- Email: [email protected]
Amphioxus neurocircuits, enhanced arousal, and the origin of vertebrate consciousness
Consciousness and Cognition xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Review article
Amphioxus neurocircuits, enhanced arousal, and the origin of vertebrate consciousness Thurston Lacalli Biology Department, University of Victoria, Victoria, BC V8W-3N5, Canada
A R T IC LE I N F O
ABS TRA CT
Keywords: Consciousness Brain evolution Dien-mesencephalon DiMes Midbrain Enhanced arousal Qualia
Gene expression studies have recently identified the amphioxus homolog of a domain comprising the combined caudal diencephalon plus midbrain, regions implicated in locomotory control and some forms of primary consciousness in vertebrates. The results of EM-level reconstructions of the larval brain of amphioxus, reviewed here, highlight the importance of inputs to this region for light and physical contact, both of which impinge on the same synaptic zone. The neural circuitry provides a starting point for understanding the organization and evolution of locomotory control and arousal in vertebrates, and implies that one of the tasks of midbrain-based consciousness, as it first emerged in vertebrates, would have been to distinguish between light and physical contact, probably sharp pain in the latter case, by assigning different qualia to each. If so, investigating midbrain circuitry more fully could lead to a better understanding of the neural basis of some forms of sensory experience.
1. Introduction The cephalochordate amphioxus (Branchiostoma) is now considered the closest living proxy for the ancestral chordate condition, and is hence an organism of considerable comparative and evolutionary interest (Gee, 2018 Holland, 2015). It is especially relevant to studies on the organization and circuitry of the central nervous system (CNS), as the expression patterns of developmentally important genes are substantially conserved between the amphioxus CNS and that of vertebrates (Holland, 2009; Holland et al., 2013). This makes amphioxus especially valuable as a model for early CNS development, and means the development of early brain circuits can, in principle, be studied in a context that is far less complex than in any vertebrate. In comparing the regional subdivision of amphioxus and vertebrate brains, a persistent problem has been to identify amphioxus homologs of the telencephalon and midbrain. Amphioxus lacks the expanded dorsal structures (e.g. cortices) that make these regions anatomically distinctive in vertebrates, as well as the sense organs whose input such structures evolved to serve. In young amphioxus larvae there are in fact no neurons in the dorsal nerve cord forward of the lamellar body, a pineal homolog, which is where any counterpart of telencephalon would be located. In older larvae, however, this same region receives dorsally directed branches from the rostral nerves (Lacalli, 2002a, 2004), a possible indication of a late-developing olfactory pathway. The olfactory system and telencephalon are generally considered vertebrate innovations (Satoh, 2005), but amphioxus has a complement of olfactory-related genes (Churcher & Taylor, 2010). A rudimentary homolog of the olfactory bulb may thus be present in older amphioxus larvae and
Abbreviations: ADBs, anterior group of dorsal bipolar neurons; C, enteropneust collar; CNS, central nervous system; Di, caudal diencephalon (thalamus and pretectum); DN, anterodorsal nerve; ESCs, epithelial sensory cells; inf, infundibular cells; IsO, isthmic organizer; JCs, Joseph cells; LPNs, large paired neurons (pairs 1 and 3); LMB, lamellar body; Mes, midbrain; MLR, mesencephalic locomotor region; PMC, primary motor center; psz, primary synaptic zone; RN, rostral nerve; TEM, transmission electron microscopy; Zli, zona limitans intrathalamica E-mail address: [email protected]. https://doi.org/10.1016/j.concog.2018.03.006 Received 26 February 2018; Received in revised form 14 March 2018; Accepted 16 March 2018 1053-8100/ © 2018 Elsevier Inc. All rights reserved.
Please cite this article as: Lacalli, T., Consciousness and Cognition (2018), https://doi.org/10.1016/j.concog.2018.03.006
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
Fig. 1. Evolutionary continuity of the dien-mesencephalon. A comparison of comparable neural domains in developing stages of a hemichordate, amphioxus and an amniote vertebrate, showing the dien-mesencephalon (in pink), which, for vertebrates, corresponds with the domain lying between the zona limitans interthalamica (Zli, vertical red line) and isthmic organizer (IsO, vertical blue line) and, in hemichordates, their homologs. The nervous system in hemichordates is largely intraepithelial, so the comparable domain there is the collar epithelium (C). In amphioxus, this zone is also the location of populations of dopaminergic neurons (green) corresponding to those found in the caudal diencephalon (Di) and midbrain (Mes) of vertebrates, and also (not shown) in the collar of hemichordates. There is currently no evidence in amphioxus for homologs of vertebrate hypothalamic dopaminergic nuclei (orange). It is also not known whether the absence of a fully internalized CNS in hemichordates is a secondary simplification, or reflects the ancestral condition, with internalization occurring early in the chordate lineage as shown here. Other abbreviations: 1–4, character states; CNS, central nervous system. Modified from Zieger et al. (2017). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
adults, but if so, it develops too late to play a role in early-stage larval behavior. In contrast to the telencephalon, a good deal more is known about the amphioxus homolog of the midbrain. From the expression patterns of genes required for CNS regionalization, this is now considered part of a larger domain, the dien-mesencephalic primordium (Albuixechi-Crespo et al., 2017), roughly equivalent to the posterior cerebral vesicle as defined anatomically. This domain includes the amphioxus homologs of the thalamus, pretectum and midbrain, and extends from the cluster of infundibular cells marking the junction between the anterior and posterior cerebral vesicle to approximately the end of somite 1. The corresponding region in vertebrate brain is the zone between the zona limitans intrathalamica and the isthmic organizer (Fig. 1), and though markers for these two signaling centers are absent in amphioxus, they are present in hemichordates (Pani et al., 2012), indicating the dienmesencephalon homolog there is the collar epithelium, which is neurogenic. The implication is that the dien-mesencephalon represents an identifiable subdomain within the nervous system of deuterostomes that predates the origin of chordates, and is retained in amphioxus despite the absence of markers for the signaling centers that define it in other taxa. Identifying the key functions this region plays in behavior is then important for what it tells us of the division of labor in deuterostome nervous systems generally, and in the brains of chordates more specifically. What we know of the neural circuitry and probable function of the amphioxus dien-mesencephalon comes largely from serial reconstructions of the larval CNS (see Wicht & Lacalli, 2005 for a summary). These identify the post-infundibular (=tegmental) neuropile as the principal integrative center for the animal at the stage where it first begins to swim. This accords with what is known of vertebrates, where a comparable part of the anterior brainstem, including structures in the basal diencephalon and midbrain, has a long evolutionary history as decision-making center, especially in relation to the control of locomotory responses (Grillner, Robertson & Stephenson-James, 2013; O’Connell & Hofmann, 2011). This is also where some forms of primary consciousness are thought to reside (e.g. Barron & Klein, 2016; Mashour & Alkire, 2013; Merker, 2007), which begs the question of whether amphioxus has anything useful to tell us about the evolutionary origin of vertebrate consciousness. The purpose of this paper is to explore this question, especially in relation to the evolutionary scenario laid out by Feinberg and Mallatt (2016a,b), who postulate a close association between consciousness and the emergence of the vertebrate optic tectum as a processing center for visual input. Consciousness, for the purposes of this review, refers to the most basic form of primary, or sensory, consciousness: subjective experiences created by the brain and characterized by specific sensations, or qualia, for each category of sensory input. Two issues are then considered. The first is that there is a plausible case to be made that midbrain-based consciousness may have evolved as part of an enhanced arousal system for predator avoidance (e.g. see Feinberg & Mallatt, 2016a, Chaps. 4 & 5), but this could have happened 2
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
Fig. 2. Escape circuits in young amphioxus larvae. Shows a schematic left side view of the anterior nerve cord of an amphioxus larva, from TEM reconstructions of a 12-day specimen of B. floridae. The cells of the escape circuit are highlighted in color, as well as the infundibular cells (inf, dark blue), which are important landmarks, separating the amphioxus homolog of the eye rudiment plus hyopothalamus from the dien-mesencephalon, as shown by the black horizontal bars. The primary synaptic zone, located in the latter (psz, circled in red, and paired at this stage, as there would be a corresponding set of inputs on the right side of the cord), is where the axons of epithelial sensory cells (ESCs) entering the nerve cord via the first pair of dorsal nerves (DN, yellow) first encounter and synapse to dendrites belonging to the principle interneurons (LPNs, green) of the primary motor center (PMC), which are a major source of excitatory input to the escape response. Inputs are also received here from the anterior group of bipolar cells (ADBs, also yellow), which are intramedullary sensory neurons located on either side of the lamellar body (LMB, light blue) about halfway along its length, in a region that is probably somewhere near the front of the midbrain in terms of homology. Horizontal blue bars indicate the axial extent of selected structures in the mid-larval phase, and (dashed lines) their expansion during later development. The Joseph Cells, for example, first develop caudal to somite 1, but extend forward to the front of the lamellar body in the adult, at least in B. lanceolatum, an observation as yet unconfirmed for B. floridae. The transition point between somites 1 and 2 coincides approximately with the caudal end of the lamellar body as indicated by the blue bar. Modified from Lacalli (2017). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
as image-forming eyes and tectal processing evolved, in basal vertebrates that were themselves predators, or before, at a time when vertebrates were more like amphioxus in being filter feeders rather than predators, with only rudimentary eyes and a correspondingly less developed tectum. Consciousness perception of light, if it did evolve that early, would most likely have been used for distinguishing light from its absence, which leaves open the question of how, from such a modest beginning, the ability to construct a consciously perceived visual field evolved. Besides photoreceptors, the other main source of input to the amphioxus post-infundibular neuropile is from mechanoreceptors, which implies, as a second point, that a mechanism would have been needed as consciousness emerged in vertebrates to ensure that light and contact stimuli (e.g. sharp pain) were subjectively experienced in different ways. If so, this would simplify the problem of understanding how different qualia are generated at a circuitry level and made to differ, as we would then need to focus on only these two, and the neural pathways associated with them. Technical advances in imaging and CNS reconstruction can now provide the kind of detailed knowledge of synaptic circuitry needed to investigate such issues in model vertebrate systems (Ahrens, Orger, Robson, Li, & Keller, 2013; Friedrich, Genoud & Wanner, 2013; Hildebrand et al., 2017), so it is relevant also to briefly address the question of how, in practice, neural correlates of consciousness might reveal themselves. 2. Ventral dien-mesencephalon: a core integrative center The principle zone of neuropile in young amphioxus larvae occupies the ventral portion of the posterior half of the cerebral vesicle, a region extending from the infundibular cells and the front of the lamellar body to approximately the end of somite 1 (Fig. 2). Hence it is post-infundibular and lies largely within the confines of the dien-mesencephalon as defined by molecular markers. Based on the larva so far most fully reconstructed, ca. 12 days in age (section numbers, where mentioned, refer to this specimen), inputs to the post-infundibular neuropile are from the frontal eye, from an assortment of sensory-type neurons of the preinfundibular (hypothalamic) part of the anterior cerebral vesicle, including the balance organ, from the lamellar body, and from rostral sensory cells whose axons enter the nerve cord via the paired rostral and anterodorsal nerves (Lacalli, 1996, 2002a; Lacalli, Holland & West, 1994; Lacalli & Kelly, 2003a). Escape behaviors are induced most easily by mechanical stimulation of the rostrum, and sensory fibers originating there converge on the primary synaptic zone (the psz, which is also paired when it first forms, see Fig. 2), along with fibers from the frontal eye and synaptic terminals belonging to the anterior group of dorsal bipolar cells (ADBs), all of which synapse to dendrites belonging to the large paired interneurons (LPNs) of the primary motor center (PMC). The psz occupies a region ca. 15 µm in length near the midpoint of the neuropile (centered on section 900), though synaptic inputs from the fibers in the rostral nerve continue caudally for some distance. Input from the frontal eye is less from the photoreceptors themselves, i.e. rows 1 and 2, which have comparatively short axons and rather irregular terminals (Lacalli, 1996; Vopalensky et al., 2012), than from the row 4 interneuron on the left side, cell 4L. From what is known of this cell, from one specimen, its synapses are few in number but large, 3
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
with clearly defined targets, principally the forward-projecting dendrites of the left LPN1 and LPN3, which travel forward from the PMC in the left ascending bundle (synapses are between sections 770 and 890, see Figs. 5B and 6D in Lacalli, 2002a, and Fig. 5B, C in Lacalli & Kelly, 2003a), and the dendrites of one of the putative dopaminergic neurons located in this same region (ca. section 860, see Fig. 4 in Zieger, Lacalli, Pestarino, Schubert, & Candiani, 2017). To reach these, the 4L axon projects caudally to the level of the psz and, along with the rostral nerve tract, is diverted dorsally to bring it in line with the ascending bundle. Frontal eye output via 4L thus has features in common with the retinotectal pathway of vertebrates, in projecting dorsally, though to a rather limited degree, at approximately midbrain level. Axons from cells in the lamellar body, in contrast, project ventrally to the center of the neuropile, bypassing the psz, and make few if any direct contacts with the interneurons responsible for locomotory control (Bozzo et al., 2017). Early pineal projections in vertebrates are to more anterior targets in the basal forebrain (Ross, Parrett & Easter, 1992), and they also bypass the main visual pathway. Judging by the complexity of the circuitry, and the numerous paracrine inputs to it, probably modulatory in nature, the locomotory control system in young amphioxus larvae is more than a simple set of parallel reflex pathways. My own observations on larval behavior support this view, in that the response to external stimuli of otherwise similar larvae in culture is highly variable. This accords also with published observations on amphioxus behavior, which are often inconclusive for the same reason (Pergner & Kozmik, 2017). The post-infundibular neuropile would seem, therefore, to be acting as a kind of motivational center, establishing a basal level of responsiveness to sensory input that varies a good deal between individuals. Because the population of epithelial sensory cells increases as the larvae grow, it is also important that an appropriate balance be maintained between excitation and inhibition throughout development. This is the problem of gain control (Priebe & Ferster, 2002), and implies a continuing role for negative feedback mechanisms to adjust the response of the system to prevent its being overwhelmed by increasing levels of unmodulated input. There seems also to be a role during neuropile development for positive feedback mechanisms, acting via excitatory dopaminergic neurons (Zieger et al., 2017). In short, basal levels of activity within the post-infundibular neuropile are probably set and maintained by an assortment of feedback loops and modulatory inputs. The principal output is to the PMC, whose location, in the caudal dien-mesencephalon, implies homology with the vertebrate mesencephalic locomotory region (MLR), which performs an essentially similar function in lamprey (Sirota, Di Prisco & Dubuc, 2000). It is also consistent with midbrain-based locomotory centers in chordates being evolutionarily older than those in hindbrain, though the latter are clearly important in activating the escape response in both tunicates and vertebrates (Ryan, Lu & Meinertzhagen, 2017). Amphioxus has conspicuously large interneurons in its hindbrain homolog (the giant Rohde cells, see Castro, Becerra, Manso, & Anadon, 2015; Wicht & Lacalli, 2005), and though these play a coordinating role in locomotion in older larvae and adults, they are absent in young larvae. Besides the frontal eye and lamellar body, both of which use ciliary photoreceptors, amphioxus has two sets of rhabdomeric photoreceptors. Best known are the dorsal ocelli, distributed along the nerve cord beginning at the level of somite 3. The first of these to differentiate, located in somite 5, provides input to the slow twitch motoneurons responsible for migratory (i.e. non-escape) swimming (Lacalli, 2002b), but seems not to be involved in the escape response. The other set of rhabdomeric receptors are the Joseph Cells, which are flattened and covered on their undersides with photoreceptive microvilli. The Joseph Cells form a contiguous pavement over much of the dorsal surface of the anterior nerve cord in juveniles and adults, but their function is not known, nor whether they are primary photoreceptors, with axons (provisionally reported by Castro et al., 2015), or secondary receptors, without. The fact that the anterior-most of the series overlap with the lamellar body implies an association between light reception and a region homologous, at least in part, with the vertebrate pretectum and midbrain. This could represent the ancestral condition for chordates, but the only supporting evidence, other than from amphioxus, is the presence of a rhabdomeric, midbrain-level cerebral eye in salps, a group of pelagic tunicates (Lacalli & Holland, 1998; Braun & Stach, 2017). If the Joseph Cells are ancestral in chordates rather than being an amphioxus innovation, the role the midbrain plays in visual processing in vertebrates could more easily be explained, as having begun as an association between a similar set of now vanished rhabdomeric photoreceptors and the neurons located immediately beneath (Lacalli, 2018). The problem, however, even if this were the case, is that there is nothing obvious in the anatomy of the dorsal part of the midbrain homolog in either amphioxus or salps to indicate how the optic tectum originated. The apparent apico-basal inversion of amphioxus ADBs suggests these cells could be precursors of neurons populating the cortical structures in vertebrate brains, including the tectum (Lacalli 1996, 2017), but we are still a long way from having a convincing evolutionary scenario to account for the diverse cell types and neuroanatomical complexity seen in any of these layered structures. Whether the combination of caudal diencephalon and mesencephalon into a single domain is ancestral for chordates or a secondary specialization of amphioxus, the key issue is to understand the functions carried out within this domain. The evidence from amphioxus larvae is that this is the main site for sensory integration, with overall levels of locomotory activity being modulated by multiple inputs. This argues for similar functions being carried out by this same domain in ancestral chordates, including in immediate ancestors of vertebrates, and for the presence of a counterpart of the psz. The amphioxus dien-mesencephalon lacks identifiable subdivisions, and the lamellar body is not an entirely reliable marker for the amphioxus counterpart of the diencephalon, so it is difficult to place the vertebrate homolog of the psz more precisely than to say that it should lie somewhere in the basal part of either the caudal diencephalon or rostral midbrain. 3. Amphioxus and the origin of consciousness It is comparatively uncontroversial to suppose that other primates, and indeed other mammals, very likely have conscious experiences qualitatively similar to our own. This supposition is based on the similar structure of those regions of the brain generally accepted as the seat of consciousness in mammals, i.e. the cerebral cortex and thalamus (Butler, 2008a,b; Meyer 2011). Avian and mammalian forebrains are sufficiently similar that birds also can be supposed to have conscious experiences of a kind we would 4
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
recognize, but the situation in amphibians and fish is harder to assess. To the degree that consciousness depends on anatomical structures and cell types derived from the neocortex, fish and amphibians, which lack these structures, might be expected to have a much reduced or, at the very least, qualitatively different awareness of sensory experience than mammals and birds. Circumstantial evidence implies that the midbrain may itself be a source of conscious sensations (Barron & Klein, 2016; Merker, 2005), including a visual awareness of the surrounding environment that aids in guiding locomotion and other muscular actions. The case for midbrainbased consciousness is most fully developed in a recent book and companion article by Feinberg and Mallatt (2016a,b), who argue first, for recognizing the diverse forms consciousness takes, whereby some (e.g. vision) are associated with isomorphic sensory maps, while others (affective feelings) are not. They then summarize the arguments in support of conscious sensations being generated in brain structures other than thalamo-cortical ones. For fish and amphibians, the optic tectum is the structure that most closely approaches mammalian cortex in terms of complexity, neural connectivity and hierarchical structure. This implies it may have similar functional capabilities, including the ability to generate a consciously perceived visual display, and the visual search behaviors of fish and amphibians are at least consistent with this supposition. To properly assess the role the midbrain may play in consciousness, a better understanding is needed of the structure and circuitry of the optic tectum than is currently available, including of pathways from the tectum to more basal centers. There are reciprocal links between these, not unlike those between thalamus and cortex (McHaffie, Stanford, Stein, Coizet, & Redgrave, 2005), which is, at the very least, suggestive. The operative question, from an amphioxus perspective, relates more specifically to how dorsal inputs to basal midbrain centers differ between vertebrates and amphioxus, and whether those differences reveal anything about how conscious sensations are produced. Here the evolutionary context becomes relevant. Chordates diversified in the early to middle Cambrian at a time of increasingly intense predation, notably from large pelagic arthropods with grasping mouthparts, and this is probably what drove improvements to ancestral chordate musculature, body support systems and swimming ability (Feinberg & Mallatt, 2016a, Chap. 4; Lacalli, 2012; McAllister 2003). To this, vertebrates added improved sense organs and sensory processing systems (Feinberg & Mallatt, 2016a, Chap. 5; Gans, 1989; Gee, 2018, Chaps. 10 & 15; Khaner, 2007; Lacalli, 2001; Shimeld & Holland, 2000), but to influence locomotion, these would presumably still have had to act through existing control centers, which could have included the vertebrate equivalent of the amphioxus psz. The postulate to consider here then, is that consciousness may have evolved as part of an enhanced arousal mechanism, enabling a preexisting, amphioxus-type synaptic center to produce a behavioral response that was either more rapid, more focused (e.g. via better suppression of competing behaviors), or more selective (via more effective sensory filtering) than could be achieved using the ancestral non-conscious reflex pathways alone. All of these functions can be achieved without consciousness, but conscious awareness of one or more of the relevant sensory inputs has the potential to be a significant further enhancement. The emergence of consciousness would then be expected to correlate in some way with the appearance of novel types of circuits in the inputs to the psz homolog, dedicated to selective signal amplification or involving a degree of integrative complexity beyond that needed simply to coordinate an existing set of reflexes. Linking consciousness to the evolution of the vertebrate tectum implies a link also to the evolution of image-forming eyes. Amphioxus is not particularly informative here, because its frontal eye is too small and simple in structure to form an image. In the absence of evidence that the frontal eye is secondarily simplified, it would seem to represent something close to the basal condition in chordates before image-forming eyes evolved. The key distinction that then needs to be made is between the earliest chordates, which were filter feeders that probably lacked paired eyes, and ancestral vertebrates with paired eyes, some of which, e.g. conodonts, were clearly predators (McAllister 2003; Purnell, 2001a,b). The ability to visually track prey in three dimensions while swimming is highly advantageous for an active predator (e.g. Andrew, Tommasi, & Ford, 2000), and midbrain-based consciousness could have emerged solely to assist in this function. However, predators are always at risk from yet larger predators, including larger con-specifics, so in this respect they face the same problem that non-predatory invertebrate chordates and early vertebrates would have faced, and would benefit just as much from mechanisms that enhanced their ability to detect and respond rapidly to predator attack. In summary, if midbrain-based consciousness emerged very early in the vertebrate lineage, there are two scenarios to consider: that this happened (1) in parallel with the evolution of image-forming eyes and tectal processing in predatory vertebrates, most likely to aid in tracking prey, or (2) as part of an enhanced arousal mechanism in yet earlier chordates, either predatory or non-predatory, with much simpler visual systems. In the latter case there would likely have been no perception of a visual field, but only of light, or its absence, or of changes in light intensity. With option (1), we face the more difficult task of explaining how a complex sensory construct, in the form of a consciously perceived visual field, first came into being. 4. Neural correlates of consciousness, and a strategy for investigating qualia The involvement of midbrain centers in processing visual inputs is a shared feature of vertebrates, and so, if the midbrain acts as a source of conscious sensations, this also should be a feature shared by all living vertebrates. From a research perspective, the midbrain of lower vertebrates is then potentially as informative a subject as mammalian forebrain for investigating the neural basis of consciousness. Lower vertebrates also offer significant advantages as research subjects, the zebrafish larva being a prime example, whose small size and transparency make it especially suitable for both 3D EM-level brain reconstruction (Hildebrand et al., 2017) and the optogenetic methods used to image the activity of neural circuits in real time (Ahrens et al., 2013; Friedrich et al., 2013). Mapping the synaptic circuitry of significant part of the larval zebrafish brain in full detail then becomes an achievable goal, at least in principle. The task of analysis is likely to be daunting, even with a complete synaptic map in hand, and the involvement of nonsynaptic interactions that lack clear morphological correlates may complicate matters further. These are practical difficulties, however, rather than reasons to deny the possibility of obtaining, perhaps in the not-too-distant future, a dataset that contains all the information necessary to explain consciousness. 5
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
For anyone dealing directly with neurocircuitry data from brain regions known or supposed to generate conscious sensations, the operative question is how one would identify neural correlates of consciousness if indeed they were present. Data from simple invertebrate systems, including from the nematode Caenorhabditis, the best-studied example where the neural morphology is fully known, show that reflex circuits consisting of no more than tens or hundreds of neurons can be extraordinarily difficult to analyze (Smith, 2017; Yan et al., 2017). The smallest of vertebrate brains have orders of magnitude more neurons, and if the neural circuits responsible for consciousness are fully embedded in reflex pathways of correspondingly greater complexity, distinguishing the former from the latter could be an insurmountable task. One insight from amphioxus is that this is only the worst-case scenario in a range of possibilities. Young amphioxus larvae roughly match Caenorhabditis larvae in size and numbers of neurons, but a subset of neural pathways in amphioxus have proven comparatively easy to analyze. The activating pathway for the escape response is an example, where a high degree of redundancy yields a strong signal regarding connectivity patterns, overriding the noise inherent in incomplete data and variation between cells and individuals. In contrast, the dorsal modulatory circuits in even young amphioxus larvae involve a diverse assortment of translumenal interneurons (Lacalli & Kelly, 2003b), for which it is far more difficult to identify consistent patterns of connectivity. The problem increases with development, as the translumenal component of the brain itself increases (Castro et al., 2015), which implies that a full analysis of circuits in the juvenile and adult brain is likely to be a decidedly non-trivial task. So, how difficult will it be to identify the circuits responsible for vertebrate consciousness? My guess is very difficult indeed, if, as above, they are deeply embedded in circuits that are already highly complex. On the other hand, if they are highly redundant add-ons to existing pathways, they might announce themselves in ways that make their identity obvious by inspection. Even if neural correlates of consciousness can eventually be identified and their functional properties understood, this in itself will not necessarily yield a conceptually satisfying explanation for how conscious sensations are produced. The issue here is whether subjective experience can in principle be understood in material terms (e.g. Chalmers, 1995; Levine, 1983), which, though a serious concern, is somewhat beyond the scope of this account. Whether or not midbrain circuitry in an organism like the zebrafish larva will, by itself, reveal a mechanism for generating conscious sensations, any circuitry differences that vary depending on the type of sensory input are potentially important for what they may reveal about the nature of qualia, i.e. the way a particular sensory input is subjectively experienced. Explaining how qualia are generated is one of the classical hard problems of consciousness, complicated by the richness of sensory experience, and the myriad variations within each sensory category, e.g. of different colors or a range of sounds, all of which needs explaining. The insight from amphioxus is that the problem may be much simpler, because there could well have been only two sensory inputs of importance when midbrain-based consciousness first emerged in chordates, for light and for physical contact. In the latter case, we would most likely be dealing with sharp pain of the kind produced by a wound or injury rather than sustained deeper forms of pain, as the neural correlates of the former in lower vertebrates (fibers types in this instance) are a better match to the amniote condition. Feinberg and Mallatt (2016a, Chap. 8) deal at some length with the question of whether fish feel pain in the same way as amniotes. The case that they do is less compelling than for vision because, in contrast to visual processing, pain signals are not associated in the same way with a dedicated integrative center organized similarly across vertebrate taxa (see Feinberg & Mallatt, 2016a, Table 7.2 and references therein). It is consequently more difficult to construct a plausible argument as to where, in the vertebrate lineage, nonconscious nociception was replaced by a conscious perception of pain comparable to our own, or what role the midbrain played in this transition. However, even if the conscious perception of both light and sharp pain are not of equal antiquity, it is a valid investigative strategy to focus on these two sensory modalities in the first instance, so long as there is reason to suppose that both emerged before other forms of consciousness. One would then be looking for differences in the midbrain pathways for light and sharp pain that might account for the different way light and pain are perceived. As to the qualia themselves, it is possible they were identical or very similar at first and diverged subsequently. Conversely, perhaps they differed from the start, with sharp pain being, say, a negative affect and light a positive one. It is worth noting that, for the first option, assuming increased light intensity both increased the risk of predator attack and correlated consistently with the kind of pain caused by such attack, there would have been less reason to distinguish between light and pain stimuli. Either would produce the required effect, of activating the arousal pathway, which could explain features common to the experience of intense pain and an intense flash of light, for example, the way both are similarly able to momentarily monopolize one’s complete attention. Perhaps by this means we experience, at first hand, something of what it would have been like to be an early vertebrate, where what mattered was for one signal to dominate over all others, and a broader spectrum of additional qualia had yet to evolve.
5. Conclusions We are entering a potentially very fruitful period in neuroscience, with methods emerging that promise to reveal neural connectivity and real-time function at the cellular level in exquisite detail. Obtaining the data necessary to understand the neural basis of consciousness is a daunting task, given the size and complexity of even the smallest vertebrate brain, but it is no longer an unrealistic goal in principle. The insight from amphioxus is that there are specific brain regions in vertebrates where there is a better prospect than elsewhere for identifying the innovations in neural circuitry that made conscious sensation possible, namely in the pain and visual pathways impinging on the locomotory control centers of the basal midbrain. Whether this will fully explain consciousness as a phenomenon remains to be determined, but there are reasons to be optimistic that currently available methods may be sufficient to solve at least some of the hard problems of consciousness.
6
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
Declaration of interest None. Acknowledgements The author thanks Jon Mallatt for his comments on an earlier draft of this review, and Heather Down for assisting with the preparation of figures. Source of finding This work was supported by the Lionel G. Harrison Research Trust. References Andrew, R. J., Tommasi, L., & Ford, N. (2000). Motor control by vision and the evolution of cerebral lateralization. Brain and Language, 73, 220–235. Ahrens, M. B., Orger, M. B., Robson, D. N., Li, J. M., & Keller, P. J. (2013). Whole-brain functional imaging at cellular resolution using light-sheet microscopy. Nature Methods, 10, 413–430. Albuixechi-Crespo, B., Irimia, M., Burguera, D., Maeso, I., Arrones, L. S., Moreno-Bravo, J. A., ... Ferran, J. L. (2017). Molecular regionalization of the amphioxus neural tube challenges major partitions of the vertebrate brain. PLoS Biology, 15, 4. Barron, A. B., & Klein, C. (2016). What insects can tell us about the origins of consciousness. Proceedings of the National Academy of Sciences USA, 113, 4900–4908. Bozzo, M., Macri, S., Calzia, D., Sgarra, R., Manfioletti, G., Ramoino, P., ... Candiani, S. (2017). The HMGA gene family in chordates: Evolutionary perspectives from amphioxus. Development, Genes and Evolution, 227, 201–211. Braun, K., & Stach, T. (2017). Structure and ultrastructure of the eyes and brains of Thalia democratica. Journal of Morphology, 278, 1421–1437. Butler, A. (2008a). Evolution of brains, cognition and consciousness. Brain Research Bulletin, 75, 442–449. Butler, A. (2008b). Evolution of the thalamus: A morphological and functional review. Thalamus and Related Systems, 4, 35–58. Castro, A., Becerra, M., Manso, M. J., & Anadon, R. (2015). Neuronal organization in the brain in the adult amphioxus (Branchiostoma lanceolatum): A study with acetylated tubulin immunohistochemistry. Journal of Comparative Neurology, 523, 2211–2232. Chalmers, D. J. (1995). Facing up to the problem of consciousness. Journal of Consciousness Studies, 2, 200–219. Churcher, A. M., & Taylor, J. S. (2010). The antiquity of chordate odorant receptors is revealed by the discovery of orthologs in the cnidarian Nematostella vectensis. Genome Biology and Evolution, 3, 36–43. Feinberg, T. E., & Mallatt, J. M. (2016a). The ancient origins of consciousness. Cambridge, MA: MIT Press. Feinberg, T. E., & Mallatt, J. M. (2016b). The nature of primary consciousness. A new synthesis. Consciousness and Cognition, 43, 113–127. Friedrich, R. W., Genoud, C., & Wanner, A. A. (2013). Analyzing the structure and function of neuronal circuits in zebrafish larvae. Frontiers in Neural Circuits, 7, 71. Gans, C. (1989). Stages in the origin of vertebrates: Analysis by means of scenarios. Biological Reviews, 64, 221–268. Gee, H. (2018). Across the bridge: Understanding the origin of vertebrates. Chicago: University of Chicago Press. Grillner, S., Robertson, B., & Stephenson-James, M. (2013). The evolutionary origin of the vertebrate basal ganglia and its role in action selection. Journal of Physiology, 591, 5425–5431. Hildebrand, D. G. C., Cicconet, M., Torres, R. M., Choi, W., Quan, T. M., Moon, J., ... Engert, F. (2017). Whole-brain serial-section electron microscopy in larval zebrafish. Nature, 545, 345–349. Holland, L. Z. (2009). Chordate roots of the vertebrate nervous system: Expanding the molecular toolkit. Nature Reviews Neuroscience, 10, 736–746. Holland, L. Z. (2015). Genomics, evolution and development of amphioxus and tunicates: The Goldilocks principle. Journal of Experimental Zoology (Molecular Development and Evolution), 324, 342–352. Holland, L. Z., Carvalho, J. E., Escriva, H., Laudet, V., Schbert, M., Shimeld, S. M., & Yu, J. K. (2013). Evolution of the bilaterian central nervous system: A single origin? EvoDevo, 4, 27. Khaner, O. (2007). Evolutionary innovations of the vertebrates. Integrative Zoology, 2, 60–67. Lacalli, T. C. (1996). Frontal eye circuitry, rostral sensory pathways and brain organization in amphioxus larvae: Evidence from 3D reconstruction. Philosophical Transactions of the Royal Society of London, Series B, 351, 243–263. Lacalli, T. C. (2001). New perspectives on the evolution of protochordate sensory and locomotory systems, and the origin of brains and heads. Philosophical Transactions of the Royal Society of London, Series B, 356, 1565–1572. Lacalli, T. C. (2002a). Sensory pathways in amphioxus larvae I. Constituent fibers of the rostral and anterodorsal nerves, their targets and evolutionary significance. Acta Zoologica, 83, 149–166. Lacalli, T. C. (2002b). The dorsal compartment locomotory control system in amphioxus larvae. Journal of Morphology, 252, 227–237. Lacalli, T. C. (2004). Sensory systems in amphioxus: A window on the ancestral chordate condition. Brain Behavior and Evolution, 64, 148–162. Lacalli, T. C. (2012). The Middle Cambrian fossil Pikaia and the evolution of chordate swimming. EvoDevo, 3, 12. Lacalli, T. C. (2017). Interpreting amphioxus, and thoughts on ancestral mouths and brains. International Journal of Developmental Biology, 61, 649–654. Lacalli, T. C. (2018). Amphioxus, motion detection, and the evolutionary origin of the vertebrate retinotectal map. EvoDevo, 9, 6. Lacalli, T. C., Holland, N. D., & West, J. E. (1994). Landmarks in the anterior central nervous system of amphioxus larvae. Philosophical Transactions of the Royal Society of London, Series B, 344, 165–185. Lacalli, T. C., & Holland, L. Z. (1998). The developing dorsal ganglion of the salp Thalia democratica, and the nature of the ancestral chordate brain. Philosophical Transactions of the Royal Society of London, Series B, 353, 1943–1967. Lacalli, T. C., & Kelly, S. J. (2003a). Ventral neurons in the anterior nerve cord of amphioxus larvae I. An inventory of cell types and synaptic patterns. Journal of Morphology, 257, 190–211. Lacalli, T. C., & Kelly, S. J. (2003b). Sensory pathways in amphioxus larvae II. Dorsal tracts and translumenal cells. Acta Zoologica, 84, 1–13. Levine, J. (1983). Materialism and qualia: The explanatory gap. Pacific Philosophical Quarterly, 64, 354–361. Mashour, G. A., & Alkire, M. T. (2013). Evolution of consciousness: Phylogeny, ontogeny, and emergence from general anesthesia. Proceedings of the National Academy of Sciences USA, 110(Suppl. 2), 10357–10364. McAllister, J. (2003). Predation of fishes in the fossil record. In P. H. Kelley, M. Kowalewski, & T. A. Hansen (Eds.). Predator-prey interactions in the fossil record (pp. 303–324). Berlin: Springer. McHaffie, J. G., Stanford, T. R., Stein, B. E., Coizet, V., & Redgrave, P. (2005). Subcortical loops through the basal ganglia. Trends in Neurosciences, 28, 401–407. Merker, B. (2005). The liabilities of mobility: A selection pressure for the transition to consciousness in animal evolution. Consciousness and Cognition, 14, 89–114. Merker, B. (2007). Consciousness without a cerebral cortex: a challenge for neuroscience and medicine. Behavior and Brain Sciences, 30, 63-81, discussion 81-134. Meyer, K. (2011). Primary sensory cortices, top-down projections and conscious experience. Progress in Neurobiology, 94, 408–417. O’Connell, L. A., & Hofmann, H. A. (2011). The vertebrate mesolimbic reward system and social behavior network: A comparative synthesis. Journal of Comparative Neurology, 519, 3599–3639.
7
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
Pani, A. M., Mullarkey, E. E., Aronowicz, J., Assimacopoulos, S., Grove, E. A., & Lowe, C. J. (2012). Ancient deuterostome origins of vertebrate brain signalling centres. Nature, 483, 289–294. Pergner, J., & Kozmik, Z. (2017). Photoreceptors of amphioxus – Insights into evolution of vertebrate opsins, vision and circadian rhythmicity. International Journal of Developmental Biology, 61, 665–681. Priebe, N. J., & Ferster, D. (2002). A new mechanism for neuronal gain control (or how the gain in brains has mainly been explained). Neuron, 35, 602–604. Purnell, M. A. (2001a). Feeding in conodonts and other early vertebrates. In D. E. G. Briggs, & P. R. Crowther (Eds.). Palaeobiology II (pp. 401–404). Oxford: Blackwell Science. Purnell, M. A. (2001b). Scenarios, selection and the ecology of early vertebrates. In P. E. Ahlberg (Ed.). Major events in early vertebrate evolution (pp. 187–208). London: Taylor & Francis. Ross, L. S., Parrett, T., & Easter, S. S. (1992). Axonogenesis and morphogenesis in the embryonic zebrafish brain. Journal of Neuroscience, 12, 467–482. Ryan, K., Lu, Z., & Meinertzhagen, I. A. (2017). Circuit homology of decussating pathways in the Ciona larval CNS and the vertebrate startle-response pathway. Current Biology, 27, 721–728. Satoh, G. (2005). A trajectory of increasing activity and the elaboration of chemosensory modality: A new perspective on vertebrate origins. Zoological Science, 22, 613–626. Shimeld, S. M., & Holland, P. W. H. (2000). Vertebrate innovations. Proceedings of the National Academy of Sciences, USA, 97, 4449–4452. Sirota, M. G., Di Prisco, G. V., & Dubuc, R. (2000). Stimulation of the mesencephalic locomotor region elicits controlled swimming in semi-intact lampreys. European Journal of Neuroscience, 12, 4081–4092. Smith, K. (2017). The brain-circuit challenge. Nature, 548, 150–152. Vopalensky, P., Pergner, J., Liegertova, M., Benito-Guitierrez, E., Arendt, D., & Kozmik, Z. (2012). Molecular analysis of the amphioxus frontal eye unravels the evolutionary origin of the retina and pigment cells of the vertebrate eye. Proceedings of the National Academy of Sciences USA, 109, 15383–15388. Wicht, H., & Lacalli, T. C. (2005). The nervous system of amphioxus: Structure, development, and evolutionary significance. Canadian Journal of Zoology, 83, 122–150. Yan, G., Vertes, P. E., Towlson, E. K., Chew, Y. L., Walker, D. S., Schafer, W. R., & Barabasi, A. L. (2017). Network control principles predict neuron function in the Caenorhabditis elegans connectome. Nature, 550, 519–523. Zieger, E., Lacalli, T. C., Pestarino, M., Schubert, M., & Candiani, S. (2017). The origin of dopaminergic systems in chordate brain: Insights from amphioxus. International Journal of Developmental Biology, 61, 749–761.
8
Contents lists available at ScienceDirect
Consciousness and Cognition journal homepage: www.elsevier.com/locate/concog
Review article
Amphioxus neurocircuits, enhanced arousal, and the origin of vertebrate consciousness Thurston Lacalli Biology Department, University of Victoria, Victoria, BC V8W-3N5, Canada
A R T IC LE I N F O
ABS TRA CT
Keywords: Consciousness Brain evolution Dien-mesencephalon DiMes Midbrain Enhanced arousal Qualia
Gene expression studies have recently identified the amphioxus homolog of a domain comprising the combined caudal diencephalon plus midbrain, regions implicated in locomotory control and some forms of primary consciousness in vertebrates. The results of EM-level reconstructions of the larval brain of amphioxus, reviewed here, highlight the importance of inputs to this region for light and physical contact, both of which impinge on the same synaptic zone. The neural circuitry provides a starting point for understanding the organization and evolution of locomotory control and arousal in vertebrates, and implies that one of the tasks of midbrain-based consciousness, as it first emerged in vertebrates, would have been to distinguish between light and physical contact, probably sharp pain in the latter case, by assigning different qualia to each. If so, investigating midbrain circuitry more fully could lead to a better understanding of the neural basis of some forms of sensory experience.
1. Introduction The cephalochordate amphioxus (Branchiostoma) is now considered the closest living proxy for the ancestral chordate condition, and is hence an organism of considerable comparative and evolutionary interest (Gee, 2018 Holland, 2015). It is especially relevant to studies on the organization and circuitry of the central nervous system (CNS), as the expression patterns of developmentally important genes are substantially conserved between the amphioxus CNS and that of vertebrates (Holland, 2009; Holland et al., 2013). This makes amphioxus especially valuable as a model for early CNS development, and means the development of early brain circuits can, in principle, be studied in a context that is far less complex than in any vertebrate. In comparing the regional subdivision of amphioxus and vertebrate brains, a persistent problem has been to identify amphioxus homologs of the telencephalon and midbrain. Amphioxus lacks the expanded dorsal structures (e.g. cortices) that make these regions anatomically distinctive in vertebrates, as well as the sense organs whose input such structures evolved to serve. In young amphioxus larvae there are in fact no neurons in the dorsal nerve cord forward of the lamellar body, a pineal homolog, which is where any counterpart of telencephalon would be located. In older larvae, however, this same region receives dorsally directed branches from the rostral nerves (Lacalli, 2002a, 2004), a possible indication of a late-developing olfactory pathway. The olfactory system and telencephalon are generally considered vertebrate innovations (Satoh, 2005), but amphioxus has a complement of olfactory-related genes (Churcher & Taylor, 2010). A rudimentary homolog of the olfactory bulb may thus be present in older amphioxus larvae and
Abbreviations: ADBs, anterior group of dorsal bipolar neurons; C, enteropneust collar; CNS, central nervous system; Di, caudal diencephalon (thalamus and pretectum); DN, anterodorsal nerve; ESCs, epithelial sensory cells; inf, infundibular cells; IsO, isthmic organizer; JCs, Joseph cells; LPNs, large paired neurons (pairs 1 and 3); LMB, lamellar body; Mes, midbrain; MLR, mesencephalic locomotor region; PMC, primary motor center; psz, primary synaptic zone; RN, rostral nerve; TEM, transmission electron microscopy; Zli, zona limitans intrathalamica E-mail address: [email protected]. https://doi.org/10.1016/j.concog.2018.03.006 Received 26 February 2018; Received in revised form 14 March 2018; Accepted 16 March 2018 1053-8100/ © 2018 Elsevier Inc. All rights reserved.
Please cite this article as: Lacalli, T., Consciousness and Cognition (2018), https://doi.org/10.1016/j.concog.2018.03.006
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
Fig. 1. Evolutionary continuity of the dien-mesencephalon. A comparison of comparable neural domains in developing stages of a hemichordate, amphioxus and an amniote vertebrate, showing the dien-mesencephalon (in pink), which, for vertebrates, corresponds with the domain lying between the zona limitans interthalamica (Zli, vertical red line) and isthmic organizer (IsO, vertical blue line) and, in hemichordates, their homologs. The nervous system in hemichordates is largely intraepithelial, so the comparable domain there is the collar epithelium (C). In amphioxus, this zone is also the location of populations of dopaminergic neurons (green) corresponding to those found in the caudal diencephalon (Di) and midbrain (Mes) of vertebrates, and also (not shown) in the collar of hemichordates. There is currently no evidence in amphioxus for homologs of vertebrate hypothalamic dopaminergic nuclei (orange). It is also not known whether the absence of a fully internalized CNS in hemichordates is a secondary simplification, or reflects the ancestral condition, with internalization occurring early in the chordate lineage as shown here. Other abbreviations: 1–4, character states; CNS, central nervous system. Modified from Zieger et al. (2017). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
adults, but if so, it develops too late to play a role in early-stage larval behavior. In contrast to the telencephalon, a good deal more is known about the amphioxus homolog of the midbrain. From the expression patterns of genes required for CNS regionalization, this is now considered part of a larger domain, the dien-mesencephalic primordium (Albuixechi-Crespo et al., 2017), roughly equivalent to the posterior cerebral vesicle as defined anatomically. This domain includes the amphioxus homologs of the thalamus, pretectum and midbrain, and extends from the cluster of infundibular cells marking the junction between the anterior and posterior cerebral vesicle to approximately the end of somite 1. The corresponding region in vertebrate brain is the zone between the zona limitans intrathalamica and the isthmic organizer (Fig. 1), and though markers for these two signaling centers are absent in amphioxus, they are present in hemichordates (Pani et al., 2012), indicating the dienmesencephalon homolog there is the collar epithelium, which is neurogenic. The implication is that the dien-mesencephalon represents an identifiable subdomain within the nervous system of deuterostomes that predates the origin of chordates, and is retained in amphioxus despite the absence of markers for the signaling centers that define it in other taxa. Identifying the key functions this region plays in behavior is then important for what it tells us of the division of labor in deuterostome nervous systems generally, and in the brains of chordates more specifically. What we know of the neural circuitry and probable function of the amphioxus dien-mesencephalon comes largely from serial reconstructions of the larval CNS (see Wicht & Lacalli, 2005 for a summary). These identify the post-infundibular (=tegmental) neuropile as the principal integrative center for the animal at the stage where it first begins to swim. This accords with what is known of vertebrates, where a comparable part of the anterior brainstem, including structures in the basal diencephalon and midbrain, has a long evolutionary history as decision-making center, especially in relation to the control of locomotory responses (Grillner, Robertson & Stephenson-James, 2013; O’Connell & Hofmann, 2011). This is also where some forms of primary consciousness are thought to reside (e.g. Barron & Klein, 2016; Mashour & Alkire, 2013; Merker, 2007), which begs the question of whether amphioxus has anything useful to tell us about the evolutionary origin of vertebrate consciousness. The purpose of this paper is to explore this question, especially in relation to the evolutionary scenario laid out by Feinberg and Mallatt (2016a,b), who postulate a close association between consciousness and the emergence of the vertebrate optic tectum as a processing center for visual input. Consciousness, for the purposes of this review, refers to the most basic form of primary, or sensory, consciousness: subjective experiences created by the brain and characterized by specific sensations, or qualia, for each category of sensory input. Two issues are then considered. The first is that there is a plausible case to be made that midbrain-based consciousness may have evolved as part of an enhanced arousal system for predator avoidance (e.g. see Feinberg & Mallatt, 2016a, Chaps. 4 & 5), but this could have happened 2
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
Fig. 2. Escape circuits in young amphioxus larvae. Shows a schematic left side view of the anterior nerve cord of an amphioxus larva, from TEM reconstructions of a 12-day specimen of B. floridae. The cells of the escape circuit are highlighted in color, as well as the infundibular cells (inf, dark blue), which are important landmarks, separating the amphioxus homolog of the eye rudiment plus hyopothalamus from the dien-mesencephalon, as shown by the black horizontal bars. The primary synaptic zone, located in the latter (psz, circled in red, and paired at this stage, as there would be a corresponding set of inputs on the right side of the cord), is where the axons of epithelial sensory cells (ESCs) entering the nerve cord via the first pair of dorsal nerves (DN, yellow) first encounter and synapse to dendrites belonging to the principle interneurons (LPNs, green) of the primary motor center (PMC), which are a major source of excitatory input to the escape response. Inputs are also received here from the anterior group of bipolar cells (ADBs, also yellow), which are intramedullary sensory neurons located on either side of the lamellar body (LMB, light blue) about halfway along its length, in a region that is probably somewhere near the front of the midbrain in terms of homology. Horizontal blue bars indicate the axial extent of selected structures in the mid-larval phase, and (dashed lines) their expansion during later development. The Joseph Cells, for example, first develop caudal to somite 1, but extend forward to the front of the lamellar body in the adult, at least in B. lanceolatum, an observation as yet unconfirmed for B. floridae. The transition point between somites 1 and 2 coincides approximately with the caudal end of the lamellar body as indicated by the blue bar. Modified from Lacalli (2017). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
as image-forming eyes and tectal processing evolved, in basal vertebrates that were themselves predators, or before, at a time when vertebrates were more like amphioxus in being filter feeders rather than predators, with only rudimentary eyes and a correspondingly less developed tectum. Consciousness perception of light, if it did evolve that early, would most likely have been used for distinguishing light from its absence, which leaves open the question of how, from such a modest beginning, the ability to construct a consciously perceived visual field evolved. Besides photoreceptors, the other main source of input to the amphioxus post-infundibular neuropile is from mechanoreceptors, which implies, as a second point, that a mechanism would have been needed as consciousness emerged in vertebrates to ensure that light and contact stimuli (e.g. sharp pain) were subjectively experienced in different ways. If so, this would simplify the problem of understanding how different qualia are generated at a circuitry level and made to differ, as we would then need to focus on only these two, and the neural pathways associated with them. Technical advances in imaging and CNS reconstruction can now provide the kind of detailed knowledge of synaptic circuitry needed to investigate such issues in model vertebrate systems (Ahrens, Orger, Robson, Li, & Keller, 2013; Friedrich, Genoud & Wanner, 2013; Hildebrand et al., 2017), so it is relevant also to briefly address the question of how, in practice, neural correlates of consciousness might reveal themselves. 2. Ventral dien-mesencephalon: a core integrative center The principle zone of neuropile in young amphioxus larvae occupies the ventral portion of the posterior half of the cerebral vesicle, a region extending from the infundibular cells and the front of the lamellar body to approximately the end of somite 1 (Fig. 2). Hence it is post-infundibular and lies largely within the confines of the dien-mesencephalon as defined by molecular markers. Based on the larva so far most fully reconstructed, ca. 12 days in age (section numbers, where mentioned, refer to this specimen), inputs to the post-infundibular neuropile are from the frontal eye, from an assortment of sensory-type neurons of the preinfundibular (hypothalamic) part of the anterior cerebral vesicle, including the balance organ, from the lamellar body, and from rostral sensory cells whose axons enter the nerve cord via the paired rostral and anterodorsal nerves (Lacalli, 1996, 2002a; Lacalli, Holland & West, 1994; Lacalli & Kelly, 2003a). Escape behaviors are induced most easily by mechanical stimulation of the rostrum, and sensory fibers originating there converge on the primary synaptic zone (the psz, which is also paired when it first forms, see Fig. 2), along with fibers from the frontal eye and synaptic terminals belonging to the anterior group of dorsal bipolar cells (ADBs), all of which synapse to dendrites belonging to the large paired interneurons (LPNs) of the primary motor center (PMC). The psz occupies a region ca. 15 µm in length near the midpoint of the neuropile (centered on section 900), though synaptic inputs from the fibers in the rostral nerve continue caudally for some distance. Input from the frontal eye is less from the photoreceptors themselves, i.e. rows 1 and 2, which have comparatively short axons and rather irregular terminals (Lacalli, 1996; Vopalensky et al., 2012), than from the row 4 interneuron on the left side, cell 4L. From what is known of this cell, from one specimen, its synapses are few in number but large, 3
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
with clearly defined targets, principally the forward-projecting dendrites of the left LPN1 and LPN3, which travel forward from the PMC in the left ascending bundle (synapses are between sections 770 and 890, see Figs. 5B and 6D in Lacalli, 2002a, and Fig. 5B, C in Lacalli & Kelly, 2003a), and the dendrites of one of the putative dopaminergic neurons located in this same region (ca. section 860, see Fig. 4 in Zieger, Lacalli, Pestarino, Schubert, & Candiani, 2017). To reach these, the 4L axon projects caudally to the level of the psz and, along with the rostral nerve tract, is diverted dorsally to bring it in line with the ascending bundle. Frontal eye output via 4L thus has features in common with the retinotectal pathway of vertebrates, in projecting dorsally, though to a rather limited degree, at approximately midbrain level. Axons from cells in the lamellar body, in contrast, project ventrally to the center of the neuropile, bypassing the psz, and make few if any direct contacts with the interneurons responsible for locomotory control (Bozzo et al., 2017). Early pineal projections in vertebrates are to more anterior targets in the basal forebrain (Ross, Parrett & Easter, 1992), and they also bypass the main visual pathway. Judging by the complexity of the circuitry, and the numerous paracrine inputs to it, probably modulatory in nature, the locomotory control system in young amphioxus larvae is more than a simple set of parallel reflex pathways. My own observations on larval behavior support this view, in that the response to external stimuli of otherwise similar larvae in culture is highly variable. This accords also with published observations on amphioxus behavior, which are often inconclusive for the same reason (Pergner & Kozmik, 2017). The post-infundibular neuropile would seem, therefore, to be acting as a kind of motivational center, establishing a basal level of responsiveness to sensory input that varies a good deal between individuals. Because the population of epithelial sensory cells increases as the larvae grow, it is also important that an appropriate balance be maintained between excitation and inhibition throughout development. This is the problem of gain control (Priebe & Ferster, 2002), and implies a continuing role for negative feedback mechanisms to adjust the response of the system to prevent its being overwhelmed by increasing levels of unmodulated input. There seems also to be a role during neuropile development for positive feedback mechanisms, acting via excitatory dopaminergic neurons (Zieger et al., 2017). In short, basal levels of activity within the post-infundibular neuropile are probably set and maintained by an assortment of feedback loops and modulatory inputs. The principal output is to the PMC, whose location, in the caudal dien-mesencephalon, implies homology with the vertebrate mesencephalic locomotory region (MLR), which performs an essentially similar function in lamprey (Sirota, Di Prisco & Dubuc, 2000). It is also consistent with midbrain-based locomotory centers in chordates being evolutionarily older than those in hindbrain, though the latter are clearly important in activating the escape response in both tunicates and vertebrates (Ryan, Lu & Meinertzhagen, 2017). Amphioxus has conspicuously large interneurons in its hindbrain homolog (the giant Rohde cells, see Castro, Becerra, Manso, & Anadon, 2015; Wicht & Lacalli, 2005), and though these play a coordinating role in locomotion in older larvae and adults, they are absent in young larvae. Besides the frontal eye and lamellar body, both of which use ciliary photoreceptors, amphioxus has two sets of rhabdomeric photoreceptors. Best known are the dorsal ocelli, distributed along the nerve cord beginning at the level of somite 3. The first of these to differentiate, located in somite 5, provides input to the slow twitch motoneurons responsible for migratory (i.e. non-escape) swimming (Lacalli, 2002b), but seems not to be involved in the escape response. The other set of rhabdomeric receptors are the Joseph Cells, which are flattened and covered on their undersides with photoreceptive microvilli. The Joseph Cells form a contiguous pavement over much of the dorsal surface of the anterior nerve cord in juveniles and adults, but their function is not known, nor whether they are primary photoreceptors, with axons (provisionally reported by Castro et al., 2015), or secondary receptors, without. The fact that the anterior-most of the series overlap with the lamellar body implies an association between light reception and a region homologous, at least in part, with the vertebrate pretectum and midbrain. This could represent the ancestral condition for chordates, but the only supporting evidence, other than from amphioxus, is the presence of a rhabdomeric, midbrain-level cerebral eye in salps, a group of pelagic tunicates (Lacalli & Holland, 1998; Braun & Stach, 2017). If the Joseph Cells are ancestral in chordates rather than being an amphioxus innovation, the role the midbrain plays in visual processing in vertebrates could more easily be explained, as having begun as an association between a similar set of now vanished rhabdomeric photoreceptors and the neurons located immediately beneath (Lacalli, 2018). The problem, however, even if this were the case, is that there is nothing obvious in the anatomy of the dorsal part of the midbrain homolog in either amphioxus or salps to indicate how the optic tectum originated. The apparent apico-basal inversion of amphioxus ADBs suggests these cells could be precursors of neurons populating the cortical structures in vertebrate brains, including the tectum (Lacalli 1996, 2017), but we are still a long way from having a convincing evolutionary scenario to account for the diverse cell types and neuroanatomical complexity seen in any of these layered structures. Whether the combination of caudal diencephalon and mesencephalon into a single domain is ancestral for chordates or a secondary specialization of amphioxus, the key issue is to understand the functions carried out within this domain. The evidence from amphioxus larvae is that this is the main site for sensory integration, with overall levels of locomotory activity being modulated by multiple inputs. This argues for similar functions being carried out by this same domain in ancestral chordates, including in immediate ancestors of vertebrates, and for the presence of a counterpart of the psz. The amphioxus dien-mesencephalon lacks identifiable subdivisions, and the lamellar body is not an entirely reliable marker for the amphioxus counterpart of the diencephalon, so it is difficult to place the vertebrate homolog of the psz more precisely than to say that it should lie somewhere in the basal part of either the caudal diencephalon or rostral midbrain. 3. Amphioxus and the origin of consciousness It is comparatively uncontroversial to suppose that other primates, and indeed other mammals, very likely have conscious experiences qualitatively similar to our own. This supposition is based on the similar structure of those regions of the brain generally accepted as the seat of consciousness in mammals, i.e. the cerebral cortex and thalamus (Butler, 2008a,b; Meyer 2011). Avian and mammalian forebrains are sufficiently similar that birds also can be supposed to have conscious experiences of a kind we would 4
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
recognize, but the situation in amphibians and fish is harder to assess. To the degree that consciousness depends on anatomical structures and cell types derived from the neocortex, fish and amphibians, which lack these structures, might be expected to have a much reduced or, at the very least, qualitatively different awareness of sensory experience than mammals and birds. Circumstantial evidence implies that the midbrain may itself be a source of conscious sensations (Barron & Klein, 2016; Merker, 2005), including a visual awareness of the surrounding environment that aids in guiding locomotion and other muscular actions. The case for midbrainbased consciousness is most fully developed in a recent book and companion article by Feinberg and Mallatt (2016a,b), who argue first, for recognizing the diverse forms consciousness takes, whereby some (e.g. vision) are associated with isomorphic sensory maps, while others (affective feelings) are not. They then summarize the arguments in support of conscious sensations being generated in brain structures other than thalamo-cortical ones. For fish and amphibians, the optic tectum is the structure that most closely approaches mammalian cortex in terms of complexity, neural connectivity and hierarchical structure. This implies it may have similar functional capabilities, including the ability to generate a consciously perceived visual display, and the visual search behaviors of fish and amphibians are at least consistent with this supposition. To properly assess the role the midbrain may play in consciousness, a better understanding is needed of the structure and circuitry of the optic tectum than is currently available, including of pathways from the tectum to more basal centers. There are reciprocal links between these, not unlike those between thalamus and cortex (McHaffie, Stanford, Stein, Coizet, & Redgrave, 2005), which is, at the very least, suggestive. The operative question, from an amphioxus perspective, relates more specifically to how dorsal inputs to basal midbrain centers differ between vertebrates and amphioxus, and whether those differences reveal anything about how conscious sensations are produced. Here the evolutionary context becomes relevant. Chordates diversified in the early to middle Cambrian at a time of increasingly intense predation, notably from large pelagic arthropods with grasping mouthparts, and this is probably what drove improvements to ancestral chordate musculature, body support systems and swimming ability (Feinberg & Mallatt, 2016a, Chap. 4; Lacalli, 2012; McAllister 2003). To this, vertebrates added improved sense organs and sensory processing systems (Feinberg & Mallatt, 2016a, Chap. 5; Gans, 1989; Gee, 2018, Chaps. 10 & 15; Khaner, 2007; Lacalli, 2001; Shimeld & Holland, 2000), but to influence locomotion, these would presumably still have had to act through existing control centers, which could have included the vertebrate equivalent of the amphioxus psz. The postulate to consider here then, is that consciousness may have evolved as part of an enhanced arousal mechanism, enabling a preexisting, amphioxus-type synaptic center to produce a behavioral response that was either more rapid, more focused (e.g. via better suppression of competing behaviors), or more selective (via more effective sensory filtering) than could be achieved using the ancestral non-conscious reflex pathways alone. All of these functions can be achieved without consciousness, but conscious awareness of one or more of the relevant sensory inputs has the potential to be a significant further enhancement. The emergence of consciousness would then be expected to correlate in some way with the appearance of novel types of circuits in the inputs to the psz homolog, dedicated to selective signal amplification or involving a degree of integrative complexity beyond that needed simply to coordinate an existing set of reflexes. Linking consciousness to the evolution of the vertebrate tectum implies a link also to the evolution of image-forming eyes. Amphioxus is not particularly informative here, because its frontal eye is too small and simple in structure to form an image. In the absence of evidence that the frontal eye is secondarily simplified, it would seem to represent something close to the basal condition in chordates before image-forming eyes evolved. The key distinction that then needs to be made is between the earliest chordates, which were filter feeders that probably lacked paired eyes, and ancestral vertebrates with paired eyes, some of which, e.g. conodonts, were clearly predators (McAllister 2003; Purnell, 2001a,b). The ability to visually track prey in three dimensions while swimming is highly advantageous for an active predator (e.g. Andrew, Tommasi, & Ford, 2000), and midbrain-based consciousness could have emerged solely to assist in this function. However, predators are always at risk from yet larger predators, including larger con-specifics, so in this respect they face the same problem that non-predatory invertebrate chordates and early vertebrates would have faced, and would benefit just as much from mechanisms that enhanced their ability to detect and respond rapidly to predator attack. In summary, if midbrain-based consciousness emerged very early in the vertebrate lineage, there are two scenarios to consider: that this happened (1) in parallel with the evolution of image-forming eyes and tectal processing in predatory vertebrates, most likely to aid in tracking prey, or (2) as part of an enhanced arousal mechanism in yet earlier chordates, either predatory or non-predatory, with much simpler visual systems. In the latter case there would likely have been no perception of a visual field, but only of light, or its absence, or of changes in light intensity. With option (1), we face the more difficult task of explaining how a complex sensory construct, in the form of a consciously perceived visual field, first came into being. 4. Neural correlates of consciousness, and a strategy for investigating qualia The involvement of midbrain centers in processing visual inputs is a shared feature of vertebrates, and so, if the midbrain acts as a source of conscious sensations, this also should be a feature shared by all living vertebrates. From a research perspective, the midbrain of lower vertebrates is then potentially as informative a subject as mammalian forebrain for investigating the neural basis of consciousness. Lower vertebrates also offer significant advantages as research subjects, the zebrafish larva being a prime example, whose small size and transparency make it especially suitable for both 3D EM-level brain reconstruction (Hildebrand et al., 2017) and the optogenetic methods used to image the activity of neural circuits in real time (Ahrens et al., 2013; Friedrich et al., 2013). Mapping the synaptic circuitry of significant part of the larval zebrafish brain in full detail then becomes an achievable goal, at least in principle. The task of analysis is likely to be daunting, even with a complete synaptic map in hand, and the involvement of nonsynaptic interactions that lack clear morphological correlates may complicate matters further. These are practical difficulties, however, rather than reasons to deny the possibility of obtaining, perhaps in the not-too-distant future, a dataset that contains all the information necessary to explain consciousness. 5
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
For anyone dealing directly with neurocircuitry data from brain regions known or supposed to generate conscious sensations, the operative question is how one would identify neural correlates of consciousness if indeed they were present. Data from simple invertebrate systems, including from the nematode Caenorhabditis, the best-studied example where the neural morphology is fully known, show that reflex circuits consisting of no more than tens or hundreds of neurons can be extraordinarily difficult to analyze (Smith, 2017; Yan et al., 2017). The smallest of vertebrate brains have orders of magnitude more neurons, and if the neural circuits responsible for consciousness are fully embedded in reflex pathways of correspondingly greater complexity, distinguishing the former from the latter could be an insurmountable task. One insight from amphioxus is that this is only the worst-case scenario in a range of possibilities. Young amphioxus larvae roughly match Caenorhabditis larvae in size and numbers of neurons, but a subset of neural pathways in amphioxus have proven comparatively easy to analyze. The activating pathway for the escape response is an example, where a high degree of redundancy yields a strong signal regarding connectivity patterns, overriding the noise inherent in incomplete data and variation between cells and individuals. In contrast, the dorsal modulatory circuits in even young amphioxus larvae involve a diverse assortment of translumenal interneurons (Lacalli & Kelly, 2003b), for which it is far more difficult to identify consistent patterns of connectivity. The problem increases with development, as the translumenal component of the brain itself increases (Castro et al., 2015), which implies that a full analysis of circuits in the juvenile and adult brain is likely to be a decidedly non-trivial task. So, how difficult will it be to identify the circuits responsible for vertebrate consciousness? My guess is very difficult indeed, if, as above, they are deeply embedded in circuits that are already highly complex. On the other hand, if they are highly redundant add-ons to existing pathways, they might announce themselves in ways that make their identity obvious by inspection. Even if neural correlates of consciousness can eventually be identified and their functional properties understood, this in itself will not necessarily yield a conceptually satisfying explanation for how conscious sensations are produced. The issue here is whether subjective experience can in principle be understood in material terms (e.g. Chalmers, 1995; Levine, 1983), which, though a serious concern, is somewhat beyond the scope of this account. Whether or not midbrain circuitry in an organism like the zebrafish larva will, by itself, reveal a mechanism for generating conscious sensations, any circuitry differences that vary depending on the type of sensory input are potentially important for what they may reveal about the nature of qualia, i.e. the way a particular sensory input is subjectively experienced. Explaining how qualia are generated is one of the classical hard problems of consciousness, complicated by the richness of sensory experience, and the myriad variations within each sensory category, e.g. of different colors or a range of sounds, all of which needs explaining. The insight from amphioxus is that the problem may be much simpler, because there could well have been only two sensory inputs of importance when midbrain-based consciousness first emerged in chordates, for light and for physical contact. In the latter case, we would most likely be dealing with sharp pain of the kind produced by a wound or injury rather than sustained deeper forms of pain, as the neural correlates of the former in lower vertebrates (fibers types in this instance) are a better match to the amniote condition. Feinberg and Mallatt (2016a, Chap. 8) deal at some length with the question of whether fish feel pain in the same way as amniotes. The case that they do is less compelling than for vision because, in contrast to visual processing, pain signals are not associated in the same way with a dedicated integrative center organized similarly across vertebrate taxa (see Feinberg & Mallatt, 2016a, Table 7.2 and references therein). It is consequently more difficult to construct a plausible argument as to where, in the vertebrate lineage, nonconscious nociception was replaced by a conscious perception of pain comparable to our own, or what role the midbrain played in this transition. However, even if the conscious perception of both light and sharp pain are not of equal antiquity, it is a valid investigative strategy to focus on these two sensory modalities in the first instance, so long as there is reason to suppose that both emerged before other forms of consciousness. One would then be looking for differences in the midbrain pathways for light and sharp pain that might account for the different way light and pain are perceived. As to the qualia themselves, it is possible they were identical or very similar at first and diverged subsequently. Conversely, perhaps they differed from the start, with sharp pain being, say, a negative affect and light a positive one. It is worth noting that, for the first option, assuming increased light intensity both increased the risk of predator attack and correlated consistently with the kind of pain caused by such attack, there would have been less reason to distinguish between light and pain stimuli. Either would produce the required effect, of activating the arousal pathway, which could explain features common to the experience of intense pain and an intense flash of light, for example, the way both are similarly able to momentarily monopolize one’s complete attention. Perhaps by this means we experience, at first hand, something of what it would have been like to be an early vertebrate, where what mattered was for one signal to dominate over all others, and a broader spectrum of additional qualia had yet to evolve.
5. Conclusions We are entering a potentially very fruitful period in neuroscience, with methods emerging that promise to reveal neural connectivity and real-time function at the cellular level in exquisite detail. Obtaining the data necessary to understand the neural basis of consciousness is a daunting task, given the size and complexity of even the smallest vertebrate brain, but it is no longer an unrealistic goal in principle. The insight from amphioxus is that there are specific brain regions in vertebrates where there is a better prospect than elsewhere for identifying the innovations in neural circuitry that made conscious sensation possible, namely in the pain and visual pathways impinging on the locomotory control centers of the basal midbrain. Whether this will fully explain consciousness as a phenomenon remains to be determined, but there are reasons to be optimistic that currently available methods may be sufficient to solve at least some of the hard problems of consciousness.
6
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
Declaration of interest None. Acknowledgements The author thanks Jon Mallatt for his comments on an earlier draft of this review, and Heather Down for assisting with the preparation of figures. Source of finding This work was supported by the Lionel G. Harrison Research Trust. References Andrew, R. J., Tommasi, L., & Ford, N. (2000). Motor control by vision and the evolution of cerebral lateralization. Brain and Language, 73, 220–235. Ahrens, M. B., Orger, M. B., Robson, D. N., Li, J. M., & Keller, P. J. (2013). Whole-brain functional imaging at cellular resolution using light-sheet microscopy. Nature Methods, 10, 413–430. Albuixechi-Crespo, B., Irimia, M., Burguera, D., Maeso, I., Arrones, L. S., Moreno-Bravo, J. A., ... Ferran, J. L. (2017). Molecular regionalization of the amphioxus neural tube challenges major partitions of the vertebrate brain. PLoS Biology, 15, 4. Barron, A. B., & Klein, C. (2016). What insects can tell us about the origins of consciousness. Proceedings of the National Academy of Sciences USA, 113, 4900–4908. Bozzo, M., Macri, S., Calzia, D., Sgarra, R., Manfioletti, G., Ramoino, P., ... Candiani, S. (2017). The HMGA gene family in chordates: Evolutionary perspectives from amphioxus. Development, Genes and Evolution, 227, 201–211. Braun, K., & Stach, T. (2017). Structure and ultrastructure of the eyes and brains of Thalia democratica. Journal of Morphology, 278, 1421–1437. Butler, A. (2008a). Evolution of brains, cognition and consciousness. Brain Research Bulletin, 75, 442–449. Butler, A. (2008b). Evolution of the thalamus: A morphological and functional review. Thalamus and Related Systems, 4, 35–58. Castro, A., Becerra, M., Manso, M. J., & Anadon, R. (2015). Neuronal organization in the brain in the adult amphioxus (Branchiostoma lanceolatum): A study with acetylated tubulin immunohistochemistry. Journal of Comparative Neurology, 523, 2211–2232. Chalmers, D. J. (1995). Facing up to the problem of consciousness. Journal of Consciousness Studies, 2, 200–219. Churcher, A. M., & Taylor, J. S. (2010). The antiquity of chordate odorant receptors is revealed by the discovery of orthologs in the cnidarian Nematostella vectensis. Genome Biology and Evolution, 3, 36–43. Feinberg, T. E., & Mallatt, J. M. (2016a). The ancient origins of consciousness. Cambridge, MA: MIT Press. Feinberg, T. E., & Mallatt, J. M. (2016b). The nature of primary consciousness. A new synthesis. Consciousness and Cognition, 43, 113–127. Friedrich, R. W., Genoud, C., & Wanner, A. A. (2013). Analyzing the structure and function of neuronal circuits in zebrafish larvae. Frontiers in Neural Circuits, 7, 71. Gans, C. (1989). Stages in the origin of vertebrates: Analysis by means of scenarios. Biological Reviews, 64, 221–268. Gee, H. (2018). Across the bridge: Understanding the origin of vertebrates. Chicago: University of Chicago Press. Grillner, S., Robertson, B., & Stephenson-James, M. (2013). The evolutionary origin of the vertebrate basal ganglia and its role in action selection. Journal of Physiology, 591, 5425–5431. Hildebrand, D. G. C., Cicconet, M., Torres, R. M., Choi, W., Quan, T. M., Moon, J., ... Engert, F. (2017). Whole-brain serial-section electron microscopy in larval zebrafish. Nature, 545, 345–349. Holland, L. Z. (2009). Chordate roots of the vertebrate nervous system: Expanding the molecular toolkit. Nature Reviews Neuroscience, 10, 736–746. Holland, L. Z. (2015). Genomics, evolution and development of amphioxus and tunicates: The Goldilocks principle. Journal of Experimental Zoology (Molecular Development and Evolution), 324, 342–352. Holland, L. Z., Carvalho, J. E., Escriva, H., Laudet, V., Schbert, M., Shimeld, S. M., & Yu, J. K. (2013). Evolution of the bilaterian central nervous system: A single origin? EvoDevo, 4, 27. Khaner, O. (2007). Evolutionary innovations of the vertebrates. Integrative Zoology, 2, 60–67. Lacalli, T. C. (1996). Frontal eye circuitry, rostral sensory pathways and brain organization in amphioxus larvae: Evidence from 3D reconstruction. Philosophical Transactions of the Royal Society of London, Series B, 351, 243–263. Lacalli, T. C. (2001). New perspectives on the evolution of protochordate sensory and locomotory systems, and the origin of brains and heads. Philosophical Transactions of the Royal Society of London, Series B, 356, 1565–1572. Lacalli, T. C. (2002a). Sensory pathways in amphioxus larvae I. Constituent fibers of the rostral and anterodorsal nerves, their targets and evolutionary significance. Acta Zoologica, 83, 149–166. Lacalli, T. C. (2002b). The dorsal compartment locomotory control system in amphioxus larvae. Journal of Morphology, 252, 227–237. Lacalli, T. C. (2004). Sensory systems in amphioxus: A window on the ancestral chordate condition. Brain Behavior and Evolution, 64, 148–162. Lacalli, T. C. (2012). The Middle Cambrian fossil Pikaia and the evolution of chordate swimming. EvoDevo, 3, 12. Lacalli, T. C. (2017). Interpreting amphioxus, and thoughts on ancestral mouths and brains. International Journal of Developmental Biology, 61, 649–654. Lacalli, T. C. (2018). Amphioxus, motion detection, and the evolutionary origin of the vertebrate retinotectal map. EvoDevo, 9, 6. Lacalli, T. C., Holland, N. D., & West, J. E. (1994). Landmarks in the anterior central nervous system of amphioxus larvae. Philosophical Transactions of the Royal Society of London, Series B, 344, 165–185. Lacalli, T. C., & Holland, L. Z. (1998). The developing dorsal ganglion of the salp Thalia democratica, and the nature of the ancestral chordate brain. Philosophical Transactions of the Royal Society of London, Series B, 353, 1943–1967. Lacalli, T. C., & Kelly, S. J. (2003a). Ventral neurons in the anterior nerve cord of amphioxus larvae I. An inventory of cell types and synaptic patterns. Journal of Morphology, 257, 190–211. Lacalli, T. C., & Kelly, S. J. (2003b). Sensory pathways in amphioxus larvae II. Dorsal tracts and translumenal cells. Acta Zoologica, 84, 1–13. Levine, J. (1983). Materialism and qualia: The explanatory gap. Pacific Philosophical Quarterly, 64, 354–361. Mashour, G. A., & Alkire, M. T. (2013). Evolution of consciousness: Phylogeny, ontogeny, and emergence from general anesthesia. Proceedings of the National Academy of Sciences USA, 110(Suppl. 2), 10357–10364. McAllister, J. (2003). Predation of fishes in the fossil record. In P. H. Kelley, M. Kowalewski, & T. A. Hansen (Eds.). Predator-prey interactions in the fossil record (pp. 303–324). Berlin: Springer. McHaffie, J. G., Stanford, T. R., Stein, B. E., Coizet, V., & Redgrave, P. (2005). Subcortical loops through the basal ganglia. Trends in Neurosciences, 28, 401–407. Merker, B. (2005). The liabilities of mobility: A selection pressure for the transition to consciousness in animal evolution. Consciousness and Cognition, 14, 89–114. Merker, B. (2007). Consciousness without a cerebral cortex: a challenge for neuroscience and medicine. Behavior and Brain Sciences, 30, 63-81, discussion 81-134. Meyer, K. (2011). Primary sensory cortices, top-down projections and conscious experience. Progress in Neurobiology, 94, 408–417. O’Connell, L. A., & Hofmann, H. A. (2011). The vertebrate mesolimbic reward system and social behavior network: A comparative synthesis. Journal of Comparative Neurology, 519, 3599–3639.
7
Consciousness and Cognition xxx (xxxx) xxx–xxx
T. Lacalli
Pani, A. M., Mullarkey, E. E., Aronowicz, J., Assimacopoulos, S., Grove, E. A., & Lowe, C. J. (2012). Ancient deuterostome origins of vertebrate brain signalling centres. Nature, 483, 289–294. Pergner, J., & Kozmik, Z. (2017). Photoreceptors of amphioxus – Insights into evolution of vertebrate opsins, vision and circadian rhythmicity. International Journal of Developmental Biology, 61, 665–681. Priebe, N. J., & Ferster, D. (2002). A new mechanism for neuronal gain control (or how the gain in brains has mainly been explained). Neuron, 35, 602–604. Purnell, M. A. (2001a). Feeding in conodonts and other early vertebrates. In D. E. G. Briggs, & P. R. Crowther (Eds.). Palaeobiology II (pp. 401–404). Oxford: Blackwell Science. Purnell, M. A. (2001b). Scenarios, selection and the ecology of early vertebrates. In P. E. Ahlberg (Ed.). Major events in early vertebrate evolution (pp. 187–208). London: Taylor & Francis. Ross, L. S., Parrett, T., & Easter, S. S. (1992). Axonogenesis and morphogenesis in the embryonic zebrafish brain. Journal of Neuroscience, 12, 467–482. Ryan, K., Lu, Z., & Meinertzhagen, I. A. (2017). Circuit homology of decussating pathways in the Ciona larval CNS and the vertebrate startle-response pathway. Current Biology, 27, 721–728. Satoh, G. (2005). A trajectory of increasing activity and the elaboration of chemosensory modality: A new perspective on vertebrate origins. Zoological Science, 22, 613–626. Shimeld, S. M., & Holland, P. W. H. (2000). Vertebrate innovations. Proceedings of the National Academy of Sciences, USA, 97, 4449–4452. Sirota, M. G., Di Prisco, G. V., & Dubuc, R. (2000). Stimulation of the mesencephalic locomotor region elicits controlled swimming in semi-intact lampreys. European Journal of Neuroscience, 12, 4081–4092. Smith, K. (2017). The brain-circuit challenge. Nature, 548, 150–152. Vopalensky, P., Pergner, J., Liegertova, M., Benito-Guitierrez, E., Arendt, D., & Kozmik, Z. (2012). Molecular analysis of the amphioxus frontal eye unravels the evolutionary origin of the retina and pigment cells of the vertebrate eye. Proceedings of the National Academy of Sciences USA, 109, 15383–15388. Wicht, H., & Lacalli, T. C. (2005). The nervous system of amphioxus: Structure, development, and evolutionary significance. Canadian Journal of Zoology, 83, 122–150. Yan, G., Vertes, P. E., Towlson, E. K., Chew, Y. L., Walker, D. S., Schafer, W. R., & Barabasi, A. L. (2017). Network control principles predict neuron function in the Caenorhabditis elegans connectome. Nature, 550, 519–523. Zieger, E., Lacalli, T. C., Pestarino, M., Schubert, M., & Candiani, S. (2017). The origin of dopaminergic systems in chordate brain: Insights from amphioxus. International Journal of Developmental Biology, 61, 749–761.
8