Cephalove

Cephalopods are great subjects for studies on vision, because they are so dependent on their vision that you can get robust behavioral effects by manipulating the visual environment of a test animal. In some new research in the October edition of the Journal of Experimental Biology, CM Talbot and J Marshall (from Queensland) investigate the visual system of the pyjama squid (S. lineolata) and two species of cuttlefish (S. plangon and S. mestrus) – specifically, to find out whether they can respond to polarized light, and in the case of S. lineolata, how photoreceptors are distributed on its retina. I’ve blogged about a study on visual perception in Nautilus before, as well as a study on the retinal topography of squid, so if you would like to see more of the same sort of research, check out those posts.

In these two papers, the authors assessed the ability of their experimental specimens to respond to polarized light by monitoring their optokinetic and optomotor responses to a rotating drum. The optokinetic response is the movement of an animals eyeballs to follow a moving object in the visual environment, while the optomotor response is the movement of the animal’s body to follow movement in the visual environment. The experimenters monitored the optokinetic response in S. lineolata, because it tends to stay motionless on the substrate, partially buried – as such, it will not exhibit an optomotor response under most circumstances. On the other hand, S. plangon and S. mestus both tend to hover in the water, and so show optomotor responses more readily.

A basic scheme of the apparatus used is shown in this figure from the S. lineolata paper:

The animal is in the tank (in this case, prevented from burying itself by being enclosing in a transparent cylinder,) while a drum is rotated around the tank. By varying the pattern on the drum, it is possible to determine the sensory abilities of the animal – assuming that animals generally don’t inhibit optokinetic or optomotor responses, the animal will respond to any pattern it can perceive. If the animal can’t perceive the pattern on the drum (for example, if the drum is visually continuous, as is the case with an all white drum,) it will not perceive any motion and the response will be absent.

The authors used a drum that consisted of alternating stripes of orthogonally oriented polarization filters – that is, the drums were striped, but the difference between adjacent stripes was only in the direction of polarized light that they transmitted. All the stripes transmitted the same total amount of light, and had the same appearance. Thus, the animals would only show an optomotor or optokinetic response to these drums if they could perceive the direction of polarization of the light.

In fact, this is what happened, in all three species. Two control drums were used, one of alternating black and white stripes (to make sure the animals had otherwise normal optokinetic and optomotor responses) and one of a uniform-direction polarization filter (to make sure that the animals weren’t responding to some other part of the drum – the tape used to hold it together, seams resulting from the drum’s construction, etc.,) making it pretty clear that the animals were responding to the alternating directions of polarization and not anything else.

This result is pretty unambiguous, but I’d like to point out a problem that this type of experiment presents in its interpretation: specifically, it’s very difficult to interpret negative results. In this case, it’s very easy to know what it means in terms of the animal’s sensory ability when it responds to a stimulus: it means the animal can detect that stimulus. But what if the cuttlefish didn’t respond (for example, as was found in a very similar study by Darmaillacq and Shashar (2008) in a different species of cuttlefish, Sepia elongata)? It’s hard to know what that means – did the animal fail to perceive the stimulus, or did the stimulus just not mean enough to generate a behavioral response? This is a general problem that crops up in studies on sensation and perception in animals, or any study that relies on an animal perceiving something and emitting a behavioral response. Many things need to happen to get any behavioral response to a stimulus, even one as apparently simple as eye movements. The animal must have a functional sensory apparatus appropriate to perceive the stimulus, it must have the energy and intact musculature to perform whatever behavior it is you’re looking for, it must be expressing no other behaviors that might mask or supress the behavior of interest, it must be motivated to perform the behavior of interest, etc. A negative result in such an experiment means that one of these many things is not the case, but because it’s so difficult to tell the difference between all of these steps between “stimulus” and “behavior”, it’s hard to say what exactly it is that the animal isn’t doing. Is it failing to sense the stimulus, is it failing to respond because the stimulus isn’t relevant, or is it failing to behave because it’s afraid, or stressed, or tired? Darmaillacq and Shashar note that S. elongata has retinal anatomy that looks like it would allow the animal to sense polarized light, and so they are (wisely) wary of claiming that their subjects could not perceive polarized light – but there’s no way to make any claim about S. elongata‘s vision at all from these results (except, of course, the most conservative assertion that S. elongata failed to show an optomotor response to a certain type of polarized-light stimulus under the experimental conditions used in that specific study.)

Fortunately, though, Talbot and Marshall found positive results, and so avoided that quagmire all together. It turns out that all three species they studied can respond to polarized-light stimuli with optokinetic or optomotor responses. They went on to examine the distribution of photoreceptor cells (also called “retinal topography”) in the S. lineolata retina. If you’ll remember from my post on squid visual ecology, it turns out that you can relate the retinal topography of cephalopods to their lifestyle – squids that live near coasts have retinas that are specialized to allow the animal to see below it clearly, whereas oceanic squids have retinas that are specialized for monitoring the water column above them. What might we expect from S. lineolata, an animal who spends much of its time buried in sand? The sensible answer is that is eyes would be specified to look up, since that’s where its predators and prey would be in most cases. Let’s take a look at what Talbot and Marshall found:

The darker the blue is, the higher photoreceptor density is in that area. It turns out that the striped pyjama squid does indeed have a high photoreceptor density in the ventral part of its retina, which probably gives it good visual acuity in the upper part of its visual field (if you don’t know why this is, check out this explanation of image formation in the eye for a primer.) This fits in neatly with what we know about the lifestyle of this squid.

I hope these studies represent the start of a trend towards the study of less “classical” cephalopod species (the “classical” ones being Loligo pealai, Octopus vulgaris, Sepia officinalis.) There’s a lot to learn from the less common species of cephalopods, due in part to the fact they we know very little about most of them.

Thanks for reading!

Talbot CM, & Marshall J (2010). Polarization sensitivity and retinal topography of the striped pyjama squid (Sepioloidea lineolata – Quoy/Gaimard 1832). The Journal of experimental biology, 213 (Pt 19), 3371-7 PMID: 20833931

Talbot CM, & Marshall J (2010). Polarization sensitivity in two species of cuttlefish – Sepia plangon (Gray 1849) and Sepia mestus (Gray 1849) – demonstrated with polarized optomotor stimuli. The Journal of experimental biology, 213 (Pt 19), 3364-70 PMID: 20833930

Here it is, finally: the post you’ve been waiting for. Having already convinced you that you should care about the possibility of consciousness in cephalopods in Part 1 and having briefly outlined the state of research on consciousness in non-human animals in Part 2, I’ll get right down to it and discuss the possibility of consciousness in cephalopods in this post. If you’re unfamiliar with the topic, I suggest reading Parts 1 and 2 of the series – in this article, I’ll be very brief with explaining some concepts that are explained in more detail there.

In this post, I’ll reference Jennifer Mather’s 2008 article (which I can’t recommend highly enough) “Cephalopod consciousness: behavioural evidence” and Edelman and Seth’s article (which is also an excellent read) “Animal consciousness: A synthetic approach”. These are both review articles, so I’ll be citing their descriptions of other people’s experiments a lot – I know this is bad practice, but hey, this is a blog – I can get away with it. I’ll cite a research study itself if I discuss it detail, but I’ll mostly be sticking to the arguments outlined in these two papers.

If you’ll remember from the last post in the series (and I’m sure you do, but I’ll summarize here anyways,) there are several methods that researchers use to get at the question of consciousness. Most directly, there is accurate self-report, whose use is limited to animals with whom we can communicate through language. This is not a useful approach with cephalopods, who (thus far) are not known to use language.

In the absence of language, animals can be trained to report on their experience (such as by performing a task for a reward when they detect a certain stimulus.) This approach is not well developed in cephalopods. Octopuses have been trained through reward and punishment to attack certain stimuli and not others in many studies; despite this, there is no protocol (that I know of) that to train octopuses with a task that would allow hypotheses about the animal’s awareness of its own experience to be tested directly in the ways that has been done for primates (for example.) Nevertheless, there is some evidence suggesting that cephalopods may be consciously aware from studies that use specific trained tasks.

Mather makes the point that the ability of cephalopods to learn a variety of tasks reliably and quickly, and then to forget them afterwards, makes them good candidates for at least primary consciousness because it implies the sort of behavioral and cognitive complexity that appears to be associated with consicousness in vertebrates. As an example of this, when experiments on visual discrimination in the octopus were done (mostly by M. J. Wells in the 1970′s and earlier,) experimenters attempted to discover the basis by which octopuses discriminated between two visual stimuli. In a sense, they were looking at how the octopus categorized stimuli. A number of hypotheses were generated to explain this within a simple, computational framework, but it was eventually concluded that octopuses (that is, individuals of the species O. vulgaris, the common octopus) don’t use a set of simple rules to categorize objects. Rather, Mather argues, they “[evaluate] a figure on several dimensions and [generate] a simple concept, where [a] concept is an abstract or general idea inferred or derived from specific instances.” Other evidence for the ability of cephalopods to exhibit learning like that taken to indicate cognitive ability (and thus the potential for consciousness) in vertebrate species comes from more complex learning tasks. The spatial learning abilities of cephalopods have been studied and it has been found, in general, that they might be capable of spatial learning to rival that of commonly used vertebrate laboratory species (such as rodents,) as long as the apparatus used is adapted to their capabilities (obviously, we cannot expect a rat and a cuttlefish to learn the same things in the same circumstance, but both can show impressive spatial learning given the right circumstances.)

Consciousness can also be suggested by non-trained behavior of an animal. As I’ll address at more length in a little bit, such evidence in cephalopods is found in accounts of their foraging behavior, their responses to novel objects in their environment, and the presence of sleep-like (and possibly REM-like) states. Most convincingly, in my mind, is the evidence suggesting the superior behavioral flexibility of cephalopods.

One of the more straight-forward tasks that is used to suggest conscious awareness in human and non-human alike is the mirror self-recognition task (MSR). What happens when you show a cephalopod a mirror – does it recognize itself, or does it treat its reflection as if it were another animal? Mather cites a personal communication suggesting that cuttlefish fail the MSR. You can see for yourself, in this great video of a cuttlefish at Epcot being shown its own image on an electronic screen. It turns very dark and pursues its image as if it were confronting another cuttlefish. The mechanics are a bit different, but it’s essentially similar to the MSR:

Mather makes a case for the cognitive abilities of cephalopods using the results of a study that looked at the strategies that octopuses use to open bivalves (which she discusses in this interview on Scientific American.) Not only do octopuses use different techniques for opening clams of different species (that is, they pry open the shells of the weaker ones, and drilling holes through the shells of the stronger ones, but they could switch strategies if one wasn’t working properly. When the experimenters took the clams that the octopus normally ate by prying open and wired them together so that they couldn’t be opened, the octopus figured this out and started drilling. This sort of behavioral flexibility, particularly the selection of one possible behavior among many on the basis of its effectiveness in a specific situation, could be attributed to some sort of centralized “executive processor” that might associated with consciousness.

Although definitions of “play” are often disagreed-upon, Mather argues that some octopuses have been observed playing with objects. While the existence of play behavior in a species is not indicative of consciousness, it suggests the possibility of consciousness; object play is, as Mather says, “something that intelligent animals do” to allow them to learn about things in their environment. (You can read my discussion of one study of octopus play at this link.)

It has also been (rather famously) argued that some octopuses have evolved the ability to use tools – specifically, one species of octopus (Amphioctopus marginatus) has been seen carrying empty coconut shells across the sea floor, which they use as mobile shelters. It can be argued that tool use is only possible when the animal using the tool has developed some rather sophisticated cognitive awareness of their surroundings that allows them to appreciate how an object can be used for a certain function. Here’s a video of this behavior, taken by one of the authors of the 2009 paper on the subject:

The comparative neuroanatomical argument for consciousness (epitomized by Panksepp’s “triangulation” approach to the problem, which recommends using affective, behavioral, and neural approaches together to infer consciousness in non-humans) is much more difficult to make for cephalopods than it is for vertebrates. The reason for this is simple: humans are vertebrates, and share many features of the anatomical and functional organization of our brains with other vertebrates. If you dissect a rat brain and a human brain side-by-side, most of the parts in one of them will show up in the other one in some form. Thus, it is rather easy to make an argument from analogy claiming that, because the brain activity and behavior of the two species in some situation are similar, it is likely that their experiences are likely to also be similar. It is harder to make this argument between people and cephalopods, because there is no direct equivalence between any of the parts of a cephalopod brain and the parts of a human brain, with the possible exceptions of the retina and primary visual processing areas of the two species and some parts of memory systems (eg. the vertical lobe system in cephalopods and the hippocampus in humans.) Even these are examples of convergent evolution (meaning they started from different places and got the same functional result,) and so the equivalences between these two brain areas in cephalopods and humans are only approximate, and based on a very limited knowledge of the functions of the cephalopod brain. Despite this difficulty, there are some overall features of the cephalopod brain that suggest consciousness, including its apparent organization as a complex integrator for sensory information, its lateralization, and its patterns of activity during sleep and wakefulness.

Edelman and Seth argue that we have a good reason to suspect that birds have some sort of consciousness, based on apparent anatomical and functional correspondence between the brains of mammals (including humans) and birds. They show this figure, which illustrates this correspondence – it shows diagrams of a human brain and a finch brain, with homologous structures colored similarly in each diagram:

As you can see, human and zebra finch brains (and indeed, mammalian and avian brains in general) have somewhat similar layouts, which allows one to make an argument for the inference of similar subjective states that correspond to certain types of neural activity in multiple vertebrate species. The basic logic is simple: if the brains are similar, and most of the output of the brain (that is, behavior) is similar in a certain situation, the rest of the output of the brain (that is, affective and/or conceptual awareness, eg. consciousness) is reasonably likely to be similar.

At the bottom of the figure, though, they show the octopus brain. Notice that it’s done in a completely different color scheme. This is because the functional or anatomical subunits of the octopus brain are not clearly equivalent to those found in vertebrate brains. A few localized functions of the octopus brain can be compared to those of vertebrate brains – for one, the vertebrate retina and the octopus optic lobe have apparently analogous structures and functions (that being the initial processing of visual information,) and the vertical lobe/medial superior frontal lobe system of the octopus is known to be involved in memory consolidation, and may have a microscopic structure that resembles that found in the mammalian hippocampus (for more info on this, check out Young, 1991, who makes the argument that the cellular structure and computational properties of the mammalian hippocampus might resemble those of the octopus memory system.)

Functionally, however, it is possible to find similarities between cephalopod brains and vertebrate brains, even if it is difficult to do so anatomically. Mather discusses the evidence for lateralized specialization of function in the cephalopod brain at length (that is, the general feature of the brain that two mirror-image halves can work somewhat independently, and may have different functions.) Lateralization is seen in humans and other primates, and seems to be one evolutionary result of the need for cortical tissue to be both locally differentiated and highly interconnected; it allows for more specialized cortical areas, because the right and left sides of the brain need not be functionally equivalent. Thus, the apparent laterality of the octopus brain (as this is already getting on in length, I’ll let you check out Mather’s article for a more complete discussion) might suggest that it has also evolved the sort of complex cognitive capacities that lateralization is associated with in mammals.

Finally, EEG-like recordings have been done in both octopus and cuttlefish, leading to the general (but very preliminary) finding that cephalopods have complex, low-frequency “background” electrical activity in some parts of their brains that seems to vary with their states of consciousness. In addition, they show sensory-evoked changes in this activity, in the same way that human EEGs do. This suggests that some of the gross functional properties of the cephalopod brain might resemble those of mammals on a system-wide level.

All of the arguments by analogy should be taken with a grain of salt, because while it is interesting to consider the possible theoretical importance of the apparent similarities between octopus and vertebrate brains, it seems premature at this point, given how little we know about them. While laterality, distributed low-amplitude electrical activity, and a certain kind of memory system architecture are found in the brains of animals who are almost definitely conscious (eg. mammals and birds,) it’s hard to say that their presence in such highly divergent nervous systems (eg. those of vertebrates and cephalopods) has the same set of functional consequences in all cases.

So there it is – these are the arguments for consciousness in cephalopods. It’s an astoundingly complicated and difficult question, and one that I’m sure I haven’t done justice to. Look for the last planned article of the series later this week, where I’ll reflect upon these arguments and figure out where I stand (and also hopefully invite discussion) on the science of cephalopod consciousness.

Thanks for reading!

P.S. Today is my first day of classes for the Fall semester. Wish me luck!

MATHER, J. (2008). Cephalopod consciousness: Behavioural evidence Consciousness and Cognition, 17 (1), 37-48 DOI: 10.1016/j.concog.2006.11.006

Edelman, D., & Seth, A. (2009). Animal consciousness: a synthetic approach Trends in Neurosciences, 32 (9), 476-484 DOI: 10.1016/j.tins.2009.05.008

Young, J. (1991). Computation in the Learning System of Cephalopods Biological Bulletin, 180 (2) DOI: 10.2307/1542389

Finn, J., Tregenza, T., & Norman, M. (2009). Defensive tool use in a coconut-carrying octopus Current Biology, 19 (23) DOI: 10.1016/j.cub.2009.10.052

In this second post of the series “Cephalopod Consciousness”, I’ll talk about the methods that scientists have used to attempt to study consciousness in animals. For perhaps the first time in the history of this blog, I’ll write about science without making any specific reference to cephalopods – I’m saving that for part 3. Here I’ll just cover enough background get a basic handle on the study of consciousness in non-humans, so that I can talk all about its application to cephalopods next time.

I’ll refer primarily to three review articles as I move through the various paradigms used to argue for or against non-human consciousness. These articles are Animal consciousness: a synthetic approach by Edelman and Seth (2009), Subjective experience is probably not limited to humans: The evidence from neurobiology and behavior by Baars (2005), and Affective consciousness: Core emotional feelings in animals and humans by Panksepp (2005). There are many good articles and books on the topic that I am not covering here, so feel free to point out what might be better/useful sources in the comments if you think I’ve missed something important.

In any case, let’s dive right in!

We have to start out assuming that the question of consciousness in non-human animals is worth investigating (eg. that my last post in the series contained at least one valid argument – I might be pushing my luck, but bear with me!) Where do we start?

The first thing to do is to operationalize consciousness. We have to determine how we will identify consciousness in non-human animals, if it exists. The classic way of studying consciousness in humans is through “accurate report”, which Edelman and Seth (2009) define as “a first-person account of what an individual is experiencing, made without the attempt to mislead.” Assuming that you believe that other humans are actually conscious (which can be argued; I won’t get into that here, though,) this is as direct a way as any to study consciousness. It is, however, very difficult to do with animals, as we for the most part lack any reliable form of verbal communication with non-humans. Notable possible exceptions to this include parrots (like Alex the Grey Parrot, who learned language well enough to pretty unambiguously demonstrate cognitive capacities such as numerical representation and the ability to categorize objects) and some chimpanzees who have been taught to use simple language (for example, Washoe, who was taught to use American Sign Language to communicate with her keepers.) Despite these exceptions, linguistic reports remain a rare and difficult-to-use tool for studying consciousness in animals.

One way of working around the inability of most animals to use language (and our inability to interpret the other ways they might be projecting information) is to allow the animals to report on their experience through some sort of trained response, such as by pressing a lever, pushing a button, or another physical activity. For example, Baars (2005) describes a study (Cowey and Stoerig, 1995) in which Macaques were trained to touch a screen where a target stimulus appeared, and then also to indicate (by pressing a button) whether they had perceived any stimulus on the screen (known as a “signal-detection task”, this is a pretty standard way to determine whether an intact animal can sense something.) After damaging parts of the cortex that process visual information in these monkeys, the experimenters found that they continued to point to the correct spot, but they not longer reported seeing a stimulus when the stimulus was in a certain part of the visual field. This parallels a phenomenon known as “blindsight” in humans, where a subject will claim not to perceive anything in a part of the visual field but will otherwise show basically normal behavioral responses to objects in that portion of the visual field. By training the monkeys in this study to report on their experience, the authors of this study were able to show that their awareness of their sensory world is separable from the at least some of the basic functionality of their sensory world, arguing that they have some sort of conscious perception of the world on top of the ability to make motor responses to sensory stimuli. By providing a way for animals to make “commentary” on their experience, Baars claims, methods like this provide a method of studying consciousness that is functionally equivalent to the method of accurate report in humans.

In some cases, animals do not need to be trained to show behavioral evidence of complex cognitive processes, which suggest (but importantly do not prove) the existence of consciousness. For example, as part of their arguments for the possibility of consciousness in birds, Edelman and Seth (2009) cite observations of birds exhibiting object constancy (which is the ability to attend to an object even though it leaves the visual field, such as when it is hidden behind another object – for example, peek-a-boo is fun because young babies do not have object constancy, and so they act as if you disappear when you are hidden from sight,) using and modifying tools, and changing their behavior based on their perceptions of being watched by other birds. They argue that these behaviors show that birds have a working memory and spatial cognition as well as “the ability to make sophisticated discriminations and to plan behaviors before executing them.”

Other behavioral experiments get at the question of whether animals have “selfhood” – that is, do animals have a sense of identity? Such a distinction between self and other is considered key to the sort of “higher-order” consciousness that humans have. The most classical method of doing this in humans and apes is by testing to see if they can recognize themselves in a mirror. This ability is rather straightforwardly called Mirror Self-Recognition (or MSR.) It has been used on many animals, and some that appear to have the ability to recognize themselves include dolphins, chimpanzees, gorillas, and (in one of my new personal favorite behavioral studies by Plotnik et al., 2006) elephants.

If you’re like me, you’re a bit troubled right now. These behavioral methods fall short of actually addressing consciousness per se, and they would never fly as an argument for consciousness in animals in and of themselves (actually, the results with macaques are a veritable one-hit KO in this argument, but only because they involve a species so closely related to humans – arguments from analogy to more distant evolutionary relatives require correspondingly more evidence to make.) Behavioral experiments do not solve the problem of identifying the internal states of animals, which is what we mean when we say “consciousness.” In a particularly lucid explanation of how this problem might be solved despite the shortcomings of behavioral evidence to inform us about internal states, Panksepp (2005) argues for a “psycho-neuro-ethological triangulation” strategy to address the problem of animal consciousness. According to this strategy, we should use neurological processes (some well-studied ones are the mobilization or production of neuroactive chemicals in the body and changes in EEG patterns) as a link between the behaviors we know to be associated with conscious states in humans (in his argument, emotional states in particular) and analogous behaviors in animals. For example, we know that humans feel pain when they are burned by a hot stove (the “psycho-” component of the strategy), and they then withdrawal from the stove and attend to the site of injury. If we watch a rat touch its paw to a hot piece of metal and get burned, we can observe the same sort of reaction (the “ethological” component of the strategy.) Finally, we can attempt to identify neural processes in the rat that correspond with this behavioral reaction in the rat and in humans, as well as neural processes that correspond specifically with the perception of the event (in this case, pain) in humans. If we find that homologous neural processes and behaviors occur in both cases, we have a good case for suggesting that analogous subjective experiences also occur.

In apparent agreement with this idea, both Baars (2005) and Edelman and Seth (2009) make a case for the identification of consciousness in non-humans through the study of neural processes that resemble those associated with human consciousness. The latter authors, in their argument for the possibility of consciousness in birds, identify the presence of human-like (or conscious-like) EEG patterns in birds and the presence of a neural circuit analogous to the thalamocortical circuit of humans (which has been shown through studies of brain-damaged patients and neuroimaging studies to be closely associated with consciousness) as evidence supporting the interpretation of bird behavior as indicative of consciousness. Baars argues that the apparent evolution of these brain structures suggests that consciousness is universal at least among all mammals. Because conscious states and phenomena (for example, wakefulness, REM sleep, and sensory perceptions) are modulated by brainstem structures and “seated in” the thalamocortical circuit, structures which have not undergone much overall structure change throughout mammalian evolution, they are likely to be conserved across all mammals. This is what he claims – I regrettably do not have the expertise in paleobiology or comparative anatomy to agree with or dispute his claims about brain evolution, but they sound like they could be disputed.

In essence, the argument for consciousness in animals remains an argument by analogy from the easily acceptable existence of consciousness in humans. It uses both behavioral and neural evidence to build this case. Critically, though, it makes use of comparative neuroscience to support the existence of consciousness in non-human vertebrates. Remember, though, that non-human mammals and birds are relatively closely related to people, and so their neuroanatomy is (arguably) suitably homologous to human neuroanatomy to make such an argument. What can we make of this line of inquiry when we try to apply it to an animal that is, evolutionarily speaking, much more distantly related to humans – say, an octopus?

Tune in next time to find out!

(For those who are interested in the topic, the journal Consciousness and Cognition put out an issue dedicated to animal consciousness in 2005. It’s very worth checking out.)

BAARS, B. (2005). Subjective experience is probably not limited to humans: The evidence from neurobiology and behavior Consciousness and Cognition, 14 (1), 7-21 DOI: 10.1016/j.concog.2004.11.002

Edelman, D., & Seth, A. (2009). Animal consciousness: a synthetic approach Trends in Neurosciences, 32 (9), 476-484 DOI: 10.1016/j.tins.2009.05.008

PANKSEPP, J. (2005). Affective consciousness: Core emotional feelings in animals and humans Consciousness and Cognition, 14 (1), 30-80 DOI: 10.1016/j.concog.2004.10.004

Plotnik JM, de Waal FB, & Reiss D (2006). Self-recognition in an Asian elephant. Proceedings of the National Academy of Sciences of the United States of America, 103 (45), 17053-7 PMID: 17075063

Cowey, A., & Stoerig, P. (1995). Blindsight in monkeys Nature, 373 (6511), 247-249 DOI: 10.1038/373247a0

Let’s take a minute to talk about connectomics.  No, not genomics.  No, not metabolomics.  Not any of the other -omics, but connectomics.  It’s a new-ish field that the computational neuroscience geek in all of us can love.

By way of introduction, the “connectome” is the “network of elements and connections forming the human brain” (according to Sporns et al, 2005).  Let’s forget the part about human brain, and (for the purposes of this post) say that a connectome is the set of all the neural connections in a nervous system.  Connectomics is the subfield of neuroscience that attempts to understand the structure and function of nervous systems by studying these connections as a whole.  The goal of this discipline is to determine the anatomy of the nervous system to a very fine scale (on the order of individual cells), and then relate this structural information to the functioning of the nervous system.  “Solving” a connectome is achieved when the connections between all the neurons in a nervous system are mapped.  This has been done for C. Elegans, a nematode who has only 302 neurons, by the use of electron microscopy – this work is summarized in White et al., 1986.  Since then, we’ve expanded our goals, with the Human Connectome Project aiming to solve the connectome of the human brain.   Let’s step back for a second, though, and ask: why do we want to know all of this?

Since the elucidation of the electrical properties of neurons (which, by the way, started with the squid giant axon – you can read one of my older posts about that topic here ), neuroscientists have been interested in the information that neurons carry.  A nervous system can be thought of as an organ that processes information to tell the rest of the body what to do.  Some stimulus might impinge upon a sensory organ (let’s say, for example, that one sees a car speeding towards one’s self,) which causes a cascade of electrical activity through the nervous system, eventually leading to such diverse effects as the movement of one’s muscles to carry one’s self out of the path of the car, the emotional distress that occurs when one is almost killed by some jerk who isn’t paying attention to the road, and the realization that one has just gotten very lucky.  Later on, one might tell the story of this near miss to her friends over dinner, exhibiting the ability of the nervous system not only to process, but to store information for later use.  Further demonstrating this ability, one might learn not to walk in the middle of the road in the future.  The ability of animals to exhibit behavior, to move, feel, and learn, is all due (according to the dogma of neuroscience) to the processing of information by cells in the nervous system.

Now, it’s relatively routine to study how a single neuron processes information.  To sum it up very briefly, information enters a “typical” neuron in the form of electrical impulses on that neuron’s dendrites.  The information flows through the neuron (as an easy-to-get analogy, imagine electronic information flowing through the wire) and then a decision is made: at any given time, if a neuron is electrically excited enough, it can discharge an action potential.  An action potential is a burst of electrical activity that will travel through the neuron’s axon to affect the activity of other neurons that the axon makes synapses with.  Thus, one can imagine a general flow of information through a neuron from the dendrites, through the cell body, and out the axon.  A nerve cell is diagrammed below, with the dendrites and axon clearly labeled, showing the flow of electrical impulses through the neuron.  Such neurons are linked together to form functional circuits that accomplish complex tasks such as recognizing objects, coordinating movement, and recognizing when food tastes good or bad (to name but a few.)

Think of it this way: each neuron works like a tiny computer processor.  At any given time, it’s integrating all of the electrical signals coming to it, and deciding whether or not to fire an action potential.  Nervous systems can process information because they have many such relatively simple processors connected together to process that information.  (Keep in mind that this is a very simplified, and thus necessarily inaccurate description of the nervous system.  It will have to do for now, however, and gets across those aspects of the function of neurons that are most relevant to the problem of the connectome well enough.)  Thus, to understand the function of a nervous system, one only has to understand the functioning of each of its neurons.  This turns out to be incredibly difficult, and at best we only achieve approximations of this goal.  I’ll come back to this later on.

To study how a single neuron processes information is relatively easy, if we’re selective about which neurons we study.  For example, we could record electrical activity from the optic nerve while we expose the eyes of the animal we’re studying to light.  In this way, we would see the way that different visual signals are encoded by neurons in the optic nerve.  In fact, this has been done countless times in studies on the visual system in cats, frogs, ferrets, and many other species.  Using this method of electrical recording, we can determine how neurons are functionally connected to each other, and how they respond to various inputs.  However, to record electrical information from a neuron requires that one physically place an electrode into the brain, and also requires that one focus on a single neuron or a small subset of neurons at a time.  To figure out how an entire brain functions in terms of the interaction of millions of individual neurons would be all but impossible using these electrophysiological methods.  In addition, this method can never be employed on humans, as it is considered unethical to put electrodes into peoples brains without an urgent medical need (of course!)

Another now-classical way of learning about the connections between neurons in a nervous system is through tracer studies.  In these studies, neurons in one part of the brain are dyed in some way.  Then, other parts of the brain (or the whole brain) can be examined to see if they have dye in them.  If they do, it’s concluded that the neurons make some sort of connection between the part of the brain where the dye was injected and the part of the brain where the dye was later seen.  This has many of the same downfalls as electrophysiological methods of figuring out neural circuits.  For one, it can only be done in a small group of neurons in any preparation, and so the connections in the nervous system must be mapped out piecemeal, a few at a time.  In addition, it is often difficult to tell what route an axon might take from the cell body to its destination, even if it is clear where each of these points are.

The difficulty that these methods have in resolving the microscopic structure of the nervous system beg for a faster, more flexible technique.  Even the reconstruction of the relatively simple nervous system of C. elegans, done using images from an electron microscope, had to be done largely by hand.  These processes are labor- and time-intensive, and do not lend themselves well to the reconstruction of nervous systems that may have millions or billions of neurons.

Enter the field of computational neuroscience.  Computational neuroscientists study nervous systems in terms of their information processing capabilities.  Standing at the junction of computer science and neuroscience, they have both the tools and the impetus to understand the details of the connectomes of whichever organisms they study.

An approach that has been taken in humans involves using the technique of diffusion tensor imaging, and MRI technique that can determine the direction that axons run in in an intact brain.  For example, the following image (by Thomas Schultz) shows a DTI-derived image of the connections that run through the midline of a living human brain:

Such images are of great potential use in studying brain lesions, doing studies on brain function, clinical diagnosis, and whole-brain level analysis of neural circuits.  However, they lack the resolution needed to map individual synapses, thus falling short (for the time being) of being able to comprehensively map the connections between neurons in a brain.  For this, we have to go to microscopy techniques that involve looking directly at neural tissue.  These can only been done in animals, because it is presently illegal to harvest brain tissue from humans for experimentals purposes (again, a no-brainer.)

By now, you’re wondering when I’ll mention a cephalopod.  After all, this is a blog about cephalopods.  You have every right to expect that I’ll mention squids, octopodes, or nautiloids at least once in each post.  Never fear!

Last week I came across a poster presentation write-up on Biomed Central called “Charting out the octopus connectome at submicron resolution using the knife-edge scanning microscope.”  As you might imagine, I was tickled.  A research team with members from Texas, Naple, Michigan, Illinois, California, and Seoul (including Graziano Fiorito, notable for his research on observational learning in the octopus) is working on reconstructing the octopus connectome using a mostly-automated 3D microscope called the knife-edge scanning microscope (KESM).  This microscope takes a block of tissue and slices it, taking a picture of each slice of the tissue as it is cut.  Then, a computer program can create a high-resolution 3D image of the tissue.  From this, the computer can (and this is the tricky part) automatically trace the paths that nerve cells take through the tissue, and – this being the goal of this research – reconstruct a detailed network showing the morphology of each neuron in the tissue.  For examples of the resulting images, you can see this gallery from the brain networks laboratory at Texas A&M.

Why the octopus?  Well, in an introduction that makes the comparative neuroanatomist in me jump for joy, the authors suggest that because “the neural architecture of this cephalopod mollusk differs markedly from that of any vertebrate… [investigating] the difference and simlarities between the neural architecture – or connectome – of the octopus and mammals, such as the mouse, may lead to deep insights into the computational principles underlying animal cognition.”  In their concluding remarks, the authors note that they “expect that this pilot study and the more detailed investigations to follow will allow fruitful comparisons of the neural circuitries of individual octopuses with different ecological life histories, as well as of animals that have been exposed to a variety of neurodegenerative insults… In sum, this approach should contribute greatly to our understanding of the computational architecture of invertebrates and ultimately provide insights into the differences between invertebrate and vertebrate cognitive capabilities.”

I’m intriguied by this article, but also a little dissappointed.  Mostly, I’m dissappointed that a more complete study isn’t out yet!  I’ll be watching these guys from now on, and I’ll cover any other publications they put out on the topic.  Hurrah for octopus connectomics!

In closing, I want to mention that a complete neuroanatomical picture of a nervous system does not actually explain its computational properties.  To understand how nervous systems process information, we need to know the physiology of each cell and the biochemistry of the interactions, a topic that is probably more complex than even the very fine-grained study of neuroanatomy represented by the studies I’ve mentioned here.  In terms of our understanding of nervous systems, however, connectomics offers and opportunity to study the relationship between the cellular structure of the nervous system and its overall capabilities – a relationship whose description has been one of the goals of neuroscience practically since its inception.

Thanks for reading!

Sporns, O., Tononi, G., & Kötter, R. (2005). The Human Connectome: A Structural Description of the Human Brain PLoS Computational Biology, 1 (4) DOI: 10.1371/journal.pcbi.0010042

Yoonsuck Choe, Louise C Abbott, Giovanna Ponte, John Keyser, Jaerock Kwon, David Mayerich, Daniel Miller, Donghyeop Han, Anna Maria Grimaldi, Graziano Fiorito, David B Edelman, & Jeffrey L McKinstry (2010). Charting out the octopus connectome at submicron resolution using the knife-edge scanning microscope BMC Neuroscience, 11 (Supplement 1), 136-137 : 10.1186/1471-2202-11-S1-P136

MAYERICH, D., ABBOTT, L., & McCORMICK, B. (2008). Knife-edge scanning microscopy for imaging and reconstruction of three-dimensional anatomical structures of the mouse brain Journal of Microscopy, 231 (1), 134-143 DOI: 10.1111/j.1365-2818.2008.02024.x

White, J., Southgate, E., Thomson, J., & Brenner, S. (1986). The Structure of the Nervous System of the Nematode Caenorhabditis elegans Philosophical Transactions of the Royal Society B: Biological Sciences, 314 (1165), 1-340 DOI: 10.1098/rstb.1986.0056

MORI, S., & ZHANG, J. (2006). Principles of Diffusion Tensor Imaging and Its Applications to Basic Neuroscience Research Neuron, 51 (5), 527-539 DOI: 10.1016/j.neuron.2006.08.012

In this post, I’ll be talking about octopus tactile sensation.  M. J. Wells and J. Z. Young did the classic experimental work on touch discrimination and learning in the octopus, although a bit of recent work has been done on the neurochemical basis of touch learning in the octopus (which I won’t get into here.)

We’ll focus on Tactile Discrimination of Surface Curvature and Shape by the Octopus (1964) by Wells.  This was one of his later papers in a series on tactile learning in the octopus.  Prior to this paper, Wells had already determined that octopus do not use proprioception to discriminate between objects (as a blindfolded person might do when trying to feel what an object is with his hand,) but rather use (almost exclusively) tactile cues about the object’s shape.  Let me explain.

It had been found that a blinded octopus could discriminate, on the basis of touch, between a sphere and a cube.  This could be explained by the presence of some sort of proprioception that monitors the relative position of the octopus’s arms in space – a system like this is known to exist in most vertebrates.  However, Wells carried out a series of experiments that show that this is, if anything, a very subtle factor contributing to the octopus’s ability to perform tactile discriminations.  He found that octopuses learn to recognize the corners of a cube as variations in texture, which are encoded in reference to the extent that the suckers contacting the object are deformed.  For example, a sucker that is on the corner of a cube will wrap around the corner, bending itself along a sharp angle.  This information is encoded as some sort of distinct textural component, and sent along to the brain where it can interact with learning centers (which I’ll discuss in a later post, hopefully) that will allow the octopus to remember what a particular texture means.  Thus, if you teach an octopus to respond to a cube (meaning that you reward it with food when it grabs the cube, and punish it with electric shock when it grabs another object, say, a sphere,) this theory would predict that it would also respond to any object which induces a similar deformation of the suckers that contact it, such as a rectangular prism, or a thin rod.  This is called a transfer experiment, because it tests the extent to which a learned task transfers to situations other than the one it was learned in.  Indeed, Wells found that he could substitute a thin rod for the cube, and the octopus will respond to it as if it is a cube, presumably because the suckers contacting the rod are bent into a relatively sharp angle, as are those contacting the edges of the cube.

This evidence alone didn’t quite clear up the question of how octopus performed touch discriminations, though – specifically, Wells’ experiment with the cube, sphere, and rod did not use enough variations of form and dimension to really probe the mechanism of touch discrimination.  Thus, Wells decided to conduct a number of transfer experiments between differently sized and textured cylinders in order to figure out the characteristics that octopuses use to identify objects by touch.  The stimuli he used are shown here:

The numbers under the cylinder cross-sections indicate their diameter in millimeters.  Wells notes that the octopus he is working with have suckers that are 10mm or less in diameter.  Knowing this, one can gauge the approximate deformation of a sucker that the different cylinders would produce.  For example, a 6mm wide cylider would induce a significant curvature in the sucker, whereas the 38mm cylinder would produce a very slight curvature, and thus would appear essentially “flat” to the octopus, if Wells’ theory is correct.

Wells quantified this difference, and generally found that the greater the difference in curvature between two cylinders, the easier the discriminate was.  This is great, but it doesn’t rule out the proprioception theory.  What if the octopus was actually “feeling” the position of the arm as it bent around the cylinder? 

To solve this problem, Wells used the two cylinders shown at the bottom of Figure 1, those labeled 8* and 6*.  These are “composite cylinders” were made of 7 small cylinders attached together, parallel to each other.  If the sucker-distortion hypothesis is correct, then these objects should be treated as equivalent to small cylinders, because they create equivalent deformation of the suckers contacting them.  If there is some mechanism that determines the shape or position of the grasping arm as a whole, then they should be treated as equivalent to the large cylinders, as they would require the same arm position and curvature to grasp as the 24mm and 18mm “simple” cylinders, respectively.  In fact, this is what Wells found, although the experiments with compound cylinders did not adhere quite as closely to his proposed model regarding differences in curvature as did those with the simple cylinders.  This might be expected, as the actual curvature experienced by the suckers is more variable with a more complex object.

Wells tested his idea further, by offering already trained octopuses P1 (which was grooved) and P4 (which was smooth.)  Other than their texture, these objects did not differ at all.  If sucker deformation is the basis of discrimination, we would predict that P1 feels most like a small-diameter rod to an octopus, as it would deform the suckers touching it greatly.  P4, on the other hand, would feel like a large-diameter cylinder, because, well, it is.  In fact, this is what Wells found – octopuses who were trained to take the larger diameter cylinder transfered this learning to the P1/P4 discrimination, and tended to take the smooth one.  Animals who were trained to take the smaller diameter cylinder tended to take the grooved one.

Wells goes on to consider discrimination using a cube with rounded corners (which proves difficult for an octopus) and a cube/rectangular prism discrimination (which is also difficult,) but I’ll let him tell you about those, as the point is amply made already.

What about the neuroanatomy of this system?  Wells provides us with a figure showing the cross-sectional structure of a single sucker, including the receptors that putatively monitor mechanical distortion of the sucker (in the area labeled “2″ at the rim of the sucker, towards the bottom of the diagram.

 

These receptors detect the mechanical forces from the object deforming the rim of the sucker, and then send this information to the ganglia of the arm.  It seems likely (although I don’t know that it has been tested) that these mechanoreceptors don’t send their information the whole way to the central nervous system, but rather input into some processing system in the nervous system of the arms first.  It would be interesting to know the minimum number of steps that information from the suckers might go through before it gets to the brain, because this would give a rough idea of how ”
processed” the sensation is before it gets to brain areas involved in learning.  J. Z. Young’s “Anatomy of the nervous system of Octopus vulgaris” didn’t seem to have a clear answer for this question, so for the time being, I’ll assume that it’s unanswered (though, if I’m wrong, please point me to the literature.)

All in all, this might seem like a poor way to distinguish two things from each other.  When you keep in mind the fact that octopus can’t discriminate objects based on weight, either, even though it can adjust its posture and muscle tone to hold a heavy object, it would seem that the octopus has a sort of crappy tactile sensory system.  We should ask, then: what does the octopus use this for?

When octopuses hunt, they often use a “blind” foraging strategy.  They will pounce on an area where prey is likely to be with their arms and web spread open and then feel for prey.  Alternatively, in rocky areas, an octopus might feel around in cracks for prey items.  If the octopus feels a prey item, she grabs it, moves it towards her mouth, and eats it.  It seems likely to me that the sort of touch discrimination that Wells trained octopuses with is not anything like what is demanded of them under ecological conditions.  For one, it is likely that octopuses are sensitive to movement as well, as they must be able to discriminate between rocks and prey, both of which might be similarly textured.  While hunting, an octopus also has other sensory systems to rely on.  They’re not primarily visual predators, but they can be, spotting prey and then attacking it (as they do when shown a live crab in an aquarium.)  They also probably have chemoreceptors on their arms which could help them identify objects under their web.  It doesn’t seem to me as if lacking proprioceptive input to the central nervous system is at all a deficit to the octopus in its natural habitat.

Thanks for reading!

M. J. Wells (1964). Tactile Discrimination of Surface Curvature and Shape by the Octopus Journal of Experimental Biology, 41, 433-445