Modern Sciences is a premier science journal that bridges the gap between science and its application to society.
LATEST STORIES

Detailed shape characterization reveals complementary ecomorphological patterns in brain shape and size evolution across the avian radiation

Talia Lowi-Merri12* and Akinobu Watanabe23

    1Department of Organismic and Evolutionary Biology, Harvard University, Cambridge Massachusetts, USA

    2Department of Anatomy, New York Institute of Technology College of Osteopathic Medicine, Old Westbury New York, USA

    3Division of Paleontology, American Museum of Natural History, New York, New York, USA

    *Corresponding author Corresponding author Email: talialowimerri{at}gmail.com

      bioRxiv preprint DOI: https://doi.org/10.1101/2025.10.14.682392

      Posted: October 14, 2025, Version 1

      Copyright: This pre-print is available under a Creative Commons License (Attribution 4.0 International), CC BY 4.0, as described at http://creativecommons.org/licenses/by/4.0/

      ABSTRACT

      Birds have exceptionally large brains for their body sizes, which is thought to have facilitated their ability to adapt and survive following the end-Cretaceous mass extinction, and has been associated with many of their diverse ecological adaptations today. However, size metrics are limited in their ability to fully characterize neuroanatomical variation, since the brain and its regions can have divergent morphologies across taxa while maintaining the same volume. Here, we employ high-density geometric morphometric (GM) shape data to capture localized anatomical differences in avian brain endocasts and compare how evolutionary patterns differ between shape and size. We find that the majority of shape variation falls along an ‘elongated-globular’ spectrum, where terrestrial and aquatic birds generally have anteroposteriorly elongated brains while more aerial birds typically have more rounded, globular brains. Optic lobe shape contained the greatest amount of ecological signal, while rates of shape evolution in the cerebrum were the most uniform across the avian phylogeny. A supermajority (>90%) of endocranial shape variation is independent of evolutionary allometry, demonstrating that a geometric morphometric approach characterizes key aspects of brain anatomy that are missing from volumetric data. Broadly, our results suggest that neuroanatomical diversity in birds is not driven by the dominance of any particular factors, but rather through shape divergence in localized brain structures from the global allometric and ecomorphological patterns.

      INTRODUCTION

      The evolutionary transition from non-avian dinosaurs to crown birds (Neornithes; hereby ‘birds’) marks a major transformation in vertebrate evolution, giving rise to one of the most diverse and successful groups of terrestrial vertebrates following a drastic change in body plan (1–3). One such change was the evolution of a highly encephalized brain, which may have contributed to the survivorship of crown bird lineages across the end-Cretaceous mass extinction event, and the ability to diversify and adapt to different environments (4–6). These exceptionally large brains, especially in crows and parrots (7), are often accompanied by unique cognitive abilities such as innovativeness and tool-use (8–12), behavioural flexibility to environmental variation (4, 13–15), greater ‘social intelligence’ characterized by strong social bonds (16–18), and song-learning shaped by neurons that resemble those of language-learning in humans (19, 20). As such, the avian brain provides a compelling comparative system to mammals for the evolution of encephalized brains capable of higher cognitive functions.

      In previous macroevolutionary studies, adaptive selection has been explored as a driving factor of neuroanatomical diversity in birds using brain size metrics (4, 5, 15, 21). For example, birds developed proportionally large brains in part as a result of body size reduction occurring more rapidly than brain size reduction along the dinosaur-bird lineage, which presumably relates to reducing body mass for flight (5). Additionally, ecological traits that reflect sensory abilities, such as vision and spatial orientation, are strongly correlated with both global and regional brain size metrics (22–25). Ecological and behavioral correlates of neuroanatomy often follow the ‘principle of proper mass’ (26), which in an evolutionary perspective, implies that relatively larger brain structures reflect greater demand and selection for the functions associated with those regions. This holds true in several aspects of the avian brain: an enlarged Wulst is correlated with binocular depth perception (27), and an enlarged nucleus connecting the associative pallium with the cerebellum is correlated with complex foraging behaviours (28).

      While size measurements are informative, they are naturally limited in characterization of brain morphology. For instance, brains with equivalent relative and absolute sizes could diverge in brain morphologies, especially in macroevolutionary studies that encompass large interspecific variation. In addition, functional ability is often associated with structures across multiple regions, or manifests as subtle, localized differences within a region. One aspect of neuroanatomical variation that is not captured by size is the spectrum from elongated to rounded, or ‘globular’, brain shapes (29–33). Globularization is characterized by a restructuring of brain composition, such as the centralization of the thalamus, sometimes explained through the ‘spatial packing’ hypothesis (32, 33), in which large brains are accommodated within the cranial cavity by becoming more rounded in shape and through flexion of the cranial base (34). While spatial packing is generally associated with large brains, the degree to which globularization is correlated with brain or body size, and whether additional ecological or evolutionary variables are associated with this axis of variation, is also unclear (31, 32). While extensive 3D characterization and ecomorphological analyses have proceeded in avian skeletal systems (35–37), investigations into brain and endocranial shape variation have been limited to either relatively simple shape characterization with few landmarks (35, 38, 39) or have been applied to a sparse taxonomic sampling given the speciose avian diversity (25, 40–42). This leaves a major gap in our understanding of the ecological, evolutionary, and anatomical correlates of avian brain shape.

      Applying a series of phylogenetically-informed methods on 300 crown bird species and six non-avialan dinosaurs, we employ a high-density geometric morphometric (GM) approach to examine the macroevolutionary and ecomorphological patterns in avian brain shape, using digital reconstructions of the brain cavity within the cranium as a proxy for brain anatomy (hereby referred to as ‘endocasts’), which is highly correlated with brain morphology in birds (40, 41, 43, 44). Overall shape is compared with a wide breadth of ecological variables, including diet, foraging behaviors, locomotor lifestyle, and developmental mode (altriciality and precociality), to explore how ecological factors are associated with brain shape. To investigate how both brain size and shape provide complementary insights, we conducted analyses on both modalities of neuroanatomical characterization. With these data, we examine 1) major axes of brain shape variation in extant birds; 2) trends in rates of brain shape evolution across the avian phylogeny; 3) ecomorphological signal in brain morphology; 4) potential link between brain morphology and developmental mode; and 5) unifying insights into avian brain evolution revealed through combined analysis of brain shape and size.

      Institutional abbreviations

      FMNH, Field Museum of Natural History, Chicago, IL, USA; IGM, Institute of Geology, Mongolian Academy of Sciences, Ulaanbaatar, Mongolia; IVPP, Institute of Vertebrate Paleontology and Paleoanthropology, Beijing, China; NHMUK PV OR, Natural History Museum, Palaeontology Vertebrates, Ornithology, London, UK.

      RESULTS

      Morphometric variation

      To visualize major aspects of variation in endocranial shape, we produced a phylomorphospace through principal component analysis (PCA) on GM data (Fig. 1). In the dataset comprising extant birds and non-avian theropod dinosaurs, the first two principal components (PCs) accounted for 55.9% of the cumulative shape variance. Subsequent axes each accounted for < 10% of the shape variance and thus will not be discussed. PC1 (30.9% of variation with fossils; 30.1% without fossils) clearly separates the non-avian dinosaurs from crown birds, as shown in previous studies (40, 42, 45), accounting for the relative size and anterodorsal position of the cerebrum to the other brain regions. Positive PC1 values are associated with a proportionately larger, dorsally positioned cerebrum and an anteroposteriorly short cerebellum. Taxa with the most extreme positive PC1 values include passerine birds such as corvids (dietary generalists), furnariids Scytalopus and Campyloramphus (insectivores), as well as the soaring parrot Nestor (herbivore) and the kiwi Apteryx (insectivore). Negative PC1 values are associated with a relatively smaller, rounder, and anteriorly positioned cerebrum, an anteroposteriorly longer cerebellum and a dorsoventrally longer optic lobe. Extreme negative PC1 values are occupied by the fossil taxa, aquatic pursuit hunters like the Sulidae (anhingas and cormorants), Galliformes (landfowl), and Caprimulgiformes (nightjars). PC2 (25.04% of variation with fossils; 25.3% without fossils) is characterized by degree of dorsoventral flexion of the brain, especially the cephalic flexion between the cerebrum and midbrain. Positive PC2 values correspond with a dorsoventrally curved, anteroposteriorly compressed cerebrum, a ventrally angled cerebellum and brainstem, and a dorsoventrally longer optic lobe. The most extreme positive PC2 values are occupied by both apodiform groups, the hummingbird Archilochus (nectarivore) and the swift Hemiprocne (aerial insectivore), which are also clustered by diet. Negative PC2 values are associated with more elongated brains, with a dorsoventrally flatter cerebrum, cerebellum, and brainstem, with extreme values being occupied by Tyrannosaurus rex, the most basally divergent fossil in the dataset, and the aquatic predators Anhinga and Phalacrocorax.

      Fig. 1:

      Phylomorphospace of the first two principal component (PC) axes highlighting the major aspects of endocranial shape variation in the extant and fossil dataset. Points are colored by discrete dietary category. Inset images within the plot illustrates the landmark scheme used in this study on the endocast of Picoides pubescens (FMNH 290218). Inset images along the axes depict shape changes associated with PC1 and PC2 axes, colored by region (green; cerebrum; red: optic lobe; blue: cerebellum; yellow: brainstem)

      The phylomorphospace shows some phylogenetic structure in morphological variation, most clearly with Apodiformes. To directly assess the amount of phylogenetic signal in endocranial shape data, we calculated the phylogenetic signal of the shape and each brain region separately in the extant dataset using Blomberg’s multivariate K statistic (Kmult) (46). All brain regions exhibited lower phylogenetic signal for shape than expected under Brownian motion (Kmult < 1.0). Overall brain shape showed a Kmult of 0.62 (p < 0.001) with optic lobe and cerebrum shape exhibiting similar values (Kmult = 0.53 and Kmult = 0.31, respectively), whereas phylogenetic signal was lower in the brainstem (Kmult = 0.27) and cerebellum (Kmult = 0.14) (Table S1).

      Size and shape allometry

      To test for the relationship between brain and body size, we regressed log-transformed endocast centroid size, a proxy for brain size, against log-transformed cube-rooted body mass. As expected, endocast size was strongly correlated with body mass (R2 = 0.89, slope = 0.62, p < 0.005) (Fig. 2), with Palaeognathae (e.g., Eudromia, Struthio, Rhea) and Galliformes (e.g., Tetrao, Gallus, Bonasa) exhibiting smaller brain sizes relative to body than expected, while Psittaciformes (e.g., Nestor, Psittacus, Deroptyus) exhibited larger brain sizes than the avian-wide trend. Allometric scaling of shape was visualized by plotting the regression score representing overall shape against both centroid size and body size (Fig. 2B–D). Endocast shape showed statistically significant correlation with both size metrics, with similar coefficients of variation with brain centroid size (R2 = 0.064, p = 0.001) and body size (R2 = 0.063, p = 0.001). This result indicates that size accounts for proportionately small amount of total shape variation, supported visually by the substantial scatter surrounding the main trend (Fig. 2).

      Fig. 2:

      Plots of shape variable against size metrics colored by discrete dietary categories. A. Log-transformed cube-rooted body mass plotted against log-transformed centroid size. B. Log centroid size against regression score, a metric representing overall shape captured by a PGLS regression, of the extant-only dataset. C. Log-transformed cube-rooted body mass against regression score. D. Log centroid size against regression score of the dataset including both extant and fossil taxa.

      Evolutionary rates

      To assess the evolutionary tempo and mode of brain shape, we fit multiple variable-rates models of evolution to the extant + fossil shape dataset. These analyses used PC axes accounting for 99% of shape variance, and endocast shape was analyzed both globally and regionally. A λ model of trait evolution was the best-supported evolutionary model for each brain partition. We then mapped these rates of morphological change onto time-calibrated phylogenies to visualize how evolutionary rates of differ across taxonomic groups (Fig. 3; Fig. S1). Elevated rates are concentrated at the base of the tree among non-avialan dinosaurs for all brain partitions. For the whole endocast and within crown birds (Fig. S1), high rates are observed in the branches leading to Apteryx (kiwi), Anserinae (swans, geese), Apodiformes (nectarivorous hummingbirds and aerial insectivorous swifts), Rallidae (rails), independently across several diving birds, Aegypiinae (Old World vultures), Psittaciformes (parrots), and corvids (crows, ravens). Evolution of the cerebrum also shows elevated rates within branches of Psittaciformes and within Sulidae and its phylogenetically and ecologically close group Threskiornithidae (ibises, spoonbills), as well as in the branch leading to Apteryx (Fig. 3A). Evolution of the optic lobe was more variable throughout the tree, with high rates seen at the base of Apodiformes, the charadriiform clade containing Stercorariidae + Alcidae (auks and jaegers), within Sphenisciformes (penguins), and within Accipitrimorphae (aerial vertebrate predatory hawks, eagles, and vultures), and the branch leading to Passeriformes, the hyperdiverse clade of perching birds (Fig. 3B). Evolutionary rates remained low among crown birds for the cerebellum, with higher rates observed at the base of Anatidae (ducks, geese and swans) as well as at the base of the passerine family Ptilonorhynchidae (bowerbirds) (Fig. 3C). The brainstem shows elevated rates at the base of Caprimulgiformes (nightjars) as well as throughout Apodiformes, the Piciformes suborder Galbuli (jacamars and puffbirds), and especially accelerated rates at the base of Corvidae (Fig. 3D).

      Fig. 3:

      Estimated rates of endocast shape evolution mapped onto time-calibrated phylogeny of birds under best-supported model of trait evolution (λ in all cases), in the cerebrum (A), optic lobe (B), cerebellum (C), and brainstem (D). Color gradient on branches indicates the rate of shape evolution as denoted by the inset histograms. Taxonomic naming from ref. (124). Species names on a phylogenetic tree are given in Fig. S1.

      Ecological covariation

      To test for patterns of ecological covariation with brain morphology, we performed phylogenetically-informed MANOVAs on endocranial size and shape with ecological traits (Fig. 4, S3; Table S1). Analyses were conducted on whole endocast shape and its subdivisions (cerebrum, optic lobe, cerebellum, brainstem) against ecological categories (diet, foraging, locomotion, developmental mode) recorded as percentages (diet, foraging) and discrete data (diet, foraging, locomotion, developmental mode) based on previous studies (37, 47). Neither shape nor size of any endocranial region were found to be significantly correlated with developmental mode (p ≥ 0.2). All analyses of shape were corrected for size as part of the multiple linear regression model, which was significantly correlated with shape in each analysis.

      Fig. 4:

      Summary heatmap from ecomorphological analyses on dietary guild and locomotor lifestyle based on Z-scores, where darker gradient indicates greater values, thus relatively stronger association between endocast shape. Gradient is scaled to values along each row.

      Whole Endocast

      After correcting for size, endocast shape was significantly associated with discrete dietary category (R2 = 0.04, Z = 2.14, p = 0.02), as was relative endocranial size (R2 = 0.097, Z = 3.44, p = 0.002). Across both analyses that were based on Type I sum of squares (only one ecological variable considered at a time) and Type II sum of squares (all variables considered together and non-sequentially), whole endocast shape was most significantly associated with terrestrial locomotion, in which brains of terrestrial birds are dorsoventrally flatter (Fig. 4; Table S1). Scavenging, which is associated with a flatter cerebrum than hindbrain, was more significant in Type II than in Type I analyses, indicating a potential interaction between variables in the model. Aquatic predation was also significantly associated with endocranial shape in Type II but not Type I analyses. Endocast centroid size, used as a metric for absolute brain size, was significantly associated with vertivory, invertivory, scavenging, soaring, and arboreality across both Type I and Type II analyses. Granivory was significantly associated with endocast centroid size in Type I but not Type II analyses, suggesting an equivocal association when other variables are accounted for, while frugivory was significant in Type II but not Type I analyses. Endocast size corrected by body size, or brain size relative to body size, was significantly associated with vertivory and scavenging across both Type I and II analyses. Invertivory and aroboreality were significant in Type I analyses but not Type II, while herbivory was significant in Type II but not Type I analyses (Fig. 4; Table S1). Results based on foraging guild were generally compatible with those of diet and locomotion, and thus are described in the Supplementary Text.

      Cerebrum

      Individual brain regions were compared with ecological variables to assess regionalized ecomorphological signal. Several sensory pathways are contained within the cerebrum, including the dorsal ventricular ridge that contains auditory, visual, somatosensory, and navigational systems (48, 49), which is possibly homologous with the mammalian neocortex (22, 50). One of three visual pathways in the avian brain, the thalamofugal pathway, terminates in the cerebrum as a protuberance called the ‘Wulst’ on the dorsal surface of each cerebral hemisphere, with various associated functions including binocular vision, motion perception, and spatial orientation (22, 24, 51–53). Cerebrum shape was significantly associated with discrete dietary category (R2 = 0.034, Z = 1.79, p = 0.036), while cerebrum size was not. Vertivory and aquatic locomotion were weakly associated with cerebrum shape across both Type I and II analyses, while aquatic predation and scavenging were significantly associated with cerebrum shape in Type II analyses but not Type I. Neither cerebrum centroid size nor proportional size were significantly associated with any dietary guild or locomotor variables (Fig. 4; Table S1).

      Optic lobe

      The optic lobe, a midbrain structure related to processing visual stimuli, and is the terminus of the primary visual system, the tectofugal pathway (22). While there is functional overlap with the thalamofugal pathway of the cerebrum, the tectofugal pathway is also involved in object recognition (54). Optic lobe shape was significantly associated with discrete dietary category (R2 = 0.045, Z = 2.68, p = 0.002), as was absolute size (R2 = 0.22, Z = 3.32, p = 0.002) and proportional size (R2 = 0.12, Z= 3.9, p = 0.001). Across both Type I and Type II analyses, optic lobe shape was significantly associated with invertivory, aquatic predation, scavenging, aquatic locomotion, and terrestriality, and weakly with arboreality. Granivory was significant in Type I but not Type II analyses (Fig. 4; Table S1). Optic lobe centroid size was significantly associated with granivory, invertivory, aquatic predation, scavenging, and weakly with aquatic locomotion and soaring across both Type I and II analyses. Frugivory and vertivory were significant in Type II but not Type I analyses (Fig. 4; Table S1). Proportional optic lobe size was significantly associated with granivory, invertivory, scavenging, and aquatic locomotion across both Type I and Type II analyses. Vertivory and aquatic predation were significant in Type I but not Type II analyses, while herbivory was significant in Type II but not Type I analyses (Fig. 4; Table S1).

      Cerebellum

      The cerebellum, among other functions, processes input from both the visual and vestibular systems to control motor functions associated with balance and initiation of muscular movement (55, 56). Despite occupying a small proportion of the avian brain mass, it contains nearly half of all neurons in the brain, and is characterized by substantial infolding (57). Cerebellum shape was weakly associated with discrete dietary category (R2 = 0.04, Z = 1.59, p = 0.06), while cerebellum size was not. Across both Type I and Type II analyses, cerebellum shape was significantly associated with arboreality. Frugivory, invertivory, and aquatic predation were significantly associated with shape in Type II but not Type I analyses, indicating a potential interaction between variables, which is corroborated by the significant association between cerebellum shape and foraging for aquatic invertebrates (Table S2; Fig. S3). Cerebellum centroid size was weakly correlated with granivory, aquatic predation, and terrestriality across both types of analyses. Invertivory was weakly significant in Type II analyses but not Type I (Fig. 4; Table S1). Proportional cerebellum size was significantly associated with terrestriality, as well as weakly with granivory across both types of analyses. Aquatic predation and invertivory were weakly associated with proportional cerebellum size in Type II but not Type I analyses (Fig. 4; Table S1).

      Brainstem

      The brainstem connects the brain to the spinal cord, and is less intrinsically involved in cognitive functions but critical for unconscious processes, such as respiration, auditory processing, and moderating heart rate (56, 58, 59). Brainstem centroid size (R2 = 0.06, Z = 1.72, p = 0.05) and proportional size (R2 = 0.07, Z = 2.12, p = 0.02) were significantly associated with discrete dietary category while brainstem shape was not. Brainstem shape was significantly associated with granivory, and weakly with aquatic predation, across both Type I and II analyses (Fig. 4; Table S1). Vertivory was significant, while herbivory, scavenging, and arboreality were weakly significant, in Type I but not Type II analyses. Brainstem centroid size was significantly associated with herbivory, arboreality, and weakly with aquatic locomotion across both Type I and II analyses (Fig. 4; Table S1). Vertivory and aquatic predation were significant in Type I but not Type II analyses. Proportional brainstem size was significantly associated with herbivory, aquatic locomotion, and arboreality across both Type I and II analyses. Vertivory and invertivory were significant in Type I but not Type II analyses (Fig. 4; Table S1).

      Discriminant analysis

      Cross validation through phylogenetic functional discriminant analysis (pFDA) results in misclassification error of 70% for dietary category, suggesting that despite a significant correlation, endocranial shape is a poor predictor of dietary category. All three locomotor variables showed approximately 50% misclassification error. Over 100 replicates, aquatic lifestyle showed the lowest misclassification rate (mean: 43%) and the greatest variance in predicted error (25% – 53%), suggesting the fewest misclassifications across all replicates. Misclassification was similar for terrestriality (47% (43% – 50%)) and arboreality (48% (44% – 50%).

      DISCUSSION

      Brain Shape Variation and Evolution

      We find that the greatest amount of variation in avian brain shape falls along the elongate-globular spectrum, which is consistent with the spatial packing hypothesis of brain evolution (33) and largely independent of variation in size metrics (Fig. 2). This flexion of the cranial base, reorganizing the brain into a more globular structure, is accompanied by a transition to a more ventrally oriented foramen magnum observed across both birds and mammals (32–34, 36, 60, 61). Our results imply that head posture differences drive the majority of brain shape variation in crown birds. This outcome agrees with a study by Kawabe and colleagues (35), which found that birds with a posteriorly oriented foramen magnum also possess more anteroposteriorly elongated brains accompanied by elongated and laterally-oriented orbits, whereas birds with globular brains and orbits have a ventrally positioned foramen magnum and a more upright posture. Likewise, Kulemeyer and colleagues (62) suggest that birds with a more posteriorly oriented foramen magnum and anteroposteriorly elongated cranium (and thereby endocast) experience reduced drag during sustained locomotion. These elongated brain morphotypes are characterized by a relatively smaller and more anteriorly positioned cerebrum with reduced dorsal curvature, a feature convergent on the brain morphotypes of the fossil taxa in our study, which also have a posteriorly positioned foramen magnum. Our results demonstrate that rounded brain regions are associated with a ventrally angled hindbrain (cerebellum and brainstem), indicating a more globular brain in birds with a ventrally positioned foramen magnum and an upright posture. These globular and elongated morphotypes are partly associated with aerial and aquatic foragers, respectively, which have adapted for an upright, perching stance in the former and a hydrodynamic head shape in the latter.

      Evolutionary rates analyses reveal accelerated evolution among non-avian dinosaurs and highly heterogeneous trends across avian taxa and among brain regions, complementing previous macroevolutionary studies on avian brain evolution (e.g., (5)). While this rates analysis differs from identifying distinct shifts in brain size parameters, elevated rates of brain shape evolution mirrors many of the phylogenetic positions where distinct allometric trends were identified previously (5). These include the lineage leading to paravian dinosaurs, Apteryx (kiwi), Apodiformes (hummingbirds and swifts), Strigiformes (owls), Picidae (woodpeckers), Psittaciformes (parrots), Ptilonorhychidae (bowerbirds), and Corvidae (crows, ravens). Especially noteworthy are the elevated rates of morphological evolution in two exemplary passerine groups, the bowerbirds and the corvids. Bowerbirds are notorious for their complex nest-building and courtship strategies (63) and previous studies found a positive relationship between the complexity of their bowers and brain size, specifically cerebellum size (64, 65). Our analyses show accelerated evolutionary rates in cerebellum shape in bowerbirds, where the cerebellum is anteroposteriorly shortened, dorsoventrally rounded, and ventrally angled. Together, these results suggest an association between the cognitive traits that are associated with a proportionately larger and arched cerebellum, such as motor planning and control, procedural learning, postural displays in bower construction behaviour, and a greater motor than visual cognitive load (65). Similarly, cerebrum, optic lobe, and brainstem shapes show elevated evolutionary rates in the lineage leading to the highly encephalized Corvidae (66). The brains of corvids have been extensively investigated in relation to their cognitive abilities, with several studies on their exceptionally large brains (11, 67), and specifically on the size of associative and motor-related structures in the forebrain associated with tool use and problem-solving skills in New Caledonian crows (10, 68–70). The rapid evolution towards a shorter brainstem in corvids could be explained by a number of factors, including a more ventral foramen magnum position and thus postural stance that facilitates controlled motor behaviors (62), or spatial packing of a large cerebrum leading to restructuring of the remaining brain regions (36).

      Taxa exhibiting the two extremes of the elongated-globular spectrum also show some of the strongest associations with ecology. Aerial foraging specialists like Apodiformes (hummingbirds and swifts) and the closely related Caprimulgiformes (nightjars and nighthawks) show the most globular and anteroposteriorly shortened forebrains, and ventrally angled hindbrains. These closely related groups are uniquely adept at intricate maneuvers in flight to either catch insect prey (71) or hover over nectar-bearing flowers (72, 73), and further display elevated rates of shape evolution in the brainstem (59). Caprimulgiforms also show distinct convergent traits with owls, notably their wide binocular visual fields along with extremely large Wulsts (27, 74, 75), however owls do not show nearly as globular brains as caprimulgiforms do. Therefore, these patterns of globularization are clearly neither driven by enlarged Wulsts nor relative brain size. The other end of the spectrum is occupied by the aquatic predatory Pelecaniformes, particularly the anhinga and cormorant, which showed anteroposteriorly elongated and straight endocasts. All aquatic predators have this brain shape to some extent, likely due to the necessity for a hydrodynamic body profile, including of the skull, in an aquatic predatory lifestyle (37, 76–79). As such, the elongated brain profile of aquatic predators may be driven by adaptations in the skull for an aquatic lifestyle.

      Several bursts of brain shape evolution co-occur with major ecological transitions. While both MANOVA and pFDA indicate that avian-wide associations between ecology and brain shape are weak overall, analysis of evolutionary rates points to clade-specific evolutionary patterns that could be strongly linked to their specific ecologies. For instance, the lineage leading to Strisores (‘nightbirds’) exhibit fast evolution in the optic lobe which could reflect adaptation for maintaining visual activity under nocturnal behaviors. Within Strisores, Apodiformes (hummingbirds, swifts), which engage in more diurnal habits, show subsequent accelerated rates of evolution in the optic lobe. Indeed, the endocasts of most non-apodiform Strisores that we sampled show more quadrangular optic lobes that extend more posteriorly along the ventral margin of the cerebrum, whereas apodiform taxa possess optic lobes that are more spherical and positioned anteroventrally to the cerebrum, as is more typical in birds. Rapid evolutionary changes in the optic lobe also occurred towards the nocturnal Apteryx (kiwi), but in contrast to the nocturnal Strisores species (e.g., nightjars), their optic lobes are highly reduced in favor of reliance on their olfactory sense (80). Similarly, the optic lobes underwent rapid evolution independently among raptors which rely on vision for their predatory behavior. Notably, the cerebrum shape also underwent rapid changes independently in the polyphyletic Old and New World vultures, which are likely attributed to enlarged olfactory bulbs that are continuous with the anterior portion of the cerebrum. Accelerated changes to the cerebrum are observed for Threskiornithidae (spoonbills, ibises) with hypersensitive beaks, that could be associated with increased sensory perception through the beak in this clade. Their cerebra show a hypertrophied anterior portion near the olfactory bulbs that receives somatosensory signals, including those from trigeminal nerves that provide sensory innervation to the beak (81, 82). This result is consistent with larger principal sensory nucleus of trigeminal nerve in the brainstem observed in birds that rely on specialized sensory abilities using their beak or tongue (83). The cerebellum underwent moderately fast shape evolution in the lineage leading to Rallidae (rails), a clade with multiple secondary losses of flight. However, the shape changes are unlikely to be due to flightlessness because the sampled rallid species are volant and previous studies have shown poor correlation between brain morphology and flightlessness (45, 84, 85). Interestingly, omnivory showed high rates of evolution in cerebellum and second highest in cerebrum which is consistent with previous studies that have shown correlation between larger brain with more generalist diet (86).

      Of all the ecological variables we tested, developmental mode conspicuously showed lack of significant correlations with shape in any brain region. Substantial evidence has supported the hypothesis that altricial birds tend to have larger brains, as a longer growth period and greater parental investment allows for further brain development after hatching (21, 87–89). Further, skull shape is also not significantly correlated with developmental mode (47), suggesting that while a longer post-hatching growth period allows for greater brain volumes to develop, brain shape is not uniformly associated with timing of ontogenetic development across birds.

      Similarities and Differences in Brain Shape and Size Evolution

      Brain size accounts for a very small percentage (∼6%) of brain shape variation. Therefore, the majority of brain morphology is unaccounted for when size or volumetric data alone are considered in a phylogenetic context. Ksepka et al. (5) identified parrots and crows as having independently enlarged relative brain sizes, followed closely by owls, woodpeckers, and bowerbirds. Our analyses of shape evolution revealed elevated rates in some of these groups for certain brain regions, including the whole brain and cerebrum in parrots, which possess an inflated cerebrum with enlarged Wulsts, the cerebellum in bowerbirds, which is anteroposteriorly shortened and rounded, and the brainstem in crows, which is craniocaudally shortened and dorsoventrally flattened. Groups with substantial evolutionary shifts in brain shape that did not correspond with associated size evolution included the two extremes of the elongated-globular spectrum: the overall highly globular brain and laterally rounded optic lobe of the Apodiformes (hummingbirds and swifts), and the elongated brains with flatter optic lobes in the aquatic predators in Pelecaniformes. Interestingly, optic lobe shape evolution, which showed the highest correlation with ecology, did not result in evolutionary shifts that corresponded with brain size enlargement. This reinforces the ecological significance not only of the elongated-globular spectrum, but of the optic lobe itself, which is rounder and dorsoventrally elongated in more aerial birds (e.g, those with arboreal and aerial foraging ecologies), while being flatter (less convex) and laterally oriented in terrestrial and aquatic birds. The optic lobe is also enlarged in long-distance migratory birds (90), while being reduced in waterfowl, owls, and parrots despite no evidence of reduced visual ability through the tectofugal pathway (91). Between the two visual pathways that occupy space on the brain surface, the optic lobe shape appears to show significantly more functional and ecological signal than the Wulst. Therefore, while relative size loosely correlates with this elongated-globular spectrum of shape (Fig. 2), crows and parrots – which have been touted as the birds with the largest relative brain sizes and greatest cognitive ability – do not fall under either extreme of elongated or globular brains. For example, parrot brains have distinctly quadrangular brain shapes and a pronounced interhemispheric sulcus in larger parrot taxa (92), suggested to accommodate a bony interhemispheric septum between their olfactory bulbs, which may provide protection to the brain during feeding and locomotion (93). However, parrots did not display any consistent shape traits associated with ecology, and their large and specialized beaks are more strongly associated with allometry and integration rather than with diet (94). This reinforces that encephalization alone does not capture the full spectrum of complexity in avian neuroanatomy (38, 95, 96), and additional ecological and phylogenetic factors substantially contribute to shape variation.

      While there is significant overlap in brain shape across clades, our complementary analyses of endocast size and shape reveal large-scale patterns in avian brain evolution: 1) after accounting for evolutionary allometry, species with primarily aerial-based ecologies have more globular brain shapes, while terrestrial and aquatic birds have more elongated brains; 2) optic lobe shape has the greatest number of ecological covariates and the greatest variation in evolutionary rates; 3) phylogenetic signal in brain shape is low; and 4) despite statistically significant correlations, predictive power of brain shape for ecology is limited when applied across birds. The poorly predicted form-and-function link does not demonstrate the absence of a strong connection between brain morphology and functional abilities—rather, that the relationship is complex across the entire avian phylogeny. The weak association across macroevolutionary timescales could be due to many-to-one or one-to-many relationships between neuroanatomy and their ecologically relevant functions (97). For example, the optic lobe had the greatest number of covariates among brain regions, possibly because the optic lobe has a more targeted function than the other regions characterized in this study. In contrast, cerebrum shape which includes the Wulst, an important structure for visual cues (98), and involves many other ecologically relevant functions, shows the weakest relationship with ecological variables despite being one of the major drivers of differentiation in brain shape between avian groups. Through our comprehensive study, we find that the remarkable neuroanatomical diversity of birds today emerged not through the dominance of any particular factors, but rather through independent divergence of brain structures and avian clades from the global allometric and ecomorphological patterns.

      METHODS

      Endocranial Sampling

      We compiled endocranial reconstructions for 300 extant species spanning the avian tree of life that encompass their phylogenetic, ecological, and morphological breadth. Of the 300 extant samples, 124 endocast models were created from CT imaging of avian specimens by the authors (mostly skeletal, but fluid specimens were used when skeletal were not available), 96 endocasts models were newly constructed from CT image stacks or mesh files available on Morphosource (99), and 80 endocast mesh files were created and shared by colleagues. For anchoring the ancestral condition, we also included endocasts from six coelurosaurian dinosaur fossils from previous studies (100, 101): Tyrannosaurus rex (FMNH PR2081), Incisivosaurus gauthieri (IVPP V 13326), Citipati osmolskae (IGM 100/973), Khaan mckennai (IGM 100/973), an unnamed troodontid IGM 100/1126, and Archaeopteryx lithographica (NHMUK PV OR 37001). Multiple methods were used for segmentation of endocranial cavity (see Supplementary Text for additional information). Due to the broad phylogenetic scope of this study, any systemic differences in shape data due to different imaging, segmentation, and mesh creation procedures were considered to be negligible compared to interspecific variation. We refer to endocranial regions by the name of the brain structure that left the impression (e.g., ‘cerebrum’ for the impression of cerebrum on the endocast).

      Morphometric data

      To quantify shape of the endocast, we collected high-density GM data building off the landmark scheme of (41) (Fig. S4; Table S3). We used Stratovan Checkpoint (102) to digitally place landmarks and curve semilandmarks on the right side of the endocasts, capturing the major brain regions, including the cerebrum, optic lobe, cerebellum, and brainstem. Additional sliding semilandmarks on the surfaces of these brain regions were semiautomatically placed by implementing the ‘placePatch’ function from the R package ‘Morpho’ version 2.12 (103). This function uses a template, in this case from the model avian species Gallus gallus, consisting of an endocast mesh and corresponding landmark and semilandmark curve coordinates, along with a constellation of evenly distributed surface semilandmarks, and maps these surface semilandmarks for each endocast mesh. To limit biases produced by performing Procrustes alignment on one-sided data of bilaterally symmetrical structures (104, 105), the right sided landmark configuration was mirrored along the median plane to produce a bilaterally symmetrical dataset of endocast landmarks using the ‘mirrorfill’ function in the ‘paleomorph’ R package (106). Using the ‘slider3d’ function in the package ‘Morpho’(103), all curve and surface semi-landmarks were then slid along their corresponding mesh surface, while minimizing total bending energy (107, 108). Generalized Procrustes alignment (GPA) was performed using the ‘gpagen’ function in the ‘geomorph’ R package to generate Procrustes shape coordinates and extract centroid sizes (109). Aligned coordinates from the mirrored left side of the brain were then removed, and subsequent analyses were conducted only on coordinates from the midline and right side of the brain to avoid overweighting of bilateral landmarks through duplicated shape information. This procedure resulted in a total of 241 landmarks, of which there were 13 fixed landmarks, 130 curve semilandmarks and 98 surface semilandmarks. For regional analyses, we extracted a subset of globally aligned Procrustes coordinates that characterize each brain region and ran another round of GPA to locally align the coordinates of each region. In addition to centroid size, proportional regional size of each brain region was calculated as the centroid size of the region in question divided by the total endocast centroid size. Body mass data were taken from (110).

      Ecological data

      To test ecomorphological hypotheses, we compiled a set of ecological traits. Following (37, 111), we compiled discrete dietary categories, dietary guild scores and foraging guild scores, the latter two both as percentages, to broadly characterize the trophic niche of each taxon. Discrete dietary categories (frugivore, nectarivore, granivore, herbivore, vertivore, invertivore, aquatic predator, scavenger, omnivore) were assigned based on whether each taxon obtained greater than or equal to 60% of its resources from any one food category, with remaining taxa being classified as ‘omnivores’. Foraging behaviour was scored to encompass both food resource and the method by which the species obtains the resource and thus is not independent from the dietary scores. These percentage data were tested in Type I analyses (see ‘Ecomorphological Analysis’ below), and these categories were then converted to binary presence/absence data for multivariate (Type II) analyses, to avoid redundancy of variables. Further detail on dietary and foraging scores are listed in (37). In addition to diet, locomotor categories were included to encompass the full spectrum of avian lifestyle ecology, scored as binary presence/absence data on volancy, aquatic lifestyle, soaring, terrestriality, and arboreality (112). Binary data on developmental modes (altriciality, precociality) were compiled from (47).

      Phylogeny

      Phylogenetic comparative analyses were conducted using a composite extant phylogeny from (113), which combines the genus-level resolution from the supertree distribution of the (114) backbone, and scales the age estimates and branch lengths according to (115). We incorporated the fossil taxa to the base of the extant tree using an equal branching method (1) based on the middle age of the occurrence estimates from the Paleobiology Database (paleobiodb.org).

      Analysis

      Morphometric variation

      We conducted comparative morphometric analyses using the R package ‘geomorph’ (109) unless indicated otherwise, performed in R version 4.3.2 (116). To visualize patterns of morphological variation in the avian brain of the extant + fossil dataset, major axes of shape variation were visualized using a principal component analysis (PCA) on the Procrustes coordinates, using the ‘gm.prcomp’ function. Phylogeny was mapped onto the PCA to create a ‘phylomorphospace’, to aid in visualization of the phylogenetic structure in shape variation. We calculated phylogenetic signal of endocranial shape for the entire endocast, as well as each region, as Blomberg’s K statistic using the ‘physignal’ function (46).

      We evaluated allometric patterns of variation for both endocranial shape and size in the extant dataset alone, as well as in the extant + fossil dataset, with centroid size of the endocast as a proxy for brain size (117). Allometry of brain size was analyzed by comparing endocast centroid size with body mass from (110) of each taxon from the extant dataset, analyzed with a phylogenetic generalized least squares (PGLS) regression using the ‘gls’ function from the R package ‘nlme’ version 3.1-166 (118). We analyzed allometry of endocast shape by performing a phylogenetic generalized least squares (pGLS) analysis on three variable combinations: 1) shape against centroid size of the extant-only dataset, 2) shape against log-cube rooted body mass of the extant-only dataset, and 3) shape against centroid size of the extant + fossil dataset. These allometric relationships were visualized by plotting the given size variable against regression score coefficients, a projection of the data that expresses the covariation between shape and size as standardized shape scores, accounting for the maximum correlation between shape changes predicted by the regression (119, 120). Since brain shape was significantly correlated with brain size (see results), centroid size was incorporated in linear models used in subsequent ecomorphological analyses.

      Rates of evolution

      To estimate evolutionary rates for the overall brain and its partitions, we used BayesTraits V4 (121) to fit variable-rates evolutionary models to the extant + fossil dataset. Rates were calculated on PC axes accounting for 99% of the total brain shape variation were analyzed. All available evolutionary models were compared, including BM, OU, δ, κ, and λ models, and the best-supported model was determined using Bayes Factor comparisons using functions from the BTRtools R package (source code archived at [github.com/tiago-simoes/BTprocessR2]). Rates analyses were performed on the total endocast shape, and then on each individual brain region captured by the endocast (cerebrum, optic lobe, cerebellum, brainstem). Further, we compared relative evolutionary rates within each brain partition to examine how they differ between extant birds with different dietary categories. This was implemented with the ‘compare.evol.rates’ function in ‘geomorph’, and then rates were divided by the highest rate in the brain region to standardize rate values and compare relative rates for each dietary category.

      Ecomorphological analysis

      We tested for ecological correlates of extant endocast shape using distance-based phylogenetic generalized least squares (PGLS) on the Procrustes coordinates. These were conducted using the ‘procD.pgls’ function using both the Type I (sequential) sum of squares for analyses for each single ecological variable separately, and Type II (hierarchical) sum of squares on a set of ecological variables (122). Separate analyses were conducted for each brain region: whole endocast, cerebrum, optic lobe, cerebellum, and brainstem. Shape and size for each region were compared against discrete dietary category, and then further sets of regression models were implemented for each set of multivariate data: 1) dietary guild data, 2) foraging guild data, and 3) locomotor ecology data (Table S1, S2). For dietary guild and foraging guild, Type I analyses used percentage score data and included one variable per analysis, while Type II analyses incorporated all variables into a single model, and converted these data into presence and absence of each category. If a variable was found to be statistically significant in a Type I analysis but not Type II, then the correlation is considered to be equivocal; alternatively, if the variable is significant in Type II but not Type I, then there is likely an interaction or strong covariation between ecological variables in the model that are associated with the morphological variable when considered together.

      Discriminant analysis

      To test for the predictive power of endocranial shape for dietary category, we implemented a phylogenetic flexible discriminant analysis (pFDA) on the Procrustes coordinates of the extant dataset using the ‘phylo.FDA’ function developed by (123). Similar to linear discriminant analysis, pFDA uses categorical data, such as dietary category, and multivariate shape data to estimate the posterior probability of ecological traits for each taxon while accounting for phylogenetic non-independence in the multivariate dataset. Since there is unequal representation of each dietary category in the dataset, we subsampled the dataset at random to reach an equal number of taxa exhibiting each trait before running the pFDA and repeated this for 100 replicates to obtain robust estimates for the strength of affinity for each ecological trait.

      Supporting information

      Supplementary Text[supplements/682392_file02.pdf]

      FUNDING

      U.S. National Science Foundation (DBI 1828305 for NYIT CT imaging; IOS 2045466 to Akinobu Watanabe).

      ACKNOWLEDGMENTS

      We would like to thank the following: Amy Balanoff, Federico DeGrange, Catherine Early, Ryan Felice, Todd Green, Anthony Herrel, Soichiro Kawabe, Daniel Ksepka, Chris Milensky, Adam Smith, Stig Walsh, and Larry Witmer, who contributed endocranial data in the form of either CT scans or surface data; Peter Capainolo, Paul Sweet, and Tom Trombone (American Museum of Natural History) for providing access to specimens; Ben Marks (Field Museum of Natural History), Chris Milensky (Smithsonian National Museum of Natural History), and Kristof Zyskowski (Yale Peabody Museum) for facilitating specimen loans; Roger Benson, Sharon Grant, David Blackburn, Serina Brady, Casey Dillman, Daniel Field, Jessie Maisano, Nelson Rios, Mark Robbins, and Brett Benz for providing access to specimen data via MorphoSource; Kelsi Hurdle (New York Institute of Technology) for providing access and assistance with NYIT scanning facility; Morgan Chase and Phoebe Fu (American Museum of Natural History) for access and assistance with AMNH scanning facility.

      Funder Information Declared

      U.S. National Science Foundation, https://ror.org/021nxhr62, IOS 2045466

      REFERENCES

      1. 1. S. L. Brusatte, G. T. Lloyd, S. C. Wang, M. A. Norell, Gradual assembly of avian body plan culminated in rapid rates of evolution across the dinosaur-bird transition. Curr. Biol. 24, 2386–2392 (2014).
      2. 2. V. Allen, K. T. Bates, Z. Li, J. R. Hutchinson, Linking the evolution of body shape and locomotor biomechanics in bird-line archosaurs. Nature 497, 104–107 (2013).
      3. 3. S. Macaulay, et al., Decoupling body shape and mass distribution in birds and their dinosaurian ancestors. Nat. Commun. 14, 1575 (2023).
      4. 4. D. Sol, T. Székely, A. Liker, L. Lefebvre, Big-brained birds survive better in nature. Proc. R. Soc. B Biol. Sci. 274, 763–769 (2007).
      5. 5. D. T. Ksepka, et al., Tempo and Pattern of Avian Brain Size Evolution. Curr. Biol. 30, 2026–2036.e3 (2020).
      6. 6. C. R. Torres, M. A. Norell, J. A. Clarke, Bird neurocranial and body mass evolution across the end-Cretaceous mass extinction: The avian brain shape left other dinosaurs behind. Sci. Adv. 7, 1–10 (2021).
      7. 7. N. J. Emery, Cognitive ornithology: The evolution of avian intelligence. Philos. Trans. R. Soc. B Biol. Sci. 361, 23–43 (2006).
      8. 8. L. Lefebvre, S. M. Reader, D. Sol, Brains, innovations and evolution in birds and primates. Brain. Behav. Evol. 63, 233–246 (2004).
      9. 9. L. Lefebvre, P. Whittle, E. Lascaris, A. Finkelstein, Feeding innovations and forebrain size in birds. Anim. Behav. 53, 549–560 (1997).
      10. 10. J. Mehlhorn, G. R. Hunt, R. D. Gray, G. Rehkämper, O. Güntürkün, Tool-making new caledonian crows have large associative brain areas. Brain. Behav. Evol. 75, 63–70 (2010).
      11. 11. J. Cnotka, O. Güntürkün, G. Rehkämper, R. D. Gray, G. R. Hunt, Extraordinary large brains in tool-using New Caledonian crows (Corvus moneduloides). Neurosci. Lett. 433, 241–245 (2008).
      12. 12. S. E. Overington, J. Morand-Ferron, N. J. Boogert, L. Lefebvre, Technical innovations drive the relationship between innovativeness and residual brain size in birds. Anim. Behav. 78, 1001–1010 (2009).
      13. 13. D. Sol, L. Lefebvre, J. D. Rodríguez-Teijeiro, Brain size, innovative propensity and migratory behaviour in temperate Palaearctic birds. Proc. R. Soc. B Biol. Sci. 272, 1433– 1441 (2005).
      14. 14. O. Vincze, Light enough to travel or wise enough to stay? Brain size evolution and migratory behavior in birds. Evolution 70, 2123–2133 (2016).
      15. 15. F. Sayol, et al., Environmental variation and the evolution of large brains in birds. Nat. Commun. 7, 13971 (2016).
      16. 16. S. Shultz, R. I. M. Dunbar, Social bonds in birds are associated with brain size and contingent on the correlated evolution of life-history and increased parental investment. Biol. J. Linn. Soc. 100, 111–123 (2010).
      17. 17. A. N. Iwaniuk, K. E. Arnold, Is cooperative breeding associated with bigger brains? A comparative test in the Corvida (Passeriformes). Ethology 110, 203–220 (2004).
      18. 18. N. J. Emery, A. M. Seed, A. M. P. Von Bayern, N. S. Clayton, Cognitive adaptations of social bonding in birds. Philos. Trans. R. Soc. B Biol. Sci. 362, 489–505 (2007).
      19. 19. E. D. Jarvis, Learned birdsong and the neurobiology of human language. Ann. N. Y. Acad. Sci. 1016, 749–777 (2004).
      20. 20. C. I. Petkov, E. D. Jarvis, Birds, primates, and spoken language origins: Behavioral phenotypes and neurobiological substrates. Front. Evol. Neurosci. 4, 1–24 (2012).
      21. 21. P. M. Bennett, P. H. Harvey, Relative brain size and ecology in birds. J. Zool. 207, 151– 169 (1985).
      22. 22. T. Shimizu, H. J. Karten, “The avian visual system and the evolution of the neocortex” in Vision, Brain and Behavior in Birds, H. P. Zeigler, H.-J. Bischof, Eds. (1993), pp. 103–114.
      23. 23. C. Gutiérrez-Ibáñez, A. N. Iwaniuk, T. J. Lisney, D. R. Wylie, Comparative study of visual pathways in Owls (Aves: Strigiformes). Brain. Behav. Evol. 81, 27–39 (2013).
      24. 24. A. N. Iwaniuk, C. P. Heesy, M. I. Hall, D. R. W. Wylie, Relative Wulst volume is correlated with orbit orientation and binocular visual field in birds. J. Comp. Physiol. A Neuroethol. Sensory, Neural, Behav. Physiol. 194, 267–282 (2008).
      25. 25. M. M. Demmel Ferreira, F. J. Degrange, G. A. Tirao, Brain surface morphology and ecological and macroevolutionary inferences of avian New World suboscines (Aves, Passeriformes, Tyrannides). J. Comp. Neurol. 532 (2024).
      26. 26. H. Jerison, Evolution of the Brain and Intelligence (Academic Press, 1973).
      27. 27. A. N. Iwaniuk, D. R. W. Wylie, The evolution of stereopsis and the Wulst in caprimulgiform birds: A comparative analysis. J. Comp. Physiol. A Neuroethol. Sensory, Neural, Behav. Physiol. 192, 1313–1326 (2006).
      28. 28. C. Gutiérrez-Ibáñez, A. N. Iwaniuk, D. R. Wylie, Parrots have evolved a primate-like telencephalic-midbrain-cerebellar circuit. Sci. Rep. 8, 1–11 (2018).
      29. 29. A. Irurtzun, The “Globularization Hypothesis” of the language-ready brain as a developmental frame for prosodic bootstrapping theories of language acquisition. Front. Psychol. 6, 1–7 (2015).
      30. 30. C. Boeckx, A. Benítez-Burraco, The shape of the human language-ready brain. Front. Psychol. 5, 1–23 (2014).
      31. 31. S. Neubauer, J. J. Hublin, P. Gunz, The evolution of modern human brain shape. Sci. Adv. 4, 1–8 (2018).
      32. 32. V. Weisbecker, et al., Global elongation and high shape flexibility as an evolutionary hypothesis of accommodating mammalian brains into skulls. Evolution (N. Y). 75, 625– 640 (2021).
      33. 33. D. E. Lieberman, B. Hallgrímsson, W. Liu, T. E. Parsons, H. A. Jamniczky, Spatial packing, cranial base angulation, and craniofacial shape variation in the mammalian skull: Testing a new model using mice. J. Anat. 212, 720–735 (2008).
      34. 34. C. F. Ross, M. J. Ravosa, Basicranial flexion, relative brain size, and facial kyphosis in nonhuman primates. Am. J. Phys. Anthropol. 91, 305–324 (1993).
      35. 35. S. Kawabe, T. Shimokawa, H. Miki, S. Matsuda, H. Endo, Variation in avian brain shape: Relationship with size and orbital shape. J. Anat. 223, 495–508 (2013).
      36. 36. J. Marugán-Lobón, S. M. Nebreda, G. Navalón, R. B. J. Benson, Beyond the beak: Brain size and allometry in avian craniofacial evolution. J. Anat., 1–13 (2021).
      37. 37. R. N. Felice, J. A. Tobias, A. L. Pigot, A. Goswami, Dietary niche and the evolution of cranial morphology in birds. Proc. R. Soc. B Biol. Sci. 286, 20182677 (2019).
      38. 38. J. Marugán-Lobón, A. Watanabe, S. Kawabe, Studying avian encephalization with geometric morphometrics. J. Anat. 229, 191–203 (2016).
      39. 39. S. Kawabe, S. Matsuda, N. Tsunekawa, H. Endo, Ontogenetic shape change in the chicken brain: Implications for paleontology. PLoS One 10, 1–16 (2015).
      40. 40. A. Watanabe, A. M. Balanoff, P. M. Gignac, M. E. L. Gold, M. A. Norell, Novel neuroanatomical integration and scaling define avian brain shape evolution and development. Elife 10, 1–21 (2021).
      41. 41. A. Watanabe, et al., Are endocasts good proxies for brain size and shape in archosaurs throughout ontogeny? J. Anat. 234, 291–305 (2019).
      42. 42. L. King, et al., Endocranial development in non-avian dinosaurs reveals an ontogenetic brain trajectory distinct from extant archosaurs. Nat. Commun. 15, 1–10 (2024).
      43. 43. C. M. Early, A. N. Iwaniuk, R. C. Ridgely, L. M. Witmer, Endocast structures are reliable proxies for the sizes of corresponding regions of the brain in extant birds. J. Anat. 237, 1162–1176 (2020).
      44. 44. A. R. Keirnan, et al., Avian telencephalon and cerebellum volumes can be accurately estimated from digital brain endocasts. Biol. Lett. 21 (2025).
      45. 45. M. E. L. Gold, A. Watanabe, Flightless birds are not neuroanatomical analogs of non-avian dinosaurs. BMC Evol. Biol. 18, 1–11 (2018).
      46. 46. D. C. Adams, A Generalized K Statistic for Estimating Phylogenetic Signal from Shape and Other High-Dimensional Multivariate Data. Syst. Biol. 63, 685–697 (2014).
      47. 47. E. S. E. Hunt, R. N. Felice, J. A. Tobias, A. Goswami, Ecological and life-history drivers of avian skull evolution. Evolution (N. Y). 77, 1720–1729 (2023).
      48. 48. M. Stacho, et al., A cortex-like canonical circuit in the avian forebrain. Science (80-.). 369 (2020).
      49. 49. Y. Wang, A. Brzozowska-Prechtl, H. J. Karten, Laminar and columnar auditory cortex in avian brain. Proc. Natl. Acad. Sci. U. S. A. 107, 12676–12681 (2010).
      50. 50. A. Reiner, D. J. Perkel, C. V. Mello, E. D. Jarvis, Songbirds and the revised avian brain nomenclature. Ann. N. Y. Acad. Sci. 1016, 77–108 (2004).
      51. 51. P. J. Straight, P. M. Gignac, W. J. Kuenzel, A histological and diceCT-derived 3D reconstruction of the avian visual thalamofugal pathway. Sci. Rep. 14, 1–14 (2024).
      52. 52. N. Michael, S. Löwel, H. J. Bischof, Features of the retinotopic representation in the visual wulst of a laterally eyed bird, the zebra finch (Taeniopygia guttata). PLoS One 10, 1–26 (2015).
      53. 53. J. Baron, L. Pinto, M. O. Dias, B. Lima, S. Neuenschwander, Directional responses of visual wulst neurones to grating and plaid patterns in the awake owl. Eur. J. Neurosci. 26, 1950–1968 (2007).
      54. 54. H. J. Bischof, S. Watanabe, On the structure and function of the tectofugal visual pathway in laterally eyed birds. Eur. J. Morphol. 35, 246–254 (1997).
      55. 55. A. Balanoff, et al., Quantitative functional imaging of the pigeon brain: implications for the evolution of avian powered flight. Proc. R. Soc. B Biol. Sci. 291 (2024).
      56. 56. A. B. Butler, W. Hodos, Comparative vertebrate neuroanatomy: evolution and adaptation, 2nd editio (John Wiley & Sons, 2005).
      57. 57. G. F. Striedter, R. G. Northcutt, Brains through time: a natural history of vertebrates (Oxford University Press, 2019).
      58. 58. M. F. Kubke, D. P. Massoglia, C. E. Carr, Bigger brains or bigger nuclei? Regulating the size of auditory structures in birds. Brain. Behav. Evol. 63, 169–180 (2004).
      59. 59. D. H. Cohen, H. J. Karten, “The structural organization of avian brain: An overview” in Birds, Brain and Behavior, I. J. Goodman, M. W. Schein, Eds. (Academic Press, 1974), pp. 29–73.
      60. 60. V. Weisbecker, A. Goswami, Brain size, life history, and metabolism at the marsupial/placental dichotomy. Proc. Natl. Acad. Sci. U. S. A. 107, 16216–16221 (2010).
      61. 61. N. M. A. Crouch, Shared patterns of spatial accumulation of lineages across terrestrial vertebrates. J. Biogeogr. 48, 1811–1823 (2021).
      62. 62. C. Kulemeyer, K. Asbahr, P. Gunz, S. Frahnert, F. Bairlein, Functional morphology and integration of corvid skulls a 3D geometric morphometric approach. Front. Zool. 6, 1–14 (2009).
      63. 63. R. Kusmierski, G. Borgia, A. Uy, R. H. Crozier, Labile evolution of display traits in bowerbirds indicates reduced effects of phylogenetic constraint. Proc. R. Soc. B Biol. Sci. 264, 307–313 (1997).
      64. 64. J. Madden, Sex, bowers and brains. Proc. R. Soc. B Biol. Sci. 268, 833–838 (2001).
      65. 65. L. B. Day, D. A. Westcott, D. H. Olster, Evolution of bower complexity and cerebellum size in bowerbirds. Brain. Behav. Evol. 66, 62–72 (2005).
      66. 66. N. J. Emery, N. S. Clayton, The Mentality of Crows: Convergent Evolution of Intelligence in Corvids and Apes. Science (80-.). 306, 1903–1907 (2004).
      67. 67. S. Olkowicz, et al., Birds have primate-like numbers of neurons in the forebrain. Proc. Natl. Acad. Sci. U. S. A. 113, 7255–7260 (2016).
      68. 68. B. Kenward, C. Rutz, A. A. S. Weir, A. Kacelnik, Development of tool use in New Caledonian crows: inherited action patterns and social influences. Anim. Behav. 72, 1329– 1343 (2006).
      69. 69. A. M. P. von Bayern, R. J. P. Heathcote, C. Rutz, A. Kacelnik, The Role of Experience in Problem Solving and Innovative Tool Use in Crows. Curr. Biol. 19, 1965–1968 (2009).
      70. 70. C. Rutz, The ecological significance of tool use in New Caledonian crows (Science (2010) (1523)). Science (80-.). 333, 1703 (2011).
      71. 71. D. R. Warrick, M. W. Bundle, K. P. Dial, Bird Maneuvering Flight: Blurred Bodies, Clear Heads. Integr. Comp. Biol. 42, 141–148 (2002).
      72. 72. R. Dakin, P. S. Segre, D. L. Altshuler, Individual variation and the biomechanics of maneuvering flight in hummingbirds. J. Exp. Biol. 223, 1–7 (2020).
      73. 73. R. Dakin, P. S. Segre, A. D. Straw, D. L. Altshuler, Morphology, muscle capacity, skill, and maneuvering ability in hummingbirds. Science (80-.). 657, 653–657 (2018).
      74. 74. J. Wallman, J. D. Pettigrew, Conjugate and disjunctive saccades in two avian species with contrasting oculomotor strategies. J. Neurosci. 5, 1418–1428 (1985).
      75. 75. A. N. Iwaniuk, P. L. Hurd, The evolution of cerebrotypes in birds. Brain. Behav. Evol. 65, 215–230 (2005).
      76. 76. O. T. Owre, Adaptations for locomotion and feeding in the Anhinga and the Double-crested Cormorant. Ornithol. Monogr. 6, 1–138 (1967).
      77. 77. J. R. Lovvorn, G. A. Liggins, M. H. Borstad, S. M. Calisal, J. Mikkelsen, Hydrodynamic drag of diving birds: Effects of body size, body shape and feathers at steady speeds. J. Exp. Biol. 204, 1547–1557 (2001).
      78. 78. J. R. Lovvorn, G. A. Liggins, Interactions of body shape, body size and stroke-acceleration patterns in costs of underwater swimming by birds. Funct. Ecol. 16, 106–112 (2002).
      79. 79. K. E. Crandell, R. O. Howe, P. L. Falkingham, Repeated evolution of drag reduction at the air–water interface in diving kingfishers. J. R. Soc. Interface 16 (2019).
      80. 80. B. A. Moore, J. R. Paul-Murphy, A. J. D. Tennyson, C. J. Murphy, Blind free-living kiwi offer a unique window into the ecology and evolution of vertebrate vision. BMC Biol. 15, 1–3 (2017).
      81. 81. J. L. Dubbeldam, The Sensory Trigeminal System in Birds: Input, Organization and Effects of Peripheral Damage. A Review. Arch. Physiol. Biochem. 106, 338–345 (1998).
      82. 82. O. Güntürkün, The avian “prefrontal cortex” and cognition. Curr. Opin. Neurobiol. 15, 686–693 (2005).
      83. 83. C. Gutiérrez-Ibáñez, A. N. Iwaniuk, D. R. Wylie, The independent evolution of the enlargement of the principal sensory nucleus of the trigeminal nerve in three different groups of birds. Brain. Behav. Evol. 74, 280–294 (2010).
      84. 84. A. N. Iwaniuk, J. E. Nelson, H. F. James, S. L. Olson, A comparative test of the correlated evolution of flightlessness and relative brain size in birds. J. Zool. 263, 317–327 (2004).
      85. 85. T. Nakao, T. Yamasaki, N. Ogihara, M. Shimada, Relationship between flightlessness and brain morphology among Rallidae. J. Anat. 241, 776–788 (2022).
      86. 86. S. Ducatez, J. Clavel, L. Lefebvre, Ecological generalism and behavioural innovation in birds: Technical intelligence or the simple incorporation of new foods? J. Anim. Ecol. 84, 79–89 (2015).
      87. 87. A. N. Iwaniuk, J. E. Nelson, Developmental differences are correlated with relative brain size in birds: A comparative analysis. Can. J. Zool. 81, 1913–1928 (2003).
      88. 88. M. Griesser, S. M. Drobniak, S. M. Graber, C. P. van Schaik, Parental provisioning drives brain size in birds. Proc. Natl. Acad. Sci. 120, e2121467120 (2023).
      89. 89. C. J. Charvet, G. F. Striedter, Developmental modes and developmental mechanisms can channel brain evolution. Front. Neuroanat. 5, 1–5 (2011).
      90. 90. O. Vincze, C. I. Vagasi, P. L. Pap, G. Osvath, A. P. Møller, Brain regions associated with visual cues are important for bird migration. Biol. Lett. 11, 14–17 (2015).
      91. 91. A. N. Iwaniuk, C. Gutierrez-Ibanez, J. M. P. Pakan, D. R. Wylie, Allometric scaling of the tectofugal pathway in birds. Brain. Behav. Evol. 75, 122–137 (2010).
      92. 92. J. Carril, C. P. Tambussi, F. J. Degrange, M. J. Benitez Saldivar, M. B. J. Picasso, Comparative brain morphology of Neotropical parrots (Aves, Psittaciformes) inferred from virtual 3D endocasts. J. Anat. 229, 239–251 (2016).
      93. 93. A. Iwaniuk, J. Nelson, S. O’Leary, Big brains need big protection: the ‘interhemispheric septum’ in the Psittaciform braincase in Journal of Morphology: International Congress of Vertebrate Morphology, (2001), p. 1.
      94. 94. J. A. Bright, J. Marugán-Lobón, E. J. Rayfield, S. N. Cobb, The multifactorial nature of beak and skull shape evolution in parrots and cockatoos (Psittaciformes). BMC Evol. Biol. 19, 1–9 (2019).
      95. 95. R. Hooper, B. Brett, A. Thornton, Problems with using comparative analyses of avian brain size to test hypotheses of cognitive evolution. PLoS One 17, 9–11 (2022).
      96. 96. C. P. Klingenberg, “Multivariate allometry” in Advances in Morphometrics, L. Marcus, C. M, A. Loy, G. Naylor, D. Slice, Eds. (Plenum Press, 1996), pp. 23–49.
      97. 97. P. C. Wainwright, M. E. Alfaro, D. I. Bolnick, C. D. Hulsey, Many-to-one mapping of form to function: A general principle in organismal design? Integr. Comp. Biol. 45, 256– 262 (2005).
      98. 98. J. D. Pettigrew, M. Konishi, Neurons Selective for Orientation and Binocular Disparity in the Visual Wulst of the Barn Owl (Tyto alba). Science (80-.). 193, 675–678 (1976).
      99. 99. D. M. Boyer, G. F. Gunnell, S. Kaufman, T. M. McGeary, Morphosource: Archiving and Sharing 3-D Digital Specimen Data. Paleontol. Soc. Pap. 22, 157–181 (2016).
      100. 100. A. M. Balanoff, et al., Best practices for digitally constructing endocranial casts: examples from birds and their dinosaurian relatives. J. Anat. 229, 173–190 (2016).
      101. 101. A. M. Balanoff, G. S. Bever, T. B. Rowe, M. A. Norell, Evolutionary origins of the avian brain. Nature 501, 93–96 (2013).
      102. 102. S. Corporation, Stratovan Checkpoint [Software].
      103. 103. A. Stefan, G. Jefferis, D. Ian, M. S. Schlager, Package ‘Morpho’ (2025).
      104. 104. A. Cardini, K. Seetah, G. Barker, How many specimens do I need? Sampling error in geometric morphometrics: testing the sensitivity of means and variances in simple randomized selection experiments. Zoomorphology (2015) doi:10.1007/s00435-015-0253-z.
      105. 105. A. Cardini, Left, right or both? Estimating and improving accuracy of one-side-only geometric morphometric analyses of cranial variation. J. Zool. Syst. Evol. Res. 55, 1–10 (2017).
      106. 106. A. T. Lucas, A. Goswami, M. T. Lucas, Package ‘paleomorph’ (2022).
      107. 107. F. L. Bookstein, Principal Warps: Thin-Plate Splines and the Decomposition of Deformations. IEEE Trans. Pattern Anal. Mach. Intell. 11, 567–585 (1989).
      108. 108. P. Gunz, P. Mitteroecker, Semilandmarks: a method for quantifying curves and surfaces. Hystrix, Ital. J. Mammal. 24, 103–109 (2013).
      109. 109. D. Adams, M. Collyer, E. Baken, Package ‘geomorph’ (2025).
      110. 110. J. B. Dunning Jr., CRC Handbook of Avian Body Masses (2008) doi:10.1017/S0963180113000479.
      111. 111. H. Wilman, et al., EltonTraits 1.0: Species-level foraging attributes of the world’s birds and mammals BASE DE DATOS. Ecology 95, 2027 (2014).
      112. 112.
        1.  S. M. Billerman, 
        2. B. K. Keeney, 
        3. P. G. Rodewald, 
        4. T. S. Schulenberg
        , Eds., Birds of the World (Cornell Laboratory of Ornithology, 2022).
      113. 113. C. R. Cooney, et al., Mega-evolutionary dynamics of the adaptive radiation of birds. Nature 542, 344–347 (2017).
      114. 114. W. Jetz, G. H. Thomas, J. B. Joy, K. Hartmann, A. O. Mooers, The global diversity of birds in space and time. Nature 491, 444–448 (2012).
      115. 115. R. O. Prum, et al., A comprehensive phylogeny of birds (Aves) using targeted next-generation DNA sequencing. Nature 526, 569–573 (2015).
      116. 116. R. D. C. Team, R: A language and environment for statistical computing (2023).
      117. 117. M. L. Zelditch, D. L. Swiderski, H. D. Sheets, W. L. Fink, Geometric morphometrics for biologists: a primer (2004) doi:10.1016/B978-0-12-386903-6.00001-0.
      118. 118. J. Pinheiro, et al., Package “nlme”: Linear and Nonlinear Mixed Effects Models (2020).
      119. 119. A. G. Drake, C. P. Klingenberg, The pace of morphological change: Historical transformation of skull shape in St Bernard dogs. Proc. R. Soc. B Biol. Sci. 275, 71–76 (2008).
      120. 120. D. C. Adams, F. J. Rohlf, D. E. Slice, A field comes of age: Geometric morphometrics in the 21st century. Hystrix 24 (2013).
      121. 121. A. Meade, M. Pagel, BayesTraitsV4 (2024).
      122. 122. D. C. Adams, M. L. Collyer, Phylogenetic ANOVA: Group-clade aggregation, biological challenges, and a refined permutation procedure. Evolution (N. Y). 72, 1204–1215 (2018).
      123. 123. L. Schmitz, R. Motani, Nocturnality in Dinosaurs Inferred from Scleral Ring and Orbit Morphology. Science (80-.). 332, 705–708 (2011).
      124. 124. J. Stiller, et al., Complexity of avian evolution revealed by family-level genomes. Nature 629, 851–860 (2024).

      Journal of Modern Sciences

      info@modernsciences.org

      About
      Modern Sciences is a media company that aims to provide the academics, researches, and science enthusiasts with the latest news and stories in the world of Applied and Pure Science.