Barbara L. Finlay

Analyses of brain sizes across mammalian taxonomic groups reveal a consistent pattern of covariation between major brain components, including a robust inverse relationship between the limbic system and the neocortex. To find the functional basis of this relationship, we mapped the multidimensional representations of task-optimized artificial neural networks onto two-dimensional surfaces resembling the forebrain cortices. We found that networks optimized for visual, somatosensory, and auditory representations develop ordered spatiotopic maps where units draw information from localized regions of the sensory input. In contrast, networks optimized for olfactory and relational memory representations develop fractured maps with distributed patterns of information convergence. Evolutionary optimization of multimodal networks for varying task objectives results in inverse covariation between spatiotopic and disordered network components that compete for the representational space. These results suggest that the observed pattern of covariation between brain components reflects an essential computational duality in brain evolution.

The water-to-land transition in vertebrate evolution offers an unusual opportunity to consider computational affordances of a new ecology for the brain. All sensory modalities are changed, particularly a greatly enlarged visual sensorium owing to air versus water as a medium, and expanded by mobile eyes and neck. The multiplication of limbs, as evolved to exploit aspects of life on land, is a comparable computational challenge. As the total mass of living organisms on land is a hundredfold larger than the mass underwater, computational improvements promise great rewards. In water, the midbrain tectum coordinates approach/avoid decisions, contextualized by water flow and by the animal’s body state and learning. On land, the relative motions of sensory surfaces and effectors must be resolved, adding on computational architectures from the dorsal pallium, such as the parietal cortex. For the large-brained and long-living denizens of land, making the right decision when the wrong one means death may be the basis of planning, which allows animals to learn from hypothetical experience before enactment. Integration of value-weighted, memorized panoramas in basal ganglia/frontal cortex circuitry, with allocentric cognitive maps of the hippocampus and its associated cortices becomes a cognitive habit-to-plan transition as substantial as the change in ecology. This article is part of the theme issue ‘Systems neuroscience through the lens of evolutionary theory’.

Significance: There exists a curious lacuna in our understanding of the development and evolution of cortical areas in the brain. Whereas massive research effort has focused on the development of primary sensory and motor areas, few such studies of higher cortical areas have been done, particularly at the earliest stages of development. In parallel, there exists intense interest in the interpretation of higher cortical areas across species, in how new areas of larger brains emerge. Using a network model of the developing cortex, we show here that new cortical areas may be the natural outcome of regularly scaling brains, arising as necessary products of an activity-driven developmental program conserved in the evolution of the cortex.

Neurodevelopmental duration plays a central role in the evolution of the retina and neocortex in mammals. In the diurnal primate eye and retina, it is necessary to scale the number of cones versus the number of rods with different exponents to defend their respective functions of spatial acuity and sensitivity in eyes of different sizes. The order of photoreceptor precursor specification, cones specified first, rods second, couples their respective cell numbers at maturity to the kinetics of embryonic stem cell proliferation. Different durations of retinogenesis change the ratio of rods to cones produced so as to defend both functions over a range of eye diameters. In the evolution of nocturnality, the same coupling of photoreceptor specification to neurogenesis is altered to fewer cones and many more rods in nocturnal eyes, by delaying the onset of retinogenesis. Similarly, the neocortex also shows coupling of the specification of laminar position with duration of neurogenesis. Overall, duration of neurogenesis directly predicts neocortex volume in most mammalian clades. In larger brains with longer neocortical neurogenesis, its organization changes progressively, differentiating the frontal pole from the occipital pole in volume of connectivity and number of neurons per unit column. This permits greater, hierarchically organized information abstraction with increasing neocortex volume. Exceptions do exist, however, in species of three separate taxa, marsupials, naked mole rats, and bats, which break the correlation of neurodevelopmental duration and brain size. Naked mole rats and bats both have small brains and unusual longevity, coupled with neurodevelopmental periods characteristic of much bigger‐brained animals, raising the possibility that developmental duration and lifespan have some genetic or mechanistic control in common. The role of duration of development in mediating between the mechanistic levels of construction of retinal and cortical organization, and the different life histories associated with larger brains, such as duration of parental care, learning and overall longevity are discussed.

Research in the neuroscience of pain perception and visual perception has taken contrasting paths. The contextual and the social aspects of pain judgements predisposed pain researchers to develop computational and functional accounts early, while vision researchers tended to simple localizationist or descriptive approaches first. Evolutionary thought was applied to distinct domains, such as game-theoretic approaches to cheater detection in pain research, versus vision scientists' studies of comparative visual ecologies. Both fields now contemplate current motor or decision-based accounts of perception, particularly predictive coding. Vision researchers do so without the benefit of earlier attention to social and motivational aspects of vision, while pain researchers lack a comparative behavioural ecology of pain, the normal incidence and utility of responses to tissue damage. Hybrid hypotheses arising from predictive coding as used in both domains are applied to some perplexing phenomena in pain perception to suggest future directions. The contingent and predictive interpretation of complex sensations, in such domains as ‘runner's high’, multiple cosmetic procedures, self-harm and circadian rhythms in pain sensitivity is one example. The second, in an evolutionary time frame, considers enhancement of primary perception and expression of pain in social species, when expressions of pain might reliably elicit useful help. This article is part of the Theo Murphy meeting issue ‘Evolution of mechanisms and behaviour important for pain’.

The question of how complex human abilities evolved, such as language or face recognition, has been pursued by means of multiple strategies. Highly specialized non-human species have been examined analytically for formal similarities, close phylogenetic relatives have been examined for continuity, and simpler species have been analyzed for the broadest view of functional organization. All these strategies require empirical evidence of what is variable and predictable in both the modeled and the model species. Turning to humans, allometric analyses of the evolution of brain mass and brain components often return the interesting, but disappointing answer that volumetric organization of the human brain is highly predictable seen in its phylogenetic context. Reconciling this insight with unique human behavior, or any species-typical behavior, represents a serious challenge. Allometric analyses of the order and duration of mammalian neural development show that, while basic neural development in humans is allometrically predictable, conforming to adult neural architecture, some life history features deviate, notably that weaning is unusually early. Finally, unusual deviations in the retina and central auditory system in the laboratory mouse, which is widely assumed to be “generic,” as well as severe deviations from expected brain allometry in some mouse strains, underline the need for a deeper understanding of phylogenetic variability even in those systems believed to be best understood.

The widely held belief that the human cortex is exceptionally large for our brain size is wrong, resulting from basic errors in how best to compare evolving brains. This misapprehension arises from the comparison of only a few laboratory species, failure to appreciate differences in brain scaling in rodents versus primates, but most important, the false assumption that linear extrapolation can be used to predict changes from small to large brains. Belief in the exceptionalism of human cortex has propagated itself into genomic analysis of the cortex, where cortex has been studied as if it were an example of innovation rather than predictable scaling. Further, this belief has caused both neuroscientists and psychologists to prematurely assign functions distributed widely in the brain to the cortex, to fail to explore subcortical sources of brain evolution, and to neglect genuinely novel features of human infancy and childhood.

Comparison of neurodevelopmental sequences between species whose initial period of brain organization may vary from 100 days to 1,000 days, and whose progress is intrinsically non-linear presents large challenges in normalization. Comparing adult timelines when lifespans stretch from 1 year to 75 years, when underlying cellular mechanisms under scrutiny do not scale similarly, presents challenges to simple detection and comparison. The question of adult hippocampal neurogenesis has generated numerous controversies regarding its simple presence or absence in humans versus rodents, whether it is best described as the tail of a distribution centered on early neural development, or is several distinct processes. In addition, adult neurogenesis may have substantially changed in evolutionary time in different taxonomic groups. Here, we extend and adapt a model of the cross-species transformation of early neurodevelopmental events which presently reaches up to the equivalent of the third human postnatal year for 18 mammalian species (www.translatingtime.net) to address questions relevant to hippocampal neurogenesis, which permit extending the database to adolescence or perhaps to the whole lifespan. We acquired quantitative data delimiting the envelope of hippocampal neurogenesis from cell cycle markers (i.e., Ki67 and DCX) and RNA sequencing data for two primates (macaque and humans) and two rodents (rat and mouse). To improve species coverage in primates, we gathered the same data from marmosets (Callithrix jacchus), but additionally gathered data on a number of developmental milestones to find equivalent developmental time points between marmosets and other species. When all species are so modeled, and represented in a common time frame, the envelopes of hippocampal neurogenesis are essentially superimposable. Early developmental events involving the olfactory and limbic system start and conclude possibly slightly early in primates than rodents, and we find a comparable early conclusion of primate hippocampal neurogenesis (as assessed by the relative number of Ki67 cells) suggesting a plateau to low levels at approximately 2 years of age in humans. Marmosets show equivalent patterns within neurodevelopment, but unlike macaque and humans may have wholesale delay in the initiation of neurodevelopment processes previously observed in some precocial mammals such as the guinea pig and multiple large ungulates.