Minds: 600 million to 4 million years ago

Animals form my second major chapter in the history of life after the eukaryotes established the dominant evolutionary mechanisms of complex life. Plants and fungi have evolved some highly specialized features of their own, but animals are unique among multicellular life forms in having to meet the challenges of mobility. Being able to move through the environment introduces many more opportunities and risks than sessile forms face. While sessile lifeforms can largely just sit tight and focus on molecular-level environmental interactions, animals must continuously decide what to do next. By “decide,” I only mean that they must commit their bodies to being in one place at a time, and also must coordinate their extremities appropriately as well. Bodies and limbs are, of course, much larger structures than cells, so they shift the whole scope of the control problem to a much higher level. Top-level animal control mechanisms must address the survival needs of the animal from the top down. To do this, animals developed brains which collect information about the body and the environment from the bottom up, but then select actions for the body to take based on top-down considerations because they must decide, first and foremost, where they are going to be, which then has logical implications for all their other activities. So mobility creates, more than for any other kind of organism, a need to solve logic problems. Logically, a body necessarily acts as an agent in the world whose activities must be subdivided into discrete functions if the animal is to able to prioritize activities so that they can be completed with conflicting resources (most notably with one body). No matter the size of the brain or the animal, it must have a means of distinguishing its possible activities by function or it will oscillate between functions ineffectively like Dr. Doolittle’s pushmi-pullyu. This in no way mandates conscious awareness of such logical distinctions; I will address consciousness separately later. Also, before we get into reasoning out how the brain works, let’s look closer at what kinds of animals evolved.

The last animalian common ancestor is called the urmetazoan and is thought to have been a flagellate marine creature. The urmetazoan is important because, like the LUCA and LECA before it, an unknown but perhaps significant amount of animal evolution went into making the urmetazoan and an unknown but perhaps significant number of competing multicellular mobile forms were squeezed out by the metazoans (aka animals). Although we only now see what got through this bottleneck, animals have differentiated into many branches and have a wide variety of forms, so I will climb up through the animal family tree.

Sponges are the most primitive animals from a control standpoint, having no neurons or indeed any organs or specialized cells. But they have animal-like immune systems and some capacity for movement in distress.1. Cnidarians (like jellyfish, anemones, and corals) come next and feature diffuse nervous systems with nerve cells distributed throughout the body without a central brain, but often featuring a nerve net that coordinates movements of a radially symmetric body. Although jellyfish move with more energy efficiency than any other animal, a radial body design provides limited movement options, which may explain why all higher animals are bilateral (though some, like sea stars and sea urchins, have bilateral larvae but radial adults). Nearly all creatures that seem noticeably “animal-like” to us do so because of their bilateral design which features forward eyes and a mouth. This group is so important that we have to think of the features of the urbilaterian, the first bilateral animal about 570-600 million years ago. As I mentioned above, we now have evidence that the urbilaterian did have eyes. While the exact order in which the features of animals first appeared is still unknown, a functional principle that developed in many bilateral animals was a centralized control center that can make high-level decisions leveraging a variety of sensory information.

A few exceptions to centralized control exist among the invertebrates, most notably the octopus (a mollusk), which has a brain for each arm and a central brain that loosely administers them. Having independent eight-way control of its arms comes in handy for an octopus because the arms can often usefully pursue independent tasks. Octopus arms are vastly more capable than those of any other animals, and they use them in amazingly coordinated ways, including to “bounce-walk” across the sea floor and to jump out of the water’s edge to capture crabs.

Why, then, don’t animals all have separate brains for each limb and organ? The was function evolves is always a compromise between logical need and physical mechanism. To some degree, historical accident has undoubtedly shaped and constrained evolution, but, on the other hand, where logical needs exist, nature often finds a way, which sometimes results in convergent evolution of the same trait through completely different mechanisms. In the case of control, it seems likely that it was physically feasible for animals to either localize or centralize control according to which strategy was more effective. An example of decentralized control in the human body is the enteric nervous system, or “gut brain”, which lines the gut with more than 100 million nerve cells. This is about 0.1% of the 100 billion or so nerves in the human brain. Its main role is controlling digestion, which is largely an internal affair that doesn’t require overall control from the brain.2 However, the brain and gut brain do communicate in both directions, and the gut brain has “advice” for the brain in the form of gut feelings. Much of the information sent from the gut to the brain is now thought to arise from our microbiota. The microbes in our gut can weigh several pounds and comprise hundreds of times more genes than our own genome. So gut feelings are probably a show of “no digestion without representation” that works to both parties’ benefit.34 The key point in terms of distributed control is that if the gut has information relevant to the control of the whole animal, it needs to convey that information in a form that can impact top-level control, and it does this through feelings and not thoughts.

So let’s consider how control of the body is accomplished in the other two families of highly mobile, complex animals, namely the arthropods and vertebrates. The control system of these animals is most broadly called the neuroendocrine system, as the nervous and endocrine systems are complementary control systems that work together. The endocrine system sends chemical messages using hormones traveling in the blood while the nervous system sends electrochemical messages through axons, which are long, slender projection of nerve cells, aka neurons, and then between neurons through specialized connections called synapses. Endocrine signals are generally slower to begin than nervous signals and tend to last longer. Both arthropods and vertebrates have endocrine glands in the brain and about the body, including the ovaries and testes. Hormones regulate both physiology and behavior of bodily functions like digestion, metabolism, respiration, tissue function, sensory perception, sleep, excretion, lactation, stress, growth and development, movement and reproduction. Hormones also affect our conscious mood, which encompasses a range of slowly-changing subjective states that can influence our behavior.

While the endocrine system focuses on control of specific functions, the nervous system provides overall control of the body, which includes communication to and from the endocrine system. In addition to the enteric nervous system (gut brain) discussed before, the body has two other peripheral systems called the somatic and autonomic nervous systems that control movement and regulation of the body somewhat independently from the brain. The central nervous system (CNS) comprises the spinal cord and the brain itself. Nerve cells divide into sensory or afferent neurons that send information from the body to the CNS, motor or efferent neurons that send information from the brain to the body, and interneurons which comprise the brain itself.

The functional capabilities of brains have developed quite differently in arthropods and vertebrates, but, for my purposes, even more significantly in certain kinds of vertebrates. Moving through the vertebrates on the way to Homo sapiens, first

fish branch off, then
amphibians, and then
amniotes, which enclose embryos with an amniotic sac that provides a protective environment and makes it possible to lay eggs on land. Amniotes divide into
reptiles, from which derive
birds (by way of dinosaurs), and
mammals. And mammals then divide into
monotremes (like the duck-billed platypus), then
marsupials, and then
placentals, which gestate their young internally to a relatively late stage of development. There are eighteen orders of placentals, one of which is
primates, to which
humans belong.

Evolution did not reach its apotheosis in Homo sapiens, even though it seems that way to us. We know it is technically incorrect to say some species are “more evolved” than others because all species have had the same amount of time to evolve. However, brain power unquestionably tends to increase as one moves through the above branches toward humans, and new brain structures appear along the way that help to account for that increase in power tend to appear. Why would this happen if evolution is not directed? It is because evolution is directed; not toward more complexity but toward greater functionality. In animals, that functionality is most critically driven by the power of the top-level control center, which is the brain. When one reviews an evolutionary tree from the perspective of the organisms with the greatest range of functionality, humans in this case, one will usually see members of earlier branches that, despite having had just as much time to evolve in new directions, seem to be functionally unchanged in the fossil record since the time of the branching. This is because the ecological niches they filled then often still need to be filled, and they are still the best-equipped to fill those niches. Fish are a prime example; all the other lines above moved to land (and a few, like cetaceans, later moved back to water). There are no doubt varieties of fish which have evolved some impressive new functions since that branching, but no fish became as smart as mammals. This is mostly because evolution is inclined to keep things the same so long as they are working, so relatively unchanging aquatic environments have led to fish keeping their basic body plan, lifestyle, and brain.

But it is also worth noting that fish are cold-blooded. Warm-blooded animals need much more food but can engage in a more active lifestyle. This gives them some advantages over their less active peers, but not enough to unseat the dinosaurs during the Mesozoic era from 250 to 66 million years ago. An ingenious hypothesis from Arturo Casadevall is that warm-bloodedness evolved as a near-perfect protection against fungal diseases, which plague cold-blooded animals. The protection this offers may not have been the driving reason, but it definitely helps. Casadevall further speculated that an asteroid strike 66 million years ago amplified this advantage: “Deforestation and proliferation of fungal spores at cretaceous-tertiary boundary suggests that fungal diseases could have contributed to the demise of dinosaurs and the flourishing of mammalian species.”56 Still, most evidence suggests dinosaurs were themselves warm-blooded or close to it78, so this may not have been the advantage that led to the Age of Mammals (the Cenozoic).

My point here is that although physical differences can be suggestive of function, they are not conclusive. All the changes that distinguish fish brains from ours don’t prove that fish don’t have deep thoughts just like us. However, scientific evidence supported by evolutionary theory, physiological evidence, and behavioral studies all consistently and strongly suggest that brain complexity is necessary, if not sufficient, for more complex brain functions. Arthropods do engage in specific functionally-distinct tasks, as I said they logically must as agents, but the way they perform them appears to be entirely instinctive and not taught to them or learned from experience. That said, this is not entirely black and white as bees, for example, learn much from their environment and can adapt their behavior accordingly.9 But one part of the brain in particular, the vertebrate cerebrum, is thought to be the principal control center of more complex behavior. Let’s just consider what some measures of brain and cerebral differences among the animals suggest.

  • By total number of neurons, humans have the most (about 86 billion), except for elephants (about 250 billion) and whales, who likely have several times more as well.10
  • By total number of cerebral cortex neurons, humans have the most (about 16 billion), except for some whales who have the same number or more. But, again, whales are much larger. Elephants have about 6 billion, which is only bested by other primates and sea mammals.11
  • By brain-to-body mass ratio, ants have the most brains at 1:7, then shrews at 1:10, then small birds at 1:12, then mice and humans at 1:40, with other larger birds and mammals having smaller ratios.12
  • By Encephalization Quotient (EQ). Because some brain control functions seem to be independent of body size, larger animals need comparatively less brains than smaller animals. The EQ attempts to correct for size but also gives primates a numerical advantage (which arguably takes their larger cerebrums into account). By this measure, humans are 66% smarter than dolphins and three times smarter than chimps and ravens, with all other animals further down the list.1314

The above suggests that larger brains with more cortical neurons are generally more capable, but of course gross sizes are just the tip of the iceberg of genetic differences between species, so this tells us nothing specific.

It turns out that the mind as we think of it arose because of the logical problems the brain faces. Animals (and all organisms) have the long-term function of propagating their gene line, also called the “selfish gene” perspective, but it is really the cooperative and ratcheting effort of functions in the gene line to preserve their function and hence to survive. Perhaps individuals have their own (possibly selfish) reasons for behaving as they do, which is a subject I will broach later, we can definitely say that evolution statistically protects the gene line. The life cycle of animals requires that they eat, mate, and maintain their bodies in other appropriate ways, including breathing, sleeping, cleaning, etc. These activities usually depends on a number of subsidiary activities which also need to be done appropriately. For example, animals must search for food and must sometimes also take steps to prepare it for consumption. Mating does not happen often and so usually has a variety of formalized subtasks to ensure the suitability of a mate.

All of these sorts of activities must be performed to completion to deliver functionality, and the steps to complete them need to plan their use of resources, most importantly what the body will be doing. Since this is fundamentally a logic problem, the brain needs to think of these activities and the units they break down to as logical units or references and have ways of manipulating them logically. This doesn’t mean the logic needs to be overt or general purpose the way we think of deductive logic. It can also be managed by fixed instinctive approaches or using inductive trial-and-error learning. But however it is done, on some appropriate level the activities must be mapped to logical references. I’m not saying how the animal should do it, just that the problems exist and must be solved.

Living organisms are homeostatic, meaning they must maintain their internal conditions in a working state at all times. Animals had to evolve a homeostatic control system, meaning that it had to be able to adjust its supervision on a continuous basis. But it still needs to be able to fulfill tasks smoothly and efficiently and not in a herky-jerky panic. Karl Friston was the first to characterize these requirements of a homeostatic control system through his free energy principle, which could more accurately be called the variational bounding principle, as it does not really relate to energy or anything physical at all.15 This principle says that a homeostatic control system must minimize its surprise, meaning that it should proceed calmly with its ongoing actions so long as all incoming information falls within the expected range. Any unexpected information is a surprise, which should be bumped up in priority and dealt with until it can be brought back into an expected range itself. In order to follow this principle, the control system must know what to expect, but, more than that, it must act so as to minimize the chance that those inputs will go outside the expected range. Animals must follow this principle simply because it is maladaptive not to. Unlike human machines, which are not homeostatic or homeostatically controlled, animals must have a holistic reaction strategy that can deal with control issues fractally, that is, at potentially any level of concern.

Simple animals have simple expectations. Even a single-cell creature, like yeast, can sense its environment in some ways and respond to it. Simple creatures evolve a very limited range of expectations and fixed responses to them, but animals developed a broader range of senses, which made it possible to develop a broader range of expectations and responses. In a control arms race, animals have ratcheted up their capacity to respond to an increasing array of sensory information to develop ever more functional responses. But it all starts with the idea of real-time information, which is, of course, the specialty of sensory neurons. These neurons bring signals to the brain, but what the brain needs to know about each sense has to be converted logically into an expected range. Information within the range is irrelevant and can be ignored. Information outside the range requires a response. This requirement to translate the knowledge into a form usable for top-level control created the mind as we know it.

From a logical standpoint, here is what the brain does. First, it monitors its internal and external environment using a large number of sensory neurons, which are bundled into specific functional channels. The brain reprocesses each channel using a logical transformation and feeds it to a subprocess called the mind that maintains an “awareness” state over the channel that it ignores. These channels are kept open because a secondary process in the brain called an “attention” process evaluates each channel to see if it falls outside the expected range. When a channel does that, the attention process focuses on that channel, which moves the mind subprocess from an aware (but ignoring) state to a focused (attentive) state. The purpose of the mind subprocess is to collect incoming information that has been converted into a logical form that is relevant to tasks at hand so that it can prioritize and act so as to minimize future surprise. Of course, its more complex reactions complete necessary functions, and that is its “objective” if we view the problem deductively, but the brain doesn’t have to operate deductively or understand that it has objectives. All it needs to be able to do is convert sensory information into expected ranges and have ways of keeping them there.

Relatively simpler animal brains, like those of insects, use entirely instinctive strategies to make this happen. But you can still tell from observing them that, from a logical standpoint, they are operating with both awareness and attention. This alone doesn’t make their mind subprocess comparable to ours in any way we can intuitively identify with, but it does mean that they have a mind subprocess. They are very capable at shifting their focus when inputs fall outside expected ranges, and they then select new behaviors to deal with the situation. Do they “go back” to what they were dong once a minor problem has been stabilized? The free energy principle doesn’t answer questions like that directly, but it does indirectly. Once a crisis has been averted, the next most useful thing the animal can do to avoid a big surprise in its future is to return to what it was doing before. But for very simple animals it may be sufficiently competitive to just continually evaluate current conditions to decide what to do next rather than to devise longer-term plans. After all, current conditions can include desires for food or sex, which can then be prioritized to devise a new plan on a continuous basis. Insects have very complex instinctive strategies for getting food which often depend on monitoring and remembering environmental features. So even though their mind subprocess is simple compared to ours, it must be capable of bringing things to attention, considering remembered data, and consulting known strategies to prioritize its actions to choose an effective logical sequence of steps to take.

People usually consider the ability to feel pain as the most significant hallmark of consciousness. Insects don’t have nociceptors, the sensory neurons that transmit pain to the brain, so they don’t suffer when their legs are pulled off. It is just not sufficiently helpful or functional for insects to feel pain. More complex animals make a larger investment in each individual and need them to be able to recover from injuries, and pain provides its own signal which is interpreted within an expected range to let the mind subprocess know whether it should ignore or act. Every sensory nerve (a bundle of sensory neurons) creates its own discrete and simultaneous channel of awareness in the mind. If you have followed my argument so far, you can see that what we think of as our first-person awareness or experience of the world is just the mind subprocess doing its job. Minds don’t have to be aware of themselves, or have compression or metacognition, to feel things. Feelings are, at their lowest level, just the way these nerve channels are processed for the mind subprocess. Feelings in this sense of being experienced sensations are called qualia. We distinguish red from green as very different qualia, but we could never describe the difference to a person who has red-green color blindness. The feelings themselves are indescribable experiences; words can only list associations we may have with them.

We don’t count each sensory nerve as its own quale (pronounced kwol-ee, singular of qualia), even though we can tell it apart from all others. Instead, the brain groups the sensory nerves functionality into a fixed number of categories, and the feeling of each quale as we experience it is exactly the same regardless of which nerve triggered it. Red looks the same to me no matter which optic nerve sends it. The red we experience is a function of perception and is not part of nature itself, which deals only in wavelengths, so our experience seems like magic as it is supernatural. But it isn’t really magic because there is a natural explanation: outside of our conscious awareness in the mind subprocess, the brain has done some information processing and presented a data channel to the mind in the form of a quale. The most important requirement of each sensory nerve is that we can distinguish it from all others, and the second most important requirement is that we can concurrently categorize it into a functional group, its quale. The third most important requirement is that we monitor each channel for being what we expect, and that unexpected signals then demand our attention. These requirements of qualia must all hold from ant minds to human minds, and, in an analogous way, for senses in yeast. But the detective and responsive range in yeast is much simpler than in ants, and that in ants is much simpler than in people. As we will see, the differences that arise are not just quantitative, but also qualitative as they bring new kinds of function to the table.

  1. Matthew Cobb, “Why sponges are animals“, Why Evolution is True blog by Jerry Coyne, 2009
  2. The enteric nervous system can function pretty well even if the vagus nerve connecting it to the central nervous system is severed, see Function of the Vagus Nerve
  3. When Gut Bacteria Changes Brain Function, David Kohn, June 24, 2015, The Atlantic
  4. Pochu Ho and David A. Ross, More Than a Gut Feeling: The Implications of the Gut Microbiota in Psychiatry, Biol Psychiatry. 2017 Mar 1; 81(5): e35–e37. doi: 10.1016/j.biopsych.2016.12.018
  5. Arturo Casadevall,Fungal virulence, vertebrate endothermy, and dinosaur extinction: is there a connection?, Fungal Genet Biol. 2005 Feb;42(2):98-106. Epub 2005 Jan 5.
  6. Arturo Casadevall,Fungi and the Rise of Mammals, August 16, 2012, PLoS Pathog 8(8): e1002808
  7. Dinosaurs weren’t warm-blooded, study suggests
  8. Dinosaurs really were warm-blooded, study suggests
  9. Bee learning and communication, Wikipedia
  10. List of animals by number of neurons, Wikipedia
  11. List of animals by number of neurons, Wikipedia
  12. brain-to-body mass ratio, Wikipedia
  13. Encephalization Quotient (EQ), Wikipedia
  14. Osvaldo Cairό, External Measures of Cognition, Front Hum Neurosci. 2011; 5: 108.
  15. Friston, Karl, The free-energy principle: a rough
    guide to the brain?
    , The Wellcome Trust Centre for Neuroimaging, University College London, Queen Square, London WC1N 3BG, UK

Leave a Reply