Background: By examining species-specific innate behaviours, neuroethologists have characterized unique neural strategies and specializations from throughout the animal kingdom. Simultaneously, the field of evolutionary developmental biology (informally, “evo-devo”) seeks to make inferences about animals’ evolutionary histories through careful comparison of developmental processes between species, because evolution is the evolution of development. Yet despite the shared focus on cross-species comparisons, there is surprisingly little crosstalk between these two fields. Insights can be gleaned at the intersection of neuroethology and evo-devo. Every animal develops within an environment, wherein ecological pressures advantage some behaviours and disadvantage others. These pressures are reflected in the neurodevelopmental strategies employed by different animals across taxa. Summary: Vision is a system of particular interest for studying the adaptation of animals to their environments. The visual system enables a wide variety of animals across the vertebrate lineage to interact with their environments, presenting a fantastic opportunity to examine how ecological pressures have shaped animals’ behaviours and developmental strategies. Applying a neuroethological lens to the study of visual development, we advance a novel theory that accounts for the evolution of spontaneous retinal waves, an important phenomenon in the development of the visual system, across the vertebrate lineage. Key Messages: We synthesize literature on spontaneous retinal waves from across the vertebrate lineage. We find that ethological considerations explain some cross-species differences in the dynamics of retinal waves. In zebrafish, retinal waves may be more important for the development of the retina itself, rather than the retinofugal projections. We additionally suggest empirical tests to determine whether Xenopus laevis experiences retinal waves.

A major and long-standing mission for the field of neuroscience is the linking of nervous structure to behavioural function. This aim motivated even the earliest neuroscience experiments, as when Luigi Galvani reported in the late-18th century that electrically stimulating the nerves of a frog was sufficient to make its legs twitch [1]. In more recent years, methodological innovations have enabled fine-grained manipulation of neuronal states to an unprecedented extent. One might think that these advances afford neuroscientists ever greater confidence to infer causal relationships between brain states and behaviours, yet paradoxically, as neuroscience tends towards ever-greater reductionism in its models and philosophies, behaviour itself often takes a back seat in the study of its causes.

This notion is compellingly argued by Krakauer and colleagues in their influential 2017 article, “Neuroscience needs behavior: correcting a reductionist bias” [2]. The neural implementation of a behaviour, Krakauer et al. [2] assert, is best probed only after the careful decomposition and analysis of the behaviour itself. Yet where behaviour was once so often at the centre of neuroscience, it has now lost its role of primacy. The deemphasis of behaviour has contributed to a seemingly insurmountable gap between neuroethologists, who prioritize the naturalistic behaviours of the animals they study, and mainstream neurophysiologists, who tend towards reductionistic views of animal behaviour and its substrates. Yet this gap should not exist. As Jenny Read remarks with beautiful succinctness: “Evolutionarily, all that is required is useful behaviour” [3]. To understand evolution, we must understand behaviour. Furthermore, the rich world of animal behaviour is valuable to all sorts of neurophysiologists – even those who may not consider themselves to be working on questions of evolution.

To demonstrate this, we take as an example the neuroscience of vision. Many vertebrates and invertebrates rely on visual cues to navigate and interact with their environments. Thus, the visual system presents a fantastic window into how ecological pressures have shaped different animals’ behaviours and their developmental strategies. We turn our attention specifically to the development of the visual system, including the retina and its downstream targets in the brain, because evolution is the evolution of development [4, 5]. Evolution must act on the developing organism because adult animals do not transform into other adult animals. As such, to study how visual systems have evolved, we must study the development of visual systems across species.

In the field of visual development, a great deal of research has gone into the phenomenon of spontaneous retinal waves, which are bursts of action potentials that propagate across the retina early in development absent any external sensory input. Spontaneous activity, including spontaneous retinal waves, is known to serve a variety of functions in early development, including the establishment of neuronal circuits that shall enable sensory processing later in life. In this review, we look particularly at retinal waves as one type of spontaneous activity in CNS development known to guide the development of the visual system. Despite the importance of spontaneous retinal waves, little work has characterized inter-species differences in the presence or function of retinal waves, let alone considered how retinal waves relate to an animal’s lifestyle. Implementing a neuroethological perspective, we develop a new theory of retinal waves across the vertebrate lineage, demonstrating the value of neuroethology for understanding development and evolution.

Early in development, in some species, bursts of action potentials spontaneously initiate and travel across the retina in a wave-like fashion. These spontaneous, correlated action potentials, termed “retinal waves,” are known to play an important role in several processes of mammalian visual development, such as regulating the formation of gap junctions between retinal ganglion cells (RGCs) [6]; modulating inter-retinal coordinated activity [7]; influencing the receptive fields of neurons in visual cortex, superior colliculus, and lateral geniculate nucleus (reviewed in [8]); and guiding refinement of retinotopic maps [9, 10]. Despite their profound importance for mammalian visual development, and despite the early seminal work on spontaneous retinal activity carried out in turtles – first, with multiunit electrophysiological recordings [11, 12], then with the direct observation of waves via calcium imaging [13] – few researchers have systematically addressed the presence and function of spontaneous retinal waves in the non-mammalian vertebrate lineage.

How Widespread Are Retinal Waves?

Since their first observation in rats, spontaneous retinal waves have been studied in the development of a variety of species, including mice, rabbits, ferrets, cats, non-human primates, chicks, turtles, and zebrafish (Table 1). These species comprise a moderate diversity of the vertebrate family tree (Fig. 1), demonstrating the presence of retinal waves across myriad mammalian species as well as two species of sauropsids (the chick and the turtle), and most recently, a ray-finned fish (the zebrafish).

Table 1.

Summary of phylogenetic classification and retinal wave presence in species studied, to date

SpeciesTaxonomic cladeRetinal waves observedAge at which retinal waves studiedMethodFoundational citation(s)
Macaque (Macaca fascicularisPrimates Yes Embryonic days 60–76 Multielectrode arrays on isolated retinae Warland et al. (2006) [14
Rabbit (Oryctolagus cuniculusLagomorpha Yes Postnatal days 0–6 Whole-cell patch clamp on isolated retinae Zhou (1998) [15
Mouse (Mus musculusRodentia Yes Postnatal days 0–4 Refers to unpublished data from isolated mice retinae Mooney et al. (1996) [16
Rat (Rattus norvegicusRodentia Yes Embryonic days 17–21 In vivo extracellular unit recordings from retinae Galli and Maffei (1988) [17
Ferret (Mustela furoCarnivora Yes Postnatal days 5–21 Multielectrode arrays on isolated retinae Meister et al. (1991) [18
Cat (Felis catusCarnivora Yes Embryonic day 52; postnatal day 1 Multielectrode arrays on isolated retinae Meister et al. (1991) [18
Chick (GallusgallusGalliformes Yes Embryonic days 6–11 Calcium imaging on isolated retinae Catsicas et al. (1998) [19
Turtle (Trachemys scripta elegansTestudines Yes Stage 22 (approx. 3 weeks to hatching) – stage 26 (time of hatching) In vivo extracellular unit recordings from retinae (1995). Calcium imaging on isolated retinae (2003) Sernagor and Grzywacz (1995) [20
Sernagor et al. (2003) [13
Frog (Xenopus laevisAnura No Retinal waves not found from stage 39 (approx. 2.5 days post-fertilization) – stage 60 (approx. 46 days post-fertilization) Multielectrode arrays on isolated retinae Demas et al. (2012) [21
Zebrafish (DanioreroCypriniformes Yes 2.5–3.5 days post-fertilization In vivo two-photon calcium imaging; whole-cell patch clamp recording Zhang et al. (2016) [22
Fruit fly (Drosophila melanogasterDiptera No Retinal waves not found from 40 h after puparium formation (hAPF) – adulthood In vivo two-photon calcium, voltage, and glutamate imaging Akin et al. (2019) [23
SpeciesTaxonomic cladeRetinal waves observedAge at which retinal waves studiedMethodFoundational citation(s)
Macaque (Macaca fascicularisPrimates Yes Embryonic days 60–76 Multielectrode arrays on isolated retinae Warland et al. (2006) [14
Rabbit (Oryctolagus cuniculusLagomorpha Yes Postnatal days 0–6 Whole-cell patch clamp on isolated retinae Zhou (1998) [15
Mouse (Mus musculusRodentia Yes Postnatal days 0–4 Refers to unpublished data from isolated mice retinae Mooney et al. (1996) [16
Rat (Rattus norvegicusRodentia Yes Embryonic days 17–21 In vivo extracellular unit recordings from retinae Galli and Maffei (1988) [17
Ferret (Mustela furoCarnivora Yes Postnatal days 5–21 Multielectrode arrays on isolated retinae Meister et al. (1991) [18
Cat (Felis catusCarnivora Yes Embryonic day 52; postnatal day 1 Multielectrode arrays on isolated retinae Meister et al. (1991) [18
Chick (GallusgallusGalliformes Yes Embryonic days 6–11 Calcium imaging on isolated retinae Catsicas et al. (1998) [19
Turtle (Trachemys scripta elegansTestudines Yes Stage 22 (approx. 3 weeks to hatching) – stage 26 (time of hatching) In vivo extracellular unit recordings from retinae (1995). Calcium imaging on isolated retinae (2003) Sernagor and Grzywacz (1995) [20
Sernagor et al. (2003) [13
Frog (Xenopus laevisAnura No Retinal waves not found from stage 39 (approx. 2.5 days post-fertilization) – stage 60 (approx. 46 days post-fertilization) Multielectrode arrays on isolated retinae Demas et al. (2012) [21
Zebrafish (DanioreroCypriniformes Yes 2.5–3.5 days post-fertilization In vivo two-photon calcium imaging; whole-cell patch clamp recording Zhang et al. (2016) [22
Fruit fly (Drosophila melanogasterDiptera No Retinal waves not found from 40 h after puparium formation (hAPF) – adulthood In vivo two-photon calcium, voltage, and glutamate imaging Akin et al. (2019) [23

Species: common and binomial scientific name. Taxonomic clade: phylogenetic order of the species. Retinal waves observed: indication of whether retinal waves have been observed in this species. Age at which retinal waves studied: the developmental time points at which retinal waves were studied in the foundational studies. Method: what method was used to study retinal waves in the foundational citation(s). Foundational citation(s): exemplary, seminal literature regarding the observation or absence of spontaneous retinal activity in this species.

Fig. 1.

Cladogram depicting the study of spontaneous retinal waves across species. A green silhouette and check indicates retinal waves have been observed in this phylogenetic group. A red silhouette and cross indicates that observations have failed to detect retinal waves in this group. To our knowledge, retinal waves have not been studied in any taxonomic group besides those indicated here.

Fig. 1.

Cladogram depicting the study of spontaneous retinal waves across species. A green silhouette and check indicates retinal waves have been observed in this phylogenetic group. A red silhouette and cross indicates that observations have failed to detect retinal waves in this group. To our knowledge, retinal waves have not been studied in any taxonomic group besides those indicated here.

Close modal

Notably, studies have failed to observe spontaneous retinal waves in two species: the fruit fly (Drosophila melanogaster) [23, 24] and the African clawed frog (Xenopus laevis) [21]. First, we will consider the fruit fly. Spontaneous activity serves a role in the development of the Drosophila visual system [23, 25], though this activity likely originates in the brain rather than the retina, as it has been observed even in mutants whose retinal function has been abolished [23]. It seems likely, then, that retinal waves are a uniquely vertebrate phenomenon, though outstanding questions remain about the role and character of spontaneous activity in invertebrate CNS development.

Next, we will consider X. laevis, the African clawed frog. Within the vertebrate lineage, it was theorized prior to the observation of retinal waves in zebrafish that retinal waves occur exclusively in amniotes (i.e., animals that develop with an amnion – mammals, birds, and reptiles) [26‒28]. Retinal waves had been documented in a diversity of amniotic species and were believed to be absent in X. laevis – an anamniote – supporting a theory that amniotes evolved retinal waves as a strategy to achieve visual system refinement while relatively deprived of visual stimuli, as would be the case in ovo or in utero [27, 28]. However, the observation of retinal waves in zebrafish [22, 29] now offers a compelling justification to revisit this apparent demarcation between amniotes and anamniotes. If retinal waves are present in zebrafish, why might they be absent in Xenopus? Furthermore, can the environment and ecological niche of the developing zebrafish shed light upon a possible divergence in retinal wave function within the vertebrate tree?

The Ethological Role of Retinal Waves

Within both mammals and zebrafish, retinal waves have a known tendency to progress temporally-to-nasally (Fig. 2a) [22, 30‒32]. This temporal-to-nasal bias makes sense only in light of ethological considerations. The directionality of spontaneous retinal waves corresponds to the optic flow of sensory information across the retina as an animal moves around its environment, such as when a developing tadpole swims around its environment (Fig. 2b) or an older mouse navigates on the ground (Fig. 2c). In each case, patterned retinal activity – whether spontaneous or evoked by light – progresses from the temporal side of the retina first to the nasal side last. In other words, the directionality of spontaneous retinal waves corresponds to the directionality of visually evoked sensory information animals will receive as they navigate their environments.

Fig. 2.

Temporal-to-nasal patterned retinal activity – whether spontaneous or evoked by light. a Spontaneous retinal waves stereotypically propagate from the temporal side of the retina to the nasal side. b As a tadpole swims forward, the visual scene stimulates the retina in the same temporal-to-nasal manner. c Similarly, the visual environment stimulates the mouse’s retina temporally-to-nasally as the mouse runs forward. Figure produced by Mr. Erick Fernandez de Arteaga.

Fig. 2.

Temporal-to-nasal patterned retinal activity – whether spontaneous or evoked by light. a Spontaneous retinal waves stereotypically propagate from the temporal side of the retina to the nasal side. b As a tadpole swims forward, the visual scene stimulates the retina in the same temporal-to-nasal manner. c Similarly, the visual environment stimulates the mouse’s retina temporally-to-nasally as the mouse runs forward. Figure produced by Mr. Erick Fernandez de Arteaga.

Close modal

This is not a coincidence. The nasal-to-temporal optic flow produced by spontaneous retinal waves (Fig. 2a), swimming forward (Fig. 2b), and running forward (Fig. 2c) is important for the development of the visual system. In a provocative series of experiments, Hiramoto and Cline [33] find that exposing developmental stage 47 tadpoles to nasal-to-temporal visual stimuli – i.e., opposite to the naturalistic direction of optic flow – results in disrupted refinement of the retinotectal projection. Ge and colleagues [34] report a strikingly similar phenomenon in mice: manipulating the directionality of retinal waves in newborn mice (postnatal day 8–11) impairs neuronal responses to visual motion in the superior colliculus. In whole, these experiments demonstrate a persistent pattern: the temporal-to-nasal directionality of optic flow is important, playing a key role in the development of a visual system that will meet the needs of an animal navigating its environment primarily through forward locomotion. Behaviour is the key to understanding this pattern: it is only by considering the behaviours of these animals – for instance, the tendency for mice to run forward and tadpoles and fish to swim forward – that the importance of directionality in retinal waves can be understood.

Retinal Waves in the Broader Context of Zebrafish Development

While it has been theorized that amniotes (i.e., mammals, birds, and reptiles) evolved spontaneous retinal waves as a strategy to achieve visual system refinement under conditions of relative visual deprivation [27, 28], the observation that zebrafish experience spontaneous retinal waves [22] casts doubt upon this theory. Fish, like frogs, are anamniotes (i.e., they do not develop in an amnion); developing in the open water, zebrafish are exposed to naturalistic visual stimuli throughout their development. However, just as the environment of the developing zebrafish differs substantially from that of amniotes, zebrafish retinal waves also seem to differ markedly from those observed in amniotes.

While mammals and reptiles experience distinct types of retinal waves at different time points, spanning weeks of protracted development [13, 35‒37], zebrafish retinal waves appear to be far more homogeneous and occur within a narrow window of 24 h, from 2.5 to 3.5 days post-fertilization [22, 29]. In light of these profound differences, a natural question is whether retinal waves serve the same function between these species.

Among other functions, retinal waves in mammals simulate optic flow and guide the refinement of retinotopic maps [10, 34], thus enabling a high degree of plasticity in mammalian visual development. In contrast, it is possible that spontaneous retinal waves in zebrafish follow a different time course and show distinct dynamics because they do not serve the same functions in zebrafish development. To evaluate this possibility, it is important to consider the broader patterns of visual development in zebrafish and the ecological environment of the developing zebrafish.

The importance of visual experience for the development of the zebrafish visual system has been hotly contested for years (e.g. [38‒44]). Though it seems beyond debate that visual experience plays some role in the development of zebrafish vision, the deficits resulting from early visual deprivation are often subtle.

Of particular note, Niell and Smith [38] report that dark rearing zebrafish until 84 h post-fertilization (i.e., 3.5 days post-fertilization) produces small deficits in spatial acuity, while receptive field size, topography, and direction selectivity are wholly spared. The timing of this manipulation corresponds to the timing of spontaneous retinal waves in zebrafish; where spontaneous retinal waves cease at 3.5 days post-fertilization [22], this is also the stage at which Niell and Smith [38] ended their visual deprivation for zebrafish. Hence, retinal waves are occurring at a time when visual experience does not seem to play much of a role in zebrafish visual development. It is important to note that while dark-rearing zebrafish for short periods early in life produces only subtle deficits, longer periods of dark-rearing produce more substantial and behaviourally relevant visual deficits. For instance, Avitan et al. [41] report that dark-rearing zebrafish until 6 days post-fertilization is sufficient to produce long-lasting deficits in prey-capturing behaviours. Altogether, these experiments indicate that while visual experience plays a role in zebrafish visual development, retinal waves occur at an early time point when visual experience is less important. How does this compare with the timeline of retinal waves in amniotes, like mammals?

Zebrafish develop in open water under a constant threat of predation. Their sensory systems must develop in a rapid and stereotyped fashion, to allow for the quick deployment of sensory-driven behaviours. In contrast, amniotes develop in the relative safety of a womb or protective egg. This relative safety allows for a more protracted and experience-dependent development of sensory systems, including the visual system. Indeed, illustrating the protracted nature of amniotic sensory development, stage I retinal waves – the earliest retinal waves – initiate at around embryonic day 17 in mice, while stage III retinal waves – the final stage of retinal waves – terminate at around the time of eye opening on postnatal day 14 [35, 36, 45]. In contrast, zebrafish retinal waves occur entirely within a 24-h period from 2.5 to 3.5 days post-fertilization [22, 29]. The role of retinal waves in mammalian visual development is also far more dynamic and experience-dependent than what has been observed in zebrafish. As discussed in the paragraphs that follow, the characteristics of later retinal waves in mammals are influenced by both internal states (i.e., sleep and wakefulness) and the external environment (i.e., light exposure through the eyelid).

In mice, eye opening typically occurs around postnatal day 14 [46, 47]. In the days immediately before eye opening, during which stage III retinal waves occur [45], sudden changes in luminance (e.g., flashes of light) are sufficient to produce responses in visual cortical neurons [46]. While the majority of visual cortical activity is driven by retinal waves at this age [48], recent work by Tiriac et al. [48] has demonstrated that visual stimulation through the eyelid can trigger retinal waves with indistinguishable dynamics to spontaneous retinal waves. Furthermore, retinal waves interact with sleep in the developing mouse; bouts of wakefulness from postnatal days 8–12 dramatically suppress visual cortical activity by suppressing retinal wave-driven activity [48]. Altogether, the most recent evidence suggests that the dynamics and effects of amniotic retinal waves are experience-dependent, modulated by both external stimuli (i.e., light) and internal states (i.e., wakefulness).

The literature paints a compelling picture: through the extended period during which retinal waves drive refinement within the mammalian visual system, there is a complex interplay between three stages of retinal waves, light exposure, and sleep over the course of weeks. Zebrafish, in contrast, experience one type of retinal wave from 2.5 to 3.5 days post-fertilization, at a time when sensory experience plays a lesser role in development of the visual system. Altogether, these studies suggest that very early development of the zebrafish visual system may be less dependent upon visual experience than that of mammals, at least during the developmental stages at which retinal waves are known to occur in each species.

Now, turning our attention to other anamniotic species, it is important to note that the experience-independent nature of early zebrafish visual development parallels the experience-independent development of visually guided behaviours in the African clawed frog (X. laevis). Just as early dark-rearing produces only subtle deficits in zebrafish visual development, in Xenopus, dark-rearing does not impede the development of early visually guided, adaptive behaviours such as colour-guided phototaxis [49, 50]. Just as dark-rearing zebrafish for longer periods of time produces deficits in prey-capturing behaviours [41], the visually guided behaviours of older Xenopus tadpoles (i.e., stage 43–44 and older; approximately 3.5–4 days post-fertilization) are influenced by visual experience [51, 52]. These similarities in early visual development might be explained, in part, by the similar ecological niches of these animals. Both larval zebrafish and Xenopus tadpoles, as aquatic species, develop in the danger of open water with the threat of predation. The environment necessitates a rapid development of sensory and motor systems to adequately respond to this threat. Might the similar ecological niche between zebrafish and tadpoles be reflected in the early development of their sensory systems?

Revisiting Retinal Waves in Frogs

While zebrafish are more distantly related to amniotes than amphibians, retinal waves have been observed in zebrafish yet are thought to be absent in amphibians (Fig. 1). This poses an interesting problem. Were retinal waves a convergent evolution among amniotes and at least some ray-finned fishes? Did a common ancestor exhibit retinal waves, which were subsequently lost in the amphibian lineage? Or do amphibians experience retinal waves, but in a manner unlike those experienced by amniotes? To make sense of this conundrum, it is important to revisit the experiments looking for retinal waves in amphibians, bearing in mind the commonalities between zebrafish and amphibian visual development.

In the only study – to our knowledge – examining the question of retinal waves in frogs, Demas et al. [21] used a microelectrode array to record from the RGC layer of retinae isolated from X. laevis tadpoles staged 39–40 (approximately 2.5 days post-fertilization) and older. Subsequent to the formation of the inner plexiform layer at stage 35–37 (Fig. 3a), stage 39–40 is when the first RGC axons begin to innervate the optic tectum, forming the retinotectal circuit, the main retinofugal pathway in the frog (Fig. 3b) [54]. When Demas and colleagues [21] recorded from stage 39–40 tadpole retinae, they could not detect any spontaneous retinal waves. The retina is still responsive – light exposure reliably evokes responses in these isolated retinae [21] – yet there is no evidence of spontaneous retinal waves or other spontaneous correlated activity. The observations reported by Demas and colleagues [21] yield three possible interpretations: (a) X. laevis tadpoles do not experience spontaneous retinal waves, (b) retinal waves occur in X. laevis prior to developmental stage 39, or (c) retinal waves cannot reliably be recorded in the RGC layer of X. laevis retinae (perhaps because they occur only within the inner plexiform layer). We note that interpretations (b) and (c) are not mutually exclusive.

Fig. 3.

Development of the Xenopus visual system. a At stage 35–37, approximately 2 days post-fertilization, the IPL first forms. At this time, bipolar cells begin to synapse onto RGCs. It is unknown whether retinal waves occur in any layer at this stage. Schematic shows the organization of the Xenopus retina, including the IPL. b At stage 39–40, approximately 2.5–3 days post-fertilization, synaptogenesis continues in the IPL. The first retinofugal synapses begin to form between RGCs and tectal cells. Retinal waves do not occur in the RGC layer, per experiments reported by Demas et al. [21]. Xenopus illustrations, stage 37–38 (a) and 40 (b) © Natalya Zahn (2022) [53], via Xenbase (www.xenbase.org RRID:SCR_003280). Scale bars: 1 mm.

Fig. 3.

Development of the Xenopus visual system. a At stage 35–37, approximately 2 days post-fertilization, the IPL first forms. At this time, bipolar cells begin to synapse onto RGCs. It is unknown whether retinal waves occur in any layer at this stage. Schematic shows the organization of the Xenopus retina, including the IPL. b At stage 39–40, approximately 2.5–3 days post-fertilization, synaptogenesis continues in the IPL. The first retinofugal synapses begin to form between RGCs and tectal cells. Retinal waves do not occur in the RGC layer, per experiments reported by Demas et al. [21]. Xenopus illustrations, stage 37–38 (a) and 40 (b) © Natalya Zahn (2022) [53], via Xenbase (www.xenbase.org RRID:SCR_003280). Scale bars: 1 mm.

Close modal

The first interpretation outlined above (i.e., that X. laevis does not experience retinal waves) has generally been preferred, supporting the theory of an amniote/anamniote division in retinal wave evolution. However, the observation of retinal waves in zebrafish offers a compelling reason to revisit this interpretation.

Intriguingly, if the second interpretation is correct (i.e., X. laevis retinal waves occur prior to stage 39), that would mean that retinal waves occur before the retinotectal circuit forms [54]. In other words, if Xenopus tadpoles experience retinal waves at all, their timing would be restricted such that they cannot be involved in the refinement of the retinotectal visual circuit (Fig. 3) – in stark contrast to the important role retinal waves play in the refinement of mammalian retinofugal pathways.

Surprising as this may be, it would not be entirely unprecedented to discover that zebrafish retinal waves are particularly important for the development of the retina, but not for retinofugal pathways. While it is reported by Zhang et al. [22] that spontaneous retinal waves in zebrafish always originate with bipolar cell activity, in mammals, studies in both ferret [55, 56] and mouse [57, 58] have shown that bipolar cells contribute only to the initiation of Stage III glutamatergic retinal waves. This cross-species difference in the intra-retinal dynamics of wave activity hints that retinal waves in zebrafish might play a different role with respect to the development of the retina itself. Further supporting this idea is the fact that, at postnatal day 2.5, when zebrafish retinal waves begin, RGC axons have only reached about half of the regions of the zebrafish optic tectum which will eventually be innervated [59]. This is consistent with the retinal waves playing a more important role for the development of the zebrafish retina itself, rather than retinofugal projections.

Conversely, mammalian retinal waves are known to play an important role in the refinement of retinofugal projections as well as intra-retinal and inter-retinal development. Mammalian retinal waves not only guide the formation of connections between the retina and the lateral geniculate nucleus [60, 61], but they also regulate the formation of gap junctions between RGCs within the retina [6]. Intriguingly, retinal waves additionally modulate inter-retinal activity in mammals, with monocular enucleation influencing the dynamics of retinal waves in the remaining eye [7]. These observations could be explained, at least in part, by a transient retino-retinal projection that may serve to synchronize retinal wave activity between the two eyes [10]. This emphasizes, once again, the complex character of retinal waves in mammalian visual development, which does not seem to be present in zebrafish. If X. laevis retinal waves occur entirely prior to stage 39, then they likely play a role in the development of the retina, entirely before the formation of retinofugal projections. This would be consistent with the limited time scale of retinal waves in zebrafish which experience retinal waves only at 2.5–3.5 days post-fertilization.

An alternative, though not exclusive, interpretation is that retinal waves occur primarily outside of the RGC layer in the developing X. laevis retina. Correlated activity was not found when recording from the RGC layer of isolated Xenopus retinae [21], but as stated earlier, retinal waves in zebrafish were found to always originate with bipolar cell activity in the inner plexiform layer (IPL) and then propagate to the somata of only some RGCs [22]. The X. laevis IPL forms at stage 35 and refines through stage 42 (Fig. 3a) [62]. If X. laevis retinal waves similarly initiate in the IPL, they might not reliably propagate to the RGC layer, as is the case in zebrafish. This would explain why Demas and colleagues [21] could not detect retinal waves in the RGC layer. This would once again point to a more important role for retinal waves in the formation of the retina itself; if these waves do not reliably propagate to RGC somata, they would be unlikely to play much of a role in the formation and refinement of retinofugal projections and likely play a more important role in the formation of the retinae themselves. Thus, while previous work has demonstrated that X. laevis retinofugal pathways are unlikely to be guided by retinal wave activity, it is highly possible that X. laevis tadpoles experience a more limited form of retinal wave activity, like those observed in zebrafish.

The Origin of Retinal Waves

Before their observation in zebrafish [22], it was thought that spontaneous retinal waves were evolved by amniotes to compensate for the visual deprivation that occurs in ovo or in utero. However, it is time to revisit this theory. It seems unlikely that zebrafish would develop retinal waves where no other anamniotes have. Instead, we propose the possibility of a relatively simple ancestral retinal wave, which has been expanded upon by amniotes who leverage the relatively safe environment in ovo or in utero. Enjoying the safety of a womb or egg, amniotes tend to undergo a more protracted development of sensorimotor systems, utilizing greater experience-dependent plasticity in the early development of sensory systems.

Unlike amniotes, anamniotes – including zebrafish – develop in an external environment and often require a functioning visual system that develops quickly and reliably. The environment necessitates the rapid development of stereotypical, visually guided behaviours, precluding much of the experience-dependent plasticity and protracted development seen in amniotic sensory system development.

The key to understanding the evolutionary origin of retinal waves lies in the environment. In turtles, spontaneous retinal waves play a part in the development of a functional visual system which can aid in sea-finding (i.e., the process of crawling from the egg to the water). Sea-finding is a complex, visually guided process which relies upon multiple kinds of visual information. Turtle hatchlings are known to integrate the brightness of visual inputs from multiple angles, in order to determine the brightest direction [63‒65]; to prefer some spectral wavelengths over others, irrespective of light intensity [66]; and to use shape/form perception to orient away from high silhouettes [63, 67, 68]. (The intricacies of turtle hatchling sea-finding are reviewed excellently by Lohmann et al. [69].) It is highly possible that the spontaneous retinal waves which occur during sea turtle embryogenesis are a necessary step for the development of a retina which can make comparisons on the basis of luminance, spectral intensity, and form vision to guide hatchling turtles to the sea. In other words, the turtle’s ecological niche likely demands retinal waves as a means to form the visual system which can subserve visually guided decisions from shortly after hatching.

With the discovery of spontaneous retinal waves in zebrafish, retinal waves are now known to exist in both amniotes and anamniotes, yet these waves seem to differ markedly between the two groups. A definitive reason behind this difference remains to be established, but the most promising explanation to date comes from considering the environment and the behaviours required of many anamniotes early in life. We propose that the protection offered by wombs and eggs has enabled amniotes to build upon the retinal waves experienced by many anamniotes. Put another way, in anamniotes, retinal waves are important for the development of the retinae, but in amniotes, retinal waves take on an additional significance for the wiring of retinofugal pathways.

The finding of retinal waves in zebrafish opens a new and exciting chapter in the study of how the visual system has evolved across the vertebrate lineage. Since adult animals do not transform into other adult animals, evolution is the evolution of development [4, 5]. It is by considering the environment of a developing fish, amphibian, reptile, bird, or mammal, and the most adaptive behavioural repertoires for these respective environments, that we arrive at a novel theory of retinal wave evolution. Yet this is far from the first time that behaviour has driven advances in neurophysiology.

The Roads Paved by Behaviour

Behaviour has long paved the roads for discoveries in neurophysiology. As one brief example of this long-standing pattern, we consider Hodgkin and Huxley’s Nobel Prize-winning work on the initiation and propagation of action potentials in the giant squid axon [70‒72]. Decades prior to the work of Hodgkin and Huxley, it was John Zachary Young [73, 74] who described how the squid’s giant motor nerve fibres enable muscular contractions that expel powerful jets of water from the squid’s body. The incredible speed of this elegant escape mechanism is enabled by the enormous size of the giant squid axon; greater axon diameter is associated with higher conduction velocity, as demonstrated early on by Hursh [75] and others. The squid’s ecological surroundings drove the evolution of a speedy escape mechanism for the squid, by selecting for those squids which could most competently escape predation. By acting on the level of behavioural fitness, evolutionary processes selected for the giant axon that subserves this unique behaviour.

The giant squid axon enabled Hodgkin and Huxley to dissect the physiological basis of the action potential in the 1950s, but this outstanding discovery was enabled only by the earlier work of others, like Young, who described the physiological basis for the squid’s naturalistic behaviours. Fortuitous as it was that Young had done this work decades prior, this is a repeating pattern in the history of physiology: behavioural understanding often precedes physiological understanding.

Despite the important role that behavioural understanding plays in the discovery of physiological mechanisms, it would – in general – be impractical for a neurophysiologist to take on the additional roles of comparative physiologist, zoologist, ecologist, and animal behaviourist. So what might be done?

The Future of Behaviour

To the great fortune of physiologists, plenty of experimental work and theory has been done already in the study of naturalistic animal behaviours. Much has already been described across a wide-ranging variety of species, and this work needs only to be applied by physiologists.

Indeed, in the 1920s, it was the physiologist August Krogh who joked about the animals that were seemingly created to answer specific questions in physiology, just as the giant squid axon was “created” to study the action potential in the 1950s. Krogh writes [76]:

I have no doubt that there is quite a number of animals which are… “created” for special physiological purposes, but I am afraid that most of them are unknown to the men for whom they were “created,” and we must apply to the zoölogists to find them and lay our hands on them.

Bearing in mind the century-old wisdom of Krogh, neurophysiologists should once more begin to appreciate the incredible diversity of species that inhabit this world. To this end, we have considered the influence of ecological niche on visual system development. Specifically, we examined the emergence of retinal waves within the vertebrate lineage from a behavioural point of view. Using the specialization of animals to their environments as our guide, we have developed a novel theory of the emergence of retinal waves among vertebrates. From this theory, we predicted the presence of retinal waves across a larger swathe of the vertebrate lineage than previously appreciated, and we made predictions regarding the function of retinal waves in the context of broader patterns of visual development.

Most immediately, future work should investigate the possibility of retinal waves prior to stage 39 and in the IPL of X. laevis. However, our work predicts the presence of retinal waves in other anamniotes as well. By considering not only the environments of developing animals but also the behaviours and adaptations required by these environments, neuroethological considerations have the power to bring forward new explanations and identify unappreciated patterns in neurophysiology.

The authors would like to thank the attendees of the 2nd Meeting on Spontaneous Activity in Brain Development for their invaluable feedback. We are indebted to Mr. Erick Fernandez de Arteaga for producing Figure 2 and Mr. Caio César Santos Santana for helpful revisions to Figure 1. The authors would additionally like to thank Mr. John R. Bruno, Mr. Uwemedimo G. Udoh, and members of the Cognitive Neuroecology Lab for valuable discussions.

The authors declare no conflict of interest.

This work is funded by a Wellcome Trust Doctoral Studentship in Neuroscience (222368/Z/21/Z) to J.E.H. The funder had no role in the design or writing of this report, nor in the collection, analysis, and interpretation of data.

J.E.H. drafted the manuscript and designed the figures, except where noted in the Acknowledgments section. K.G.P. and Z.M. assisted in the editing of the manuscript and the design of the figures.

Additional Information

The authors’ ORCID numbers are as follows: J.E.H.: 0000-0002-6547-0463. K.G.P.: 0000-0002-6743-4757. Z.M.: 0000-0002-6852-6004.

1.
Galvani
L
.
De viribus electricitatis in motu musculari
.
Commentarius De Bonoiensi Scientiarum et Artium Intituo atque Academie Commentarii
.
1791
;
7
:
363
418
. doi: .
2.
Krakauer
JW
,
Ghazanfar
AA
,
Gomez-Marin
A
,
MacIver
MA
,
Poeppel
D
.
Neuroscience needs behavior: correcting a reductionist bias
.
Neuron
.
2017
;
93
(
3
):
480
90
. doi: .
3.
Read
JCA
.
Stereopsis without correspondence
.
Philos Trans R Soc Lond B Biol Sci
.
2023
;
378
(
1869
):
20210449
. doi: .
4.
García-Moreno
F
,
Molnár
Z
.
Chapter 25: the impact of different modes of neuronal migration on brain evolution
. In:
Rubenstein
J
,
Rakic
P
,
Chen
B
,
Kwan
KY
,
Kolodkin
A
,
Anton
E
, editors.
Cellular migration and formation of axons and dendrites
. 2nd Ed.
Academic Press
;
2020
. p.
555
76
. Available from: https://www.sciencedirect.com/science/article/pii/B9780128144077000250.
5.
Horváth
TL
,
Hirsch
J
,
Molnár
Z
.
Chapter 1: Zoltán Molnár: the developing brain
. In:
Horváth
TL
,
Hirsch
J
,
Molnár
Z
, editors.
Body, brain, behavior
.
Academic Press
;
2022
. p.
1
66
. Available from: https://www.sciencedirect.com/science/article/pii/B9780128180938000057.
6.
Arroyo
DA
,
Kirkby
LA
,
Feller
MB
.
Retinal waves modulate an intraretinal circuit of intrinsically photosensitive retinal ganglion cells
.
J Neurosci
.
2016
;
36
(
26
):
6892
905
. doi: .
7.
Failor
SW
,
Ng
A
,
Cheng
HJ
.
Monocular enucleation alters retinal waves in the surviving eye
.
Neural Dev
.
2018
;
13
(
1
):
4
. doi: .
8.
Thompson
A
,
Gribizis
A
,
Chen
C
,
Crair
MC
.
Activity-dependent development of visual receptive fields
.
Curr Opin Neurobiol
.
2017
;
42
:
136
43
. doi: .
9.
Huberman
AD
,
Feller
MB
,
Chapman
B
.
Mechanisms underlying development of visual maps and receptive fields
.
Annu Rev Neurosci
.
2008
;
31
(
1
):
479
509
. doi: .
10.
Murcia-Belmonte
V
,
Coca
Y
,
Vegar
C
,
Negueruela
S
,
de Juan Romero
C
,
Valiño
AJ
, et al
.
A retino-retinal projection guided by Unc5c emerged in species with retinal waves
.
Curr Biol
.
2019
;
29
(
7
):
1149
60.e4
. doi: .
11.
Sernagor
E
,
Grzywacz
NM
.
Spontaneous activity in developing turtle retinal ganglion cells: pharmacological studies
.
J Neurosci
.
1999
;
19
(
10
):
3874
87
. doi: .
12.
Sernagor
E
,
Mehta
V
.
The role of early neural activity in the maturation of turtle retinal function
.
J Anat
.
2001
;
199
(
Pt 4
):
375
83
. doi: .
13.
Sernagor
E
,
Young
C
,
Eglen
SJ
.
Developmental modulation of retinal wave dynamics: shedding light on the GABA saga
.
J Neurosci
.
2003
;
23
(
20
):
7621
9
. doi: .
14.
Warland
DK
,
Huberman
AD
,
Chalupa
LM
.
Dynamics of spontaneous activity in the fetal macaque retina during development of retinogeniculate pathways
.
J Neurosci
.
2006
;
26
(
19
):
5190
7
. doi: .
15.
Zhou
ZJ
.
Direct participation of starburst amacrine cells in spontaneous rhythmic activities in the developing mammalian retina
.
J Neurosci
.
1998
;
18
(
11
):
4155
65
. doi: .
16.
Mooney
R
,
Penn
AA
,
Gallego
R
,
Shatz
CJ
.
Thalamic relay of spontaneous retinal activity prior to vision
.
Neuron
.
1996
;
17
(
5
):
863
74
. doi: .
17.
Galli
L
,
Maffei
L
.
Spontaneous impulse activity of rat retinal ganglion cells in prenatal life
.
Science
.
1988
;
242
(
4875
):
90
1
. doi: .
18.
Meister
M
,
Wong
R
,
Baylor
D
,
Shatz
C
.
Synchronous bursts of action potentials in ganglion cells of the developing mammalian retina
.
Science
.
1991
;
252
(
5008
):
939
43
. doi: .
19.
Catsicas
M
,
Bonness
V
,
Becker
D
,
Mobbs
P
.
Spontaneous Ca2+ transients and their transmission in the developing chick retina
.
Curr Biol
.
1998
;
8
(
5
):
283
6
. doi: .
20.
Sernagor
E
,
Grzywacz
NM
.
Emergence of complex receptive field properties of ganglion cells in the developing turtle retina
.
J Neurophysiol
.
1995
;
73
(
4
):
1355
64
. doi: .
21.
Demas
JA
,
Payne
H
,
Cline
HT
.
Vision drives correlated activity without patterned spontaneous activity in developing Xenopus retina
.
Dev Neurobiol
.
2012
;
72
(
4
):
537
46
. doi: .
22.
Zhang
R
,
Li
XQ
,
Kawakami
K
,
Du
JL
,
Dulin
J
.
Stereotyped initiation of retinal waves by bipolar cells via presynaptic NMDA autoreceptors
.
Nat Commun
.
2016
;
7
(
1
):
12650
. doi: .
23.
Akin
O
,
Bajar
BT
,
Keles
MF
,
Frye
MA
,
Zipursky
SL
.
Cell-type-Specific patterned stimulus-independent neuronal activity in the Drosophila visual system during synapse formation
.
Neuron
.
2019
;
101
(
5
):
894
904.e5
. doi: .
24.
Hiesinger
PR
,
Zhai
RG
,
Zhou
Y
,
Koh
TW
,
Mehta
SQ
,
Schulze
KL
, et al
.
Activity-independent prespecification of synaptic partners in the visual map of Drosophila
.
Curr Biol
.
2006
;
16
(
18
):
1835
43
. doi: .
25.
Choi
BJ
,
Chen
YCD
,
Desplan
C
.
Building a circuit through correlated spontaneous neuronal activity in the developing vertebrate and invertebrate visual systems
.
Genes Dev
.
2021
;
35
(
9–10
):
677
91
. doi: .
26.
Ackman
JB
,
Crair
MC
.
Role of emergent neural activity in visual map development
.
Curr Opin Neurobiol
.
2014
;
24
:
166
75
. doi: .
27.
Pratt
KG
,
Hiramoto
M
,
Cline
HT
.
An evolutionarily conserved mechanism for activity-dependent visual circuit development
.
Front Neural Circuits
.
2016
;
10
:
79
. doi: .
28.
Kutsarova
E
,
Munz
M
,
Ruthazer
ES
.
Rules for shaping neural connections in the developing brain
.
Front Neural Circuits
.
2017
:
10
. doi: .
29.
Zhang
RW
,
Du
WJ
,
Prober
DA
,
Du
JL
.
Müller glial cells participate in retinal waves via glutamate transporters and AMPA receptors
.
Cell Rep
.
2019
;
27
(
10
):
2871
80.e2
. doi: .
30.
Stafford
BK
,
Sher
A
,
Litke
AM
,
Feldheim
DA
.
Spatial-temporal patterns of retinal waves underlying activity-dependent refinement of retinofugal projections
.
Neuron
.
2009
;
64
(
2
):
200
12
. doi: .
31.
Elstrott
J
,
Feller
MB
.
Direction-selective ganglion cells show symmetric participation in retinal waves during development
.
J Neurosci
.
2010
;
30
(
33
):
11197
201
. doi: .
32.
Ackman
JB
,
Burbridge
TJ
,
Crair
MC
.
Retinal waves coordinate patterned activity throughout the developing visual system
.
Nature
.
2012
;
490
(
7419
):
219
25
. doi: .
33.
Hiramoto
M
,
Cline
HT
.
Optic flow instructs retinotopic map formation through a spatial to temporal to spatial transformation of visual information
.
Proc Natl Acad Sci U S A
.
2014
;
111
(
47
):
E5105
13
. doi: .
34.
Ge
X
,
Zhang
K
,
Gribizis
A
,
Hamodi
AS
,
Sabino
AM
,
Crair
MC
.
Retinal waves prime visual motion detection by simulating future optic flow
.
Science
.
2021
;
373
(
6553
):
eabd0830
. doi: .
35.
Gnuegge
L
,
Schmid
S
,
Neuhauss
SCF
.
Analysis of the activity-deprived zebrafish mutant macho reveals an essential requirement of neuronal activity for the development of a fine-grained visuotopic map
.
J Neurosci
.
2001
;
21
(
10
):
3542
8
. doi: .
36.
Saszik
S
,
Bilotta
J
.
Constant dark-rearing effects on visual adaptation of the zebrafish ERG
.
Int J Dev Neurosci
.
2001
;
19
(
7
):
611
9
. doi: .
37.
Dixon
LJ
,
McDowell
AL
,
Houchins
JD
,
Bilotta
J
.
Effects of restricted spectral rearing on the development of zebrafish retinal physiology
.
Doc Ophthalmol
.
2004
;
109
(
1
):
17
33
. doi: .
38.
Niell
CM
,
Smith
SJ
.
Functional imaging reveals rapid development of visual response properties in the zebrafish tectum
.
Neuron
.
2005
;
45
(
6
):
941
51
. doi: .
39.
Nevin
LM
,
Taylor
MR
,
Baier
H
.
Hardwiring of fine synaptic layers in the zebrafish visual pathway
.
Neural Dev
.
2008
;
3
(
1
):
36
. doi: .
40.
Xie
J
,
Jusuf
PR
,
Bui
BV
,
Goodbourn
PT
.
Experience-dependent development of visual sensitivity in larval zebrafish
.
Sci Rep
.
2019
;
9
(
1
):
18931
. doi: .
41.
Avitan
L
,
Pujic
Z
,
Mölter
J
,
Van De Poll
M
,
Sun
B
,
Teng
H
, et al
.
Spontaneous activity in the zebrafish tectum reorganizes over development and is influenced by visual experience
.
Curr Biol
.
2017
;
27
(
16
):
2407
19.e4
. doi: .
42.
Feller
MB
,
Kerschensteiner
D
.
Retinal waves and their role in visual system development
.
Synapse development and maturation
.
Elsevier
;
2020
. p.
367
82
. Available from: https://linkinghub.elsevier.com/retrieve/pii/B9780128236727000168.
43.
Maccione
A
,
Hennig
MH
,
Gandolfo
M
,
Muthmann
O
,
van Coppenhagen
J
,
Eglen
SJ
, et al
.
Following the ontogeny of retinal waves: pan-retinal recordings of population dynamics in the neonatal mouse
.
J Physiol
.
2014
;
592
(
7
):
1545
63
. doi: .
44.
Kerschensteiner
D
.
Glutamatergic retinal waves
.
Front Neural Circuits
.
2016
;
10
:
38
. doi: .
45.
Rochefort
NL
,
Narushima
M
,
Grienberger
C
,
Marandi
N
,
Hill
DN
,
Konnerth
A
.
Development of direction selectivity in mouse cortical neurons
.
Neuron
.
2011
;
71
(
3
):
425
32
. doi: .
46.
Shen
J
,
Colonnese
MT
.
Development of activity in the mouse visual cortex
.
J Neurosci
.
2016
;
36
(
48
):
12259
75
. doi: .
47.
Mukherjee
D
,
Yonk
AJ
,
Sokoloff
G
,
Blumberg
MS
.
Wakefulness suppresses retinal wave-related neural activity in visual cortex
.
J Neurophysiol
.
2017
;
118
(
2
):
1190
7
. doi: .
48.
Tiriac
A
,
Smith
BE
,
Feller
MB
.
Light prior to eye opening promotes retinal waves and eye-specific segregation
.
Neuron
.
2018
;
100
(
5
):
1059
65.e4
. doi: .
49.
Hunt
JE
,
Bruno
JR
,
Pratt
KG
.
An innate color preference displayed by Xenopus tadpoles is persistent and requires the tegmentum
.
Front Behav Neurosci
.
2020
;
14
:
71
. doi: .
50.
Bruno
JR
,
Udoh
UG
,
Landen
JG
,
Osborn
PO
,
Asher
CJ
,
Hunt
JE
, et al
.
A circadian-dependent preference for light displayed by Xenopus tadpoles is modulated by serotonin
.
iScience
.
2022
;
25
(
11
):
105375
. doi: .
51.
Bertolesi
GE
,
Debnath
N
,
Atkinson-Leadbeater
K
,
Niedzwiecka
A
,
McFarlane
S
.
Distinct type II opsins in the eye decode light properties for background adaptation and behavioural background preference
.
Mol Ecol
.
2021
;
30
(
24
):
6659
76
. doi: .
52.
Moriya
T
,
Kito
K
,
Miyashita
Y
,
Asami
K
.
Preference for background color of the Xenopus laevis tadpole
.
J Exp Zool
.
1996
;
276
(
5
):
335
44
. doi: .
53.
Zahn
N
,
James-Zorn
C
,
Ponferrada
VG
,
Adams
DS
,
Grzymkowski
J
,
Buchholz
DR
, et al
.
Normal Table of Xenopus development: a new graphical resource
.
Development
.
2022
;
149
(
14
):
dev200356
. doi: .
54.
Holt
CE
,
Harris
WA
.
Order in the initial retinotectal map in Xenopus : a new technique for labelling growing nerve fibres
.
Nature
.
1983
;
301
(
5896
):
150
2
. doi: .
55.
Feller
MB
,
Butts
DA
,
Aaron
HL
,
Rokhsar
DS
,
Shatz
CJ
.
Dynamic processes shape spatiotemporal properties of retinal waves
.
Neuron
.
1997
;
19
(
2
):
293
306
. doi: .
56.
Wong
ROL
,
Chernjavsky
A
,
Smith
SJ
,
Shatz
CJ
.
Early functional neural networks in the developing retina
.
Nature
.
1995
;
374
(
6524
):
716
8
. doi: .
57.
Blankenship
AG
,
Ford
KJ
,
Johnson
J
,
Seal
RP
,
Edwards
RH
,
Copenhagen
DR
, et al
.
Synaptic and extrasynaptic factors governing glutamatergic retinal waves
.
Neuron
.
2009
;
62
(
2
):
230
41
. doi: .
58.
Tian
N
,
Copenhagen
DR
.
Visual stimulation is required for refinement of on and off pathways in postnatal retina
.
Neuron
.
2003
;
39
(
1
):
85
96
. doi: .
59.
Burrill
JD
,
Easter
SS
Jr
.
Development of the retinofugal projections in the embryonic and larval zebrafish (Brachydanio rerio)
.
J Comp Neurol
.
1994
;
346
(
4
):
583
600
. doi: .
60.
Butts
DA
,
Kanold
PO
,
Shatz
CJ
.
A burst-based “Hebbian” learning rule at retinogeniculate synapses links retinal waves to activity-dependent refinement
.
PLoS Biol
.
2007
;
5
(
3
):
e61
. .
61.
Lee
H
,
Brott
BK
,
Kirkby
LA
,
Adelson
JD
,
Cheng
S
,
Feller
MB
, et al
.
Synapse elimination and learning rules co-regulated by MHC class I H2-Db
.
Nature
.
2014
;
509
(
7499
):
195
200
. doi: .
62.
Grant
P
,
Rubin
E
,
Cima
C
.
Ontogeny of the retina and optic nerve in Xenopus laevis. I. Stages in the early development of the retina
.
J Comp Neurol
.
1980
;
189
(
4
):
593
613
. doi: .
63.
Salmon
M
,
Wyneken
J
,
Fritz
E
,
Lucas
M
.
Seafinding by hatchling sea turtles: role of brightness, silhouette and beach slope as orientation cues
.
Behav
.
1992
;
122
(
1–2
):
56
77
. doi: .
64.
Verheijen
FJ
,
Wildschut
JT
.
The photic orientation of hatchling sea turtles during water finding behaviour
.
Neth J Sea Res
.
1973
;
7
:
53
67
. doi: .
65.
Witherington
BE
.
Sea-finding behavior and the use of photic orientation cues by hatchling sea turtles [Ph.D. Dissertation]
.
United States, Florida
:
University of Florida
;
1992
. Available from: https://www.proquest.com/docview/303980507/abstract/F08706079D3746D4PQ/1.
66.
Sella
KN
,
Salmon
M
,
Witherington
BE
.
Filtered streetlights attract hatchling marine turtles
.
Chelonian Conserv Biol
.
2006
;
5
(
2
):
255
61
. doi: .
67.
Limpus
C
,
Kamrowski
RL
.
Ocean-finding in marine turtles: the importance of low horizon elevation as an orientation cue
.
Behav
.
2013
;
150
(
8
):
863
93
. doi: .
68.
Tuxbury
SM
,
Salmon
M
.
Competitive interactions between artificial lighting and natural cues during seafinding by hatchling marine turtles
.
Biol Conservation
.
2005
;
121
(
2
):
311
6
. doi: .
69.
Lohmann
KJ
,
Witherington
BE
,
Lohmann
CMF
,
Salmon
M.
.
Orientation, navigation, and natal beach homing in sea turtles
.
The biology of sea turtles
.
CRC Press
;
1996
.
Vol. I
.
70.
Hodgkin
AL
,
Huxley
AF
,
Katz
B
.
Measurement of current-voltage relations in the membrane of the giant axon of Loligo
.
J Physiol
.
1952
;
116
(
4
):
424
48
. doi: .
71.
Hodgkin
AL
,
Huxley
AF
.
Currents carried by sodium and potassium ions through the membrane of the giant axon of Loligo
.
J Physiol
.
1952
;
116
(
4
):
449
72
. doi: .
72.
Hodgkin
AL
,
Huxley
AF
.
The components of membrane conductance in the giant axon of Loligo
.
J Physiol
.
1952
;
116
(
4
):
473
96
. doi: .
73.
Young
JZ
.
The functioning of the giant nerve fibres of the squid
.
J Exp Biol
.
1938
;
15
(
2
):
170
85
. doi: .
74.
Young
JZ
.
Fused neurons and synaptic contacts in the giant nerve fibres of cephalopods
.
Phil Trans R Soc Lond B
.
1939
;
229
(
564
):
465
503
. .
75.
Hursh
JB
.
Conduction velocity and diameter of nerve fibers
.
Am J Physiology-Legacy Content
.
1939
;
127
(
1
):
131
9
. doi: .
76.
Krogh
A
.
The progress of physiology
.
Am J Physiol
.
1929
;
90
(
2
):
243
51
. doi: .