Brain size predicts bees' tolerance to urban environments

royalsocietypublishing.org/journal/rsbl Research Cite this article: Lanuza JB, Collado MÁ, Sayol F, Sol D, Bartomeus I. 2023 Brain size predicts bees’tolerance to urban environments. Biol. Lett. 19: 20230296. https://doi.org/10.1098/rsbl.2023.0296 Received: 24 June 2023 Accepted: 13 November 2023 Subject Category: Global change biology Subject Areas: ecology, environmental science, neuroscience, evolution, cognition, behaviour Keywords: relative brain size, habitat occupancy, Apoidea, urbanization, pollinators Author for correspondence: Jose B. Lanuza e-mail: barr[email protected]om Electronic supplementary material is available online at https://doi.org/10.6084/m9.figshare. c.6935821. Brain size predicts bees’tolerance to urban environments Jose B. Lanuza 1,2 , Miguel Á. Collado 1,3 , Ferran Sayol 4 , Daniel Sol 4,5 and Ignasi Bartomeus 1 1 Estación Biológica de Doñana (EBD-CSIC), 41092 Seville, Spain 2 Spatial Interaction Ecology, German Centre for Integrative Biodiversity Research (iDiv) Halle-Jena-Leipzig, Leipzig, Saxony, Germany 3 Departamento de Ciencias de la Computación e Inteligencia Artificial, Universidad de Sevilla, Seville, Spain 4 Centre for Ecological Research and Forestry Applications (CREAF), Bellaterra, Catalonia, Spain 5 Department of Ecology, CSIC, Spanish National Research Council, CREAF-UAB, Bellaterra, Catalonia, Spain JBL, 0000-0002-0287-409X; MÁC, 0000-0002-4216-317X; FS, 0000-0003-3540-7487; DS, 0000-0001-6346-6949; IB, 0000-0001-7893-4389 The rapid conversion of natural habitats to anthropogenic landscapes is threatening insect pollinators worldwide, raising concern regarding the negative consequences on their fundamental role as plant pollinators. However, not all pollinators are negatively affected by habitat conversion, as certain species find appropriate resources in anthropogenic landscapes to persist and proliferate. The reason why some species tolerate anthropogenic environments while most find them inhospitable remains poorly understood. The cognitive buffer hypothesis, widely supported in vertebrates but untested in insects, offers a potential explanation. This theory suggests that species with larger brains have enhanced behavioural plasticity, enabling them to confront and adapt to novel challenges. To investigate this hypothesis in insects, we measured brain size for 89 bee species, and evaluated their association with the degree of habitat occupancy. Our analyses revealed that bee species mainly found in urban habitats had larger brains relative to their body size than those that tend to occur in forested or agricultural habitats. Additionally, urban bees exhibited larger body sizes and, consequently, larger absolute brain sizes. Our results provide the first empirical support for the cognitive buffer hypothesis in invertebrates, suggesting that a large brain in bees could confer behavioural advantages to tolerate urban environments. 1. Introduction Pollinators deliver a fundamental ecosystem service on which the Earth’s vegetation and human economy depend [1]. Regrettably, there is increasing evidence of recent declines in pollinator populations [2–4]. One of the main contributing factors to the current pollinator declines is the alteration and loss of their habitat due to human activity [5,6]. Anthropogenic landscapes present new challenges for the survival and reproduction of organisms, increasing their risk of extinction by maladaptation [7–9]. Yet, not all pollinator species are negatively affected by land use change. Indeed, some bee species are able to tolerate human-altered environments [10–12] or even to thrive [13,14]. Human-dominated habitats, notably cities, drastically modify the ancestral conditions where pollinators evolved, but can also offer unique ecological opportunities in the form of new nesting spots, shelter from phytosanitary products, reduced predation pressure and high food availability associated with nonindigenous plants [7,12,15]. The question is then why are only some species able to tolerate and exploit urban environments. The ‘cognitive buffer hypothesis’provides an explanation for this conundrum, suggesting that in novel environments the chancesto successfullysurvive and reproduce depends on enhanced cognition togather,storeandplasticallyreacttonewinformation[16,17].Whilethe‘cognitive © 2023 The Authors. Published by the Royal Society under the terms of the Creative Commons Attribution License http://creativecommons.org/licenses/by/4.0/, which permits unrestricted use, provided the original author and source are credited. Downloaded from https://royalsocietypublishing.org/ on 05 May 2025 buffer hypothesis’receives ample support from studies of vertebrates [18–21], similar evidence is lacking for insects. Insects have long been associated with the most basic types of learning due to their miniature brain [22,23]. However, recent studies have shown that bees and other insects have sophisticated cognition that goes beyond simple associative learning or conditioning [24–26]. Some of these more complex forms of cognition involve the use of tools, attention, social learning or metacognitive processes [22,23]. In addition, there is also evidence for substantial variation across species in brain size, both in absolute terms and relative to body size, with species that have larger brains also exhibiting enhanced cognitive performance, at least for some tasks such as conditional learning [27]. Therefore, we can ask whether the varying success of insects in human-altered habitats might be explained by variation in brain size. Here, we report the first test of the ‘cognitive buffer hypothesis’in insects. Our test is based on a unique dataset of brain measures for 89 European and North American bee species. By means of detailed georeferenced information of species occurrences, we characterize the degree of habitat occupancy for all the species, and use a phylogenetically informed comparative analysis to assess whether bees that proliferate in human-altered habitats have enlarged brains compared to those that avoid them. 2. Methods (a) Brain measurements Our dataset contains measurements of brain and body size for bee specimens captured on flowers by hand netting in different areas of the East Coast of the United States and northern Central Europe. These specimens were collected opportunistically mostly in semi-natural habitats. The dataset includes information of 335 female individuals from 89 species that represent six families and 31 genera. We considered only female specimens because they engage in a wider variety of tasks, facing greater environmental pressures, and possess distinct brain structures and functions compared to males [28,29]. Brain size was measured as the weight of fixed brains [30] and body size as intertegular span ([31]; electronic supplementary material, text S1). Given that brain size scales allometrically with body size [30], we considered for our analysis the size of the brain relative to the body. Following Sayol et al. [30], we estimated relative brain size as the residuals of a log–log phylogenetic linear model of brain mass against body size, using the Bayesian approximation implemented in the function brm from the package brms v. 2.18 [32]. The phylogeny was built with the help of a previously published genus-level phylogeny [33] and it was processed with the help of the packages ape 5.6-2 [34], phytools 1.2-0 [35] and MCMCglmm v. 2.33 [36]. High values of relative brain size indicate larger brains than expected for their body size, while low values denote smaller brains. Our bee dataset showed a strong allometric relationship between brain and bodysize (Bayesian R 2 =0.9) that was constrained by the species’evolutionary history (phylogenetic signal of relative brain size, λ=0.6; p< 0.01). However, we found considerable variability in relative brain size within and across taxonomic groups (electronic supplementary material, figure S1; figure 1a). (b) Habitat occupancy We downloaded occurrence information for all the measured species from the Global Biodiversity Information Facility (GBIF; http://www.gbif.org/) for North America and Europe. The data were downloaded via the R programming language with the help of the function occ_download from the package rgbif v. 3.7.3 [37]. We selected the states or countries with the highest density of records for our set of species. For North America, we selected states located on the East Coast of the United States (electronic supplementary material, figure S2a), covering an approximate area of 136 937 km 2 . For Europe, we selected countries located on the north and centre of the continent (electronic supplementary material, figure S2b), representing a total area of 600 497 km 2 .We only included species with a minimum number of 50 records and whose geographical distribution was larger than the sampled area (i.e. excluding species at the edge of their distributions). We optimized the match between species occurrence and the land cover data by only using georeferenced records obtained between 1990 and 2022 with a minimum of two decimals of latitude/longitude coordinates. We assigned a habitat type to each GBIF occurrence by merging land cover information with the georeferenced records of species occurrences. The land cover classification was obtained from the 2006 online inventories of the National Land Cover Database (NLCD) for the United States and the Corine Land Cover (CLC) for Europe. After downloading these inventories as raster files, we used the functions rast and extract from the Terra package v. 1.6-41 [38] to read and obtain the cover classification of the different georeferenced records, respectively. To simplify the interpretation and conduct a joint analysis for both regions, we divided the resulting cover classes into three single categories: (i) natural, (ii) agricultural and (iii) urban (see electronic supplementary material, tables S1 and S2 for details). With this information we can build an occurrence matrix with species in rows, habitats in columns and cells depicting the number of occurrences per species–habitat combination. The degree of habitat occupancy was estimated by assessing whether the occurrences of species in a given habitat were more frequent than expected by chance. Although previous work has referred to this metric as ‘habitat preference’[12], we have opted to use ‘degree of habitat occupancy’as it avoids the connotation of active choice by the different species. To calculate habitat occupancy, we generated 10 000 randomized matrices based on the occurrence matrix with the function nullmodel from the package bipartite v. 2.16 [39]. We used the method ‘r2dtable’, which maintains constant row and column sums by using Patefield’s algorithm [40]. This maintains constant the proportional dominance of the species and habitats but reshuffles their associations. We then estimated the percentage of simulated occurrences per species and habitat that were under the observed ones (i.e. percentile). Lastly, we considered that a bee species exhibited a ‘high habitat occupancy’for any of the three studied habitats when the number of occurrences observed in this habitat exceeded the 80th percentile of the values obtained from the simulations. On the contrary, we considered a ‘low habitat occupancy’for the habitat when the observed occurrences were below the 20th percentile. To better understand if our findings are affected by the evolutionary history of the species, we estimated the phylogenetic signal of relative brain size and the degree of habitat occupancy across habitats for our set of species with the help of the function phylosig from the phytools package. (c) Diet specialization To ensure diet is not acting as a confounding factor we investigated how diet specialization is associated with the degree of habitat occupancy and brain size. For this, we used existing diet information (see [30]), where bees were classified as oligolectic when they use a single plant family to feed their brood or polylectic when they use several. We tested for statistical differences in diet across the degree of habitat occupancy per habitat type and brain size by using Wilcoxon’s test. We found 2 royalsocietypublishing.org/journal/rsbl Biol. Lett. 19: 20230296 Downloaded from https://royalsocietypublishing.org/ on 05 May 2025 a low number of specialist species in our dataset with no clear associations with the degree of habitat occupancy or brain size that can explain the observed relationship between the two (electronic supplementary material, figure S3). (d) Analysis To evaluate how the association between the degree of habitat occupancy and relative brain size per species changed by habitat type, we used a phylogenetic Bayesian approach to model their association. For this, we first joined the resulting values of the degree of habitat occupancy per species with their respective average relative brain sizes. Our macro-ecological approach is justified by: (i) independent data ensuring robust and generalizable ecological patterns and (ii) minimal intraspecific variation in brain and body size compared to interspecific variation (see electronic supplementary material, text S2 and figure S4). The resulting distribution of the degree of habitat occupancy for each of the habitats analysed followed a 0–1 inflated beta distribution (electronic supplementary material, figure S5), indicating that there were high frequencies of habitat occupancy close to 0 or 1 but low frequencies of intermediate values between 0 and 1. In other words, simulated occurrences per species and habitat were consistently lower or higher than the observed ones. Hence, in our analyses we take a conservative approach and only modelled the extremes of the distribution (i.e. species classified as high or low habitat occupancy). Because we assessed the degree of habitat occupancy as binary (‘low’or ‘high’), we specified a Bernouilli distribution where the degree of habitat occupancy was the response variable and relative brain size the predictor. To control for the shared evolutionary history of the species, we included the phylogeneticcovariance matrix asa random factor. Finally, we replicated the analyses for the United States and Europe independently with analogous models. 0 0.1 0.2 0.3 (a)(b) Dasypoda hirtipes Stelis breviuscula Anthidium oblongatum Anthidium manicatum Megachile ligniseca Megachile willughbiella Megachile centuncularis Megachile texana Megachile pugnata Megachile rotundata Osmia cornuta Osmia leaiana Osmia caerulescens Osmia bicornis Osmia lignaria Osmia atriventris Osmia pumila Osmia cornifrons Anthophora quadrimaculata Anthophora plumipes Nomada succincta Nomada cressonii Nomada luteoloides Xylocopa violacea Xylocopa virginica Ceratina strenua Ceratina calcarata Bombus jonellus Bombus muscorum Bombus hypnorum Bombus pratorum Bombus hortorum Bombus pascuorum Bombus terrestris Bombus griseocollis Bombus impatiens Colletes thoracicus Hylaeus mesillae Hylaeus modestus Augochloropsis metallica Augochlora pura Augochlorella aurata Agapostemon sericeus Agapostemon virescens Halictus confusus Halictus rubicundus Halictus ligatus Lasioglossum sexstrigatum Lasioglossum sexnotatum Lasioglossum punctatissimum Lasioglossum calceatum Lasioglossum malachurum Lasioglossum fulvicorne Lasioglossum zonulum Lasioglossum coriaceum Lasioglossum leucozonium Calliopsis andreniformis Panurgus banksianus Panurgus calcaratus Andrena gravida Andrena carantonica Andrena nigriceps Andrena varians Andrena tibialis Andrena labiata Andrena barbilabris Andrena fucata Andrena nigroaenea Andrena pilipes Andrena angustior Andrena ovatula Andrena semilaevis Andrena subopaca Andrena fulva Andrena bicolor Andrena haemorrhoa Andrena nitida Andrena chrysosceles Andrena flavipes Andrena cineraria Andrena dorsata Andrena carlini Andrena dunningi Andrena vicina Andrena miserabilis Andrena milwaukeensis Andrena nubecula Andrena wilkella Andrena hirticincta natural agricultural urban data Europe USA USA and Europe relative brain size 0.50 0.25 0 0.25 0.50 0 (low) 0.5 habitat occupancy 1.0 (high) Figure 1. Phylogenetic relationship and degree of habitat occupancy for the selected bee species with brain weight and intertegular distance information (N= 89). (a) Phylogenetic tree at the genus level. Tree branches are coloured based on the geographical location of the different bee species (northern Central Europe, East Coast of the United States or from both regions). The deviation of the brain size in relation to the body (i.e. residuals) is represented with filled circles of proportional area at the end of the tip branches. Larger circles indicate larger brains in proportion to their body size and vice versa. (b) Heat map showing the degree of habitat occupancy for each bee species. The columns delimit the habitat type (i.e. natural, agricultural and urban) and the rows the different bee species. 3 royalsocietypublishing.org/journal/rsbl Biol. Lett. 19: 20230296 Downloaded from https://royalsocietypublishing.org/ on 05 May 2025 All our models were run with 4000 iterations with previous 1000 warm-up iterations, using non-informative or weakly informative priors [32]. Further, the different posterior predictive checks were conducted with the function pp_check, also from the brms package. All our analyses were undertaken in R v. 4.0.5 [41], and all data processing and graphics were done with the set of packages from the tidyverse v. 1.3.0 [42]. 3. Results The degree of habitat occupancy varied substantially across species (figure 1b) and showed moderate phylogenetic signal λ= 0.38; p= 0.02). In general, most bee species have high occupancy in one or two habitat types but more rarely occurred indistinctly in the three (figure 1b). The habitats with the highest occurrences in comparison with random expectations were the agricultural and natural ones, with 49 and 48 species over 80% of the values from null models, respectively (figure 1b and electronic supplementary material, figure S5). By contrast, most species showed a low degree of occupancy in urban habitats (56 species under the 20th percentile) and just 28 species showed high occupancy in this habitat type (figure 1band electronic supplementary material, figure S5). We found that relative brain size was associated with the degree of occupancy across the different habitat types (figure 2a; Bayesian R 2 = 0.11). Specifically, we found that bees with larger relative brains showed a higher occupancy in urban habitats than bees with smaller ones (figure 2a). Contrarily, bee species with smaller relative brains showed higher occupancy in natural and agricultural habitats than bees with larger relative brains (figure 2a). The models of the association between absolute brain size and intertegular span with habitat occupancy also showed a marked difference between habitat types (figure 2b,c; Bayesian R 2 = 0.25; Bayesian R 2 = 0.23, respectively). Specifically, we found that bees with larger brains and body sizes appear more in urban habitats and bees with smaller brains and body sizes are more common in natural and agricultural ones (figure 2b,c). These findings were consistent with the analogous analyses by geographical regions (United States and Europe; electronic supplementary material, figure S6). 4. Discussion By analysing 89 bee species from two different continents, we observed that bee species inhabiting urban habitats tend to have larger brains relative to their body while those found in forest or agricultural habitats exhibit relatively smaller brains. These results are in line with the cognitive buffer hypothesis [17], which predicts that a large relative brain should provide enhanced behavioural plasticity to tolerate novel environments. Since urban environments exhibit dynamic challenges, including novel resources and changing human disturbances [43], a large brain may provide the cognitive flexibility to exploit these new resources while avoiding risks. Although the exact mechanisms remain unclear, we expect cognitive flexibility to be essential in a variety of contexts such as the use of human-made materials for the nest [44] or the use of exotic flowers [45]. In line with the cognitive buffer hypothesis [46], urban bees also showed bigger bodies and larger absolute brains. Larger species require longer development times and obtain greater net benefits from exploring and learning, especially in heterogeneousenvironments[47,48].Forexample,thelargecarpenter bees of the genus Xylocopa can live up to 2 years and are frequent urban dwellers, while small Andrena forest specialists complete their adult life cycle in a few weeks. The patchiness of urban resources also seems to favour larger body sizes, as low high –0.4 0 0.4 0.8 relative brain size habitat occupancy low high 0246 brain weight (mg) low high 246 intertegular distance (mm) habitat natural agricultural urban (a)(b) (c) Figure 2. Association between relative brain size (a), brain weight (b) and intertegular distance (c) with the degree of habitat occupancy by habitat type (i.e. natural, agricultural and urban). The shaded and coloured areas by habitat type represent 95% credible intervals. Coloured circles indicate the raw values of relative brain size for the resulting levels of habitat occupancy. 4 royalsocietypublishing.org/journal/rsbl Biol. Lett. 19: 20230296 Downloaded from https://royalsocietypublishing.org/ on 05 May 2025 suggestedbytheobservationthat foragingdistanceispositively associated with body size in bees[31,49].There is also some evidence that associate increased tolerance to urbanization with wider ecological niches [50–52], and both larger bodies and relative brains are thought to be features that can facilitate broader diets and niche expansion [20,53,54]. In our study, we examined only a few bee specialists, highlighting the need for further research on how different diets and life strategies relate to cognition and urban adaptation. Our findings support and extend upon previous evidence in vertebrates that having larger brains can facilitate tolerance to urban environments [55,56], highlighting that a cognitive buffer is possible even with tiny brains. The use of the entire brain size as a proxy for cognitive performance is not exempt from criticism [24,57]. We primarily analysed brain size due to data availability, supported by prior findings showing larger brains improving certain cognitive aspects in bees, such as learning [27]. Moreover, brain size is less subject to measurement error or context-dependent biases in comparison with other experimental measures of cognition [24,57]. However, future finer measures, such as neuropil size or mushroom bodies (the suspected centres of cognitive processes in bees; [58,59,60]), are likely to enhance our understanding of bee cognition. Downscaling our analysis to the individual specimen level where brain size, habitat use and cognitive performance can be tracked through their lifespan would be a challenging but promising next step. Our findings highlight the importance of cognition for understanding the dynamics of insect populations in altered environments and stresses the need to avoid viewing them as passive agents of external pressures. Ethics. This work did not require ethical approval from a human subject or animal welfare committee. Data accessibility. Data available from the Dryad Digital Repository: https://doi.org/10.5061/dryad.zw3r228dr [61]. Electronic supplementary material is available online [62]. Declaration of AI use. We have not used AI-assisted technologies in creating this article. Authors’contributions. J.B.L.:conceptualization,datacuration,formalanalysis, visualization, writing—original draft, writing—review and editing; M.Á.C.: conceptualization, data curation, formal analysis, writing— original draft, writing—review and editing; F.S.: conceptualization, data curation, writing—review and editing; D.S.: conceptualization, writing—review and editing; I.B.: conceptualization, formal analysis, supervision, writing—original draft, writing—review and editing. All authors gave final approval for publication and agreed to be held accountable for the work performed therein. Conflict of interest declaration. We declare we have no competing interests. Funding. This work was supported by the Ministerio de Economía y Competitividad, Gobierno de España (grant no. CGL2013-47448-P). Acknowledgements. We thank Fransico P. Molina, Ivo Reamakers and Parker Gambino for collecting some of the specimens. References 1. Ollerton J. 2017 Pollinator diversity: distribution, ecological function, and conservation. Ann. Rev. Ecol. Evol. Syst. 48, 353–376. (doi:10.1146/annurevecolsys-110316-022919) 2. Bartomeus I, Stavert J, Ward D, Aguado O. 2019 Historical collections as a tool for assessing the global pollination crisis. Phil. Trans. R. Soc. B 374, 20170389. (doi:10.1098/rstb.2017.0389) 3. Potts SG, Biesmeijer JC, Kremen C, Neumann P, Schweiger O, Kunin WE. 2010 Global pollinator declines: trends, impacts and drivers. Trends Ecol. Evol. 25, 345–353. (doi:10.1016/j.tree.2010. 01.007) 4. Scheper J, Reemer M, van Kats R, Ozinga WA, van der Linden GT, Schaminée JH, Siepel H, Kleijn D. 2014 Museum specimens reveal loss of pollen host plants as key factor driving wild bee decline in the Netherlands. Proc. Natl Acad. Sci. USA 111, 17 552–17 557. (doi:10.1073/pnas. 1412973111) 5. Harrison T, Winfree R. 2015 Urban drivers of plant– pollinator interactions. Funct. Ecol. 29, 879–888. (doi:10.1111/1365-2435.12486) 6. Harrison T, Gibbs J, Winfree R. 2018 Forest bees are replaced in agricultural and urban landscapes by native species with different phenologies and lifehistory traits. Glob. Change Biol. 24, 287–296. (doi:10.1111/gcb.13921) 7. Winfree R, Bartomeus I, Cariveau DP. 2011 Native pollinators in anthropogenic habitats. Annu. Rev. Ecol. Evol. Syst. 42,1–22. (doi:10.1146/annurevecolsys-102710-145042) 8. Brown MJ, Paxton RJ. 2009 The conservation of bees: a global perspective. Apidologie 40, 410–416. (doi:10.1051/apido/2009019) 9. Harrison T, Gibbs J, Winfree R. 2019 Anthropogenic landscapes support fewer rare bee species. Landsc. Ecol. 34, 967–978. (doi:10.1007/s10980- 017-0592-x) 10. Biesmeijer JC et al. 2006 Parallel declines in pollinators and insect-pollinated plants in Britain and the Netherlands. Science 313, 351–354. (doi:10.1126/science.1127863) 11. Bartomeus I, Ascher JS, Gibbs J, Danforth BN, Wagner DL, Hedtke SM, Winfree R. 2013 Historical changes in northeastern US bee pollinators related to shared ecological traits. Proc. Natl Acad. Sci. USA 110, 4656–4660. (doi:10.1073/pnas.1218503110) 12. Collado MÁ, Sol D, Bartomeus I. 2019 Bees use anthropogenic habitats despite strong natural habitat preferences. Divers. Distrib. 25, 924–935. (doi:10.1111/ddi.12899) 13. Rasmont P, Coppée A, Michez D, De Meulemeester T. 2008 An overview of the Bombus terrestris (L. 1758) subspecies (Hymenoptera: Apidae). Ann. Soc. Entomol. Fr. 44, 243–250. (doi:10.1080/00379271. 2008.10697559) 14. Theodorou P et al. 2020 Urban areas as hotspots for bees and pollination but not a panacea for all insects. Nat. Commun. 11, 576. (doi:10.1038/ s41467-020-14496-6) 15. Cane JH, Minckley RL, Kervin LJ, Roulston TH, Williams NM. 2006 Complex responses within a desert bee guild (Hymenoptera: Apiformes) to urban habitat fragmentation. Ecol. Appl. 16, 632–644. (doi:10.1890/1051-0761(2006)016[0632: CRWADB]2.0.CO;2) 16. Shettleworth SJ. 2009 Cognition, evolution, and behavior. New York, NY: Oxford University Press. 17. Sol D. 2009 Revisiting the cognitive buffer hypothesis for the evolution of large brains. Biol. Lett. 5, 130–133. (doi:10.1098/rsbl.2008.0621) 18. Sol D, Bacher S, Reader SM, Lefebvre L. 2008 Brain size predicts the success of mammal species introduced into novel environments. Am. Nat. 172, S63–S71. (doi:10.1086/588304) 19. Amiel JJ, Tingley R, Shine R. 2011 Smart moves: effects of relative brain size on establishment success of invasive amphibians and reptiles. PLoS ONE 6, e18277. (doi:10.1371/journal.pone. 0018277) 20. Sayol F, Maspons J, Lapiedra O, Iwaniuk AN, Székely T, Sol D. 2016 Environmental variation and the evolution of large brains in birds. Nat. Commun. 7, 13971. (doi:10.1038/ncomms13971) 21. Howell KJ, Walsh MR. 2023 Transplant experiments demonstrate that larger brains are favoured in high-competition environments in Trinidadian killifish. Ecol. Lett. 26,53–62. (doi:10.1111/ ele.14133) 22. Giurfa M. 2015 Learning and cognition in insects. WIREs Cogn. Sci. 6, 383–395. (doi:10.1002/wcs. 1348) 23. Perry CJ, Barron AB, Chittka L. 2017 The frontiers of insect cognition. Curr. Opin. Behav. Sci. 16, 111–118. (doi:10.1016/j.cobeha.2017.05.011) 5 royalsocietypublishing.org/journal/rsbl Biol. Lett. 19: 20230296 Downloaded from https://royalsocietypublishing.org/ on 05 May 2025 24. Chittka L, Niven J. 2009 Are bigger brains better? Curr. Biol. 19, R995–R1008. (doi:10.1016/j.cub. 2009.08.023) 25. Avargues-Weber A, Deisig N, Giurfa M. 2011 Visual cognition in social insects. Annu. Rev. Entomol. 56, 423–443. (doi:10.1146/annurev-ento-120709-144855) 26. Sheehan MJ, Tibbetts EA. 2011 Specialized face learning is associated with individual recognition in paper wasps. Science 334, 1272–1275. (doi:10. 1126/science.1211334) 27. Collado MÁ, Montaner CM, Molina FP, Sol D, Bartomeus I. 2021 Brain size predicts learning abilities in bees. R. Soc. Open Sci. 8, 201940. (doi:10.1098/rsos.201940) 28. Streinzer M, Kelber C, Pfabigan S, Kleineidam CJ, Spaethe J. 2013 Sexual dimorphism in the olfactory system of a solitary and a eusocial bee species. J. Comp. Neurol. 521, 2742–2755. (doi:10. 1002/cne.23312) 29. Roselino AC, Hrncir M, da Cruz Landim C, Giurfa M, Sandoz J-C. 2015 Sexual dimorphism and phenotypic plasticity in the antennal lobe of a stingless bee, Melipona scutellaris.J. Comp. Neurol. 523, 1461–1473. (doi:10.1002/cne.23744) 30. Sayol F et al. 2020 Feeding specialization and longer generation time are associated with relatively larger brains in bees. Proc. R. Soc. B 287, 20200762. (doi:10.1098/rspb.2020.0762) 31. Kendall LK et al. 2019 Pollinator size and its consequences: robust estimates of body size in pollinating insects. Ecol. Evol. 9, 1702–1714. (doi:10.1002/ece3.4835) 32. Bürkner P-C. 2017 Brms: an R package for bayesian multilevel models using Stan. J. Stat. Softw. 80, 1–28. 33. Hedtke Shannon M, Patiny Sébastien, Danforth Bryan N. 2013 The bee tree of life: a supermatrix approach to apoid phylogeny and biogeography. BMC Evol. Biol. 13, 138. (doi:10.1186/1471-2148- 13-138) 34. Paradis E, Blomberg S, Bolker B, Brown J, Claude J, Cuong HS, Desper R, Didier G. 2019 Package ‘ape’. Anal. Phylogenetics Evol. 2, 47. 35. Revell LJ. 2012 Phytools: an R package for phylogenetic comparative biology (and other things). Methods Ecol. Evol. 3, 217–223. (doi:10. 1111/j.2041-210X.2011.00169.x) 36. Hadfield JD. 2010 MCMC methods for multiresponse generalized linear mixed models: the MCMCglmm R package. J. Stat. Softw. 33,1–22. (doi:10.18637/jss.v033.i02) 37. Chamberlain S, Barve V, Mcglinn D, Oldoni D, Desmet P, Geffert L, Ram K. 2023 rgbif: Interface to the Global Biodiversity Information Facility API. R package version 3.7.8. See https://CRAN.R-project. org/package=rgbif. 38. Hijmans RJ, Bivand R, Forner K, Ooms J, Pebesma E, Sumner MD. 2022 Package ‘terra’. See https://CRAN. R-project.org/package=terra. 39. Dormann CF et al. 2008 Prediction uncertainty of environmental change effects on temperate European biodiversity. Ecol. Lett. 11, 235–244. (doi:10.1111/j.1461-0248.2007.01142.x) 40. Patefield W. 1981 Algorithm AS 159: an efficient method of generating random R × C tables with given row and column totals. J. R. Stat. Soc. C (Appl. Stat.)30,91–97. (doi:10.2307/2346669) 41. R Core Team. 2021 R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. 42. Wickham H et al. 2019 Welcome to the tidyverse. J. Open Source Softw. 4, 1686. (doi:10.21105/joss. 01686) 43. Wheater CP. 2002 Urban habitats. London, UK: Routledge. 44. Allasino ML, Marrero HJ, Dorado J, Torretta JP. 2019 Scientific note: first global report of a bee nest built only with plastic. Apidologie 50, 230–233. (doi:10. 1007/s13592-019-00635-6) 45. Lowenstein DM, Matteson KC, Minor ES. 2019 Evaluating the dependence of urban pollinators on ornamental, nonnative, and weedy floral resources. Urban Ecosyst. 22, 293–302. (doi:10.1007/s11252- 018-0817-z) 46. Ducatez S, Sol D, Sayol F, Lefebvre L. 2020 Behavioural plasticity is associated with reduced extinction risk in birds. Nat. Ecol. Evol. 4, 788–793. (doi:10.1038/s41559-020-1168-8) 47. Bateman PW, Fleming PA. 2012 Big city life: carnivores in urban environments. J. Zool. 287, 1–23. (doi:10.1111/j.1469-7998.2011.00887.x) 48. Theodorou P, Baltz LM, Paxton RJ, Soro A. 2021 Urbanization is associated with shifts in bumblebee body size, with cascading effects on pollination. Evol. Appl. 14,53–68. (doi:10.1111/ eva.13087) 49. Greenleaf SS, Williams NM, Winfree R, Kremen C. 2007 Bee foraging ranges and their relationship to body size. Oecologia 153, 589–596. (doi:10.1007/ s00442-007-0752-9) 50. Geslin B, Gauzens B, Thébault E, Dajoz I. 2013 Plant pollinator networks along a gradient of urbanisation. PLoS ONE 8, e63421. (doi:10.1371/ journal.pone.0063421) 51. Buchholz S, Egerer MH. 2020 Functional ecology of wild bees in cities: towards a better understanding of trait–urbanization relationships. Biodivers. Conserv. 29, 2779–2801. (doi:10.1007/s10531-020- 02003-8) 52. Prendergast KS, Dixon KW, Bateman PW. 2022 A global review of determinants of native bee assemblages in urbanised landscapes. Insect Conserv. Divers. 15,385–405. (doi:10.1111/icad.12569) 53. Ducatez S, Clavel J, Lefebvre L. 2015 Ecological generalism and behavioural innovation in birds: technical intelligence or the simple incorporation of new foods? J. Anim. Ecol. 84,79–89. (doi:10.1111/ 1365-2656.12255) 54. Sol D, Sayol F, Ducatez S, Lefebvre L. 2016 The lifehistory basis of behavioural innovations. Phil. Trans. R. Soc. B 371, 20150187. (doi:10.1098/rstb.2015.0187) 55. Santini L, González-Suárez M, Russo D, Gonzalez- Voyer A, von Hardenberg A, Ancillotto L. 2019 One strategy does not fit all: determinants of urban adaptation in mammals. Ecol. Lett. 22, 365–376. (doi:10.1111/ele.13199) 56. Sayol F, Sol D, Pigot AL. 2020 Brain size and life history interact to predict urban tolerance in birds. Front. Ecol. Evol. 8, 58. (doi:10.3389/fevo.2020.00058) 57. Lihoreau M, Latty T, Chittka L. 2012 An exploration of the social brain hypothesis in insects. Front. Physiol. 3, 442. (doi:10.3389/fphys.2012.00442) 58. Barth M, Heisenberg M. 1997 Vision affects mushroom bodies and central complex in Drosophila melanogaster.Learn. Mem. 4, 219–229. (doi:10. 1101/lm.4.2.219) 59. Withers G, Fahrbach S, Robinson G. 1995 Effects of experience and juvenile hormone on the organization of the mushroom bodies of honey bees. J. Neurobiol. 26, 130–144. (doi:10.1002/neu. 480260111) 60. Fahrbach SE, Robinson GE. 1995 Behavioral development in the honey bee: toward the study of learning under natural conditions. Learn. Mem. 2, 199–224. (doi:10.1101/lm.2.5.199) 61. Lanuza J, Collado M, Ferran S, Sol D, Bartomeus I. 2023 Data from: Brain size predicts bees’tolerance to urban environments. Dryad Digital Repository. (doi:10.5061/dryad.zw3r228dr) 62. Lanuza Jb, Collado MÁ, Sayol f, Sol D, Bartomeus I. 2023 Brain size predicts bees’tolerance to urban environments. Figshare. (doi:10.6084/m9.figshare.c. 6935821) 6 royalsocietypublishing.org/journal/rsbl Biol. Lett. 19: 20230296 Downloaded from https://royalsocietypublishing.org/ on 05 May 2025