Climatic-environmental influences on hominin brain size over the last 5 million years

Climatic-environmental influences on hominin brain size over the last 5 million years | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Climatic-environmental influences on hominin brain size over the last 5 million years View ORCID Profile Samuel L. Nicholson , Thomas A. Püschel , Joanna Baker , Chris Venditti doi: https://doi.org/10.1101/2024.09.09.611970 Samuel L. Nicholson 1 Climate Geochemistry, Max Planck Institute for Chemistry , Mainz, Germany 2 Ecology and Evolutionary Biology, University of Reading , Reading, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Samuel L. Nicholson For correspondence: samuel.nicholson{at}mpic.de c.d.venditti{at}reading.ac.uk Thomas A. Püschel 2 Ecology and Evolutionary Biology, University of Reading , Reading, UK 3 Institute of Human Sciences, School of Anthropology and Museum Ethnography, University of Oxford , UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site Joanna Baker 2 Ecology and Evolutionary Biology, University of Reading , Reading, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site Chris Venditti 2 Ecology and Evolutionary Biology, University of Reading , Reading, UK Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: samuel.nicholson{at}mpic.de c.d.venditti{at}reading.ac.uk Abstract Full Text Info/History Metrics Supplementary material Preview PDF Abstract A large brain relative to body mass is considered a distinguishing hominin trait. It has frequently been related to a suite of social, behavioral, technological, and other cognitive adaptations that differentiate humans from other species. The processes underlying large brain size evolution have therefore been a subject of rigorous scientific debate. Many hypotheses have been proposed to explain how climate and environment drive the selection of larger brain sizes, but monotonic influences of climate-environmental selective pressures are often assumed and rarely have between- and within-species effects been considered. Here, we apply Bayesian phylogenetic comparative techniques to the hominin fossil record to test the effect of climatic and environmental pressures (C-E) on brain size evolution, whilst simultaneously accounting for body mass and chronological age. We find that colder and more variable temperatures have a positive within-species effect on brain size evolution, likely related to biological adaptations to mitigate against hypothermia. However, in Homo, the strength of this effect diminishes over time suggesting that in later species ( Homo sapiens and Homo neanderthalensis ) brain sizes were less affected by C-E conditions. Introduction Relative brain size is a particularly important trait as it is often used as a proxy for cognitive abilities 1 – 4 . It is widely reported that over the last few million years, relative brain size in hominins has increased, culminating in the iconic large brains of our own species 5 . However, relative brain size increase across hominin evolution arose from gradual change within individual species over time 6 . Hence, we must take a different approach from much previous research which only seeks patterns across species to truly understand the ecological drivers of brain size increase in hominins 4 , 7 , 8 . Climatic and environmental (C-E) pressures have long been assumed to have played a crucial role on human encephalisation 3 , 4 , 9 – 14 . Consequently, multiple hypotheses have been proposed to explain the role of C-E variables (e.g., precipitation, temperature, vegetation) on hominin brain size evolution 3 . However, these hypotheses have been traditionally framed in ambiguous terms, leaving it unclear how they could be tested and with which data. More recently, Will et al. 3 explicitly outlined hypotheses and associated expectations of how cranial capacity could be predicted from suites of so-called bioclimatic variables summarizing temperature and precipitation as well as a variable describing vegetation (henceforth net primary productivity, NPP). Briefly: The environmental stress hypothesis suggests that resource-deficient environments may induce stress-related brain size increases 3 , 15 whereas the contrary environmental constraints hypothesis suggests that resource-rich environments are more likely to support an expensive, larger brain 3 . The environmental stress and environmental constraints hypotheses specifically predict opposing effects of temperature, precipitation and NPP on brain size. The environmental variability hypothesis predicts that increased cognitive abilities at short-time scales (or adaptive flexibility at longer timescales) are required to tolerate fluctuating environments 12 whereas the environmental consistency hypothesis argues that climatic and environmental stability are more suited to maintaining large and metabolically costly brains 3 , 8 . The environmental consistency and environmental variability hypotheses make opposing predictions for variation in rainfall, temperature and NPP. All four hypotheses clearly outline the importance of either low/fluctuating resources or high/stable resources on varying timescales and make explicit predictions based on C-E data. Whilst different studies find support for different hypotheses across the hominin radiation 3 , 4 , 8 , 15 , the data underlying the expectations of all hypotheses are not independent from one another (e.g., 16 ; Fig. S1), preventing clear conclusions from being drawn. Although bioclimatic variables and NPP are commonly used in studies of past environments and ecologies of extinct species, it is not possible to separate the effects of certain aspects owing to high levels of collinearity 17 . For instance, recent work has demonstrated that temperature, NPP, and precipitation all have similar impacts on body mass 3 . In short, the data underlying the predictions of current hypotheses seeking to explain the environmental drivers of hominin encephalisation are inextricably linked. It is necessary to understand this further before it is even possible to derive new, unbiased, and unambiguous environmental hypotheses of hominin brain size evolution. Here, we take a unified approach to resolve outstanding questions about the influence of the climate and environment on hominin encephalisation. We reconstruct C-E data describing the quantity and variation of NPP, rainfall and temperature (including 17 bioclimatic variables; see SI and methods for full list of variables) for the given age and coordinates for each of 284 hominin specimens with cranial capacity using a climatic emulator spanning the last 5 million years of hominin evolution (see methods). This allowed us to specifically associate every analysed specimen with the C-E conditions of where they lived. To avoid collinearity among our predictors and reduce the dimensionality of our dataset we generate independent variables using principal component analysis (PCA). We then use our new orthogonal axes of C-E variation to formulate new hypotheses that can be tested in a Bayesian phylogenetic generalised linear mixed modelling (PGLMM) framework. To date, previous tests of the C-E drivers of hominin encephalisation have failed to 1) account for shared ancestry, 2) incorporate alternative phylogenies ( 6 ; Fig. 1B ), 3) allow slopes to vary both between and within species, and 4) simultaneously account for within- and between-species effects of body mass and chronological age 6 . Our approach combines a reformulation of the C-E hypotheses and bioclimate proposed to drive hominin brain size evolution. We provide a robust novel perspective on encephalisation in hominins and the C-E conditions in which this occurred, whilst accounting for numerous sources of uncertainty (see methods). Download figure Open in new tab Figure 1. (A) Map of cranial capacity specimens used in this study and (B) Phylogenetic trees of the hominin radiation over the last 5 Ma. Thin lines represent a random sample (500) of alternative topologies, whilst the thicker topology corresponds to the hominin maximum credibility phylogenetic tree (MCC). Homo is colored in red, while the rest of the hominin genera are in blue. Results Climate principal component analysis We reconstructed monthly mean temperature, precipitation and NPP for the last 5 Ma of hominin evolution (at 1-kyr intervals with a 0.5° resolution, see Methods) using the PALEO-PGEM climatic emulator 18 . From precipitation and temperature data, so-called “bioclimatic variables” - which describe the annual distribution and variation of rainfall, temperature and are independent of calendar months (for example, Bio-5 represents the temperature of the warmest month, rather than the temperature of August) and are widely used in ecological studies - were calculated following the ANUCLIM classification 19 . Additionally, to study long-term variance, we calculated the coefficient of variation for annual mean temperature, precipitation and NPP using a window of 23-kyrs to represent precession-scale variability (see SI for a complete list of these C-E variables). We then used point sampling procedures to extract these variables for each one of the hominins under study. We refer to all our data including the bioclimatic variables, NPP, and coefficients of variation as climatic-environmental (C-E) variables collectively. We conducted PCA on a dataset comprising the C-E variables for 284 hominin specimens with cranial capacities (see methods). This process was repeated over a sample of 1,000 datasets, which allowed us to incorporate uncertainties related to body mass estimates, temporal range, and taxonomic assignment ( 6 ; see methods). Our first axis, PC1, explaining 36% of the variation in the data ( Fig. 2A ), is predominantly associated with temperature and can broadly be interpreted as a spectrum: low values describing warm environments with reduced temperature variability and range but with precipitation variability at both seasonal and long-term scales; up to high values, describing colder environments with greater temperature variability and range but with both annual and long-term precipitation stability ( Fig. 2B ). The second axis, PC2, accounts for ∼ 27% of the variance and is weighted more heavily towards precipitation: higher PC2 values indicate greater precipitation and NPP whereas lower values indicate drier environments. Download figure Open in new tab Figure 2. (A) Variance of the data explained by PCs. (B) Loadings of the bioclimatic variables on PC1 and PC2. Numbers relate to their respective bioclimate variables (see Tab. S1 for full list of variables). All values represent the mean over 1000 datasets. (C) Conceptual illustration of our new hypotheses for C-E influencers of brain size: i) the cold-variable hypothesis (where colder or more variable temperatures influence encephalisation), ii) the warm-stable hypothesis (where warmer or more stable temperatures influence encephalisation), iii) the mesic hypothesis (where wetter and more productive environments influence encephalisation), and iv) the xeric hypothesis (where drier and less productive environments influence encephalisation). Seasonal variation is unimportant for PC2. Only these two first PCs were used in the subsequent analysis based on a parametric bootstrap method that was applied to determine the number of PCs to be kept 20 . The Climatic-Environmental drivers of hominin encephalisation To test the expectations of the hypotheses we outline above, we conduct a phylogenetic generalised linear mixed model (PGLMM) testing the effect of each of our two PC variables on cranial capacity and whether this has been different within individual species. Our PGLMM approach enables us to account for body mass and chronological age whilst also allowing us to understand whether patterns differ both across and within individual species, taking into consideration phylogenetic relatedness. We find a significant positive relationship between cranial capacity within-species PC1 ( Fig. 3 ), whilst there is no significant effect of PC2 on cranial capacity. Neither PC1 nor PC2 predicts brain size increase across species. In all models studied, we find a high phylogenetic signal (h2 = 0.88; see methods) and an overall good fit (R2marginal = 0.66; R2conditional = 0.95; see methods). Download figure Open in new tab Figure 3. Fig. 3 . (A) Within-species slopes (color lines) and overall, within-species slope (black line) for Scaled log10 cranial capacity vs. PC1. (B) Mean within-species slope and error (vertical lines) and species temporal durations (horizontal lines) through time for Homo species. Only species with >9 cc specimens are displayed. (C) PCA biplot of hominin cranial capacity specimens. Points represent the mean PC value over 1,000 datasets. H. naledi and H. floresiensis are not plotted on B as these do not have sufficient known temporal spans nor geographic information to accurately estimate their slopes. Discussion The influence of predictor variables on hominin cranial capacity can invoke different modes of evolution (Fig. S2). Age and body mass have recently been studied and found that whereas body mass has a between-species effect, species age has a variable within-species. In other words, brain size increases occur at speciation events coeval to body mass increases and variably throughout the duration of a species (Fig S3: e,j and o). Here, we show that PC1 mostly has a positive effect on cranial capacity but there is a significant relationship within species ( Fig. 3A ). This reveals a generally consistent trend supporting the predictions of the cold-variable hypothesis: within individual species, colder and/or more variable environments drive brain size increase, owing to stress. Furthermore, as between-species PC1 did not predict relative cranial capacity, we show that step changes in species average PC1 values do not influence brain size. Instead, it is the colder temperatures experienced in a species own range that enforce intraspecific increases in relative brain size. Variable within-species slopes for PC1 show that there is no monotonic influence of PC1 on relative brain size across the whole phylogeny: different species are affected differently by cold stress. Indeed, inversely, Au. africanus and H. neanderthalensis have seemingly negative slopes, suggesting warmer temperatures influence relative brain size in these species ( Fig. 3A ). Like previous findings 6 , between-species body mass and within-species age (with variable random slopes) also remain significant predictors of cranial capacity in hominins in our analysis. Another pertinent observation is that age remains significant, even after accounting for body mass and C-E conditions, implying that other processes throughout the duration of a species’ existence also influence relative cranial capacity increases. In other words, C-E conditions are not the only means to increase hominin cranial capacities. Unknown factors not studied here have additional influence and future investigation will be required to elucidate these. Together, these reveal a pattern whereby increases of cranial capacity aligned with body mass increases occur speciation events (Fig. S2: j); subsequent within-species encephalisation occurred due to colder conditions ( Fig. 3A ) and unknown factors consistent with age. But how do low/variable temperatures induce brain size increases? Previous attempts to directly associate colder temperatures and/or temperature variability with encephalisation in primates and hominins have focused on biological 8 , 21 – 23 , social 23 and techno-cultural solutions to prevent hypothermia. It is tempting to link this to Bergmann’s rule, where larger organisms are typically related to higher latitudes. Indeed, temperature correlates with latitude 24 . However, Bergmann’s rule is related to total body mass, which was explicitly accounted for in our analysis. Furthermore, we ran an independent test for relative latitude (i.e., degrees distance from the equator) and, contrary to other studies 8 , it was not a significant predictor of hominin brain size within- or between-species (Tab. S1). Moreover, most attempts to link temperature to encephalisation have also assumed that temperature controls “resources”, where colder/variable environments are more resource-scarce/variable, thus creating selective pressure for enhanced cognitive capabilities 3 . However, we found a low correlation between temperature indices and NPP (i.e., a proxy for productive environments; Fig. S1); NPP neither loads highly on PC1, nor does PC2 (with a high NPP loading) predict hominin brain size. Instead, NPP is highly correlated with precipitation indices (Fig. S1). Our results suggest the direct effects of ambient absolute temperature/variability influence brain size evolution. One potential mechanistic explanation for the observed relation between relative brain size and PC1 values representing temperature can be advanced by the so-called “radiator hypothesis” 21 . This hypothesis stipulates that a network of emissary veins evolved to cool the brain under conditions of hyperthermia, and diploic veins that provide a mechanism for cooling have been associated with larger brains. Colder environments typically have greater seasonal temperature oscillations; fluctuations between hyper- and hypothermic conditions within a species’ environment may have provided a mechanism for encephalisation by demanding an enhanced diploic vein system for effective thermoregulation 8 . Whilst the radiator hypothesis is heavily linked to the evolution of bipedal posture, climate may have influenced its evolution by providing a stimulus for more efficient thermoregulation 21 , particularly for those species with negative PC1 slopes. But the radiator hypothesis primarily describes the prevention of hyperthermic conditions 21 – what about colder temperatures? Recent studies have shown that vertebrates typically have larger brains in colder conditions, even when accounting for body mass 25 . Additionally, large brains in Asian colobine monkeys have been shown to have evolved during colder events and resultant cold adaptations may have influenced social organisation 23 . In H. sapiens larger brain sizes seemingly occurred with globally cooler periods 15 . If colder events influence brain size, we must establish the limits at which temperature stress could occur. Past studies have provided minimum sustainable temperatures for Homo erectus of 11.67°C and 6.27°C 26 , 27 . With the exemption of tropical latitudes, this is well within the temperature variation experienced during cold events (Fig. S2). Assuming, based on smaller body masses, this value was higher in earlier species, cold may have been a long-term stress throughout hominin evolution. Thus, this explanation would work within-species, as it would take effect within any species’ own geographic range where colder areas are more likely to induce species-specific stress. We also show that global mean annual temperature did not predict brain size (Tab S2), reinforcing our suggestion for an influence of temperature in local environments. Yet, further information on the minimum sustainable temperatures across the hominin radiation is required to truly elucidate the mechanisms of cold stress-induced encephalisation. Previous studies 3 found that, whereas body mass evolved as a phenotypic adaptation to cope with short-lived low temperatures, brain size was related to variations in precipitation and NPP. Our results however, which explicitly accounted for the correlation between cranial capacity and body mass, rather show that relative brain size also evolved to cope with stress induced by short-lived low temperatures, but this effect varied between species. The exact processes of how colder-variable temperatures influence brain size is not clear and likely complicated. Owing to the high correlation of temperature variables, a simple “cold causes larger brains” does not sufficiently explain the within-species variance of the effect and its influence on even low-latitude hominin species. One stark observation that we make is that, within later species of Homo, PC1 is having a lesser impact on encephalisation. There is a significant correlation (r = 0.95, p = 0.02) between median age (ka) and the median slope estimation. The median slope estimates decline from ∼0.06 in Homo ergaster to ∼0.01 in Homo sapiens , and fall to ∼-0.01 in Homo neanderthalensis . However, H. sapiens and H. neanderthalensis have a more rapid encephalisation throughout their temporal duration when compared to earlier species 6 . One potential climatic explanation for this encephalisation is that increased temperature variability related to greater glacial-interglacial amplitudes, following the Mid-Bruhnes event (∼460 ka) could have enhanced climatic pressures on later Homo species 28 . Indeed, broad narratives of global Quaternary climate change include a transition from orbitally-paced 41-kyr “obliquity” to 100-kyr “eccentricity” interglacial-glacial cycles between 1.2-0.7 Ma 29 – 32 , followed by increased amplitude and duration of glacials following the Mid-Bruhnes event 28 . These changes are seemingly coeval to increases in maximum hominin brain sizes. We therefore conducted further PGLMM analyses using the moving Coefficient of Variation on orbital cycles (precession [long]: CoV23-kyr; obliquity: CoV41-kyr; eccentricity: CoV100-kyr) for Bio-1 (Mean Annual Temperature), Bio-12 (Annual Precipitation) and NPP (Net Primary Productivity) (methods). Neither the CoV of precipitation and NPP on precession (∼23-kyrs), obliquity (∼41-kyrs) and eccentricity (∼100-kyrs) timescales (Fig. S3) predicted relative brain size (Tab. S3). On the other hand, Bio-1 CoV100-kyr was found to be significant. Additional testing, however, found a strong correlation with PC1 (0.75) and these effects cancelled each other when modelled in tandem (Tab. S3). The effects of PC1 and Bio-1 CoV100-kyr are therefore indistinguishable, and we cannot relate this Mid-Late Pleistocene rapid encephalisation to enhanced glacial-interglacial climatic variability in temperature, precipitation or NPP. It is likely that alternative factors are influencing encephalisation within these later hominin species. One consideration is that these later Homo species have large geographic ranges 33 . Preceding species, such as H. ergaster (southern and eastern Africa) and H. erectus s . s . (south-eastern and eastern Asia) appear to have more restricted regional distributions. Conversely, depending on controversial species assignments, H. heidelbergensis may have been distributed across Africa, Europe, and Asia. Likewise, Neanderthal specimens were recovered in Europe and south-western Asia and H. sapiens specimens were on every continent (except Antarctica) by the end of the Pleistocene (11.7 ka). Both Neanderthals and H. sapiens plot more widely on PC1 and PC2 ( Fig. 3C ) and reflect greater C-E variation. Such a range of C-E conditions may have abated the importance of any one climatic metric, suggesting that the multiplicity of climates and environments may have contributed to Mid-Late Pleistocene encephalisation. Another consideration is that techno-cultural (i.e., fire, clothing, shelter-building, other survival/subsistence strategies 26 , 34 – 36 ) and socio-behavioral innovations (i.e., language, symbolism, group-size changes 37 ) may have provided mechanisms (e.g., niche construction) that abated the C-E effects influencing encephalisation, or perhaps were alternative inducers of encephalisation.Understanding this trend should be a target for future research, where the effect of biogeographic and C-E range throughout a phylogeny can be tested against brain size. As shown here in our work, countless climatic hypotheses can be envisioned but only a more limited number of them can be statistically assessed using the current evidence from the fossil record 3 , 4 , 8 . From our reformulation of C-E encephalisation hypotheses, our analysis supports a scenario in which absolute temperature and temperature variability within a species’ own geographic range influenced encephalisation. Importantly, the differential effects of climate at the species level mitigate against broad narratives of climatic impacts on hominin evolution, such as previously proposed hypotheses 4 , 11 – 13 , 38 , 39 . Our results show a diminishing impact of climate within the later species of large-bodied Homo, including H. sapiens , as total geographic ranges expanded and technological/behavioral innovations increased 34 . We, therefore, emphasise against a simple determinist view of climatically driven cognitive advances for these species. Our findings contrast recent studies, and we highlight the necessity of studying climate impacts on cranial capacity whilst accounting for between- and within-species effects, phylogenetic relatedness, body mass and chronological age within a single inclusive framework. Materials and Methods Hominin sample and phylogenies We used the fossil dataset of hominin specimens with cranial capacity from Püschel et al. 6 (supp. data 1). However, we excluded S. tchadensis , as the PALEO-PGEM climate emulator does not extend back to its reputed age of ∼7.2 Ma 40 . This resulted in a sample size of 284 hominin specimens ranging from ∼ 4.4 Ma to the end of the Pleistocene. We associated every specimen with cranial capacity to a body mass value and chronometric age using well-defined criteria (see 6 for details), which allowed us to consider uncertainties related to both body mass estimates, taxonomic assignment, and temporal range. This process was repeated 1,000 times resulting in 1,000 unique datasets. We also provided geographical coordinates, recorded as longitude and latitude 6 , for each one of the hominin specimens under study. A sample of 1,000 phylogenetic trees was obtained from Püschel et al. 6 (supp. data 2). These trees were randomly sampled from a posterior distribution of phylogenies obtained using a ‘combined-evidence’ Bayesian phylogenetic reconstruction of hominin species using stratigraphic, molecular, and morphological data. These phylogenetic trees were used in our subsequent analyses (see PGLMM section). Climate-Environmental Data We reconstructed environmental data for the last 5 Ma starting with climatic variables generated using the PALEO-PGEM climatic emulator. PALEO-PGEM applies a Gaussian process emulation to singular value decomposition of ensemble runs of the intermediate complexity atmosphere-ocean general circulation model PLASIM-GENIE21,42. Boundary condition forcings include atmospheric CO2 (after 800 ka 41 ; before 800 ka 42 ), sea-level (as a proxy for ice-sheet) 42 and orbit (obliquity and precession-modulated eccentricity) 43 . Orography is fixed. The native-resolution (5°) emulations have previously been validated against of four discrete periods: the mid-Holocene, the Last Glacial Maximum, the Last Interglacial, and the mid-Pliocene warm period 18 , 44 . The PALEO-PGEM climate emulator was run via R (R version 4.1.2 45 ) on the default gaussian-exponential setting, following the same procedures as Raia et al. 44 . In this setting, the PALEO-PGEM uses a power exponential covariance function, the mean GP prediction and ten principal components. Spatial fields of temperature, precipitation and annual Net Primary Productivity are then emulated at 1,000-year intervals, driven by time series of scalar boundary condition forcing, and assuming the climate is in quasi-equilibrium 18 , 44 . However, unlike Raia et al. 44 – who ascertained minimum seasonal temperature, maximum seasonal temperature, minimum seasonal precipitation, and maximum seasonal precipitation – we calculated monthly temperature and precipitation values. PALEO-PGEM was run ten times for each monthly temperature and precipitation variable and a mean was taken for the analysis. Following this, outputs were converted to 1-kyr rasters of 5° resolution using a Thin Plate Spline interpolation (e.g., 46 ). These were then downscaled to 0.5° resolution and bias-corrected using the respective climate rasters averaged over the 1981-2010 period available from the CHELSA database48–50 47 – 49 . Downscaling of temperature, precipitation and NPP followed previously established procedures 18 , 44 , where an additive correction was used for temperature and a conditional hybrid (additive or multiplicative) correction was used for precipitation and NPP (see 18 , 44 for full description) using the following formulae: Additive correction: Multiplicative correction: Where Emt is the pre-downscaled emulator output at time (t), Em0 is the pre-downscaled emulator at 0 ka, Cm0 is the climate variable at 0 ka by which downscaling should take place, and Cmt is the downscaled climate variable at t. For the hybrid procedure, an additive correction is employed if Cm0>Em0; whereas a multiplicative procedure is employed if Cm0<Em0. This downscaling procedure aims to minimize negative precipitation values in hyper-arid (deserts) areas and unrealistically high values in hyper-humid areas (tropical rainforests). However, where we identified instances of negative precipitation values in deserts, these were manually adjusted to 0. Raw outputs from the emulator are measured as Celsius and mm day -1 for monthly temperature and precipitation, respectively, and kg C m2 s -1 (kilograms of carbon per square meter per second) for annual NPP. These were transformed to monthly Kelvin (temperature), monthly absolute mm (which in this case equates to mm month-1) for precipitation, and g C m 2 yr -1 (grams of carbon per square meter per year) for NPP. Following downscaling, we calculated 17 of the 19 bioclimatic variables defined in the ANUCLIM classification 19 . Two bioclimatic variables, Bio-2 (Mean Diurnal Range) and Bio-3 (isothermality) 19 , were not calculated as these require monthly minimum and maximum temperatures and presently PALEO-PGEM can only predict monthly mean temperatures. Additionally, for bioclimatic variables where monthly minimum or maximum temperature are given (Bio-5: maximum temperature of the warmest month; Bio-6: minimum temperature of the coldest month; Bio-7: annual temperature range (Bio-5 – Bio-6), these are instead mean temperatures of the hottest/coldest month. Next, we calculated the moving Coefficient of Variation (CoV = standard deviation/mean) for annual mean temperature and precipitation, as well as NPP using a window of 23-kyrs to represent long precession-scale variability. This resulted in a total of 21 variables that were used to characterize past environments. We then used coordinate and age data to sample the appropriate location on each 1-kyr map of each of the 21 environmental variables, which allowed us to link every specimen from the 1,000 unique datasets to the conditions of where they lived. This procedure resulted in 1,000 datasets which included each specimen with a chronometric age (constrained by the specimen age range) with associated cranial capacity, body mass and environmental data (supp. data 3 and supp. data 4). Cranial capacity, body mass and all environmental variables (excluding CoV) were log10 transformed. Finally, all data was standardized (i.e., centered and scaled) prior to analysis. Environmental PCA We carried out a PCA using all the environmental variables to reduce the dimensionality of our dataset and avoid predictor collinearity using the R (version 4.1.2) statistical software package 45 . This PCA was performed using the correlation matrix of the 17 bioclimatic variables, NPP and CoV23-kyr for Bio-1, Bio-12 and NPP (Fig. S1), which allowed us to generate new orthogonal axes of variation (PCs) that were subsequently used as independent environmental variables in our analyses. To determine the number of PCs to be retained in our further analytical steps, we applied a full parametric bootstrap method recommended for standardized variables 20 . All these procedures were repeated 1,000 times using the 1,000 datasets previously mentioned. Bayesian phylogenetic generalised linear mixed modelling Each of the 1,000 datasets were analysed using Bayesian phylogenetic generalised linear mixed models (PGLMMs) to test the relationship between cranial capacity and the within- and between-species effect of body mass, chronological age, and environment (i.e., PC1 and PC2). A PGLMM is like a conventional phylogenetic generalised least squares (PGLS) method but goes beyond merely estimating the variance of the phylogenetic effect: It also incorporates an additional residual error term which allows for the inclusion of other factors such as intraspecific variance, environmental effects, measurement error, and more. PGLMM incorporates shared ancestry information by including a phylogenetic random effect that is assumed to follow a normal distribution with a variance that considers correlation among phylogenetic effects based on a phylogenetic variance-covariance (or correlation) matrix. In our case, we computed these matrices using the 1,000 hominin phylogenies previously mentioned. To assess the role of inter- and intra-specific variability, we used a technique known as within-group centering 50 . This technique partitions the predictor variables into two different components: the group-level mean (i.e., the mean of that predictor for each species, resulting in the between-species differences) and the within-group variation (i.e., the difference of each subject from its within-group mean, resulting in the within-species differences). We also accounted for possible slope differences per species (i.e., a random slope model) by adding different random effects (i.e., species-specific random effects for the within-group variability of both body mass, time and environment represented by PC1 and PC2). For each bioclimatic variable and PC, we conducted PGLMM analyses with a set random-effect structure (supplementary files). As our previous work has demonstrated that between-species body mass and within-species chronological age are significant predictors of hominin cranial capacities 6 , all regressions included between- and within-species body mass, chronological age, and PCs as fixed effects (supplementary files). Within-species predictors were also included as random effects. In addition, we ran all bioclimatic variables following this structure (Tab. S1). Results for other significant climate variables (Bio-1, -4, -6, -7 and -11), as well as the results of non-significant climate variables are included in the supplementary files (Tab. S1). All reported values from our modelling approach correspond to grand means (i.e., average of means) computed for the estimated effects obtained by the models ran using our 1,000 unique datasets (each using an alternative phylogenetic tree). PGLMM analyses were conducted using the MCMCglmm v.2.33 R package 51 . For the fixed effects, we used a diffuse, normally distributed, prior centered around zero (μ=0) with very large variance (σ2=10^8). Inverse-Gamma prior distributions with shape (α) and scale (β) parameters equal to 0.01 were applied for the random effects. All regressions were conducted using a total of 510,000 iterations, with a thinning interval of 250, and the first 10,000 iterations were removed as burn-in. Markov chain convergence and mixing were visually assessed by looking at the trace plots of each one of the fixed and random effects and we checked that effective sample sizes were > 1,000. Significance was assessed by the proportion of the posterior distribution crossing zero (Px). Where Px < 0.05, less than five percent of the posterior distribution overlaps with zero, and we consider a predictor to be different from zero. Acknowledgements This work was supported by Leverhulme Research Leadership Award RL-2019-012 Footnotes Competing Interest Statement: The authors declare no conflicts of interest. References 1. ↵ Isler , K. & Van Schaik , C. P. How humans evolved large brains: Comparative evidence . Evolutionary Anthropology: Issues, News, and Reviews 23 , 65 – 75 ( 2014 ). OpenUrl 2. ↵ Dunbar , R. I. M. The social brain hypothesis . Evolutionary Anthropology: Issues, News, and Reviews 6 , 178 – 190 ( 1998 ). OpenUrl 3. ↵ Will , M. , Krapp , M. , Stock , J. T. & Manica , A. Different environmental variables predict body and brain size evolution in Homo . Nat Commun 12 , 4116 ( 2021 ). OpenUrl 4. ↵ Shultz , S. & Maslin , M. Early Human Speciation, Brain Expansion and Dispersal Influenced by African Climate Pulses . PLoS One 8 , e76750 ( 2013 ). OpenUrl CrossRef PubMed 5. ↵ Rightmire , G. P. Brain size and encephalization in early to Mid-PleistoceneHomo . Am J Phys Anthropol 124 , 109 – 123 ( 2004 ). OpenUrl CrossRef PubMed Web of Science 6. ↵ Püschel , T. A. , Nicholson , S. L. , Baker , J. , Barton , R. A. & Venditti , C. Hominin brain size increase has emerged from within-species encephalization . bioRxiv 582715 , ( 2024 ). 7. ↵ DeSilva , J. M. , Traniello , J. F. A. , Claxton , A. G. & Fannin , L. D. When and Why Did Human Brains Decrease in Size? A New Change-Point Analysis and Insights From Brain Evolution in Ants . Front Ecol Evol 9 , ( 2021 ). 8. ↵ Ash , J. & Gallup , G. G. Paleoclimatic Variation and Brain Expansion during Human Evolution . Human Nature 18 , 109 – 124 ( 2007 ). OpenUrl CrossRef PubMed 9. ↵ Maslin , M. A. , Shultz , S. & Trauth , M. H. A synthesis of the theories and concepts of early human evolution . Philosophical Transactions of the Royal Society B: Biological Sciences 370 , 20140064 ( 2015 ). OpenUrl CrossRef PubMed 10. Potts , R. Environmental hypotheses of hominin evolution . Am J Phys Anthropol 107 , 93 – 136 ( 1998 ). OpenUrl CrossRef PubMed 11. ↵ Maslin , M. A. et al. East African climate pulses and early human evolution . Quat Sci Rev 101 , 1 – 17 ( 2014 ). OpenUrl CrossRef 12. ↵ Potts , R. Variability selection in hominid evolution . Evol Anthropol 7 , 81 – 96 ( 1998 ). OpenUrl CrossRef 13. ↵ Potts , R. Evolution and climate variability . Science (1979) 273 , 922 – 923 ( 1996 ). OpenUrl Abstract / FREE Full Text 14. ↵ Antón , S. C. , Potts , R. & Aiello , L. C. Evolution of early Homo[]: An integrated biological perspective . Science (1979) 345 , ( 2014 ). 15. ↵ Stibel , J. M. Climate Change Influences Brain Size in Humans . Brain Behav Evol ( 2022 ) doi: 10.1159/000528710 . OpenUrl CrossRef 16. ↵ De Marco , P. & Nóbrega , C. C. Evaluating collinearity effects on species distribution models: An approach based on virtual species simulation . PLoS One 13 , e0202403 ( 2018 ). OpenUrl CrossRef 17. ↵ Freckleton , R. P. Dealing with collinearity in behavioural and ecological data: model averaging and the problems of measurement error . Behav Ecol Sociobiol 65 , 91 – 101 ( 2011 ). OpenUrl CrossRef 18. ↵ Holden , P. B. et al. PALEO-PGEM v1.0: a statistical emulator of Pliocene–Pleistocene climate . Geosci Model Dev 12 , 5137 – 5155 ( 2019 ). OpenUrl 19. ↵ Xu , T. & Hutchinson , M. F. New developments and applications in the ANUCLIM spatial climatic and bioclimatic modelling package . Environmental Modelling & Software 40 , 267 – 279 ( 2013 ). OpenUrl 20. ↵ Forkman , J. , Josse , J. & Piepho , H.-P. Hypothesis Tests for Principal Component Analysis When Variables are Standardized . J Agric Biol Environ Stat 24 , 289 – 308 ( 2019 ). OpenUrl 21. ↵ Falk , D. Brain evolution in Homo: The “radiator” theory . Behavioral and Brain Sciences 13 , 333 – 344 ( 1990 ). OpenUrl CrossRef Web of Science 22. Falk , D. & Gage , T. B. Flushing the radiator? A reply to Braga and Boesch . J Hum Evol 33 , 495 – 502 ( 1997 ). OpenUrl PubMed 23. ↵ Qi , X.-G. et al. Adaptations to a cold climate promoted social evolution in Asian colobine primates . Science (1979) 380 , ( 2023 ). 24. ↵ Beals , K. L. et al. Brain Size, Cranial Morphology, Climate, and Time Machines [and Comments and Reply] . Curr Anthropol 25 , 301 – 330 ( 1984 ). OpenUrl CrossRef Web of Science 25. ↵ Gillooly , J. F. & McCoy , M. W. Brain size varies with temperature in vertebrates . PeerJ 2 , e301 ( 2014 ). OpenUrl 26. ↵ Hosfield , R. Walking in a Winter Wonderland? Strategies for Early and Middle Pleistocene Survival in Midlatitude Europe . Curr Anthropol 57 , 653 – 682 ( 2016 ). OpenUrl CrossRef 27. ↵ van Andel , T. H. & Davies , W. Aiello , L. & Wheeler , P. Neanderthal thermoregulation and the glacial climate . in Neanderthals and modern humans in the European landscape during the last glaciation (eds. van Andel , T. H. & Davies , W. ) 147 – 166 ( MacDonald Institute for Archaeological Research , Cambridge , 2003 ). 28. ↵ Ao , H. et al. Two-stage mid-Brunhes climate transition and mid-Pleistocene human diversification . Earth Sci Rev 210 , 103354 ( 2020 ). OpenUrl 29. ↵ Tzedakis , P. C. , Crucifix , M. , Mitsui , T. & Wolff , E. W. A simple rule to determine which insolation cycles lead to interglacials . Nature 542 , 427 – 432 ( 2017 ). OpenUrl CrossRef PubMed 30. Hays , J. D. , Imbrie , J. & Shackleton , N. J. Variations in the Earth’s Orbit: Pacemaker of the Ice Ages: For 500,000 years, major climatic changes have followed variations in obliquity and precession . Science (1979) 194 , 1121 – 1132 ( 1976 ). OpenUrl Abstract / FREE Full Text 31. Imbrie , J. et al. On the Structure and Origin of Major Glaciation Cycles 1 . Linear Responses to Milankovitch Forcing. Paleoceanography 7 , 701 – 738 ( 1992 ). OpenUrl 32. ↵ Raymo , M. E. & Nisancioglu , K. H. The 41 kyr world: Milankovitch’s other unsolved mystery . Paleoceanography 18 , ( 2003 ). 33. ↵ Roberts , P. & Stewart , B. A. Defining the ‘generalist specialist’ niche for Pleistocene Homo sapiens . Nat Hum Behav 2 , 542 – 550 ( 2018 ). OpenUrl 34. ↵ Galway-Witham , J. , Cole , J. & Stringer , C. Aspects of human physical and behavioural evolution during the last 1 million years . J Quat Sci 34 , 355 – 378 ( 2019 ). OpenUrl 35. White , M. J. Things to do in Doggerland when you’re dead: Surviving OIS3 at the northwestern-most fringe of Middle Palaeolithic Europe . World Archaeol 38 , 547 – 575 ( 2006 ). OpenUrl 36. ↵ González-Forero , M. & Gardner , A. Inference of ecological and social drivers of human brain-size evolution . Nature 557 , 554 – 557 ( 2018 ). OpenUrl CrossRef PubMed 37. ↵ Bailey , D. H. & Geary , D. C. Hominid Brain Evolution . Human Nature 20 , 67 – 79 ( 2009 ). OpenUrl CrossRef 38. ↵ deMenocal , P. B. African climate change and faunal evolution during the Pliocene-Pleistocene . Earth Planet Sci Lett 220 , 3 – 24 ( 2004 ). OpenUrl 39. ↵ deMenocal , P. B. Plio-Pleistocene African Climate . Science (1979) 270 , 53 – 59 ( 1995 ). OpenUrl Abstract / FREE Full Text 40. ↵ Novello , A. et al. Diatom, phytolith, and pollen records from a 10Be/9Be dated lacustrine succession in the Chad basin: Insight on the Miocene–Pliocene paleoenvironmental changes in Central Africa . Palaeogeogr Palaeoclimatol Palaeoecol 430 , 85 – 103 ( 2015 ). OpenUrl GeoRef 41. ↵ Lüthi , D. et al. High-resolution carbon dioxide concentration record 650,000–800,000 years before present . Nature 453 , 379 – 382 ( 2008 ). OpenUrl CrossRef GeoRef PubMed Web of Science 42. ↵ Stap , L. B. , van de Wal , R. S. W. , de Boer , B. , Bintanja , R. & Lourens , L. J. The influence of ice sheets on temperature during the past 38 million years inferred from a one-dimensional ice sheet–climate model . Climate of the Past 13 , 1243 – 1257 ( 2017 ). OpenUrl 43. ↵ Berger , A. & Loutre , M. F. Parameters of the Earths orbit for the last 5 Million years in 1 kyr resolution. Supplement to: Berger, A; Loutre, M F (1991): Insolation values for the climate of the last 10 million of years . Quaternary Science Reviews , 10 ( 4 ), 297 – 317 , doi: 10.1016/0277-3791(91)90033-Q Preprint at https://doi.org/10.1594/pangaea.56040 ( 1999 ). OpenUrl CrossRef 44. ↵ Raia , P. et al. Past Extinctions of Homo Species Coincided with Increased Vulnerability to Climatic Change . One Earth 3 , 480 – 490 ( 2020 ). OpenUrl CrossRef 45. ↵ R Core Team. R . https://www.R-project.org/ ( 2021 ). 46. ↵ Fick , S. E. & Hijmans , R. J. WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas . International Journal of Climatology 37 , 4302 – 4315 ( 2017 ). OpenUrl CrossRef 47. ↵ Karger , D. N. et al. Data from: Climatologies at high resolution for the earth’s land surface areas . EnviDat https://doi.org/10.16904/envidat.228.v2.1%0A%0A ( 2018 ) x doi: 10.16904/envidat.228 . OpenUrl CrossRef 48. Karger , D. N. et al. Climatologies at high resolution for the earth’s land surface areas . Sci Data 4 , 170122 ( 2017 ). OpenUrl 49. ↵ Beck , H. E. et al. Bias Correction of Global High-Resolution Precipitation Climatologies Using Streamflow Observations from 9372 Catchments . J Clim 33 , 1299 – 1315 ( 2020 ). OpenUrl 50. ↵ van de Pol , M. & Wright , J. A simple method for distinguishing within-versus between-subject effects using mixed models . Anim Behav 77 , 753 – 758 ( 2009 ). OpenUrl CrossRef Web of Science 51. ↵ Hadfield , J. MCMC Generalised Linear Mixed Models . Preprint at https://cran.r-project.org/web/packages/MCMCglmm/MCMCglmm.pdf ( 2022 ). View the discussion thread. Back to top Previous Next Posted September 10, 2024. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Climatic-environmental influences on hominin brain size over the last 5 million years Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Climatic-environmental influences on hominin brain size over the last 5 million years Samuel L. Nicholson , Thomas A. Püschel , Joanna Baker , Chris Venditti bioRxiv 2024.09.09.611970; doi: https://doi.org/10.1101/2024.09.09.611970 Share This Article: Copy Citation Tools Climatic-environmental influences on hominin brain size over the last 5 million years Samuel L. Nicholson , Thomas A. Püschel , Joanna Baker , Chris Venditti bioRxiv 2024.09.09.611970; doi: https://doi.org/10.1101/2024.09.09.611970 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Evolutionary Biology Subject Areas All Articles Animal Behavior and Cognition (7644) Biochemistry (17728) Bioengineering (13916) Bioinformatics (42037) Biophysics (21488) Cancer Biology (18636) Cell Biology (25552) Clinical Trials (138) Developmental Biology (13401) Ecology (19940) Epidemiology (2067) Evolutionary Biology (24367) Genetics (15621) Genomics (22545) Immunology (17764) Microbiology (40475) Molecular Biology (17208) Neuroscience (88744) Paleontology (667) Pathology (2842) Pharmacology and Toxicology (4834) Physiology (7659) Plant Biology (15175) Scientific Communication and Education (2047) Synthetic Biology (4304) Systems Biology (9834) Zoology (2272)