Which Family of Birds Has the Largest Brain to Body Ratio
Body size correlates with most structural and functional components of an organism's phenotype – brain size being a prime number case of allometric scaling with animal size. Therefore, comparative studies of brain development in vertebrates rely on controlling for the scaling effects of torso size variation on brain size variation past calculating brain weight/body weight ratios. Differences in the encephalon size-torso size relationship between taxa are usually interpreted as differences in selection acting on the encephalon or its components, while selection pressures acting on trunk size, which are among the near prevalent in nature, are rarely acknowledged, leading to conflicting and confusing conclusions. We address these problems by comparing brain-body relationships from across >1,000 species of birds and non-avian reptiles. Relative brain size in birds is often causeless to be 10 times larger than in reptiles of similar body size. Nosotros examine how differences in the specific gravity of body tissues and in body design (eastward.g., presence/absence of a tail or a dense shell) between these two groups can affect estimates of relative brain size. Using phylogenetic comparative analyses, we show that the gap in relative brain size between birds and reptiles has been grossly exaggerated. Our results highlight the demand to take into account differences between taxa arising from selection pressures affecting torso size and design, and call into question the widespread misconception that reptile brains are pocket-size and incapable of supporting sophisticated behavior and cognition.
© 2019 S. Karger AG, Basel
Introduction
Our understanding of vertebrate brain evolution rests largely on comparative analyses quantifying interspecific differences in brain size and structure [Striedter, 2005]. Given that brain size scales predictably with trunk size, reporting the size of the whole brain or of its individual components as brain/body ratios and presenting comparative information graphically on bivariate brain size-body size plots are a widespread exercise [Schmidt-Nielsen 1984]. The relationship between brain and body size is allometric rather than isometric, such that on a log-log plot of brain size on torso size, the data points for a given taxonomic group fall effectually a regression line with a slope of less than one [Harvey and Pagel, 1988]. There has been much debate nigh the biological significance of the slope of the regression line and of the magnitude of the residuals from this relationship across species [Deacon, 1990; Montgomery et al., 2016]. The latter have often been interpreted as reflecting cerebral ability or adaptations to particular ecological conditions [due east.g. Emery et al., 2007; Shettleworth, 2010; Menzel and Fischer, 2011]. In general, however, models of brain evolution have focused on interpreting relative brain size based on selection pressures acting on the numerator of the brain/torso ratio, while the effects that pick exerts indirectly on brain size through its effects on torso size remain largely neglected. Hither, nosotros explore how not taking into account selection pressures on body size can misconstrue estimates of relative encephalon size, giving rise to long-standing and heavily entrenched misconceptions regarding differences between taxa [encounter Smaers et al., 2012]. Nosotros employ Reptilia every bit a case study, a highly diverse grouping that contains two major vertebrate radiations – birds and not-avian reptiles – whose brain sizes and cognitive achievements are often field of study to comparison.
Birds have brains that are as large as or even larger than those of mammals of like trunk size. Non-avian reptiles and most ectothermic vertebrates, on the other hand, have brains that are smaller, relative to their body size, than those of birds and mammals. But, how much larger are bird brains compared to reptilian brains? Published estimates range widely, only most authors state that the average difference is x-fold, i.e. a bird or a mammal accept a brain 10 times larger (heavier) than a reptile of similar body size [e.m. Martin, 1981; Hurlburt, 1996; van Dongen, 1998; Northcutt, 2011; Hurlburt et al., 2013; Dicke and Roth, 2016; Güntürkün et al., 2017; Shimizu et al., 2017]. Such big difference is puzzling because the phylogenetic relatedness between birds and reptiles and the similarities in brain organisation between these ii groups. For example, in stark contrast to mammals, the largest portion of the pallium in both birds and reptiles is subcortical, located ventral to the lateral ventricles, and gives rise to a structure known every bit the dorsal ventricular ridge (DVR, besides called the nidopallium and mesopallium in birds). The DVR receives ascending visual, auditory, and somatosensory thalamic projections, and is considered by many to be functionally convergent with the mammalian neocortex [Butler and Hodos, 2005; Jarvis, 2009; Güntürkün et al., 2017; Yamashita and Nomura, 2017; Tosches et al., 2018]. As reptile brains are similar in relative size to those of terrestrial frogs, toads, and teleost fish [Striedter, 2005; van Dongen, 1998], the large size departure between birds and reptiles has been interpreted as evidence that, during the reptile-bird transition, bird brains massively increased their size while those of non-avian reptiles barely budged.
The 10-fold figure describing the boilerplate brain size difference betwixt birds and reptiles can be traced back to the work of Harry Jerison, who conducted the first serious endeavor to compare brain size data across dissimilar vertebrate lineages. Jerison [1973] famously plotted encephalon and body weight information on a log-log scale and drew minimum convex polygons enclosing the data points for different taxonomic groups (Fig. 1). The polygons show an orderly relationship, those of birds and mammals laying above the polygons of other vertebrate radiations. Although much criticized, Jerison's polygon plots are intuitively pleasing because they suggest a progressive increment in relative brain size during vertebrate development that roughly matches the assumed intelligence rankings informally assigned to the various vertebrate groups. Jerison's polygons have been reproduced in countless publications, often used to justify the presumed cognitive superiority of birds and mammals: the polygons for birds and mammals show almost consummate overlap, yet a prominent gap separates the bird-mammal polygon from the polygon representing the remaining vertebrates (Fig. ane). Jerison [1973] fitted past eye lines with a 2/three slope to the unlike polygons and estimated the average departure in encephalon size between "higher" (mammals and birds) and "lower" (reptiles, amphibians, and fish) vertebrates to be one lodge of magnitude (ten×).
Fig. i.
Polygon plots and allometric lines depicting relative brain size beyond vertebrates. a From Jerison [1973]. b From Witmer et al. [2003]. The upper minimum convex polygon in a contains data for mammals and birds, while the lower polygon encloses the information points for reptiles, amphibians, and bony fish (Osteichthyes). Regression lines in a are visually fitted lines with slopes of 2/3, whereas those in b were calculated using not-phylogenetic (i.e., uncorrected) reduced major centrality regression. Notation the 10-fold average difference in relative brain size betwixt Jerison'south "higher" and "lower" vertebrates (0.07/0.007). The minimum convex polygons in b correspond to the ii radiations of extant Reptilia: birds and not-avian reptiles [information from Hurlburt, 1996]. Rhamphorhynchus and Anhanguera are extinct pterosaurs (brain sizes calculated from virtual endocasts obtained using X-ray computed tomography). E and P in a stand for to brain and trunk mass, respectively. In b brain mass was measured in milligrams, whereas body mass was measured in grams.
The most ordinarily used metric in comparative studies of encephalon size is mass: brain mass and torso mass. The problem with a comparing based on mass should exist obvious but has rarely been acknowledged: in a comparing between a bird and a reptile of the same torso mass, the bird tends to be considerably larger than the reptile. This is basically due to several peculiarities of bird anatomy related to flying. The evolution of avian flight was attended by several weight-saving adaptations that take been key to reduce its metabolic costs [Gill, 2007]. Birds, for example, accept an extensive organisation of air sacs extending into the viscera, muscles, and under the skin. Air sacs also opportunistically invade and hollow out the postcranial skeleton, which has the outcome of reducing skeletal mass [Wedel, 2005]. Every bit a issue, most birds have a lower body density than mammals or reptiles [Hazlehurst and Rayner, 1992]. As brain density is the same across all vertebrates [e.thou., Iwaniuk and Nelson, 2002; Domínguez Alonso et al., 2004], a comparison of relative brain size based on mass is necessarily biased and will tend to magnify the differences betwixt birds and reptiles. The same argument applies to comparisons between bird taxa subject field to divergent option pressures affecting body mass or trunk size. For example, aerial predators such as falcons (Falconidae) take a lower trunk density than ground-eating Galliformes such as turkey, grouse, pheasant, and craven [Hazlehurst, 1991; Hamershock et al., 1993]. Thus, a falcon is lighter than a gallinaceous bird of similar trunk size (volume), which introduces bias in calculations of any variable that is expressed equally a fraction of body mass.
Other potentially confounding factors have to do with differences in the body blueprint (bauplan) of different taxa, such equally birds versus non-avian reptiles. Near lizards and crocodiles have a tail that accounts for a large pct of their total body mass [due east.thou. Jagnandan et al., 2014]. The "tail" of a bird is more often than not feathers and, therefore, very light in comparison. Interestingly, many lizards are capable of shedding the tail every bit an antipredator accommodation and therefore spend function of their lives with missing or incomplete tails. As another example, many turtles are encased in a dumbo and heavy carapace, and as a effect tend to be heavier than other reptiles of like body size. This suggests that the large gap betwixt the bird and reptile polygons could, to some extent, simply reflect the fact that the bodies of birds are lighter than expected given their brain mass. Using a dataset of brain and body size for 175 species of extant reptiles and 934 species of living birds, we examine variation in brain size in birds and reptiles and inquire about the consequence of correcting for differences in trunk density and bauplan both across and within these two taxa.
Materials and Methods
We collated data on encephalon and body mass from published literature sources [Crile and Quiring, 1940; Platel, 1974, 1975, 1979; Black, 1983; Amiel et al., 2011]. Only adult individuals of either sex were considered. Literature sources that did non specify the protocol used for brain extraction, preparation (due east.g., removing the meninges), and weighing were not considered. We too excluded reports of brain mass obtained after removal of a part of the encephalon (e.grand., olfactory bulbs, brainstem), or that calculated brain mass/volume based on stereological reconstructions of brain slices. For the snakes Hierophis viridiflavus and Natrix natrix nosotros calculated a weighted average of the data provided in Crile and Quiring [1940] and Platel [1975]. In all remaining cases in which brain and body mass data for the same species were bachelor from different sources, we used the dataset with the largest sample size. Information from the literature were supplemented with unpublished information on brain and trunk mass obtained from the authors (ix species; run across acknowledgements), and with our own information for Podarcis liolepis.
Platel's dataset comprises information on brain and body mass for approximately 60 species of reptiles and has been used extensively in previous analyses [eastward.g., Hurlburt, 1996; van Dongen, 1998], just contains a number of shortcomings that limit its usefulness. Sample sizes are very uneven, ranging from 1 private (e.g., Agama agama) to 88 (Lacerta viridis). For species with small sample sizes, Platel calculated brain mass as the average of the brain masses of all the individuals included in the sample. Nonetheless, for larger samples (>30 individuals), the encephalon mass values reported by Platel are estimates based on measurements taken from a single individual considered "representative" of its species (the predicted brain mass corresponding to an "average developed" of the species) [Platel, 1974]. Nonetheless, as raw encephalon and body mass data are provided in the original papers, in most cases we were able to substitute average values for Platel'due south estimates. Likewise, some of the brain mass values for Australian lizards taken from Blackness [1983; Tables iii–7] were recalculated using all the available adult specimens listed in Appendix I of the thesis and excluding iv species of Varanus for which brain mass was estimated rather than measured.
The density (specific gravity) of fresh brain tissue is close to one in mammals, birds, and reptiles [Jerison, 1973; Hurlburt, 1996]. To calculate the brain volume of the reptile and bird species in our sample we used a common specific gravity of 1.036 g mL–1 [Padian and Chiappe, 1998; Iwaniuk and Nelson, 2002; Domínguez Alonso et al., 2004]. Body volumes for reptiles were calculated from body mass data using a conservative overall tissue specific gravity of one.025 g mL–1, which is the average of specific gravity values for eight species bachelor in the literature [Colbert, 1962; Jackson, 1969; Hurlburt, 1999; Hochscheid et al., 2003; Peterson and Gomez, 2008].
Brain and torso mass data for 934 species of birds were likewise retrieved from published accounts [Armstrong and Bergeron, 1985; Mlikovsky, 1989a–d; Rehkämper et al., 1991; Iwaniuk and Nelson, 2001, 2002; Garamszegi et al., 2002; Iwaniuk and Arnold, 2004; Mean solar day et al., 2005; Payne, 2005; Cnotka et al., 2008; Corfield et al., 2008; Galván and Møller, 2011]. For several species for which we collected more than one data signal we used averages. Bird trunk masses were transformed into volumes using bachelor data for bird body density (online suppl. Tabular array ESM1; for all online suppl. cloth, see www.karger.com/doi/10.1159/000501161). Where body density estimates were available for a given guild, the same density value was applied to all species belonging to that order. For the remaining species we used a form-broad density of 0.718 g mL–1, which is the unweighted average of all the bachelor body density values for birds.
Although previous studies comparing relative brain size in birds and reptiles have not consistently employed phylogenetic command, nosotros performed regression analyses to approximate the magnitude of the allometric scaling (i.e., the slope and intercept) using phylogenetic generalized least squares (PGLS) to account for the non-independence of species information points due to their shared evolutionary history [Symonds and Blomberg, 2014]. Bivariate PGLS regressions were performed with the R [R Cadre Team, 2014] package "caper" [Orme et al., 2012]. Slopes and intercepts were calculated separately for mass, volume, and volume minus tail data (see below). All mass and volume data were log10 transformed before analyses.
Phylogenetic copse for 584 species of birds and 151 species of reptiles were constructed based on the time-calibrated molecular supertrees provided by Burleigh et al. [2015] and Tonini et al. [2016]. As turtles and crocodiles were absent in the Tonini et al. [2016] supertree, none were included in the phylogenetic tree for reptiles. Based on the separate trees for birds and reptiles, we created a supertree that contains both lineages. We used the approach adult by Roquet et al. [2014], which joins the source trees using the R package "ape" [Paradis et al., 2004]. We first obtained the ages since divergence between reptiles and birds from the estimates provided past the TimeTree (www.timetree.org) phylogenetic database [Hedges et al., 2006], which yielded an estimate of difference of 280 mya. Once nosotros set the deviation, the assay calculates the ages of episodes since divergence for the remainder of the nodes to assemble the supertree. This phylogenetic supertree was so employed to exam for significant differences in intercepts and slopes of bird and reptile regression lines (meet below).
Jerison and others estimated the divergence in braininess betwixt birds and reptiles to be 10-fold by looking at the separation between the intercepts of the corresponding regression lines. However, earlier comparing intercepts, it has to be established that the slopes of the regression lines being compared are not significantly unlike [Sokal and Rohlf, 2012]. Therefore, we tested for the equality of intercepts and slopes between birds and reptiles using phylogenetic ANCOVA [Smaers and Rohlf, 2016] and the phylogenetic supertree described above. We compared slopes between brain-body relationships while belongings the intercept constant, so performed tests comparing the intercepts holding the slope abiding. To farther assess differences betwixt slopes, we applied a subsequent test based on a model that holds slopes abiding (and where intercepts vary), against a model in which both intercepts and slopes vary. These analyses were performed using phylogenetic tests implemented in the R packages "caper" [Orme et al., 2012], "phytools" [Revell, 2012], "nlme" [Pinheiro et al., 2018], "geiger" [Harmon et al., 2008], and "evomap" [Smaers and Mongle, 2018]. For all the analyses we show the magnitude of phylogenetic point based on Pagel'due south λ [Pagel, 1999], which estimates the extent to which correlations in traits reflect their shared evolutionary history (every bit approximated by Brownian motion).
To estimate the potential bias that reptile tails may introduce in comparisons of relative encephalon size, we obtained data from the literature on the size of the tail in several lizard species (online suppl. Table ESM2). Nosotros then recalculated the brain book-trunk volume PGLS regressions for birds and reptiles using trunk volume values for lizards excluding the tail (book minus tail). Cadger tail volumes were calculated using family-specific relative tail sizes (i.e., the unweighted average of the available data for species within that family unit; for lacertid lizards we excluded two Takydromus species as this genus exhibits extraordinarily elongate tails). For those families for which nosotros could not become information on tail size nosotros used a relative tail size of 0.22 (the average of all the available data for lizards). We ran a similar assay to approximate the upshot of the turtle shell in relative brain size calculations. In this example, we collected data from the literature on the proportion of torso mass accounted for by the carapace and plastron (online suppl. Table ESM3), and then recalculated the encephalon volume-body volume regression for reptiles using turtle trunk book estimates that did not accept these dense structures into account. Corrected turtle body volumes were calculated using a relative shell weight of 0.32 (the unweighted average of all the bachelor information). Note that nosotros did non correct body volume for Apalone ferox, the only species of turtle in our dataset with a soft shell.
Results
Our expanded dataset comprises brain and body mass data for 175 species of reptiles, which nigh doubles the sample size of previous studies (dataset available from the Damsel Digital Repository: DOI: 10.5061/dryad.pd5hh46). Lizards make upwards 84% of the sample, while snakes (15 species), turtles (9 species), and particularly crocodiles (2 species) are relatively underrepresented. Figure ii shows scatterplots and all-time-fit phylogenetically corrected (PGLS) regression lines for mass and book data. Among reptiles, the everyman relative encephalon weights are plant in turtles and snakes (notation that most of the points for these ii groups fall below the regression line in Fig. 2a). Notwithstanding, correcting the data to account for the dense shell brings the data points for turtles closer to the allometric line (Fig. 2b).
Fig. 2.
Brain-body scaling in a sample of 175 species of living reptiles using: encephalon and trunk mass (a), and brain and body volume (b). Information for brains and bodies are in different units to avoid negative intercepts. The regression lines are from phylogenetically corrected analyses using PGLS. The data for turtles have been corrected in b to account for the dense shell, except in the case of the softshell turtle (A. ferox), which is past far the turtle with the highest relative brain mass in our dataset (crimson arrow in a). Notation how turtles seem to have relatively modest brains for their body size in a (i.e., most points fall beneath the regression line), but not in b. The phylogenetically corrected regression lines in a and b are identical because our phylogenetic tree for reptiles did not include whatsoever turtles.
Figure 3a shows the minimum convex polygons and phylogenetically corrected (PGLS) regression lines for reptiles and for a sample of 934 bird species. For all body sizes bird brains are heavier than reptile brains, but in that location is considerable variation inside both groups. Absolute brain mass in reptiles ranges from 0.0045 g in the Australian skink Lerista muelleri to xv.6 thousand in the American crocodile, Crocodylus acutus. In birds, absolute brain size ranges from 0.167 g in the hummingbird Phaetornis ruber to 44.3 yard in the Emperor penguin Aptenodytes forsteri.
Fig. 3.
Polygon plots showing the distribution of relative encephalon size in birds (upper polygons; n = 934) and not-avian reptiles (lower polygons; n = 175). Phylogenetically corrected (PGLS) regression lines are shown. Note that the polygons for birds and reptiles are closer to each other using book (b) rather than mass (a). The separation of the intercepts of the regression lines corresponds to a half dozen.5-fold difference in brain mass and a 5.4-fold difference in volume between birds and reptiles. a Encephalon mass was measured in milligrams and trunk mass in grams. b Brain volume was measured in microliters and trunk volume in milliliters.
Birds besides have brains that are larger than those of reptiles relative to their torso mass. The magnitude of the departure obviously depends on the species being compared. Thus, the brain of an 80-kg ostrich weighs 41.9 g, while that of a 134-kg crocodile weighs 15.half dozen m – i.e., less than a iii-fold divergence. The crocodile brain represents a mere 0.2% of its body mass; in contrast, the encephalon of a 0.five-kg macaw accounts for 2.five% of its body mass.
Table 1 shows the results of phylogenetically corrected (PGLS) regression analyses. Although results are similar regardless of whether species are treated equally independent data points (information not shown) or if phylogeny is taken into account, the presence of a strong phylogenetic signal (λ > 0.viii) indicates that non-phylogenetic analyses are inappropriate to compare bird and reptile relative encephalon size.
Table one.
Regression parameters for log10-log10 analyses of the human relationship betwixt brain and body size in birds and non-avian reptiles using phylogenetically corrected PGLS regression
Phylogenetic tests returned strongly consistent results for differences between slopes and intercepts. Commencement, tests of the relationship between brain mass variation equally a scaling office of body mass revealed significant differences between birds and reptiles in intercepts holding slopes constant (due north = 735, λ = 0.527, t = 4.46, p < 0.0001), while no significant differences were institute when comparing slopes holding intercepts constant (due north = 735, λ = 0.526, t = 0.129, p = 0.897). A subsequent test based on a model that holds slopes abiding against a model in which both intercepts and slopes vary confirmed that no differences be between slopes (F ii, 731 = 1.12, p = 0.29). The aforementioned analyses performed on brain volume and trunk volume showed meaning differences in intercepts holding slopes constant (north = 735, λ = 0.518, t = four.02, p < 0.0001), while tests of slopes property intercepts constant showed no significant differences between birds and reptiles (north = 735, λ = 0.516, t = 0.209, p = 0.834). The examination of the model that holds slopes abiding against a model in which both intercepts and slopes vary confirmed that no differences be between slopes (F 2, 731 = 1.32, p = 0.25). Finally, the same analyses performed on brain and torso volume were replicated, but after the effect of lizard tails was removed. Consistent with the above results, these analyses showed meaning differences between intercepts holding slopes constant (northward = 735, λ = 0.571, t = iii.39, p < 0.001), while no differences in slopes holding the intercept constant were observed between both clades (n = 735, λ = 0.567, t = 0.439, p = 0.661). This finding was confirmed by a exam of the model that holds slopes abiding against a model in which both intercepts and slopes vary (F two, 731 = three.74, p = 0.054).
The difference in relative brain mass between birds and reptiles, estimated from the separation between the intercepts of the corresponding PGLS regression lines, is 6.v-fold (antilog10 (one.987–1.177) = antilog10 0.81 = 6.46). Substituting brain-trunk volume for encephalon-trunk mass yields similar results only the gap separating birds from reptiles all only disappears. Effigy 3b shows the polygons for birds and reptiles using brain and trunk volume data. Using volume rather than mass, the two polygons abut each other, and the average difference in brain size betwixt birds and reptiles shrinks to 5.4-fold. Recalculating the brain volume-body volume regressions to exclude, for lizards only, the fraction of body volume corresponding to the tail, further reduces the gap between reptiles and birds to 4.8.
Discussion
Our large-scale comparative study provides a case study that highlights a range of potential sources of bias in the interpretation of relative brain size, all stemming from ignoring option pressures that affect torso size, the denominator in most estimates of relative encephalon size.
Filling the Gap in the Evolution of Vertebrate Brain Size
Comparative studies have revealed considerable variation in relative brain size within and beyond vertebrate radiations [Northcutt, 2002; Striedter, 2005]. Birds and mammals have brains that are, all things considered, indisputably larger relative to their trunk size than those of reptiles, but the difference has traditionally been overestimated. Every bit a consequence, the reptilian brain is often stereotyped as small, primitive, and consequently incapable of supporting complex behavior and knowledge. This label is incorrect and based on outdated evidence [meet also Northcutt, 2013].
All the early on studies of brain size evolution, including Jerison's [1973], were afflicted by conclusions being drawn based on small sample sizes. Jerison [1973] used brain and body size data for a mere twenty species of reptiles. Subsequent studies have increased the coverage past calculation more than species from all four reptilian orders (62 species in Hurlburt [1996]; 74 species in van Dongen [1998]), but the bones conclusion regarding overall differences in brain size across vertebrate groups has remained unchanged and the 10-fold figure continues to be authoritatively quoted to depict the gap between birds and reptiles. Here we show that alternative analyses lead to different conclusions. Using an extended dataset and modern phylogenetic comparative methods that allow for a more accurate examination of relative brain size in reptiles nosotros prove that the actual value is 6.5-fold. Substituting the often cited 10-fold brain size departure for a more than realistic six.5-fold difference may seem a modest change, merely worth stressing because that the literature on comparative neurobiology and cognition misrepresents reptiles past describing the divergence betwixt reptiles and birds as more dichotomous and functionally of import in stereotype-consequent ways than is warranted.
Accurate estimates of the allometric equation describing the human relationship betwixt brain and trunk size in living reptiles is important because it is often used to predict levels of encephalization in extinct reptiles such as pterosaurs and dinosaurs [Witmer et al., 2003; Hurlburt et al., 2013]. Our results show that the slope of the phylogenetically corrected regression line relating encephalon and body mass in reptiles (0.579) is far from the 0.67 (ii/three) slope proposed by Jerison [1973] on theoretical grounds, only is similar to the slopes of 0.56, 0.55, and 0.53 reported past Martin [1981], Hurlburt [1996], and van Dongen [1998], respectively. That the slopes for birds and reptiles are not significantly different in whatsoever of the comparisons (mass, volume, and volume minus tail) refutes the notion that bird brains are capable of tracking increases in trunk size more accurately than reptile brains, resulting in a higher brain mass-torso mass slope [Roth, 2013].
Our analyses reveal and quantify several sources of bias in the assessment of relative reptile brain size. In particular, the utilise of overall body mass as a scaling variable in comparisons between birds and reptiles is questionable given the strong directional selection towards reduced body mass in birds [Gill, 2007]. Our results show that using brain volume and body volume instead of mass brings the polygons for birds and reptiles even closer, reducing the boilerplate difference in relative brain size to 5.4-fold (compare Fig. 3a, b). Further reductions can be obtained by correcting for constraints imposed by the dissimilar bauplan of reptiles and birds (tails in crocodiles and lizards, dense shells in turtles), highlighting the demand to take into account differences arising from selection pressures affecting body design. For example, van Dongen [1998] attributed the comparatively small brains of turtles to them being "archaic," rather than to the obvious fact that the shell increases the mass of turtles beyond what one would expect for a reptile of their body size. In fact, softshell turtles with their leathery shell – presumably much lighter than the shell of other turtles – are the only Chelonians in our dataset with a encephalon mass higher than expected for their trunk mass, and an outlier among the turtles unless a correction is made to business relationship for the vanquish of other species (Fig. 2). Taking the higher up considerations into business relationship, the traditional chasm in relative brain size between birds and non-avian reptiles shrinks considerably, which altogether paints a much more coherent and parsimonious flick of the evolution of encephalon size within Reptilia.
Interspecific variation in the option pressures affecting body size, such every bit those relative to trunk density or bauplan, are not limited to comparisons between birds and reptiles, but more often than not applicable across all vertebrates [Smaers et al., 2012] and invertebrates [Wehner et al., 2007; Polilov and Makarova, 2017]. For example, cormorants (order Suliformes) are excellent defined, reaching depths of more than forty yard thanks, among other adaptations, to a particularly depression buoyancy (i.due east., high density) for a bird [Ribak et al., 2004]. Relative brain mass places Suliformes shut to the regression line which suggests they have brains of the expected size for their body mass (Fig. 4a), but taking volume (and hence buoyancy) into account reveals that they actually have larger brains than expected (Fig. 4b). Galliformes, which also have dense bodies (come across EMS1), have relatively small brains in a comparison based on mass (i.e., most points fall below the regression line), but their brains are closer to the expected size because volume (Fig. four). Similar arguments can be fabricated for other orders of birds, and surely for many species of mammals. Giraffes, with their incredibly long necks, are an excellent example of how differences in bauplan can affect estimates of relative encephalon size. Giraffidae have only two extant species, the long-necked giraffe (Giraffa camelopardalis) and the short-necked okapi (Okapia johnstoni). While the giraffe is often described equally a mammal with extraordinarily depression relative encephalon size [Graïc et al., 2017; Raghanti et al., 2017], the brusk-necked okapi harbors a brain that is, unsurprisingly, about the expected size for an Artiodactyl of its body size. All the same, the difference between giraffe and okapi is not so much in the encephalon (their brains have roughly the same absolute mass; Fig. 5) as information technology is in the long and heavy neck, which makes up a large proportion of a giraffe'southward body mass [Simmons and Scheepers, 1996].
Fig. four.
Polygon plots showing the distribution of relative encephalon size in four orders of birds using: brain and torso mass (a), and brain and torso volume (b; body volume calculated using specific gravity information for each of the iv different orders; see online suppl. Table ESM1). The overlay polygon encloses the data points for all the birds in our dataset. The regression lines are from analyses of all the bird data. Notation that correcting for their unusually loftier body density brings the data points for Suliformes further in a higher place the allometric line, suggesting they have brains that are relatively large for their body size. a Brain mass was measured in milligrams and body mass in grams. b Encephalon volume was measured in microliters and trunk book in milliliters.
Fig. 5.
Comparison of brain and body sizes in giraffes (Grand. camelopardalis) and the only other member of the Giraffidae, the okapi (O. johnstoni). Relative brain size in giraffes is considered extraordinarily minor for a mammal. The average relative brain mass for the few specimens for which brain and body size data are bachelor is 0.097% of body mass, with a maximum recorded estimate of 0.13% [Black, 1915; Crile and Quiring, 1940; Graïc et al., 2017]. Encephalon size data for the short-necked okapi are by and large missing, with a single estimate of 466 k [Black, 1915]. Given that adult body size in okapi ranges from 200 to 350 kg, this would yield a relative brain mass of 0.13–0.23%, considerably larger than for their long-necked closest relatives.
Recommendations for Future Studies of Relative Encephalon Size
We argue that future studies should not translate differences across species or larger groups exclusively in terms of selection pressures acting on the brain just also look for differences in body density and bauplan that could potentially distort comparisons based on relative encephalon size. Comparative studies of brain evolution should be alert to peculiarities of the taxa under written report, such every bit the possession of extremely long tails or large fat deposits, which may introduce systematic biases in calculations of body size. Although big, heavy bodies unremarkably harbor large brains, the relationship between brain and body size is rather noisy (Fig. one). A consideration of the bauplan of the species involved is necessary in order to disentangle phylogenetic and other sources of variation. Most snakes and legless lizards (e.k., Anguidae), for example, have pocket-sized brain weights for their body weight (Fig. 2). However, this is likely the result of choice for a highly elongate torso form rather than choice for minor brains [van Dongen, 1998].
From Relative Brain Size to Cognition: A Cautionary Note
Variation in brain size has been notoriously hard to interpret [Healy and Harvey, 1990; Healy and Rowe, 2007; Chittka and Niven, 2009], and the link between brain size and cognition remains one of the thorniest issues in comparative neurobiology. However, many still consider that relative brain size is a robust proxy for full general cognitive ability [Pollen et al., 2007; Burkart et al., 2017; Fristoe et al., 2017; Iwaniuk, 2017]. In accordance with this hypothesis, the much cited 10-fold difference in brain size has often been used to justify the cognitive superiority of birds relative to reptiles. Birds, with their relatively large brains, are currently considered on a par with mammals every bit far equally their behavioral and cognitive complexity [Emery and Clayton, 2004]. In contrast, reptiles are widely considered, despite arable prove to the contrary, cognitive underachievers. Rather than questioning the rationale behind this conclusion, many authors take uncritically assumed that their small brains must condemn reptiles to a life of cognitive mediocrity. Snakes have been described equally incapable of integrating information from different sensory modalities [Sjölander, 1995; Gärdenfors, 2003], crocodiles equally devoid of whatsoever emotion [MacLean, 1985], and turtles as just manifestly stupid [Robin, 1973]. The unfortunate effect of this misperception is that the cognitive abilities of reptiles are rarely tested. This, in turn, reinforces the notion that sophisticated cognition is all but absent in this grouping. In a contempo review, Güntürkün and Bugnyar [2016, p. 292] ended that "although reptilian cognition should non be underestimated, nothing at the level and scope of bird cognition has been reported for this animal group so far." This is typical of much electric current thinking in comparative noesis. Nonetheless, the persistent myth of the sluggish, primitive, stupid reptile is increasingly out of pace with reports describing examples of complex behavior and sophisticated noesis in many species of reptiles [Wilkinson and Huber, 2012; Burghardt, 2013; Doody et al., 2013]. That the brains of reptiles are not as large as those of birds and mammals makes their report fifty-fifty more interesting and brings virtually the challenge to explicate how the relatively small-scale brains of reptiles are capable of supporting their sophisticated behavior and cognition, not the other way effectually.
Still, speculations regarding cognitive abilities based solely on comparative brain size data are bound to lead u.s.a. astray. Invertebrates, with their miniaturized brains, are a example in point. Many insects bear witness remarkably sophisticated beliefs and cognition, yet their brains are staggeringly pocket-size compared to those of vertebrates [Chittka and Niven, 2009]. Although the argument has been used mainly in the context of vertebrate-invertebrate comparisons, it should as apply to comparisons among vertebrate groups: cognitive achievements do non strictly depend on the possession of relatively big brains. A large encephalon is thought to confer more intelligence because more brain tissue increases the computational capacity supporting behavioral and cognitive complexity. However, in vertebrates the correlation between encephalon size and cognitive ability is weak both intraspecifically and interspecifically [Healy and Rowe, 2007; Herculano-Houzel et al., 2014]. Furthermore, recent work by Olkowicz et al. [2016] has shown that birds accept roughly twice as many neurons in their forebrain equally mammals of like encephalon mass. In fact, their written report challenges deeply ingrained notions nearly the supposed cognitive superiority of primates, which to date largely relied on data on relative encephalon size. For example, a raven has the aforementioned number of neurons in the pallium of its ten-g encephalon as a capuchin monkey in the cortex of its 39-grand brain, and a blue-and-yellow macaw packs more neurons in the pallium of its 14-one thousand encephalon than a macaque monkey in the cortex of its 70-g encephalon [Olkowicz et al., 2016]. This suggests that the packing density of neurons in some telencephalic areas, rather than brain size, may explain the sophisticated cognition found in some birds such as parrots and corvids. Unfortunately, the same analysis has non been conducted with reptiles or other vertebrates, such as fish, which are equally capable of sophisticated behavior and noesis [Brown et al., 2011].
Acknowledgements
We thank the following colleagues for generously sharing with us their unpublished data on lizard brain size: Brian J. Powell and Manuel Leal (Anolis spp.), Lauren M. Davis and Michele A. Johnson (Coleonyx brevis, Hemidactylus turcicus), and Daniel Robert Pfau and Christy Strand (Sceloporus occidentalis). The helpful comments of G. Striedter and two bearding reviewers are greatly appreciated. Matt Kramer helped with the logarithms.
Argument of Ideals
The authors have no ethical conflicts to disclose. No experiments on living animals were performed, and no animal was euthanized purposefully for the work reported here.
Disclosure Statement
The authors declare no conflicts of interest.
Enrique Font Ethology Lab, Instituto Cavanilles de Biodiversidad y Biología Evolutiva University of Valencia, APDO 22085 ES–46071 Valencia (Spain) Eastward-Mail enrique.font@uv.es Get-go-Page Preview Received: July x, 2018 Number of Print Pages: 14 ISSN: 0006-8977 (Print) For additional data: https://www.karger.com/BBE Copyright: All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or past any ways, electronic or mechanical, including photocopying, recording, microcopying, or past any data storage and retrieval system, without permission in writing from the publisher.
Author Contacts
Article / Publication Details
Accustomed: May 22, 2019
Published online: August 22, 2019
Issue release date: November 2019
Number of Figures: 5
Number of Tables: 1
eISSN: 1421-9743 (Online)
Copyright / Drug Dosage / Disclaimer
Drug Dosage: The authors and the publisher take exerted every effort to ensure that drug selection and dosage set along in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant period of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and production references in the publication is not a warranty, endorsement, or approving of the products or services advertised or of their effectiveness, quality or condom. The publisher and the editor(s) disclaim responsibleness for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.
Source: https://www.karger.com/Article/FullText/501161
0 Response to "Which Family of Birds Has the Largest Brain to Body Ratio"
Postar um comentário