 370(1684), pp. 1-8.
Finn, J., 2009. Defensive tool use in a coconut-carrying octopus. Current Biology , 19(23), pp. 1069-1070.
Fiorito, G., 1990. Problem solving ability of Octopus vulgaris lamarck (Mollusca, Cephalopoda). Behavioural and Neural Biology , 53(2), pp. 217-230.
Fiorito, G., 1992. Observational Learning in Octopus vulgaris. Science , 256(5056), pp. 545-547.
Fiorito, G., 1999. Prey-handling behaviour of Octopus vulgaris (Mollusca, Cephalopoda) on Bivalve preys. Behavioural Processes, 46(1), pp. 75-88.
Fiorito, G., 2015. Guidelines for the Care and Welfare of Cephalopods in Research –A consensus based on an initiative by CephRes, FELASA and the Boyd Group. Laboratory Animals, 49(52), pp. 1-90.
Francione, G., 2010. Animal Welfare and the Moral Value of Nonhuman Animals. Law, Culture and the Humanities, 6(1), pp. 24-36.
Gallup, G., 2002. The Mirror Test. The cognitive animal: Empirical and theoretical perspectives on animal cognition, 1(1), pp. 323-333.
Gardner, R. G. B., 1989. A Cross-Fostering Laboratory. In: G. a. Cantfort, ed. Teaching Sign Language to Chimpanzees. Albany : State University New York Press, pp. 1-26.
Garrett, S., 2012. RNA Editing Underlies Temperature Adaptation in K+ Channels from Polar Octopuses. Science, 335(6070), pp. 848-851.
Guarino, B., 2017. The Washington Post, Octopuses and squids can rewrite their RNA. Is that why they’re so smart?. [Online]
Available at: https://www.washingtonpost.com/news/speaking-of-science/wp/2017/04/06/octopuses-and-squids-can-rewrite-their-rna-is-that-why-theyre-so-smart/
[Accessed 29 11 2019].
Hanlon, R., 2018. Cephalopod Behaviour. 2nd ed. Cambridge: Cambridge University Press.
Harvey, P. & Krebs , J., 1990. Comparing brains. Science , 249(4965), pp. 140-146.
Herman, L., 1986. Cognition and Language Competencies of Bottlenosed Dolphins. In: T. W. Schusterman, ed. Dolphin Cognition and Behavior: A Comparative Approach. New Jersey: Lawrence Erlbaum Associates, pp. 221-250.
Hobbs, M. & Young , J., 1973. A cephalopod cerebellum. Brain Research , 55(2), pp. 424-430.
Hochner, B., 2008. Octopuses. Current Biology , 18(19), pp. 897-898.
Hochner, B., 2017. Embodied Organization of Octopus vulgaris Morphology, Vision, and Locomotion. Frontiers in Physiology, 8(1), pp. 164-165.
Horowitz, A., 2007. Naturalizing Anthropomorphism: Behavioral Prompts to Our Humanizing of Animals. Anthrozoös – A multidisciplinary journal of the interactions of people and animals, 20(1), pp. 23-25.
Huber, L., 2006. Technical intelligence in animals: the kea model. Animal Cognition, 9(4), pp. 295-305.
Huffard, C., 2013. Cephalopod neurobiology: an introduction for biologists working in other model systems. Invertebrate Neuroscience, 13(1), pp. 11-18.
Ikeda, Y., 2009. A perspective on the study of cognition and sociality of cephalopod mollusks, a group of intelligent marine invertebrates. Japanese Psychological Research, 51(3), pp. 146-153.
Jerison, H., 2012. Evolution of The Brain and Intelligence. 1 ed. Los Angeles: Elsevier – Academic Press.
Jones, R., 2013. Science, sentience, and animal welfare. Biology & Philosophy, 28(1), pp. 1-30.
Jung, S., 2018. A Brain Atlas of the long arm octopus. Experimental Neurobiology , 27(4), pp. 257-266.
Kamil, 1994. A synthetic approach to the Study of Animal Intelligence . In: L. Real, ed. Behavioural Mechanisms in Evolutionary Biology . Chicago: University of Chicago Press, pp. 11-45.
Kuba, M., 2003. Looking at play in Octopus vulgaris. Berliner Paläontologische Abhandlungen, 3(1), pp. 163-169.
Kuba, M., 2006. When do octopuses play? Effects of repeated testing, object type, age, and food deprivation on object play in Octopus vulgaris. Journal of Comparative Psychology , 120(3), pp. 184-190.
Kuba, M., 2014. Learning from Play in Octopus. In: L. D. J. M. Anne-Sophie Darmaillacq, ed. Cephalopod Cognition. Cambridge: Cambridge University Press, pp. 57-72.
Legg, H., 2007. Collections of Definitions of Intelligence. In: P. W. Ben Goertzel, ed. Advances in Artificial General Intelligence: Concepts, Architectures and Algorithms. Amsterdam: IOS Press, pp. 17-24.
Lin, C., 2017. Quantitative Analysis of Dynamic Body Patterning Reveals the Grammar of Visual Signals during the Reproductive Behavior of the Oval Squid Sepioteuthis lessoniana. Frontiers in Ecology and Evolution, 18(5), p. 30.
Liscovitch-Brauer, N., 2017. Trade-off between Transcriptome Plasticity and Genome Evolution in Cephalopods. Cell, 169(2), pp. 191-202.
Mann, J., 2013. Tool use by Aquatic Animals. Philosophical Transactions of the Royal Society B: Biological Sciences, 368(1630), pp. 1-11.
Marino, L., 2002. Convergence of Complex Cognitive Abilities in Cetaceans and Primates. Brain, Behaviour, and Evolution, 59(1), pp. 21-32.
Marshall, J., 2014. Unconventional colour vision. Current Biology, 24(24), pp. 1150-1154.
Mather, J., 1999. Exploration, play and habituation in octopuses (Octopus dofleini).. Journal of Comparative Psychology, 113(3), pp. 333-338.
Mather, J., 2007. Cephalopod consciousness: Behavioural evidence.. Consciousness and Cognition, 17(1), pp. 37-48.
Mather, J., 2008. To boldly go where no mollusc has gone before: Personality, play, thinking, and consciousness in cephalopods. American Malacological Bulletin, 24(2), pp. 51-58.
Mehrkam, L., 2019. A behavior-analytic approach to understanding octopus “mind”. Animal Sentience, 258(1), pp. 1-4.
Moroz, L., 2009. On the Independent Origins of Complex Brains and Neurons. Brain, Behaviour, and Evolution, 74(1), pp. 177-190.
Pepperberg, I., 1990. Some Cognitive Capacities of an African Grey Parrot (Psittacus erithacus). Advances in the Study of Behaviour, 19(1), pp. 357-409.
Pratt, S., 2010. Collective Intelligence in Social Animals. Encyclopedia of Animal Behaviour, 1(1), pp. 303-309.
Purchon, R., 1977. The Biology of Mollusca. 2nd ed. Oxford: Pergamon Press.
Quetglas, A., 2015. Trophic interactions among grouper (Epinephelus marginatus), octopus (Octopus vulgaris) and red lobster (Palinurus elephas) in the Western Mediterranean. Instituto Español de Oceanografía , 2(1), p. 1.
Radinsky, L., 1982. Some cautionary notes on making inferences about relative brain size. In: Primate brain evolution. Boston: Springer, pp. 29-37.
Richter, J., 2016. Pull or Push? Octopuses Solve a Puzzle Problem. PLoS one, 11(3).
Rogers, L., 2004. All Animals Are Not Equal: The Interface between Scientific Knowledge and Legislation for Animal Rights. In: G. Kaplan, ed. Animal Rights: Current Debates and New Direction. New York: Oxford University Press, pp. 175-202.
Roth, G., 2005. Evolution of the brain and intelligence. Trends in Cognitive Science, 9(5), pp. 250-257.
Salazar, R., 2017. Classification of biological and bioinspired aquatic systems: A review. Ocean engineering, 148(1), pp. 11-12.
Schaik, v., 2011. Social learning and evolution: the cultural intelligence hypothesis.. Philosophical Transactions of the Royal Society B: Biological Sciences, 366(1567), pp. 1008-1016.
Schnell, A., 2019. Cephalopod cognition. Current Biology , 29(15), pp. 726-732.
Sheperd, W., 1915. Tests on Adaptive Intelligence in Dogs and Cats, as Compared with Adaptive Intelligence in Rhesus Monkeys. The American Journal of Psychology, 26(2), pp. 211-216.
Shigeno, S., 2018. Cephalopod Brains: An Overview of Current Knowledge to Facilitate Comparison With Vertebrates. Fronteirs in Physiology , 9(1), pp. 952-953.
Shomrat, T., 2015. The vertical lobe of cephalopods: an attractive brain structure for understanding the evolution of advanced learning and memory systems. Journal of Comparative Physiology A, 201(9), pp. 947-956.
Shyam, G., 2015. The Legal Status of Animals: The World Rethinks its Position. Alternative Law Journal, 40(4), pp. 266-270.
Sinn, D., 2001. Early temperamental traits in an octopus (Octopus bimaculoides). Journal of Comparative Psychology,, 115(4), pp. 351-364.
Sopher, P., 2015. The Atlantic – What Animals Teach Us About Measuring Intelligence.. [Online]
Available at: https://www.theatlantic.com/education/archive/2015/02/what-animals-teach-us-about-measuring-intelligence/386330/
[Accessed 4 December 2019].
Sunstein, C., 2003. The Rights of Animals. The University of Chicago Law Review, 70(1), pp. 387-401.
Taylor, A., 2014. Corvid Cognition. WIREs Cognitive Science, 5(3), pp. 361-372.
Thomas, R., 1980. Evolution of Intelligence: an Approach to Its Assessment. Brain, Behaviour, and Evolution, 17(1), pp. 454-472.
Thorndike, E., 1898. Some Experiments on Animal Intelligence. Science, 7(1), pp. 818-824.
Thorndike, E., 1998. Animal intelligence: An experimental study of the associate processes in animals. American Psychologist, 53(10), pp. 1125-1127.
Troy, T., 1991. The ‘correct’ definition of intelligence. International Journal of Intelligence and Counter Intelligence, 5(4), pp. 433-454.
Umbers, K., 2016. Genetic structure of mourning cuttlefish (Sepia plangon Gray, 1849) in Sydney Harbour, Australia. Journal of Molluscan Studies, 82(1), pp. 187-192.
Unsworth, R., 2012. An inter-specific behavioural association between a highfin grouper (Epinephelus maculatus) and a reef octopus (Octopus cyanea). Marine Biodiversity Records, 5(1), pp. 1-2.
Vail, A., 2013. Referential gestures in fish collaborative hunting. Nature Communications, 4(1), p. 1765.
Williamson, R., 2004. Cephalopod Neural Networks. Neurosignals, 13(1), pp. 87-98.
Wynne, C., 2004. The perils of anthropomorphism. Nature, 606(1), p. 428.
Yan Wang, Z., 2019. Cephalopod Nervous System Organization. Oxford Research Encyclopedia of Neuroscience, 10(1), pp. 1-22.
Young, J., 1971. The anatomy of the nervous system of Octopus vulgaris. 1 ed. London: Oxford University Press.
Young, J., 1995. Multiple matrices in the memory system of Octopus. In: N. Abbott, ed. Cephalopod neurobiology: neuroscience studies in squid, octopus and cuttlefish. London: Oxford University Press, pp. 431-443.