Where Should Comparative Cognition Be Going? To the Invertebrates
Abstract
Presently the study of comparative cognition is mostly on “higher” vertebrates, who only make up about 1.5% of the animal species on the planet. We can extend this narrow model to include invertebrates, especially cephalopods, insects, and spiders. This allows us to ask what social and ecological influences result in intelligent animals, how different nervous system compute and store information, and even what range of activities compose cognition itself.
In 1950, Frank Beach wrote an editorial in the Journal of Comparative Psychology (JCP) called “The Snark was a Boojum” about the narrowness of the focus of comparative psychology research of the time, which he claimed was essentially on rat learning. It stimulated the area so that JCP now has a far wider focus. Nevertheless, Macri and Richter (2015) looked at the use of animals in biomedical research and concluded that we use studies of far too few animals and paradigms, that the snark hunt is still with us. This is also a problem for comparative cognition. Studying mainly mammals and birds, we rely on too few species to understand the scope of cognition and thus limit our understanding of the processes. It is only within the past couple of decades that research has uncovered the cognitive abilities and intelligence of invertebrates, but that is why the future is with them.
One proof of this narrowness is that there are few mammals and many invertebrates. Invertebrates comprise about 97% of the animals on the planet, and a hefty 70% are insects. Mammals comprise about 0.5%. A quick count of recent titles in Animal Behaviour (Mather, 2019) showed 32% of the articles on mammals, 21% on birds, and 21% on insects. The proportion on insects looks reasonable, except that there are 170 of them for every mammal species. Three percent of articles were on other invertebrates, who represent about one quarter of all animal species. Our mammal–bird focus means that comparative cognition is presently comparative vertebrate cognition, though it does not really include fish either.
The dominant reason why we should look at invertebrates to study cognition is that they provide different models that are useful for studying many hypotheses. For the foundation of these hypotheses, Roth (2015) suggested that high intelligence in any animal group is tied to having a complex brain with a multimodal center for generating integrated behavior. He suggested a parallel between the mammalian frontal lobe, the pallium of birds and fishes, the coleoid cephalopod vertical lobe, and the insect mushroom bodies. How this centralized control is carried into intelligence and cognition in these different groups is truly comparative.
On the functional side, Birch et al. (2020) believe that animals with intelligence and minds must have certain foundational characteristics. They suggested a combination of perceptual richness, evaluative richness (affectivity), temporality (linkage across time), and unity (singularity of control and output). Perceptual richness also depends on the modality of the sensory systems involved, as vision is often emphasized in vertebrate assessment (see Gallup’s mark test) and the basic sensory modality for most animals is chemical. Mather (2021b) has taken on the challenge of outlining these areas of ability in the cephalopods.
Many different examples of the comparative approach across vertebrates and invertebrates yield useful information. Sometimes the focus across different species and groups is narrowed to a specific ability. Gatto et al. (2022), interested in quantitative processing by animals, pointed out the growing interest in this ability in areas such as food choice, social cognition, and navigation. From 1990 to 2020, there was an increase in articles about vertebrates from about 10 to 135; Gatto et al. concluded that this ability is widely shared across the group. However, as the number of articles on invertebrates increases beyond the present 25, mostly in insects, Gatto et al. hope that we will gain a better definition of quantitative ability across animals in general.
Sometimes using an invertebrate group as a comparison can help us to test assumption of the source of cognitive ability. The social intelligence hypothesis, formulated on vertebrates but especially on primates, dominated our understanding of what made an intelligent animal for years. Intelligence was thought to need big brains, large social group size, a long lifespan, and a protected juvenile period. Yet the coleoid cephalopods were undeniably intelligent, with even an edited book on their cognition (Darmaillacq et al., 2014). Although cephalopods have big brains, they do not have a long lifespan and protective childhoods. Most important, the social hypothesis was contradicted by their range of abilities (Benedixen et al., 2021), as the mostly solitary octopuses were more brainy than the schooling squid. As well, insects live an even shorter lifespan, and although young social insects are often protected, insect brains are tiny (see Roth, 2015). This comparison has forced us to look at the possibilities behind brain miniaturization (see Simons & Tibbet, 2019).
Cephalopods also allow us to examine the extent to which a nervous system needs to and can act as a unit. Three fifths of the neurons in the body of an octopus are not in the brain but outside in the arms. They make chains of ganglia down the dorsal area of each arm, with each ganglion acting as a local controller for movement, particularly of a single sucker. Ganglia link to the next-door ones (Chang & Hale, 2023), and all the arm bases are linked by a suprabrachial commissure. The brain output to arms is not extensive, and arms can do the same or different coordinated actions or can act semi-independently. Grasso (2014) suggested that the arm neural system is a plexus, a set of equal interconnected units carrying out the details of specified actions. The octopus does not have “nine brains” but allocates carrying out details of centrally commanded actions to subsidiaries. Despite this allocation of subroutines and a bilateral organization, the octopus brain learns, makes decisions, and carries out actions as a unity (Mather, 2021a).
The situation of invertebrates has also forced us to look at the boundaries and definitions for cognition, which animals are or are not intelligent as well as how they are intelligent. Cross et al. (2020) advanced the case of spiders, particularly the web-building jumping spider Portia. Jumping spiders can learn to make a detour along an unseen path, to register expectancy violation, and to solve confinement problems. They concluded that the spiders are using more of the internal representations expected of a cognitive animal and less of the immediate use of information to solve problems that we would expect from an intelligent one. But this raises an interesting question: Is the decision of whether an animal “has” cognition an all-or-none judgment of having enough ability, or can they be quantitatively compared on how much they have?
If we want to study when animals truly do or do not have cognitive abilities, we can look at the understudied echinoderms. Where does intelligence start? Freas and Cheng (2022) looked at this phylum, as echinoderms do not have a brain yet have adaptive behavioral responses. Furthermore, they are phylogenetically the closest group to the vertebrates; molluscs and arthropod are more closely related to each other than to us. Echinoderms (seastars, brittle stars, sea urchins, sea cucumbers, and feather stars) have a nerve net and pentamerous (five-sided) radial symmetry. Scattered investigations suggest that some echinoderms have not only habituation but also some simple associative learning. This suggestion is important, because Ginsberg and Jablonka (2019) postulated that on the evolutionary route to higher intelligence, we would not find simple associative learning until animals had a brain and bilateral symmetry. But the echinoderms may have some of this ability, and they are on the evolutionary route to fish, mammal, and bird cognition.
A final question that investigation of invertebrates will help answer could be described as “Where is cognition situated?” Cheng (2018) challenged our assumption that the brain controls cognition and suggested that we need to look across groups at where cognition “is.” Insects may distribute different tasks and thus abilities to different castes (see Bonabeau et al., 1998) and not accumulate all of it in each individual. Octopuses may leave some embodied processing allocated outside the brain to the arms (but see Mather, 2021a, for their overall unity). Web-building spiders (Cross et al., 2020), the web of which is almost an extension of themselves, may have extended cognition in the sense that these structures are such a major perceptual tool.
The expansion of the number of groups of animals that are intelligent and mindful poses both advantages and problems. It allows us to evaluate whether there is a specific ability for intelligence/cognition, a G factor long sought for and yet widely contested in Homo sapiens (e.g., Kovacs & Conway, 2019). Burkhart et al. (2017) looked for G in “animals,” but they used only mammals; their account is heavily anthropocentric and leans on the cultural base of the social intelligence theory (and see the critiques in the same volume). What kind of brain does an animal need for cognition and intelligence? Roth (2015) has given us a start, and de Casein and Higham (2019) postulated links between ecology and development of brain sensory or spatial areas that can be compared across phyla (see Bendixen et al., 2021). What kind of social or ecological background shapes them, and are these influences separate or linked (Henke-von den Malsburg et al., 2020)? Insect models will help us find out. How much learning ability does cognition need (Ginsberg & Jablonka, 2019) and where should it be located are Cheng’s (2008) locations for “situated cognition” really doing cognition? These are fundamental questions for comparative cognition, and studying the invertebrates will help answer them.
References
Beach, F. A. (1950). The snark was a boojum. American Psychologist, 5, 115–124. https://doi.org/10.1037/h0056510
Bendixen, T., Mather, J. A., & Muthukrishna, M. (2021). The evolution of big brains. Inference, 6(1). https://doi.org/10.37282/991819.21.22
Birch, J., Schnell, A. K., & Clayton, N. (2020). Dimensions of animal consciousness. Trends in Cognitive Science, 24, 789–801. https://doi.org/10.1016/j.tics.2020.07.007
Bonabeau, E., Sobkowski, A., Theraulaz, G., & Deneubourg, J-L. (1998). Adaptive task allocation inspired by a model of division of labor in social insects (SFI Working Paper No. 1998-01-004).
Burkhart, J. M., Schubiger, M. N., & van Schaik, C. P. (2017). The evolution of general intelligence. Behavior and Brain Sciences, 2017, 1–67. https://doi.org/10.1017/S0140525X16000959
Chang, W., & Hale, M. E. (2023). Mechanosensory signal transmission in the arms and the nerve ring, an interarm connective, of Octopus bimaculatus. iScience, 26, Article 106722. https://doi.org/10.1016/j.isci.2023.106722
Cheng, K. (2018). Cognition beyond representation: Varieties of situated cognition in animals. Comparative Cognition Behavioral Reviews, 13, Article 130001. https://doi.org/10.3819/CCBR.2018.130001
Cross, F. R., Carvell, G. E., Jackson, R. R., & Grace, R. C. (2020). Arthropod intelligence? The case for Portia. Frontiers in Psychology, 11, Article 568049. https://doi.org/10.3389/fpsyg.2020.568049
Darmaillacq, A-S., Dickel, L., & Mather, J. A. (2014). Cephalopod cognition. Cambridge University Press. https://doi.org/10.1017/CBO9781139058964
de Casein, A. R., & Higham, J. P. (2019). Primate mosaic brain evolution reflects selection on sensory and cognitive specialization. Nature Ecology & Evolution, 3, 1483–1493. https://doi.org/10.1038/s41559-019-0969-0
Freas, C. A., & Cheng, K. (2022). Neuroecology beyond the brain: Learning in Echinodermata. Learning & Behavior, 50, 20–36. https://doi.org/10.3758/s13420-021-00492-3
Gatto, E., Loukola, O. J., & Agrillo, C. (2022). Quantitative abilities of invertebrates. Animal Cognition, 25, 5–19. https://doi.org/10.1007/s10071-021-01529-w
Ginsberg, S., & Jablonka, E. (2019). The evolution of the sensitive soul: Learning and the origin of consciousness. MIT Press. https://doi.org/10.7551/mitpress/11006.001.0001
Grasso, F. W. (2014). The octopus with two brains: How are distributed and central representations integrated in the octopus central nervous system? In A-S. Darmaillacq, L. Dickel, & J. A. Mather (Eds.), Cephalopod cognition (pp. 94–124). Cambridge University Press. https://doi.org/10.1017/CBO9781139058964.008
Henke-von den Malsburg, J., Kappeller, P., & Fichtel, C. (2020). Linking ecology and cognition: Does ecological specialisation predict cognitive test performance? Behavioral Ecology and Sociobiology, 74, Article 154. https://doi.org/10.1007/s00265-020-02923-z
Kovacs, K., & Conway, A. R. A. (2019). What is IQ? Life beyond “general intelligence.” Current Directions in Psychological Science, 28, 189–194. https://doi.org/10.1177/0963721419827275
Macri, S., & Richter, S. H. (2015). The snark was a boojum—reloaded. Frontiers in Zoology, 12 (Suppl.), Article 520. https://doi.org/10.1186/1742-9994-12-S1-S20
Mather, J. A. (2019). Ethics and care: For animals, not just mammals. Animals, 9, 1018–1030. https://doi.org/10.3390/ani9121018
Mather, J. A. (2021a). The case for octopus consciousness: Unity. NeuroSci, 2, 406–415. https://doi.org/10.3390/neurosci2040030
Mather, J. A. (2021b). Octopus consciousness: The role of perceptual richness. NeuroSci, 2, 276–290. https://doi.org/10.3390/neurosci2030020
Roth, G. (2015). Convergent evolution of complex brains and high intelligence. Philosophical Transactions of the Royal Society B, 370, Article 20150049. https://doi.org/10.1098/rstb.2015.0049
Simons, M., & Tibbet, E. (2019). Insects as a model for studying the evolution of animal cognition. Current Opinions in Insect Science, 34, 117–122. https://doi.org/10.1016/j.cois.2019.05.009