research paper about evolution

Loading metrics

Open Access

Essays articulate a specific perspective on a topic of broad interest to scientists.

See all article types »

Evolutionary Biology for the 21st Century

* E-mail: [email protected]

Affiliation Museum of Comparative Zoology and Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, Massachusetts, United States of America

Affiliation Department of Zoology, Oregon State University, Corvallis, Oregon, United States of America

Affiliation Departments of Developmental Biology and Computer Science, Stanford University, Stanford, California, United States of America

Affiliation Department of Biology, University of Virginia, Charlottesville, Virginia, United States of America

Affiliation Department of Biology, Clark University, Worcester, Massachusetts, United States of America

Affiliations Museum of Comparative Zoology and Department of Organismic and Evolutionary Biology, Harvard University, Cambridge, Massachusetts, United States of America, Department of Molecular and Cellular Biology, Harvard University, Cambridge, Massachusetts, United States of America

Affiliation Department of Biochemistry and Biophysics, University of California, San Francisco, California, United States of America

Affiliation Department of Ecology and Evolutionary Biology, Yale University, New Haven, Connecticut, United States of America

Affiliations Museum of Vertebrate Zoology, University of California, Berkeley, California, United States of America, The Australian National University, Canberra, Australia

Affiliation Department of Biology, University of Rochester, Rochester, New York, United States of America

Affiliation Department of Biology, Stanford University, Stanford, California, United States of America

Affiliation Department of Biology, University of Munich, Munich, Germany

Affiliation Department of Biology, University of Missouri, St. Louis, Missouri, United States of America

Affiliation Florida Museum of Natural History, University of Florida, Gainesville, Florida, United States of America

Affiliation Department of Ecology, Evolution and Marine Biology, University of California, Santa Barbara, California, United States of America

• Jonathan B. Losos, 
• Stevan J. Arnold, 
• Gill Bejerano, 
• E. D. Brodie III, 
• David Hibbett, 
• Hopi E. Hoekstra, 
• David P. Mindell, 
• Antónia Monteiro, 
• Craig Moritz, 

Published: January 8, 2013

• https://doi.org/10.1371/journal.pbio.1001466
• Reader Comments

Citation: Losos JB, Arnold SJ, Bejerano G, Brodie ED III, Hibbett D, Hoekstra HE, et al. (2013) Evolutionary Biology for the 21st Century. PLoS Biol 11(1): e1001466. https://doi.org/10.1371/journal.pbio.1001466

Copyright: © 2013 Losos et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: The workshop that led to this report was funded by the National Science Foundation. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

We live in an exciting time for biology. Technological advances have made data collection easier and cheaper than we could ever have imagined just 10 years ago. We can now synthesize and analyze large data sets containing genomes, transcriptomes, proteomes, and multivariate phenotypes. At the same time, society's need for the results of biological research has never been greater. Solutions to many of the world's most pressing problems—feeding a global population, coping with climate change, preserving ecosystems and biodiversity, curing and preventing genetically based diseases—will rely heavily on biologists, collaborating across disciplines.

Theodosius Dobzhansky famously proclaimed that “nothing makes sense in biology except in the light of evolution." Though Dobzhansky's statement is sometimes dismissed by biologists in other fields as self-promotion, recent advances in many areas of biology have shown it to be prophetic. For example, genomics, which emerged mostly from molecular biology, is now steeped in evolutionary biology. Evolutionary theory helps to explain our origins, our history, and how we function as organisms and interact with other life forms, all of which are crucial to understanding our future (e.g., [1] – [5] ). Evolutionary approaches have helped reconstruct the history of human culture, including, for example, the history of human populations and languages [6] – [11] . And the impact of evolutionary biology is extending further and further into biomedical research and nonbiological fields such as engineering, computer sciences, and even the criminal justice system.

The pervasive relevance of evolution can be seen in the 2009 report commissioned by the National Research Council of the National Academies, A New Biology for the 21 st Century [12] , which identified four broad challenges for biology: develop better crops to feed the world, understand and sustain ecosystem function and biodiversity in a changing world, expand sustainable alternative energy sources, and understand individual health. In each of these areas, the report noted, evolutionary concepts and analyses have played—and will continue to play—an integral role.

It's hard to overstate evolutionary biology's power to explain the living world and our place in it. Many applications of evolutionary theory and methods—from animal and plant breeding to vaccine development to management of biological reserves and endangered species—affect society and promote human well-being [13] , [14] . Much human activity, however, is changing Earth's climate and habitats, with uncertain but potentially severe environmental stresses on many other species [15] – [18] , and the solutions to the many resulting problems may well require understanding evolutionary interactions among species and their mutual dependencies.

Our ability to apply evolutionary concepts to a wide range of problems has never been greater. Changes in the availability of data and an emerging scientific culture that embraces rapid, open access to many kinds of data (genomic, phenotypic, and environmental), along with a computational infrastructure that can connect these rich sources of data ( [19] , Figure 1 ), will transform the nature and scale of problems that can be addressed by evolutionary biology.

• PPT PowerPoint slide
• PNG larger image
• TIFF original image

All this can be connected to the Tree of Life (phylogeny), from populations to entire clades, and is enabled by new protocols and networks in biodiversity informatics.

https://doi.org/10.1371/journal.pbio.1001466.g001

Periodically, groups of scientists meet to identify new opportunities in evolutionary biology and associated disciplines (e.g., [12] , [20] – [23] ). Rather than set a specific research agenda for the future—clearly the charge of individual investigators—the aim has been to identify new themes and research directions that are already emerging in the field and to focus on the intersection of fundamental problems with new technologies and methods. In the following sections, we briefly highlight some key applications of evolutionary biology, provide examples of emerging research areas, and identify infrastructure and training needs.

Evolutionary Applications

Evolutionary medicine.

The new field of “evolutionary medicine" [24] – [26] posits that understanding our evolutionary past can inform us of the causes of perplexing common diseases. For instance, diabetes and autoimmune diseases such as asthma may represent mismatches between evolutionary adaptation to the environments in which humans evolved and current conditions. In addition, some age-related conditions, such as cancer, can be understood as the outcome of selection for early reproduction, when humans faced dying of disease or predation at an early age. This long-term selection on the cellular machinery to optimize growth and survival through early reproduction may now explain the prevalence of cancer late in life, a modern malaise that emerges because of the recent, rapid extension of postreproductive lifespan [27] . Aside from providing explanations for the occurrence of diseases, the field of evolutionary medicine is also concerned with suggesting strategies for slowing the evolution of resistance in pathogen populations [28] – [30] ; strategies to improve public health and reduce the incidence of common diseases [31] , [32] ; prediction of diseases that may emerge from recent host-shifts to humans [33] ; discovery, design, and enhancement of drugs and vaccines (e.g., [34] ); and understanding the role of the microbiome in human health [35] .

Feeding the Human Population

Feeding the rapidly growing human population, especially with increasing stress on agricultural systems from climate change, continues to be a major challenge. The green revolution, from the 1950s onwards, rested on selective plant breeding for larger yields and was underpinned by evolutionary theory [36] . Currently, the trend is to rely on biotechnology to introduce either herbicide resistance genes or herbivore-directed toxins, such as Bt, to combat crop competitors and herbivores, respectively, and thus promote increasing yields [37] . Unfortunately, genetically modified crops are genetically uniform and so do not represent a long-term solution against the evolution of either herbicide or Bt resistance. In addition, these herbicide resistance or toxin genes can be transferred to other nontarget species through pollen-mediated hybridization, rendering them resistant or toxic as well [38] . The agriculture of the future must incorporate evolutionary thinking to reduce the evolution of resistance and the risk of pathogen outbreaks. Maintaining genetic diversity in crop and animal production systems considerably reduces these risks [38] .

Sustaining Biological Diversity

Evolutionary approaches have often been applied to the conservation of species and ecosystems [13] , [39] – [42] . Linking spatial data on phenotypes, genomes and environments in a phylogenetic context allows us to identify and name Earth's diverse life forms. This linkage, in turn, helps to provide the basic units needed to quantify taxonomic diversity and to pursue its conservation. Determining phylogenetic relationships among species allows us to identify their unique adaptations and provides the historical context to understand how they arose [43] – [45] . Evolutionary approaches also can be used to determine the origins of invasive species [46] – [48] and to help design effective remediation [49] , [50] . Collectively, understanding the distribution of current biodiversity and its evolutionary response to past environmental change is fundamental to mitigating effects of ongoing habitat loss and climate change [51] . Given the rate of anthropogenic climate change, evolutionary theory and experiments can help predict vulnerability (i.e., inability to adapt) of species and thus improve conservation strategies [52] .

Computation and Design

Models of mutation, inheritance, and selection have inspired the development of computational evolutionary algorithms that are used to solve complex problems in many fields [53] , [54] . In particular, engineering and design processes have incorporated evolutionary computation, leading to improvements in design of cars, bridges, traffic systems robots, and wind turbine energy, among other applications [55] – [59] .

Evolution and Justice

Genealogical relationships bear on many court cases. Is the defendant really the parent of the plaintiff? Does the evidence (e.g., blood, semen, or skin cells) at the crime scene tend to exonerate or implicate a suspect? Evolutionary methods, particularly population genetics, are now used frequently in forensics and court cases to test the link of crime scene evidence to individuals [60] , and phylogenetic analyses have been vetted and accepted as valid and appropriate methods for determining facts of history in the United States criminal court system [61] .

Emerging Research and Future Challenges in Evolutionary Biology

Divining the direction of future scientific research is always fraught with difficulty. Nonetheless, we feel that it is possible to identify some general themes that will be important in coming years. We also present some examples of classic research problems that remain unsolved and that might well be the focus of future work, as well as new and important questions for which evolutionary approaches may be key.

The flood of data in all areas of evolutionary biology poses important theoretical challenges: new kinds of theory are sometimes required to make sense of new kinds of data. We can already point to certain broad areas of evolutionary biology that will likely demand sustained theoretical work. These include the elaboration of more formal theories for evolutionary developmental biology (e.g., analysis of gene network evolution and modification); the more complete incorporation of the roles of epigenetics, behavior, and plasticity in models of trait evolution; analysis of units of selection; and attempts to construct a quantitative and predictive theory that describes the genetic basis of adaptation. In other areas, problems will likely be more statistical than theoretical. Indeed, the enormous quantity of genome data poses serious statistical challenges even for fields that already possess strong theoretical foundations, such as evolutionary genetics.

The Explosion and Diversity of Data

DNA sequencing can now generate whole-genome data not only for single representatives of a few species but for multiple individuals from multiple conspecific populations and even from entire communities. Such multilevel data are giving rise to whole new fields of study (e.g., population genomics and metagenomics) as well as to new theoretical, computational, and data management challenges.

One particularly exciting avenue of research afforded by new genomic technology is the possibility of directly observing the dynamics of evolution. In the last few years, genomic analyses of experimental evolution have yielded new understanding of how RNA molecules, viruses, and bacteria evolve (bacteria: [62] , [63] ; virus: [64] ; RNA molecules: [65] ). This approach is now being applied to eukaryotic model systems such as C. elegans and yeast [66] – [68] . These efforts will continue to expand and will surely involve natural systems in field settings. Past evolution, for example, can be inferred from samples derived from ancient specimens, archived material in museum collections, lake sediments, and glacier cores. Contemporary evolution can be inferred from genomic sampling across seasons and years and can be detected in response to climatic disturbances such as El Niño events and to manmade environmental changes such as oil spills. In parallel with long-term ecological data (e.g., species abundance and distributions through time), we can now track genomic variation through ecological and evolutionary time. This capability, together with the realization that evolutionary change can occur on ecological timescales [69] , provides an important new window on real-time evolution. Evolution on contemporary time scales is likely to be especially important in the context of evolving pathogens, pest resistance, and the response of organisms to rapid environmental change.

While the explosion of data on genome sequences has received the most attention, supplementing these data with information on the natural history of individuals, species, and their environments will be important. Core information from field-collected specimens always includes species identity and place and time of collection, but increasingly, this information is being enriched with links to field notes and phenotypic (e.g., images), behavioral (e.g., sounds), and genomic data in a variety of databases (e.g., Morphbank— http://www.morphbank.net/ , Barcode of Life— http://www.barcodeoflife.org/ , Macaulay Library— http://macaulaylibrary.org/ ). Precise information on place, time, and reproductive stage can be integrated with data on local environmental conditions, often obtained from remote sensing [70] . The key is to connect information across repositories, such as natural history museums and genomic databases ( Figure 2 ). Such efforts will also include observational data provided by the broader public [71] .

Photo by Jeremiah Trimble, Department of Ornithology, Museum of Comparative Zoology, Harvard University.

https://doi.org/10.1371/journal.pbio.1001466.g002

Evolutionary Processes That Shape Genomic and Phenotypic Variation

The availability of genomic data from a remarkable range of species has allowed the alignment and comparison of whole genomes. These comparative approaches have been used to characterize the relative importance of fundamental evolutionary processes that cause genomic evolution and to identify particular regions of the genome that have experienced recent positive selection, recurrent adaptive evolution, or extreme sequence conservation [72] – [75] . Yet more recently, resequencing of additional individuals or populations is also allowing genome-wide population genetic analyses within species [76] – [82] . Such population-level comparisons will allow even more powerful study of the relative importance of particular evolutionary processes in molecular evolution as well as the identification of candidate genomic regions that are responsible for key evolutionary changes (e.g., sticklebacks [83] , butterflies [84] , Arabidopsis [85] ). These data, combined with theoretical advances, should provide insight into long-standing questions such as the prevalence of balancing selection, the relative frequency of strong versus weak directional selection, the role of hybridization, and the importance of genetic drift. A key challenge will be to move beyond documenting the action of natural selection on the genome to understanding the importance of particular selective agents. For example, what proportion of selection on genomes results from adaptation to the abiotic environment, coevolution of species, sexual selection, or genetic conflict? Finally, as sequencing costs continue to drop and analytical tools improve, these same approaches may be applied to organisms that present intriguing evolutionary questions but were not tractable methodologically just a few years ago. The nonmodel systems of today may well become the model systems of tomorrow [86] .

Earth–Biosphere Interactions Over Vast Stretches of Time and Space

We are in the midst of a massive perturbation of natural communities as species respond to human-driven changes in climate and land cover. Beyond the challenge of understanding the capacity of species to respond (e.g., [51] , [87] ) and the potential for dramatic state-shifts in the biosphere [17] lies the daunting problem of understanding the many interactions between community-scale ecological dynamics and evolution of traits within populations.

We now can also ask if evolution matters for ecosystem functioning. To date, most ecosystem studies have assumed that all individuals that compose a population within a community are equivalent ecologically. But individuals within a population are variable, and this variation may lead to ecological interactions that are in a continual state of evolutionary flux as ecologically driven evolutionary change feedbacks to alter the ongoing ecological interactions [88] – [90] . This evolutionary perspective on communities is an emerging area that will require the acquisition and analysis of large, temporal samples of genomic and phenotypic data, as well as the direct measurement of fitness. Samples that include paleo/historical DNA as well as contemporary DNA might be especially valuable by providing a temporal view on such questions.

Understanding Biological Diversification

A major and urgent challenge is to improve knowledge of the identity and distribution of species globally. While we need to retain the traditional focus on phenotypes, powerful new capabilities to obtain and interpret both genomic and spatial data can and should revolutionize the science of biodiversity. Building on momentum from single-locus “barcoding" efforts, new genome-level data can build bridges from population biology to systematics [91] . By establishing a comprehensive and robust “Tree of Life," we will improve understanding of both the distribution of diversity and the nature and timing of the evolutionary processes that have shaped it.

Studies of the biodiversity of Bacteria and Archaea are complicated by the widespread occurrence of lateral gene transfer. However, the phylogeny of these organisms and their genes remains critical to understanding their scope, origins, distributions, and change over time [92] . The advent of metagenomic sequencing of environmental microbial communities has revealed greater diversity and flux of genotypes than ever imagined, defying conventional species concepts and presenting a major challenge to applying traditional evolutionary and ecological theory to understanding microbial diversity [93] , [94] . Addressing this challenge will be necessary to advance microbial ecology beyond the descriptive stage. Moreover, it is only with such understanding that a natural history of microbes can be developed, leading to more meaningful exploration of genomic structure and function, the origin of novel genes, and increased knowledge of microbial influences at both the global and individual (microbiome) levels.

In addition to documenting biodiversity, more research is needed on the processes that produce this diversity. While research on speciation has seen a resurgence over the last two decades [95] – [97] , new tools—including genomic data—can support new approaches for a number of important questions, including discovering genomic signatures underlying the evolution of prezygotic reproductive isolation, and describing how hybridization, contact between incipient species, genome reorganization, and genome duplication, affect speciation.

Understanding the diversification of species and the origin of adaptations poses a number of challenges for evolutionary biologists, including integration of the fossil record with diversification inferred from the relationships among contemporary species; determining the relationship between lineage and phenotypic diversification; understanding the factors that lead to the replacement of clades over time; understanding the occupancy of ecological niche space through evolutionary diversification, adaptive radiation, and extinction; and assessing the role that evolving species interactions play in diversification.

All evolution has an ecological context that is essential to the interpretation of diversification. Consequently, we need to incorporate analyses of the environmental context of evolution, particularly species interactions that are likely to both set limits to diversification and promote evolutionary novelty. For all these reasons, further integration of paleontology with other fields of evolutionary biology, as well as development of genetic-evolutionary research programs on clades with excellent fossil records (e.g., foraminifera, diatoms, mollusks; Figure 3 ), will be important. More generally, uniting understanding of evolutionary pattern and process will require reductionist studies on evolutionary mechanisms of species formation and phenotypic change, as well as broadly historical studies that incorporate phylogenetic, paleontological, and geological data.

Genomic sequence data for stickleback fish is now providing insight into evolutionary patterns, such as the reduction in the pelvic skeleton, manifest both in the fossil record and in extant populations [83] . Photograph courtesy Peter J. Park.

https://doi.org/10.1371/journal.pbio.1001466.g003

As we address these challenges, the importance of natural history data cannot be overemphasized. Observations on the natural history of organisms, the basic building blocks of more detailed studies of ecology and evolution, are critical if we are to preserve and understand biological diversity [98] . Though few would argue against this point in principle, natural history research is rarely encouraged or supported. Making the acquisition of natural history data an integral part of hypothesis-driven science is now more important than ever.

Logistical Issues and Opportunities

To take full advantage of technological advances, especially the availability of new data types and databases, we must confront several challenges that involve community resources and how we use them. Some challenges concern infrastructure, while others involve aspects of scientific culture. Still others involve how we train the next generation of evolutionary biologists, who will need a better grasp of diverse disciplines—from natural history to developmental biology—as well as bioinformatics skills to handle immense datasets across multiple fields (see Text S1 and also Figure S2 ).

The infrastructure challenges center on creation of new kinds of databases—for instance, ones that focus on (continuous) phenotypic and not merely (discrete) DNA sequence data—as well as on integration across databases to allow synthesis of very different kinds of data (see Text S2 ). The cultural challenges center on the need for supporting a climate of scientific openness. Maintaining openness will require evolutionary biologists to make the results of their research available rapidly and in a form that is most useful to their colleagues. The scientific community has already made great strides in this direction (for instance by requiring deposition of data as a condition for publication and by founding open access journals), but additional steps are necessary. We strongly support the movement toward open access for the scientific literature to accelerate research and allow more investigators to participate. We also encourage provision of open software, data and databases, as well as their computational reuse and distillation, as outlined by Lathrop et al. (2011) [99] . These individual and community efforts will be increasingly necessary for development of new research programs and insights.

As noted at the outset, we live in an exciting time for evolutionary biology. The field has embraced the “omic" revolution, and answers to many classic questions, which have motivated research for a century, are now within reach. The study of evolution, which in the past was often equated with changes in gene frequencies in populations, has become more holistic and integrative. Researchers are increasingly interested in exploring how interactions among genes, individuals, and environments have shaped the evolutionary process, both at micro- and macrolevels. At the same time, large challenges such as global warming, novel infectious diseases, and threats to biodiversity are increasing, and the opportunity for evolutionary biologists to contribute to their resolution has never been greater.

Realizing the full potential inherent in evolutionary biology is, however, far from assured. The task of integrating evolutionary knowledge within and across scales of biological organization, as discussed above, requires development of many comparative databases and analytical tools. We would do well to collaborate broadly, cultivating new expertise, and to watch out for the unexpected, as analyses of new kinds of data can reveal that preconceptions are unfounded.

Because most of our science is supported by limited public funds, evolutionary biologists and ecologists should support and participate in efforts to help the public understand the issues and the value of scientific understanding. Science in general and evolutionary science in particular are often politicized, exactly because of their fundamental importance to human society. The next 20 years hold the promise of a golden age for evolutionary biology. Whether we realize that promise depends in part on how effectively we communicate that message.

Cyberinfrastructure —The research environments that support advanced data acquisition, data storage, data management, data integration, data mining, data visualization, and other computing and information processing services distributed over the Internet beyond the scope of a single institution. In scientific usage, cyberinfrastructure is a technological solution to the problem of efficiently connecting laboratories, data, computers, and people.

Evolutionary developmental biology —The study of the evolution of development, often by the comparative study of gene expression patterns through the course of development in different species.

Evolutionary genetics —Population and quantitative genetics.

Gene network —A flow diagram describing the interactions among genes during development that affect a particular phenotype or set of phenotypes.

Genomics —The study of the entire complement of DNA in organisms (Genome), including is sequence and organization.

GMO —Genetically modified organisms in which the genome has been deliberately changed; transgenic organisms resulting from DNA manipulations.

Lateral (horizontal) gene transfer —Genetic transfer between species, as opposed to vertical gene transmission from parents to offspring in a lineage.

Metadata —Data associated with individual DNA sequences or organismal specimens (e.g., the date and locality where the sample originated, its ecological context, etc.).

Model organism —Organisms whose genome has been sequenced and for which sophisticated tools for genetic manipulation are available.

Natural history —The entire description of an organism, including its phenotype, genome, and ecological context (i.e., abiotic niche as well as its biotic interactions with other species).

Nonmodel organism —Organisms whose genome has not been sequenced and/or for which sophisticated tools for genetic manipulation are not available.

Ontology —The naming of categories, especially of the functions of genes.

Population genetics —The study of the evolutionary forces that change the genetic composition of a population; the discipline is often concerned with evolution at one or a few genetic loci.

Quantitative genetics —The study of the inheritance and evolution of traits that are typically affected by many genetic loci.

Transgenic tools —Tools that enable the deliberate transfer of DNA sequences from one organism to another or the deletion or modification of DNA sequences, in every cell, in one organism.

Supporting Information

An example of the enormous phylogenetic trees that soon will represent the norm in phylogenetic analyses. This is the consensus tree of the maximum likelihood phylogenies for 55,473 species of seed plants with the location of significant shifts in species diversification rates marked in red across the tree. Adapted from [4] .

https://doi.org/10.1371/journal.pbio.1001466.s001

The Phenomobile, a remote sensing field buggy, and the Blimp, for remotely imaging an entire field. The Phenomobile integrates a variety of remote sensing technologies for measuring phenotypic variables on many plants simultaneously. The buggy straddles a plot and collects measurements of plant temperature, stress, chemistry, color, size and shape, as well as measures of senescence. The Blimp is designed to image all the plants in an entire field from a height of 30–80 m using both infrared and digital color cameras. These technologies were developed by David Deery of the High Resolution Plant Phenomics Centre at the Commonwealth Scientific and Industrial Research Organisation in Australia. Photo credit: Carl Davies, CSIRO Plant Industry.

https://doi.org/10.1371/journal.pbio.1001466.s002

Training to sustain evolutionary biology.

https://doi.org/10.1371/journal.pbio.1001466.s003

Infrastructure needs and opportunities in evolutionary biology.

https://doi.org/10.1371/journal.pbio.1001466.s004

Acknowledgments

The workshop that led to this report was funded by the National Science Foundation. We thank the American Society of Naturalists, the Society for the Study of Evolution, and the Society of Systematic Biologists for organizational and planning assistance. Many thanks to Melissa Woolley for invaluable assistance with logistics and manuscript preparation and to M. Bell, J. Borewitz, D. Jablonski, P. Parks, K. Roy, S. Smith, J. Trimble, and A. Weirman for help procuring images.

• 1. Wilson EO (2002) The future of life. New York: Alfred A. Knopf.
• 2. Millennium Ecosystem Assessment (2005) Ecosystems and human well-being: synthesis. Washington, DC: Island Press.
• 3. Mindell DP (2006) The evolving world: evolution in everyday life. Cambridge, MA: Harvard University Press.
• 4. Chivian E, Bernstein A (2008) Sustaining life: how human health depends on biodiversity. Oxford; New York: Oxford University Press.
• 5. Held LI Jr (2009) Quirks of human anatomy: an evo-devo look at the human body. Cambridge: Cambridge University Press.
• View Article
• Google Scholar
• 11. Pagel M (2012) Wired for culture: origins of the human social mind. New York: W.W. Norton & Company.
• 12. National Research Council (US) Committee on a New Biology for the 21st Century: Ensuring the United States Leads the Coming Biology Revolution (2009) A new biology for the 21st century: ensuring the United States leads the coming biology revolution. Washington, DC: National Academies Press. Available: https://download.nap.edu/catalog.php?record_id=12764 . Accessed May 25, 2012.
• 20. National Science Foundation (1998) Frontiers in population biology: report of a population biology task force. Arlington, VA: National Science Foundation.
• 21. National Science Foundation (2005) Frontiers in evolutionary biology. Arlington, VA: National Science Foundation.
• 23. National Research Council. 2010. Research at the intersection of the physical and life sciences. Washington, DC: National Academies Press.
• 25. Gluckman P, Beedle A, Hanson M (2009) Principles of evolutionary medicine. Oxford; New York: Oxford University Press.
• 36. Kingsbury N (2009) Hybrid: the history and science of plant breeding. Chicago: University of Chicago Press. 512 p.
• 53. Poli R, Langdon WB, McPhee NF (2008) A field guide to genetic programming. Available: http://dces.essex.ac.uk/staff/rpoli/gp-field-guide/A_Field_Guide_to_Genetic_Programming.pdf . Accessed May 22, 2012.
• 54. Chiong R, Weise T, Michalewicz Z (2011) Variants of evolutionary algorithms for real-world applications. New York: Springer-Verlag.
• 58. Byrne J, Fenton M, Hemberg E, McDermott J, O'Neill M, et al.. (2011) Combining structural analysis and multi-objective criteria for evolutionary architectural design. In: Di Chio C, et al.. editors. Applications of evolutionary computation: EvoApplications 2011: EvoCOMNET, EvoFIN, EvoHOT, EvoMUSART, EvoSTIM, and EvoTRANSLOG, Torino, Italy, April 27–29, 2011, Proceedings, Part II.
• 95. Howard DJ, Berlocher SH, eds (1998) Endless forms: species and speciation. New York; Oxford: Oxford University Press.
• 96. Coyne JA, Orr HA (2004) Speciation. Sunderland: Sinauer Press.
• 97. Dieckmann U, Doebeli M, Metz JAJ, Tautz D, editors (2004) Adaptive speciation. Cambridge studies in adaptive dynamics. Cambridge, UK: Cambridge University Press.

Science and evolution

• February 2019
• Genetics and Molecular Biology 42(1)

• Federal University of Rio de Janeiro

• University of Brasília

Discover the world's research

• 25+ million members
• 160+ million publication pages
• 2.3+ billion citations
• PROG BIOPHYS MOL BIO

• Reut Stahi-Hitin

• Sophie Walker

• M. Pd Istiningsih
• David Geelan
• Mohamad Agung Rokhimawan
• Aulia Himmawati

• Rosivânia de Queiróz Ribeiro

• Andrew J. Petto

• Gail M. Perry‐ryder

• Michael L. Rutledge
• Melissa A. Warden
• Recruit researchers
• Join for free
• Login Email Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google Welcome back! Please log in. Email · Hint Tip: Most researchers use their institutional email address as their ResearchGate login Password Forgot password? Keep me logged in Log in or Continue with Google No account? Sign up

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

• View all journals

Evolutionary biology articles from across Nature Portfolio

Evolutionary biology is a subdiscipline of the biological sciences concerned with the origin of life and the diversification and adaptation of life forms over time.

Latest Research and Reviews

Chromosome-level genome assembly of the morabine grasshopper Vandiemenella viatica 19

• Suvratha Jayaprasad
• Octavio Manuel Palacios-Gimenez

The evolutionary theory of cancer: challenges and potential solutions

Clonal evolution is now a central theoretical framework in cancer research. In this Perspective, Laplane and Maley identify challenges to that theory such that some non-evolutionary phenomena in cancer cannot be captured by the theory. They also outline how other challenges, including non-genetic heredity, phenotypic plasticity, reticulate evolution and clone diversity, can be included in an expanded cancer evolutionary theory.

• Lucie Laplane
• Carlo C. Maley

Recurrent evolution and selection shape structural diversity at the amylase locus

The impact of structural variation on the evolution of the amylase genes is explored using human pangenome resources and ancient DNA data.

• Davide Bolognini
• Alma Halgren
• Peter H. Sudmant

Satellitome analysis on the pale-breasted thrush Turdus leucomelas (Passeriformes; Turdidae) uncovers the putative co-evolution of sex chromosomes and satellite DNAs

• Guilherme Mota Souza
• Rafael Kretschmer
• Marcelo de Bello Cioffi

Genetic patterns reveal geographic drivers of divergence in silvereyes ( Zosterops lateralis)

• Annika Radu
• Christine Dudgeon
• Dominique A. Potvin

Satellite DNAs and the evolution of the multiple X 1 X 2 Y sex chromosomes in the wolf fish Hoplias malabaricus (Teleostei; Characiformes)

• Gustavo Akira Toma
• Alexandr Sember

News and Comment

The SplitsTree App: interactive analysis and visualization using phylogenetic trees and networks

• Daniel H. Huson
• David Bryant

Loss of sex in brown algae

An analysis of phenotypic and genomic changes that accompany the loss of sex in brown algae reveals shared features with transitions to asexuality in animals and plants.

• Christoph R. Haag

A new type of non-Mendelian segregation

An analysis of whole genomes of mothers and daughters of the clonal raider ant Ooceraea biroi shows non-random segregation of chromosomes during meiosis, which enables the species to maintain heterozygosity while still recombining because crossover products are faithfully coinherited.

• Caroline Blanc
• Marie Delattre

Informed proxy consent for ancient DNA research

Embracing the underlying principles and processes of informed proxy consent or relational autonomy consent in human ancient DNA research can transform research.

• Victoria E. Gibbon
• Jessica C. Thompson
• Sianne Alves

Summer school in wartime

Amid the brutality of war around them, some Ukrainian scientists find calm in bioinformatics.

• Vivien Marx

From Mendel’s laws to non-Mendelian inheritance

In this Journal Club article, Laura Ross discusses several seminal papers that describe the discovery of germline-specific chromosomes and paternal genome elimination, striking examples of non-Mendelian genetics.

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

• Curriculum and Education
• Open access
• Published: 04 May 2022

Correcting misconceptions about evolution: an innovative, inquiry-based introductory biological anthropology laboratory course improves understanding of evolution compared to instructor-centered courses

• Susan L. Johnston 1 ,
• Maureen Knabb 1 ,
• Josh R. Auld 1 &
• Loretta Rieser-Danner 1  

Evolution: Education and Outreach volume  15 , Article number:  6 ( 2022 ) Cite this article

3763 Accesses

1 Altmetric

Metrics details

Comprehensive understanding of evolution is essential to full and meaningful engagement with issues facing societies today. Yet this understanding is challenged by lack of acceptance of evolution as well as misconceptions about how evolution works that persist even after student completion of college-level life science courses. Recent research has suggested that active learning strategies, a focus on science as process, and directly addressing misconceptions can improve students’ understanding of evolution. This paper describes an innovative, inquiry-based laboratory curriculum for introductory biological anthropology employing these strategies that was implemented at West Chester University (WCU) in 2013–2016. The key objectives were to help students understand how biological anthropologists think about and explore problems using scientific approaches and to improve student understanding of evolution. Lab activities centered on scenarios that challenged students to solve problems using the scientific method in a process of guided inquiry. Some of these activities involved application of DNA techniques. Formative and summative learning assessments were implemented to measure progress toward the objectives. One of these, a pre- and post-course evolution concepts survey, was administered at WCU (both before and after the implementation of the new curriculum) and at three other universities with more standard introductory biological anthropology curricula. Evolution survey results showed greater improvement in understanding from pre- to post-course scores for WCU students compared with students at the comparison universities (p < .001). WCU students who took the inquiry-based curriculum also had better understanding of evolution at the post-course period than WCU students who took the course prior to implementation of the new curriculum (p < .05). In-class clicker assessments demonstrated improved understanding of evolution concepts (p < .001) and scientific method (p < .05) over the course of individual labs. Two labs that involved applying DNA methods received the highest percentage ratings by students as ‘very useful’ to understanding important concepts of evolution and human variation. WCU student ratings of their confidence in using the scientific method showed greater improvement pre- to post-course during the study period as compared with the earlier, pre-implementation period (p < .05). The student-centered biological anthropology laboratory curriculum developed at WCU is more effective at helping students to understand general and specific concepts about evolution than are more traditional curricula. This appears to be directly related to the inquiry-based approach used in the labs, the emphasis on knowledge and practice of scientific method, directly addressing misconceptions about evolution, and a structure that involves continual reinforcement of correct concepts about evolution and human variation over the semester.

Understanding the reality of evolution is fundamental to science education. However, many Americans deny the theory of evolution despite overwhelming evidence and uniform support from the scientific community (Nadelson and Hardy 2015 ). In 2006, Miller et al. published an enlightening study demonstrating the low acceptance of evolution in the United States compared to 34 other countries, with the US ranking second to last in acceptance of evolution. Data from the Pew Research Center’s ( 2015 ) Religious Landscape Study show that these results had not changed very much in the intervening decade; at that time, 34% of Americans reported that they reject evolution and believe that humans arrived on earth in their present form. Recent work by Miller et al. ( 2021 ) suggests this may be changing, with increased public acceptance of evolution in the last decade. Even though acceptance of evolution increases with level of education, from 20% in high school to 52% and 65% among college or postgraduates, respectively, the rejection rate of evolution from students in introductory biology classes can reach up to 50% (Brumfield 2005 ; Rice et al. 2010 ; Paz-y-Miño-C and Espinosa 2016 ). Even college-level instruction in evolution, then, may not increase students’ acceptance of evolution.

Perhaps more surprisingly, even when acceptance of evolution is not a factor, college-level instruction does not necessarily result in full understanding of evolution either, and numerous studies identify multiple evolution-related misconceptions held by different groups of students. For example, Cunningham and Wescott ( 2009 ) identified and evaluated biological anthropology students’ misconceptions about evolution and found that, despite acceptance of evolutionary theory, students lack understanding of the process of evolution. Tran et al. ( 2014 ) also identified similar misconceptions among advanced undergraduate biology majors. And Beggrow and Sbeglia ( 2019 ) reported that despite some differences in evolutionary reasoning and in the specific types of evolution misconceptions held by biology and anthropology majors, both populations performed poorly on a measure of evolutionary knowledge (Conceptual Inventory of Natural Selection [CINS]; Anderson et al. 2002 ). Several other instruments to assess both student misconceptions about evolution and student understanding of evolution have been developed, including the Measure of the Acceptance of Evolutionary Theory (MATE; Rutledge and Sadler 2007 ) and the Inventory of Students’ Acceptance of Evolution (I-SEA; Nadelson and Southerland 2012 ) with different student populations (see also Nehm and Mead 2019 ; Furrow and Hsu 2019 ). Results of multiple studies using these instruments show that student misconceptions continue despite college-level classroom instruction (e.g., Beggrow and Sbeglia 2019 ). Use of these types of assessment instruments aids in understanding and addressing student misconceptions, but there clearly remains a need to find the most effective teaching and learning strategies for evolution education (Glaze and Goldston 2015 ).

Pobiner ( 2016 ) recently reviewed the current state of evolution teaching and learning and concluded that focusing on human examples, such as in biological anthropology courses, is an effective way to enhance student understanding and acceptance of evolution. Based on results of the "Teaching Evolution through Human Examples" project (Pobiner et al. 2015 , 2018 ), these authors suggest that the use of human examples is helpful because human examples are relevant, they increase students’ acceptance and understanding of evolution, and they help students to appreciate historical science. Numerous other investigators have supported this suggestion (e.g., see Beggrow and Sbeglia 2019 ) and some research suggests that students across multiple disciplines (majors and non-majors) actually prefer the use of human examples when learning about evolution (e.g., Pobiner et al. 2018 ; Paz-y-Miño-C and Espinosa 2016 ). However, even with a focus on human evolution, misconceptions continue to exist (e.g., Cunningham and Westcott 2009 ; Beggrow and Sbeglia 2019 ).

Some research suggests that instructor-centered pedagogy (lecture) is less successful in helping students recognize and correct their misconceptions about evolution (Bishop and Anderson 1990 ; Gregory 2009 ) compared to historically rich, problem-solving methods of instruction that appear to significantly improve student understanding of evolution (Jensen and Finley 1996 ). Nehm and Reilly ( 2007 ) directly compared pedagogical approaches using pre- and post-course tests and found that students taught using active-learning techniques performed better than those using a more traditional approach.

Pittinsky ( 2015 ) further suggests that firsthand experience with scientific methods, as well as interactions with real scientists, would help address some of the problems in teaching evolution. It seems that when students learn to think like a scientist and use the same actions that led to original discoveries, they gain insight into the strategies and techniques used by scientists studying evolution (Passmore and Stewart 2002 ). Scharmann et al. ( 2018 ) and Nelson et al. ( 2019 ) also suggest that Nature of Science (NOS) principles should be covered before even introducing the theory of evolution. Some research supports this suggestion. For example, DeSantis ( 2009 ) reported that introduction of a curriculum module that included inquiry-based activities that model the work of paleontologists increased interest in and acceptance of the theory of evolution among middle- and high-school age students. Might the inclusion of similar inquiry-based laboratory activities also reduce the evolution misconceptions held by students (at all levels)?

Other research suggests that the order in which concepts are introduced makes a difference in students’ understanding of evolution, at least among high school students. For example, Mead et al. ( 2017 ) reported that teaching genetics first (before evolution) improves student understanding of evolution. And, Alters and Nelson ( 2002 ) as well as Beggrow and Sbeglia ( 2019 ) further suggest that targeting naïve ideas about evolution should be an instructional goal, particularly in anthropology education. Research by Bishop and Anderson ( 1990 ) and Jensen and Finley ( 1996 ) support this suggestion, reporting that confronting students’ misconceptions directly before introducing correct conceptions is associated with significant gains in student understanding of evolution. Wingert et al. ( 2022 ) show that employing instructional activities that directly challenge students' teleological concepts about natural selection improves their acceptance and understanding of evolution. 

Taken together, these results support Nelson’s ( 2008 ) recommendation of three learning strategies to improve student understanding of evolution: (1) extensively using active learning strategies; (2) focusing on science as a process and way of knowing; and (3) identifying and directly addressing student misconceptions. We report on the effectiveness of an inquiry-based laboratory curriculum that incorporates all of these strategies in an undergraduate biological anthropology course.

Evolutionary theory is central to the discipline of biological anthropology, which is fundamentally about human evolution. At West Chester University (WCU), Biological Anthropology (ANT 101) is a general education, introductory course taken by majors and non-majors that had, traditionally, been taught using a teacher-centered approach. In 2010, assessment data indicated that many students retained common misconceptions about evolution after completion of the course. For example, responses to the question “What is evolution?” included replies such as: “…survival of the fittest, species do what they need to do to pass their genes on”; “the change that occurs in an environment over time from a change in species”; “the way an organism changes to survive in a changing environment.” Clearly course changes were needed to address these misconceptions, and it seemed a good idea to attempt to do so by actively engaging students in understanding the concepts of evolution as well as the tools used by researchers to solve problems using scientific methods. Based on previous work emphasizing the need to employ human examples using active, hands-on pedagogy that emphasizes the scientific process, we developed an innovative biological anthropology laboratory course that merges these three important components of effective teaching of evolution. Based on our results, this course not only improves overall performance in correcting misconceptions when compared to other biological anthropology courses, but it also significantly improves understanding in specific areas.

Introduction to Biological Anthropology (ANT 101) has been offered annually or more frequently at WCU for nearly two decades. It is a required course for anthropology majors, and for most of that time period non-majors have been permitted to take it to meet a general education distributive requirement. Until the fall semester of 2013, it was configured as a three-hour per week lecture course with no hands-on lab component, and the department had no access to laboratory classroom facilities. For several of those years, the instructor incorporated 3–5 virtual laboratory experiences over the semester using one lecture hour for each. While students said they enjoyed these experiences, assessment data indicated that they still had persistent misconceptions about evolution at course completion.

In fall 2013, a project team at WCU, including the course instructor (a biological anthropologist), a human physiologist experienced in inquiry curricula, an evolutionary biologist, and a psychologist with expertise in assessment and program evaluation were awarded a three-year TUES (Transforming Undergraduate Education in STEM) grant from the National Science Foundation (NSF). The purpose of this award was to develop an innovative, inquiry-based laboratory curriculum targeting student misconceptions about evolution, student ability to use the scientific method, and student understanding of the investigative tools used by biological anthropologists. To accommodate this new curriculum, the course was redesigned to meet four hours per week in an integrated lecture-lab format, with roughly half of that time devoted to laboratory activities and the other half to lecture and/or discussion.

The project was submitted to the West Chester University Human Subjects Committee and received expedited approval in the summer 2013. Informed consent was obtained each semester from students enrolled in the course who wished to participate. Over the period of the project, this was all but one or two students.

During each lab period, brief instruction on methodology was provided, as appropriate to the lab, and students were presented with a challenge scenario that asked them to apply the scientific process to solving that problem using the relevant method (with the challenge scenario providing a structured context in which to do so). In a standard biological anthropology lab curriculum, students might be asked to describe and identify various casts of hominin fossil skulls using characteristics they had learned about, associate these traits with dietary differences, and receive verification of their assessments by the instructor. In the inquiry-based, structured challenge approach developed at WCU, students were given a problem to solve that required them to hypothesize the likely diet of the various hominins or hominids. They were instructed in a technique that allowed them to test one of their hypotheses, then required to state their results in an organized manner, evaluate them, indicate next steps, and so on. Thus, each lab in the curriculum is configured to (1) help students understand how biological anthropologists think about and explore problems using relevant techniques and (2) gain experience with the scientific process. The lab curriculum includes some instruction and application of basic molecular techniques (e.g., constructing simple primate phylogenies based on morphological v. genetic variation and doing a DNA fingerprinting exercise to attempt to identify a hypothetical hominin fossil), since the curriculum is also designed to help students make connections between phenotypic observations and the molecular level in service of the project goal of helping students to better understand evolution. Table 1 provides a list of the labs with descriptions of the inquiry learning activities performed.

The full lab manual can be accessed at: https://digitalcommons.wcupa.edu/anthrosoc_facpub/72 .

Standard assessments, including periodic exams and laboratory reports, were utilized to measure student learning. Responses to lab challenges at multiple time points were evaluated at the end of each semester using a rubric to measure individual students’ abilities to define the problem, to develop a plan to solve the problem, to analyze and present information, and to interpret findings and solve the challenge problem. Student lab teams also developed a project that they designed and implemented (from hypothesis to interpretation) using one of the methods they learned, and gave group presentations to the class. Other, more formative, measures of student learning were also introduced. For example, during each lab, students completed a pre-post assessment tool which was a modified version of the RSQC2 (Recall, Summarize, Question, Connect, and Comment) classroom assessment technique developed by Angelo and Cross ( 1993 ). Beginning in the second year of the project, pre- and post-lab clicker questions were incorporated for rapid assessment of the lab impact.

Several global surveys were administered at the beginning of each course, prior to any instruction, and again (for all but one survey) on the last day of the course. These included a survey focusing on evolution (17 items in year one, revised to 25 items in the second year) as well as surveys assessing students’ familiarity and comfort level with the scientific process, their level of motivation, and, at the end only, their overall assessment of their course experience. The evolution survey was also administered at WCU for 2 years prior to the course reorganization and lab implementation; data from this period are used for an internal comparison with survey results obtained during the implementation of the new curriculum. Biological anthropology colleagues at three other US universities (reported here as A, B, and C) also administered the evolution concepts survey to their students in introductory courses in biological anthropology, during the grant period, for comparison purposes. All of these courses were taught with some version of a more standard laboratory curriculum for this discipline (example of a standard approach described above). University ‘A’ is a large, midwestern state school (approximately 40,000 students); University ‘B’ is a sizable state school located in the south (approximately 30,000 students). University ‘C’ is a large, northeastern state school (approximately 30,000 students). At all three, introductory biological anthropology is taught in large lecture context with smaller recitation sections that meet one hour per week (i.e., two hours lecture, one hour of recitation or lab). At A and C, these recitations were used for weekly laboratory activities throughout the semester; at B, there were seven labs during the semester. Prior to 2013, the course at University A had no lab at all—only lecture.

The current report first describes the results of the evolution concepts instrument administered at the very beginning of the course and at the end of the course at WCU and across universities. Following a presentation of the results regarding changes in misconceptions we turn our attention to an examination of the specific areas of learning that we believe may have contributed to the reduction in misconceptions, including a look at specific assessments of students’ growing understanding of science as a process throughout the course.

Evolution misconceptions

Two versions of the evolution concepts instrument were used, one prior to the start of the grant period and throughout the first year following the grant award and a revised version used beginning in fall 2014. Each version included statements that students responded to on a 5-option Likert-type scale ranging from strongly agree to strongly disagree, or having no opinion. This instrument was based on a published and freely available tool used by other researchers (Cunningham and Wescott 2009 ). For purposes of analysis, each item was agreed by the project team to be either true or false, such that strong agreement with a true statement and strong disagreement with a false statement were considered to be ‘correct’ responses. A scale ranging from + 2 to − 2, including 0 for ‘no opinion’ was constructed, and several variables were computed from these scores, including total score (pre, post), percent of total points earned (pre, post), number of items correct (pre, post), and percent of items correct (pre, post). The use of percent variables was necessitated by a revision of the survey after the first year of curriculum implementation (2013–2014). The initial version of the survey included 24 items, but a qualitative analysis by study consultants resulted in a set of only 17 items deemed usable for the purposes of our study. This initial survey was then revised for use beginning in fall 2014 to include the 17 items kept from the original survey with the addition of 8 new items, resulting in a set of 25 usable items. The 25-question survey can be found in Additional file 1 .

Several questions were addressed using the results of the evolution concepts instrument. First, we compared WCU student survey responses to responses from the three other institutions whose students completed the survey. We asked if student performance on the evolution concepts instrument improved from pre- to post-course for all institutions and whether the amount of improvement varied by institution. Second, we examined WCU student survey responses (both pre and post surveys) over time, asking if student performance on the evolution concepts instrument improved both prior to and during the grant implementation period. Next, we asked whether the degree of improvement changed following implementation of our new inquiry-based curriculum, relative to the academic years prior to implementation of the grant. Finally, in an attempt to understand the specifics of what evolution-related misconceptions might have improved and which did not, we conducted a qualitative analysis of survey items and compared student performance on sets of related items across universities.

WCU course assessments

A variety of measures were used to assess student learning throughout each semester at WCU and to evaluate the effectiveness of particular pedagogical approaches as well as the overall curriculum. Some of these measures were objective and direct measures of student learning. Some were indirect measures, student perceptions of what they learned and/or which laboratory sessions they believed were most helpful in their learning. In this report, we provide results of four of these measures—in-class clicker questions, laboratory challenges, RSQC2 responses, and student confidence ratings—to provide insights about the effectiveness of the curriculum in meeting its primary objectives.

In-class clicker questions

Students were presented with a set of true/false statements or multiple choice questions at the beginning and end of multiple laboratory sessions. Some items were tied directly to misconceptions about evolution, others to students’ understanding of the scientific method, while others were designed to measure more general understanding of the topics covered by the individual laboratory modules. Students responded, via clickers, to these statements presented visually in class. Responses served as an important source of formative assessment but also provided information on the effectiveness of each of the laboratory modules in correcting student misconceptions about evolution and student understanding of the scientific method.

Laboratory challenges

Laboratory modules included “challenge” activities, designed specifically to enable students to apply problem-solving skills within a structured context (Knabb and Misquith 2006 ). In each of these laboratory challenges, students were asked to state research questions or generate hypotheses, collect data, draw conclusions, report/graph their results, and reflect on those results. Each student completed a laboratory worksheet during each lab module and all worksheets were submitted as part of student lab notebooks at the end of each semester. Selected lab worksheets were reviewed by faculty involved with the grant project at the end of each semester using a developmental assessment screening tool developed by all project faculty. This screening tool underwent its own developmental process, resulting in a final tool that included four measures of scientific thinking (i.e., students’ ability to use the scientific method): Defining the Problem, Developing a Plan to Assess the Problem, Analyzing and Presenting Information, and Interpreting Findings and Solving the Problem. Each of these four areas was assessed on a scale of four developmental levels: beginning, developing, appropriately developed, and exemplary. A copy of this scoring rubric can be found in Additional file 2 . Developmental changes in these four areas of scientific thinking were assessed by comparing assigned developmental levels following an early semester laboratory module with assigned developmental levels following a later semester laboratory module.

RSQC2 (Revised)

A modified version of the RSQC2 classroom assessment technique (Angelo and Cross 1993 ) was completed by students during each laboratory session. Complete details about the multiple sections of this activity can be found in Additional file 3 . For the current report, we present data on one of the sections completed by students at the end of each laboratory session. Students were asked to rate the usefulness of each laboratory session in reaching learning outcomes. Ratings were made on a 4-point Likert scale: 4 = very useful; 3 = somewhat useful; 2 = minimally useful; 1 = not useful. Questions included: How useful was today’s laboratory session in helping you to understand the important concepts of evolution and human variation discussed in this course and used by biological anthropologists? How useful was today’s laboratory session in helping you to understand the tools used by biological anthropologists to understand the concepts of evolution and human variation?

Student confidence in using scientific method

WCU students completed a 10-item survey at both the beginning and the end of each semester asking them to rate their level of confidence in their abilities and/or understanding of several pieces of the scientific process. All items were rated on a 5-point Likert scale: 1 = completely doubtful; 2 = somewhat doubtful; 3 = neutral; 4 = somewhat confident; 5 = strongly confident. A copy of this survey is available in Additional file 4 .

A variety of both univariate and multivariate linear model procedures were used to address questions of interest involving all student assessments, both within and across time periods and universities (where appropriate). Specifics regarding these analyses are discussed within the Results section.

Evolution misconceptions at WCU and other institutions

WCU evolution surveys were collected across all six semesters of the grant implementation period (fall 2013 through spring 2016), with a total of 105 complete survey sets (pre- and post-course). Survey responses from students at the three other universities were provided by institution instructors whenever possible: University A provided 469 complete survey sets across five terms; University B provided 273 complete survey sets across six terms; and University C provided 200 complete survey sets across three terms. Comparisons across universities were made across only the three terms for which data was provided by each university (fall 2014, spring 2015, and fall 2015). Figure  1 shows pre-course and post-course percent items correct at each university (WCU, University A, University B, and University C), collapsed across these three semesters.

Pre-course and post-course evolution concept survey ‘percent items answered correctly’ across 4 universities: WCU (n = 43); University A (n = 308); University B (n = 143); and University C (n = 200)

Significant change from pre- to post-course percent items correct was found within institutions for each of the three terms individually [as assessed after each term] and across all terms combined. Furthermore, significant change from pre- to post-course percent items correct was found across all three terms and 4 institutions, collapsed [t (693) = 25.762, p < 0.001]. Thus, significant improvement in overall performance on the evolution misconceptions instrument occurred at every institution and during each of the three terms considered here.

While there were no significant differences by term, institution, or term x institution in pre-course percent items correct, we did note a near significant effect of institution [F (3, 690) = 2.548, p < 0.10]. An informal review revealed that WCU pre-course scores were higher than pre-course percent items correct at all three other universities. Thus, comparison of post-course percent items correct included the pre-course percent items correct scores as a covariate. ANCOVA results support a significant effect of institution on post-course percent items correct, after controlling for pre-course percent items correct [F (3, 689) = 8.345, p < 0.001]. Post-hoc tests show significant differences between post-course scores at WCU and at all three other institutions. In addition, post-course percent items answered correctly at University B was significantly lower than percent items answered correctly at University C.

Internal WCU comparisons

The results reported above support statistically significant improvement in evolution misconception scores among students at all participating universities but further suggest that post-course scores are significantly higher at WCU than at any of the other three universities, even after controlling for potential differences in pre-course scores. WCU differs from these other institutions in terms of the curriculum focus (our inquiry-based approach versus other, more standard approaches), but WCU also differs from the other institutions in terms of class size. Individual class sections are smaller at WCU, resulting in smaller sample sizes both within and across semesters. If class size is the factor that explains the difference in post-course performance across universities, it should also be the case that post-course performance at WCU would not change following the introduction of the new inquiry-based curriculum. To evaluate this possibility, we compared WCU evolution survey results for pre-grant terms to evolution survey results following implementation of the inquiry-based curricular approach. Survey results are reported here for pre-grant (fall 2011 and fall 2012, N = 22 and 26, respectively), and grant implementation (fall 2013, spring 2014, fall 2014, spring 2015, fall 2015, and spring 2016; Ns = 18, 23, 12, 12, 19, and 21, respectively) (Fig.  2 ).

WCU pre- and post-course ‘percent items answered correctly’ by project phase: pre-grant (n = 48) and post-grant (n = 105)

There were no significant differences by term in pre-course percent items correct or post-course percent items correct during the pre-grant period (fall 2011 and fall 2012) or during the grant implementation period (fall 2013 through spring 2016). Significant change from pre- to post-course percent items correct was found across the pre-grant period [t (47) = 7.387, p < 0.001] and across the grant implementation period [t (104) = 14.871, p < 0.001]. Thus, significant improvement in performance on the evolution conceptions instrument was found both prior to and during the implementation of the grant. There were no significant differences in pre-course percent items correct between pre-grant and grant implementation periods [F (1, 151) = 2.145, p = 0.145]. But, a significant group difference was found in post-course percent items correct [F (1,151) = 5.600, p < 0.05], with students answering a larger percentage of items correctly (i.e., earning full 2 points) across the grant implementation period than during the pre-grant period.

Evolution concepts

The results reported above support the conclusion that our new laboratory curriculum may be more effective in improving student understanding of evolution and evolutionary concepts and may be more effective in reducing student misconceptions of evolution than the curriculums utilized at the other universities. In addition, significantly more WCU students answered certain survey items correctly at the post-course assessment than did students at any of the other three institutions (see Table 2 ), but a clear pattern was difficult to identify. Thus, we conducted a qualitative analysis of the 25 survey items that made up the revised version of the survey (the one implemented beginning fall 2014). We examined the survey results for the three terms for which data were available for all four institutions (fall 2014, spring 2015, fall 2015). This analysis resulted in four groups of items, each addressing one broad theme: (1) understanding of basic scientific evidence and the process of science (5 items); (2) understanding of evolution (from a general or “big picture” perspective) (7 items); (3) understanding of the mechanisms of evolution (i.e., natural selection, mutation, genetic drift, gene flow) (8 items); and (4) understanding of the evidence for evolution (5 items). Table 2 provides a list of all survey items and identifies which theme each item falls into.

A significant multivariate effect of institution was found when we included the four concept scores (i.e., percent of items within each concept grouping answered correctly) in a MANOVA procedure with both pre-course scores and post-course scores included as dependent variables. Univariate follow-up tests suggest a significant institution effect for Concepts #1, and #4. In both cases, pre-course scores were higher for WCU students than for students at other institutions. Thus, a set of Analysis of Covariance (ANCOVA) procedures were conducted, one for each set of post-course concept scores (i.e., percent of items within each concept grouping answered correctly at post-course time period), with institution included as a between-subjects factor and pre-course scores for that concept included as a covariate. Results suggest a significant institution effect for three of the four concepts (#1, #2, and #3). With regard to Concept #1 (understanding of basic scientific evidence and the process of science) post-hoc tests following an overall significant effect of institution [F (3,689) = 3.919, p < 0.05] show significantly higher post-course concept scores at WCU than at any of the other three institutions. A similar result was found for Concept #2 (understanding of evolution from a general/big picture perspective) [F (3,689) = 12.899, p < 0.001]. Again, post-course scores for WCU were significantly greater than those for the other three institutions. In addition, University A post-course scores were significantly greater than those for University B. A significant effect of institution was also found for Concept #3 (understanding of the mechanisms of evolution) [F (3,689) = 7.278, p < 0.001]. Post-hoc tests reveal that WCU post-course scores are significantly greater than those of University A and University B. WCU scores are higher than those of University C but that difference did not reach statistical significance. No significant effect of institution was found for Concept #4 scores (understanding of the evidence for evolution) [F (3,689) = 1.643, p = 0.178). But, despite the lack of an overall significant effect, WCU post-course scores are greater than those of the other institutions for this concept. Descriptive statistics for the concept scores across universities can be found in Additional file 5 .

How might this inquiry-based course have aided in the reduction of evolutionary misconceptions? In an attempt to gain insight about which course components or processes were effective in this regard, we examined student responses to in-class clicker questions about evolution concepts and scientific method , their development of scientific thinking skills over the term via lab worksheets, their perceptions about each lab’s effectiveness in helping them to learn about evolution and human variation concepts, and their confidence in using the scientific method. These results are presented below.

Clicker questions were developed over the course of the second year of the grant, then revised slightly for use across the final year of the grant (Fall 2015–Spring 2016). Questions were developed for eleven laboratory modules (see Table 1 ). Some items were included within each module to measure understanding of specific laboratory content. Items measuring evolution misconceptions were also included for all modules (1, 2, or 3 items). Items measuring scientific thinking (i.e., understanding of the scientific method) were included for only three modules (1 or 2 items): Evolution and Scientific Thinking, Primate Anatomy and Locomotion, and Human Osteology and Forensics. Clicker questions were presented at the beginning and at the end of each laboratory module session. Data for the final year of grant implementation are presented here. Complete data (across all laboratory modules) were available for 24 students across both semesters.

Overall student performance (as measured by % total items answered correctly) increased significantly from 78.64% at pre-module assessment to 91.06% at post-module assessment (across all items and all laboratory modules) [t (23) = 10.89, p < 0.001]. Performance also increased within each of the laboratory modules.

Student performance also increased significantly on the items specifically designed to measure previously identified misconceptions about evolution, with percent total items answered correctly across all laboratory modules increasing from 83.85% correct to 91.93% correct (across all items and all laboratory modules) [t (23) = 4.992, p < 0.001]. Given that evolutionary misconceptions were addressed most steadily during the early part of the semester, we examined the degree to which improvement on misconception items might be different across the semester. Table 3 shows measures of student performance on evolution misconception in-class clicker items during three time periods of the semester: Early Semester (3 modules focused on basic evolutionary concepts); Mid Semester (4 modules focused on non-human primates and human evolution); and Late Semester (4 modules focused on living human biology). While some slight improvement was noted across all time periods, the only period during which a statistically significant improvement occurred was the early semester time period.

Student performance on the items specifically designed to measure student understanding of the scientific method increased significantly from 90.00% to 97.50% (across all items and all three laboratory modules that included those items) [t (23) = 2.584, p < 0.05]. When broken down by individual laboratory module, the greatest improvement in student performance appears in the later modules but is only statistically significant in the Primate Locomotion module (see Table 4 ).

Two laboratory sessions (one early- and one mid-semester) were chosen for comparison: (1) the Evolution and Scientific Thinking laboratory module was chosen for the early-semester session; and (2) the Primate Anatomy and Locomotion module was chosen for the mid-semester session. The Evolution and Scientific Thinking laboratory module was the first laboratory module students participated in and occurred during week two of the semester. The Primate Anatomy and Locomotion session occurred at about week six of the semester. Four variables were scored from the laboratory worksheets of each of these sessions across the final two semesters of the grant implementation period, fall 2015–spring 2016: Defining the Problem; Developing a Plan to Solve the Problem; Analyzing and Presenting Information; and Interpreting Findings and Solving the Problem. All were rated on a scale of 1 to 4 (Beginning, Developing, Appropriately Developed, and Exemplary). Three faculty scorers worked together to determine final scores by consensus for each variable in each laboratory worksheet. Complete data were available for a total of 42 students across both semesters (21 each semester) (see Table 5 ).

Student responses to all items of the RSQC2 classroom assessment tool were collected across the final two semesters of the grant implementation period, fall 2015–spring 2016. As outlined earlier, students were asked to rate the usefulness of each laboratory session in helping them to understand (1) the important concepts of evolution and human variation discussed in the course, and (2) the tools used by biological anthropologists to understand the concepts of evolution and human variation. Students ranked each laboratory session, as it ended, on a 4-point scale, ranging from Not Useful to Very Useful, on each of these items. Table 6 lists the laboratory session topics and the percent of students who rated each one as “Very Useful” to their understanding of the important concepts of evolution and human variation. Table 7 lists the percent of students who rated each one as “Very Useful” to their understanding of the tools used by biological anthropologists (i.e., to their understanding of the scientific method as practiced by biological anthropologists). Some differences in student ratings across the two areas of understanding are apparent.

Student ratings of their confidence in using the scientific method are reported here for the pre-grant period (fall 2011 and fall 2012 combined), and the grant implementation period (fall 2013, spring 2014, fall 2014, spring 2015, fall 2015, and spring 2016 combined) (see Table 8 ). Student ratings increased from pre- to post-course during both time periods, but improvement was greater during the grant implementation period than during the pre-grant period.

The laboratory curriculum developed and evaluated at WCU increases students’ understanding of evolution in introductory biological anthropology compared with other institutions using more standard approaches. While students taking the evolution concepts survey demonstrated improved understanding of evolution at all of the schools that employed this instrument (WCU and comparisons) from the beginning to the end of each semester, WCU students demonstrated a greater increase in percent items answered correctly from pre- to post-course (see Fig.  1 ). Significantly more WCU students answered 18 (of 25) survey items correctly at the post-course assessment than did students at any of the other three institutions (see Table 2 ). Given that WCU class sizes are smaller than those at the three comparison universities, WCU student performance on the evolution survey before the new curriculum was implemented was compared with performance during the first three years of the new, grant-funded curriculum. Students taking the survey during the grant period answered a statistically greater percentage of items correct at the post-survey than students in the pre-grant period, with pre-survey response levels showing no significant difference across these two phases (see Fig.  2 ); class sizes were comparable across the entire time frame.

Thus, we demonstrate the impact on improved student understanding of evolution is related to the new curriculum itself. In the remaining discussion, we focus on the question of what aspects of the new curriculum may be contributing to this improvement, detailing how this curriculum incorporates all three of the key learning strategies outlined by Nelson ( 2008 ): (1) extensive use of active learning approaches; (2) focus on science as a process and way of knowing; and (3) identification and direct targeting of student misconceptions.

First, the WCU curriculum is inquiry-based, engaging students actively and directly with the process of “doing science”. Active learning (also known as student-centered learning) strategies, such as problem- or inquiry-based approaches, have been shown to be superior to instructor-centered approaches (e.g., lecture) in promoting student learning about evolution (e.g., Jensen and Finley 1996 ; Nehm and Reilly 2007 ). One of the stated learning goals of this course is to help students come to understand how biological anthropologists investigate questions. We strive to accomplish this by having them learn and actually use some of the tools scientists in this field employ—both at the ‘outward’ physical (e.g., skeletal, body shape and size, etc.) and molecular/biochemical levels (e.g., gene sequence readouts, DNA fingerprinting)—in a problem-solving context. Student lab teams receive a challenge scenario and have to come up with a methodological approach (usually using techniques they have just learned, and occasionally employing techniques learned earlier in the course), collect data, and then interpret those data—in every lab. This is fundamentally different than the typical approach in an introductory biological laboratory setting, such as those used in the comparison institutions and described earlier in this paper.

We think that this bi-level approach to teaching and using relevant methods in problem-solving helps students connect the evidence for evolution and human variation with the underlying molecular basis of that variation and change over time. Student ratings of each lab on the RSQC2 question pertaining to effectiveness in helping them to learn concepts of evolution and human variation were highest for Tree-Building and Primate Classification and DNA Fingerprinting (Table 6 ). We think it telling that both of these labs involve genetic as well as phenotypic variation linked with evolution. Ratings for the question concerning lab effectiveness in helping students to learn to use the tools biological anthropologists employ to understand evolution and human variation were highest for Forensics 2: DNA Fingerprinting, followed by Human Variation: Anthropometry, Human Genetic Adaptation: ELISA, and the Tree-Building and Primate Classification labs (see Table 7 ); all but the anthropometry lab address directly both genetic/biochemical and physical traits.

Second, the WCU curriculum focuses on the scientific way of knowing and the scientific process from the first week, in both lecture and lab contexts. The first topic after the students are introduced to the discipline is the nature of science: how science seeks to understand phenomena, the meaning of ‘fact’, ‘hypothesis’, and ‘theory’ in a scientific inquiry, and how the scientific approach to understanding natural phenomena differs from others. The first lab, which occurs early in the second week, then provides an opportunity for students to try out the scientific method and to learn, in context, about generating hypotheses, developing methods, collecting data, and interpreting those observations. They also learn about bias caused by preconceptions, measurement error, and different approaches to understanding the world (e.g., science and religion). Each lab module thereafter requires students to methodically think through and structure their work using the standard methodological sequence: question/hypothesis, explication of methods, data collection and reporting, discussion, and interpretation (see Table 1 ). Further examples of how the process of science is addressed in the curriculum are described below in the discussion about addressing evolution misconceptions.

The effectiveness of this approach is supported by the qualitative evolution concepts analysis that we undertook to look for thematic patterns in the evolution survey statements (see Table 2 and associated text). Three broad concepts showed a significant effect of institution, with WCU student post-course scores being higher than those at the other institutions; the first of these was understanding of basic scientific evidence and the process of science. The in-class clicker data we analyzed (see Table 4 ) support the idea that students gained knowledge about the scientific method during lab classes. Analysis of the change in student performance on lab challenges relevant to steps of the scientific process from early to mid-semester (see Table 5 ) also supports improved student ability to develop a plan to solve the problem (Methods) and to analyze and present information (Results) from the early time point to the later one. Additionally, students’ report of their confidence in using the scientific method (see Table 8 ) indicated greater improvement from pre- to post-course during the grant implementation period than during the pre-grant period at WCU. Firsthand experience with the scientific method and opportunities to ‘think like a scientist’ have been linked with improved ability of students to understand and accept evolution (see, e.g., Pittinsky 2015 ; DeSantis 2009 ; Robbins and Roy 2007 ; Nelson 2008 ).

Third, the WCU curriculum is designed to identify and directly address student misconceptions about evolution, and it does so from early in the course (Nelson 2008 ). Students take the evolution concepts survey on the first day of class, before any instruction about evolution. This provides a baseline of their understanding, and the concepts included in the survey are among those that the curriculum proceeds to address. The order of the labs over the semester (Table 1 ) ensures that basic concepts of evolutionary theory and mechanisms, genetics, and classification/phylogeny are covered early. As part of this attention to foundational ideas, class discussions during and at the end of labs include a focus on misconceptions about evolution and, indeed, about how scientific inquiry is conducted. For example, in the Evolution and Scientific Thinking lab (the first one), students nearly always assume the male skeleton will be the taller of the two—whether or not they overtly state that as a hypothesis. This and other ideas that students mention lead to a discussion of assumption bias and how we try to avoid that in the process of “doing” science. This is followed by a dialogue (sometimes precipitated by a student-expressed view, but more often introduced by the instructor as a story) focused on the idea some people hold that the male should have one less rib than the female. We talk through whether this is a scientific hypothesis (yes, because it can be tested); how they would test it (go count the ribs); what kind of evolution mechanism this idea reflects (Lamarkism, i.e., inheritance of acquired characteristics); and what genetic assumption is also being made (that rib number is sex-linked). We also tell students that, in reality, there is a range of variation in number of rib pairs in humans. In fact, the male skeleton is shorter than the female, and this fact also fosters a framework in which to look at what kinds of factors may affect variation in height in humans, besides sex (e.g., population or individual ancestry, various environmental influences, age). In the Tree Building and Primate Classification two-part lab, we address directly the relationship among monkeys, apes, and humans. At the outset, most students think that monkeys and apes are more closely related evolutionarily than either group is to humans; this is also typically how they interpret the anatomic evidence of the comparative skulls and build their initial trees. However, when they do the counts of pairwise differences in the gene sequence for the three primate groups, they come to understand that the genetic evidence is indicating that apes and humans are more closely related than either group is to monkeys. The discussion in this lab is also focused on the conduct of science inquiry (e.g., can we say a hypothesis is “proven” based on one gene sequence or a limited set of anatomic traits?) and evolution misconceptions (e.g., that extant species differ from each other in “how evolved” or better adapted they are, based on body size or some other assumption).

In addition, we assess student understanding about common misconceptions in all labs directly via some of the in-class clicker questions administered as a formative assessment at the beginning and end of each lab module. Use of clickers allows us to assess immediately, at the conclusion of a lab module, how well students grasped the key concepts and techniques on which the lab was based, including evolution concepts. In the data presented in this paper, scores on evolution concept clicker questions improved significantly in the early lab modules analyzed as a group compared with mid-semester and later semester groupings of lab modules (see Table 3 ). In the later phases, the baseline (pre-lab) scores were higher, reflecting student mastery of evolution concepts generally over course duration. Finally, evolution misconceptions were also addressed in ‘lecture’ class discussions as well as queried on exams. In other words, the focus on correcting misconceptions occurred at multiple levels and time points in the course.

The kind of repetition and reinforcement that we describe here has been termed “spaced practice” or “varied practice” and is documented as improving student conceptual learning (Brown et al. 2014 ; Cepeda et al. 2006 ; Lang 2016 ). Spaced practice improves learning for a variety of reasons and in a variety of ways, but one thing that spaced practice supports is long-term consolidation of information; practice over time and in various forms allows for the connection of new information to existing knowledge and for the strengthening of memory traces over time (Brown et al. 2014 ; Cepeda et al. 2006 ; Goode et al. 2008 ; Moulton et al. 2006 ). We think that reinforcing on a weekly basis both the scientific method and correct general and specific concepts about evolution (including the mechanisms of evolution) represents this kind of spaced and varied practice and may well be contributing to the comparative success of this curriculum. The close integration of lecture and lab is likely also a factor.

Following the project, the course instructor, in consultation with the project team, made a number of changes to the curriculum based on the promising findings described above. The steps of the scientific method were more explicitly built into all of the lab worksheets, for emphasis. Opportunities to emphasize key evolution concepts within particular labs were enhanced during post-lab discussions. Clicker questions were revised to incorporate more statements reflecting science process and understanding, as well as additional repetitions of evolution concepts (with altered wording each time). Eventually, two new labs were developed related to human physiological adaptability. The first of these was added in spring of 2017 and focused on blood pressure response to stress; this lab, done late in the semester, then became the basis for the students’ final group project (instead of the population ancestry lab). After the students conduct a pro-forma experiment assessing cardiovascular response to a stressor using the blood pressure sensor and software, they design and conduct their own experiments, which they then present orally the following week. In fall 2017, a second physiology lab was added in the first half of the course, focused on skin temperature response to cold, and provided another, and earlier, opportunity for students to develop their own experiments once they learned the technique, with an emphasis at this early stage on hypothesizing. Students present these first ‘mini’ projects briefly (focusing on hypothesis and results) the following week. We felt that it was important to provide students with two experiences that allow them to ask and answer research questions of their own, under guidance. In fall 2017 the course topic order was also altered, bringing most of the human biology material previously covered at the end (biological variability and adaptation) into the sequence immediately after evolutionary theory and genetics—thus the relevance of a temperature adaptability lab in week 5.

Conclusions and suggestions

The student-centered biological anthropology laboratory curriculum developed at WCU is more effective at helping students to understand general and specific concepts about evolution than are more traditional curricula. We argue here that this is not just a function of small class size, but is directly related to the inquiry-based approach used in the labs, the emphasis on knowledge of science and practice applying the scientific method regularly, the very intentional confronting of misconceptions about evolution starting early in the course, and the structure that allows for ‘spaced practice’, i.e., continual reinforcement of correct concepts about evolution and human variation. Inquiry-based approaches can be incorporated in lab sections of otherwise large lecture courses (Casotti et al. 2008 ) or as small-group activities within lecture-only science courses. Evidence suggests that these student-centered approaches also work well for diverse learners (Tuan et al. 2005 ).

We encourage instructors of introductory biological anthropology and other life science courses to incorporate these key elements in their curricula to support improved student understanding about science process and evolution. Three general suggestions that might be applied fairly readily based on our study would be: (1) assess students’ level of understanding of evolution and how science proceeds right at the beginning of the course or relevant unit, and again at the end—to take stock of the impact of the curriculum on student learning; (2) provide hands-on problem-solving opportunities, such as case studies, guided challenges, or self-designed experiments, that iteratively emphasize scientific method and correct understanding of evolution; (3) use human examples where possible, and look for opportunities to help students connect the phenotypic changes reflecting evolution with the underlying genetic changes. The WCU curriculum is freely available to those who are interested in more detail or who may wish to adapt and incorporate components of what we have discussed here in their own courses—e.g., specific labs, etc.—at the following link ( https://digitalcommons.wcupa.edu/anthrosoc_facpub/72 ); inquiries or requests for additional information may be sent directly to the first author.

Availability of data and materials

The lab manual can be accessed at the open source link: https://digitalcommons.wcupa.edu/anthrosoc_facpub/72 . Other materials can be obtained from the first author on request.

Abbreviations

West Chester University

Alters B, Nelson C. Perspectives: teaching evolution in higher education. Evolution. 2002;56:1891–901.

Article   Google Scholar  

Anderson DL, Fisher KM, Norman GJ. Development and evaluation of the conceptual inventory of natural selection. J Res Sci Teach. 2002;39:952–78.

Angelo TA, Cross KP. Classroom assessment techniques: a handbook for college teachers. 2nd ed. San Francisco: Jossey-Bass; 1993.

Google Scholar  

Beggrow EP, Sbeglia GC. Do disciplinary contexts impact the learning of evolution? Assessing knowledge and misconceptions among anthropology and biology students. Evol Educ Outr. 2019. https://doi.org/10.1186/s12052-018-0094-6 .

Bishop BA, Anderson CW. Student conceptions of natural selection and its role in evolution. J Res Sci Teach. 1990;27:415–27.

Brown PC, Roediger HL III, McDaniel MA. Make it stick: the science of successful learning. Cambridge: Belknap Press; 2014.

Book   Google Scholar  

Brumfield G. Who has designs on your students’ minds? Nature. 2005;434:1062–5.

Casotti G, Rieser-Danner L, Knabb M. Successful implementation of inquiry-based physiology laboratories in undergraduate major and nonmajor courses. Adv Physiol Educ. 2008;32:286–96. https://doi.org/10.1152/advan.00100.2007 .

Article   CAS   PubMed   Google Scholar  

Cepeda NJ, Pashler H, Vul E, Wizted JT, Rohrer D. Distributed practice in verbal recall tasks: A review and quantitative synthesis. Psychol Bull. 2006;132:354–80.

Cunningham DL, Wescott DJ. Still more “fancy” and “myth” than “fact” in students’ conceptions of evolution. Evol Educ Outr. 2009;2:505–17.

DeSantis LRG. Teaching evolution through inquiry-based lessons of uncontroversial science. Am Biol Teach. 2009;71:106–11. https://doi.org/10.1662/005.071.0211 .

Furrow RE, Hsu JL. Concept inventories as a resource for teaching evolution. Evo Edu Outreach. 2019;12:2.

Glaze AL, Goldston MJUS. science teaching and learning of evolution: a critical review of the literature 2000–2014. Sci Educ. 2015;99:500–18.

Goode MK, Geraci L, Roediger HL. Superiority of variable to repeated practice in transfer on anagram solution. Psychonom Bull Rev. 2008;15:662–6.

Gregory T. Understanding natural selection: essential concepts and common misconceptions. Evol Educ Outreach. 2009;2:156–75. https://doi.org/10.1007/s12052-009-0128-1 .

Jensen MS, Finley FN. Changes in students’ understanding of evolution resulting from different curricular and instructional strategies. J Res Sci Teach. 1996;33:870–900.

Knabb M, Misquith G. Assessing inquiry process skills in the lab using a fast, simple, inexpensive fermentation model system. Am Biol Teach. 2006;68:e25–8.

Lang JM. Small teaching: everyday lessons from the science of learning. San Francisco: Jossey-Bass; 2016.

Mead R, Hejmadi M, Hurst LD. Teaching genetics prior to teaching evolution improves evolution understanding but not acceptance. PLOS Biol. 2017;15: e2002255. https://doi.org/10.1271/journal.pbio.2002255 .

Article   PubMed   PubMed Central   Google Scholar  

Miller J, Scott E, Okamoto S. Public acceptance of evolution. Science. 2006;313:765–6.

Article   CAS   Google Scholar  

Miller JD, Scott EC, Ackerman MS, Laspra B, Branch G, Polino C, Huffaker JS. Public acceptance of evolution in the United States, 1985-2020. Public Underst Sci. 2021;31:223–38.

Moulton C-AE, Dubrowki A, Mac-Rae H, Graham B, Grober E, Reznick R. What kind of practice makes perfect? Ann Surg. 2006;244:400–9.

Nadelson LS, Hardy KK. Trust in science and scientists and the acceptance of evolution. Evol Educ Outr. 2015;8:9.

Nadelson LS, Southerland S. A more fine-grained measure of students’ acceptance of evolution: development of the Inventory of Student Evolution Acceptance—I-SEA. Int J Sci Educ. 2012;34:1637–66.

Nehm RH, Mead LS. Evolution assessment: introduction to the special issue. Evo Edu Outreach. 2019;12:7.

Nehm RH, Reilly L. Biology majors’ knowledge and misconceptions of natural selection. Bioscience. 2007;57:263–72.

Nelson C. Teaching evolution (and all of biology) more effectively: strategies for engagement, critical reasoning, and confronting misconceptions. Integr Comp Biol. 2008;48:213–25.

Nelson CE, Scharmann LC, Beard J, Flammer LI. The nature of science as a foundation for a better understanding of evolution. Evol Educ Outr. 2019. https://doi.org/10.1186/s/12052-019-0100-7 .

Passmore C, Stewart J. A modeling approach to teaching evolutionary biology in high schools. J Res Sci Teach. 2002;39:185–204.

Paz-y-Miño-C G, Espinosa A. Measuring the evolution controversy: a numerical analysis of acceptance of evolution at America’s Colleges and Universities. Newcastle: Cambridge Scholars Publ; 2016.

Pew Research Center. Religious Landscape Study. http://www.pewforum.org/2015/11/03/chapter-4-social-and-political-attitudes/;2015 . Accessed 1 Feb 2017.

Pittinsky TL. American’s crisis of faith in science. Science. 2015;348:511–2.

Pobiner B. Accepting, understanding, teaching, and learning (human) evolution: Obstacles and opportunities. Am J Phys Anthropol. 2016;159:232–74. https://doi.org/10.1002/ajpa.22910 .

Pobiner B, Bertka C, Beardsley P, Watson W. The Smithsonian’s ‘Teaching evolution through human examples’ project. Themed paper set presented at the Association for Science Teacher Educator (ASTE) conference, Reno, Nevada; 2015. http://humanorigins.si.edu/education/teaching-evolution-through-human-examples

Pobiner B, Beardsley PM, Bertka CM, Watson WA. Using human case studies to teach evolution in high school AP biology classrooms. Evol Educ Outr. 2018. https://doi.org/10.1186/s12052-018-0077-7 .

Rice JW, Olson JK, Colbert JT. University evolution education: the effect of evolution instruction on biology majors’ content knowledge, attitude toward evolution, and theistic position. Evol Educ Outr. 2010;4:137–44.

Robbins JR, Roy P. Identifying and correcting non-science student preconceptions through and inquiry-based, critical approach to evolution. Am Biol Teacher. 2007;69:460–6.

Rutledge ML, Sadler KC. Reliability of the measure of acceptance of the theory of evolution (MATE) instrument with university students. Am Biol Teach. 2007;69:332–5.

Scharmann LC. Evolution and nature of science instruction: a first-person account of changes in evolution instruction throughout a career. Faculty Publications: Department of Teaching, Learning, & Teacher Education. 2018;308. http://digitalcommons.unl.edu/teachlearnfacpub/308 .

Tran MV, Weigel EG, Richmond G. Analyzing upper level undergraduate knowledge of evolutionary processes: can class discussions help? J Coll Sci Teac. 2014;43:87–97.

Tuan H, Chin CC, Tsai CC, Cheng SF. Investigating the effectiveness of inquiry instruction on the motivation of different learning styles students. Int J Sci Math Educ. 2005;3(4):41–566.

Wingert JR, Bassett GM, Terry CE, Lee J. The impact of direct challenges to student endorsement of teleological reasoning on understanding and acceptance of natural selection: an exploratory study. Evo Edu Outreach. 2022;15:4.

Download references

Acknowledgements

The support of the (then) College of Arts and Sciences and its dean, the Provost, and the Office of Sponsored Research and Faculty Development of West Chester University are gratefully acknowledged. In addition, the authors would like to acknowledge the support and counsel of the three external advisors on the project, all biological anthropologists teaching at the three other universities that provided evolution survey comparative data.

This project was supported by an NSF TUES Award (DUE-1245013) and West Chester University.

Author information

Authors and affiliations.

West Chester University, West Chester, PA, 19383, USA

Susan L. Johnston, Maureen Knabb, Josh R. Auld & Loretta Rieser-Danner

You can also search for this author in PubMed   Google Scholar

Contributions

SLJ, MK, and JA designed the curriculum. SLJ is the instructor of record for the course and was responsible for obtaining informed consent and for implementing the curriculum. LR-D served as the project evaluator and conducted all analyses. All authors read and approved the final manuscript.

Authors' information

SLJ is a biological anthropologist and Professor of Anthropology (Department of Anthropology and Sociology); MK is a physiologist and Emeritus Professor of Biology (Department of Biology); JA is an evolutionary biologist and Professor of Biology (Department of Biology); and LR-D is Professor of Psychology (Department of Psychology).

All are affiliated with West Chester University, West Chester, PA, 19383, USA.

Corresponding author

Correspondence to Susan L. Johnston .

Ethics declarations

Ethics approval and consent to participate.

As reported in Methods section.

Consent for publication

Competing interests.

The authors have no competing interests.

Additional information

Publisher's note.

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1..

Revised version of the evolution survey that includes 25 items; administered at WCU and three other universities from Fall 2014 on.

Additional file 2.

Rubric used to review student laboratory worksheets. Includes 4 measures of scientific thinking (defining the problem, developing a plan to assess the problem, analyzing and presenting information, and interpretating findings and solving the problem), with each assessed on a scale of 4 developmental levels (beginning, developing, appropriately developed, and exemplary).

Additional file 3.

A modified version of the RSQC2 classroom assessment technique (Angelo and Cross, 1993 ), completed by students during and after each laboratory module.

Additional file 4.

A 10-item survey completed by WCU students at both the beginning and the end of each semester asking them to rate their level of confidence in their abilities and/or understanding of several pieces of the scientific process. All items were rated on a 5-point Likert scale: 1 = completely doubtful; 2 = somewhat doubtful; 3 = neutral; 4 = somewhat confident; 5 = strongly confident.

Additional file 5.

Evolution survey, concept scores: descriptive statistics by institution

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/ . The Creative Commons Public Domain Dedication waiver ( http://creativecommons.org/publicdomain/zero/1.0/ ) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Cite this article.

Johnston, S.L., Knabb, M., Auld, J.R. et al. Correcting misconceptions about evolution: an innovative, inquiry-based introductory biological anthropology laboratory course improves understanding of evolution compared to instructor-centered courses. Evo Edu Outreach 15 , 6 (2022). https://doi.org/10.1186/s12052-022-00164-4

Download citation

Received : 04 February 2022

Accepted : 02 April 2022

Published : 04 May 2022

DOI : https://doi.org/10.1186/s12052-022-00164-4

Share this article

Anyone you share the following link with will be able to read this content:

Sorry, a shareable link is not currently available for this article.

Provided by the Springer Nature SharedIt content-sharing initiative

• Evolution education
• Biological anthropology
• Inquiry-based laboratory
• Misconceptions
• College level
• Student-centered learning
• Human examples

Evolution: Education and Outreach

ISSN: 1936-6434

• Submission enquiries: [email protected]

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• v.8(6); 2018 Mar

How scientists perceive the evolutionary origin of human traits: Results of a survey study

Hanna tuomisto.

1 Department of Biology, University of Turku, Turku, Finland

Matleena Tuomisto

Jouni t. tuomisto.

2 National Institute for Health and Welfare, Kuopio, Finland

Associated Data

Data are available from the Dryad Digital Repository: https://doi.org/10.5061/dryad.s9r98

Various hypotheses have been proposed for why the traits distinguishing humans from other primates originally evolved, and any given trait may have been explained both as an adaptation to different environments and as a result of demands from social organization or sexual selection. To find out how popular the different explanations are among scientists, we carried out an online survey among authors of recent scientific papers in journals covering relevant fields of science (paleoanthropology, paleontology, ecology, evolution, human biology). Some of the hypotheses were clearly more popular among the 1,266 respondents than others, but none was universally accepted or rejected. Even the most popular of the hypotheses were assessed “very likely” by <50% of the respondents, but many traits had 1–3 hypotheses that were found at least moderately likely by >70% of the respondents. An ordination of the hypotheses identified two strong gradients. Along one gradient, the hypotheses were sorted by their popularity, measured by the average credibility score given by the respondents. The second gradient separated all hypotheses postulating adaptation to swimming or diving into their own group. The average credibility scores given for different subgroups of the hypotheses were not related to respondent's age or number of publications authored. However, (paleo)anthropologists were more critical of all hypotheses, and much more critical of the water‐related ones, than were respondents representing other fields of expertise. Although most respondents did not find the water‐related hypotheses likely, only a small minority found them unscientific. The most popular hypotheses were based on inherent drivers; that is, they assumed the evolution of a trait to have been triggered by the prior emergence of another human‐specific behavioral or morphological trait, but opinions differed as to which of the traits came first.

1. INTRODUCTION

Human evolution is a topic that interests not just researchers specialized in paleoanthropology, but also other scientists and the general public. A number of conflicting hypotheses have been put forward to explain why humans have become strikingly different from other primates. Most scientists in relevant fields (such as paleoanthropology, paleontology, ecology, evolution and human biology) have never published their views on the drivers of human evolution in general, nor on which of the proposed hypotheses on the origin of specific human traits they find most substantiated. No recent summary of the mainstream view among paleoanthropologists has been published either, so there is uncertainty as to whether scientists agree on the driving forces behind human evolution or not. The idea of carrying out a survey to find out emerged when one of us was teaching a university course on human evolution, happened to check what Wikipedia had to say on the subject, and noticed that some Talk pages (especially the one behind the article “Aquatic ape hypothesis”) contained definite but unreferenced claims about what the opinions of “all scientists” or “all paleoanthropologists” are.

Humans differ from all the other 400 primate species in many respects, some of the most striking ones being that they walk fully upright on their hind legs, have unusually big brains, and have an effectively naked rather than fur‐covered skin (Figure  1 ). Other features that among primates are uniquely human include descended larynx, articulated speech and the capacity to accumulate fat in a thick subcutaneous layer.

Male and female human figures from the plaque of the Pioneer 10 and 11 spacecrafts. The pictorial message was intended to describe the origin of the probe for potential extraterrestrial life. It shows several typically human traits, such as bipedalism, nakedness, arched nose, large head, and opposable thumbs. Source: NASA ; vectors by Mysid (Public domain), via Wikimedia Commons

A number of conflicting hypotheses have been proposed to explain why these and other traits originally evolved in the lineage leading to humans but in none of the lineages leading to other extant primates. One line of argumentation is based on the widely accepted idea that animal species adapt to their environment by natural selection: Traits that give the animal a higher probability of survival and reproduction become more common over time and traits related to lower survival and reproduction rates become less common. Adaptive traits are often morphological (like long legs that increase running speed and facilitate escaping from predators, or thick fur that protects from heat loss in cold weather), but they can also be behavioral (like building a nest or being nocturnal). The corollary of viewing traits of a species as adaptations to its environment is that traits are expected to change if the environment changes, because then also the adaptive pressures change. In particular, if sister species have very different traits in spite of close genetic relatedness, the adaptationist scenario suggests that the lineages experienced different environments during their evolutionary past.

It has indeed been proposed that the ancestors of humans came to live in a different kind of environment than the ancestors of chimpanzees and gorillas, and adapted by evolving a suite of novel traits. One of the early proposals along these lines, suggested already by Lamarck and Darwin, was that human ancestors descended from the trees and moved to the open savanna (Bender, Tobias, & Bender, 2012 ; Dart, 1925 ; Domínguez‐Rodrigo, 2014 ; Leakey & Lewin, 1977 ). Because terrestrial life in the dry savanna is very different from arboreal life in wet forests, this change in habitat would have shifted the prevailing selection pressures: Traits that were adaptive in the old environment could become maladaptive in the new one, and novel morphological traits could be favored if they gave a higher probability of survival and reproduction. The ancestors of the great apes stayed in the forest and, therefore, remained more similar to other primates.

The savanna scenario has lost some of its appeal since paleoenvironmental reconstructions started to show that the environmental setting has been more complex than was originally thought. Accordingly, more recent accounts describe the environment of early human ancestors as a mosaic of woodlands, savanna, and water bodies with considerable temporal fluctuations between climatically arid and wet periods (Bender et al., 2012 ; Domínguez‐Rodrigo, 2014 ; Kingston, 2007 ; Kovarovic & Andrews, 2007 ; Maslin & Christensen, 2007 ). Environmental variability itself has also been proposed to have selected for versatility of adaptations (Potts, 1998a , b ).

There have been different views on which aspects of terrestrial life would have required the morphological changes that the human lineage has experienced, so a large number of different explanations have been put forward for each trait. For example, the origin of the bipedal gait has been attributed to (among other things) gaining better visibility over the savanna grass (Ravey, 1978 ), reaching for food on low branches (Hunt, 1994 , 1996 ), collecting small food items from the ground (Jolly, 1970 ; Kingdon, 2003 ), exposing a smaller part of the body to the scorching sun (Wheeler, 1984 , 1991 ), allowing more energy‐efficient long‐distance travel (Carrier et al., 1984 ; Pontzer, Raichlen, & Sockol, 2009 ; Rodman & McHenry, 1980 ), and freeing the hands to carry food, tools, weapons, or babies (Bartholomew & Birdsell, 1953 ; Hewes, 1961 ; Lovejoy, 1981 ; Sutou, 2012 ; Washburn, 1960 ). It has also been proposed that bipedalism originated already in the trees for hand‐supported walking on small branches too weak for brachiation (Crompton, Sellers, & Thorpe, 2010 ; Thorpe, Holder, & Crompton, 2007 ).

Another adaptationist proposal is that the human ancestors moved from the trees to the waterside, and started to adapt to a partly aquatic way of life (Hardy, 1960 ; Morgan, 1982 ; Verhaegen, Puech, & Munro, 2002 ). This would have exposed them to similar selection pressures than semi‐aquatic mammals, rather than to selection pressures typically experienced by other primates. Under this scenario, bipedal gait would have emerged because it allowed wading to deeper water and made the body more streamlined when swimming and diving for food (Kuliukas, 2002 ; Morgan, 1990 ; Niemitz, 2010 ; Verhaegen et al., 2002 ).

Not all traits need to have originated to enhance survival, however, and critical voices have been raised against interpreting all uniquely human traits as adaptations driven by natural selection (Gee, 2013 ). Sexual selection is known to have produced spectacular new traits in various animals, typically ornaments whose sole purpose is to attract the attention of the opposite sex. These confer no survival advantage or may even be harmful to the bearer. At least human bipedalism, nakedness, and subcutaneous fat layer have been explained by this mechanism (Barber, 1995 ; Giles, 2011 ; Tanner, 1981 ). Especially in small populations, traits may even emerge due to chance fixation of random variation (Sutou, 2012 ).

For someone interested in the “why” of human evolution, it is currently hard to find a comprehensive account of the scientific state of the art. Journal articles typically address only one or a few hypotheses in isolation of the others and often their focus is more on “how” than on “why” a given trait originally emerged (e.g., Crompton et al., 2010 ; Cunnane & Crawford, 2014 ; Isler & Van Schaik, 2014 ; Stout & Chaminade, 2012 ; Watson, Payne, Chamberlain, Jones, & Sellers, 2008 ; Wells, 2006 ). Only proponents of the aquatic/waterside hypotheses (collectively known as the aquatic ape hypothesis or AAH) seem to maintain that it is possible to explain most of the uniquely human traits as adaptive responses to a specific external factor (e.g., Morgan, 1997 ; Vaneechoutte, Kuliukas, & Verhaegen, 2011 ), but these views have found little resonance in paleoanthropological journals (Bender et al., 2012 ). Indeed, AAH has been fiercely opposed and criticized for being an umbrella hypothesis that attempts to explain everything, for being unparsimonious, for lacking evidence and even for being pseudoscience (Hawks, 2005 ; Langdon, 1997 ; Moore, 2012 ).

Here, we aim to find out what scientists really think about why some of the most striking human traits have emerged. We do so by analyzing the results of an online survey where scientists were directly asked for their views on the issue.

2. MATERIALS AND METHODS

2.1. survey.

A survey was performed using an online form in early 2013. Invitation to participate in the survey was sent by email to the authors of articles and review papers that had been published in a scientific journal of a relevant field during the three previous years (2010–2012). A 3‐year period was thought to be long enough for most researchers to have published at least one scientific paper, but short enough for most of the email addresses given in those papers not to have become obsolete. The focus was on journals of paleontology, zoology, ecology, evolutionary biology, and human biology. Only journals with an ISI impact factor equal to or larger than 1.0 were considered. The exact criteria used to select the journals, as well as a full list of journal names, can be found in Appendix S1 .

Almost 58,000 unique email addresses were found in the information available online for the papers published in the selected journals during the selected time period. The full address list exceeded the capacity of the online survey system (Webropol), so the addresses were sorted in alphabetical order, and an invitation to participate in the survey was sent to the first 29,000 addresses. The remaining addresses were used for a different survey, whose results will be reported elsewhere. The first page of the online survey informed participants about the purpose of the survey. The survey was performed anonymously, and all who responded did so voluntarily. After a few reminders had been sent, a total of 1,266 persons had submitted their responses to the survey.

Although the initial sample was large and can be considered representative of the scientific community in relevant fields, the proportion of invitees who answered the survey was very small (4.4%). The sample is no doubt biased toward people who have a larger than average interest in human evolution. Therefore, the obtained answers do not reflect the opinions of the entire scientific community. Nevertheless, they can indicate whether any of the hypotheses proposed to explain the evolutionary origin of a specific human trait is universally accepted or rejected. Even if this were not the case, the survey gives indication of which hypotheses are most or least popular, although conclusions in this respect remain tentative.

The survey first asked background information of the respondent, such as gender, age, the highest academic degree obtained, number of scientific publications authored (both overall and on human evolution), degree of knowledge about human evolution, and whether the respondent has taught courses on human evolution. The second part listed fifteen human traits (such as bipedalism) and asked the respondents to rate the credibility of 51 alternative hypotheses that have been proposed to explain their evolutionary origin (such as freeing the hands for tool use or seeing over tall grass). The credibility scoring was done using a five‐point scale: very unlikely, moderately unlikely, no opinion, moderately likely, and very likely. The number of alternative hypotheses considered was ten for both bipedalism and brain size, eight for hairlessness, seven for speech, four for subcutaneous fat, and three for descended larynx. In addition, there were nine traits for which only one explanation has been proposed in the literature, and this was related to the aquatic ape hypothesis. The third part asked about the respondents’ views on criticism against AAH. All questions and a summary of the answers are presented in Appendix S2 .

2.2. Data analyses

The respondents were asked for their professional field of expertise by offering 15 alternatives. For statistical analyses, these were simplified to four categories to ensure sufficient sample size in each. The group “(paleo)anthropologist” was formed by lumping the originally separate fields “paleoanthropology” and “anthropology or archaeology.” The group “biologist” was formed by lumping all the original subfields of biology (animal physiology, anatomy, or morphology; ecology; evolution; genetics or molecular biology; other) and the group “human biologist” by lumping all subfields of human biology (cardiovascular or respiratory system, musculoskeletal system, nervous system, nutrition, other aspects of human biology). The fourth group was “other,” which contained the remaining fields (geology, paleontology, other).

Overall relationships among the hypotheses were visualized by principal coordinates analysis (PCoA), where the objects were the hypotheses and the descriptors were individual respondents, with the variable of interest being the credibility score each respondent had given to each hypothesis. A Euclidean distance matrix was calculated, such that the distance between two hypotheses reflects how differently the respondents scored their credibilities. Every respondent who gave one of the hypotheses a higher score than the other increased the final distance between the hypotheses, with the overall distance between the hypotheses equaling zero if every respondent had scored both hypotheses similarly (irrespective of whether the score itself was high or low). PCoA visualizes these pairwise distances, so the closer together two hypotheses get plotted in the ordination diagram, the more similar their explanatory value is in the opinion of an average individual respondent.

The respondents themselves were plotted in the PCoA ordination space on the basis of the scores they had given to the hypotheses. Therefore, the relative positions of the respondents reflect their opinions on the hypotheses: Respondents get plotted toward the same part of the ordination space as the hypotheses they gave highest credibility scores, and far away from the hypotheses they gave lowest scores.

Relationships between the respondents’ opinions and their backgrounds were first assessed visually with the help of the ordination diagram. We then used analysis of variance to test whether there were differences in the average opinions of respondents of different backgrounds. If so, a post hoc Tukey's honest significance test was carried out to assess which aspects of the respondents’ background were associated with differences in opinion. A more detailed breakdown of the respondents’ opinions was obtained by visually comparing the distributions of the credibility scores given to the different hypotheses. This was done both to obtain an idea of which hypotheses are most popular overall, and to see if there were differences among respondents representing different scientific fields and/or having different levels of scientific experience.

R statistical software version 3.3.2 ( https://cran.r-project.org/ ) was used both to run the analyses and to produce the graphs. The vegan package (Oksanen et al., 2015 ) was used for principal coordinates analysis. The survey data and all R code used to manipulate and analyze the data are available at Opasnet web‐workspace http://en.opasnet.org/w/Evolutionary_origin_of_human_traits . The survey data are also available from the Dryad Digital Repository https://doi.org/10.5061/dryad.s9r98 .

Principal coordinates analysis revealed some clear patterns among the hypotheses proposed to explain the evolutionary origin of specific human traits. The most eye‐catching feature of the ordination diagram in Figure  2 a is that the hypotheses got divided into two elongated groups that parallel each other but are clearly separated (the abbreviations of Fig. ​ Fig.2 2 are explained in Table  1 ). The smaller group contains all the hypotheses that evoke adaptation to swimming or diving as an explanatory factor for the emergence of a trait, and the larger group contains all other hypotheses, whether they refer to adaptation to a specific environment or to needs that emerge from a specific behavior. Because all the hypotheses in the smaller group refer to locomotion in water and have been included in the aquatic ape hypothesis (AAH), this group will be referred to as the water‐related or AAH group. For lack of a better unifying term, the larger group will be referred to as the dryland group.

Principal coordinates analysis ( PC oA) of different hypotheses proposed to explain the evolutionary origin of specific human traits. Distances between hypotheses are based on scores given by (a) all respondents, or only respondents whose main field of expertise is (b) anthropology or paleoanthropology, (c) biology, (d) human biology, or (e) other. Each colored point corresponds to one hypothesis, and the color indicates which of the traits listed in the inset the hypothesis aims to explain. Points are scaled to reflect the average credibility score given to the corresponding hypothesis by the respondents of the mentioned expertise group. The hypothesis name abbreviations are explained in Table  1 . Each gray point in (a) corresponds to one respondent, whose position within the ordination space reflects the scores given to the hypotheses. For example, respondents plotted toward the bottom left part of the respondent cloud found the hypotheses plotted toward the bottom left of the hypothesis cloud more credible than the hypotheses at the top, and vice versa. More details on the respondent ordination are shown in Figure  3

The hypotheses on the evolutionary origin of human traits that were included in an online survey to find out how popular they are among scientists. The abbreviations are used in the figures, and the full text is copied verbatim from the survey. If ambiguous, the abbreviated hypothesis is followed by a letter depicting the trait: B = bipedalism, E = encephalization (big brain), F = subcutaneous fat, N = nakedness, L = descended larynx, S = speech, O = other

Within each of the two groups, the hypotheses got sorted by their popularity, with the average credibility score increasing toward the bottom left in Figure  2 a. A tight cluster at the extreme left of the dryland group was formed by five hypotheses with high average credibility scores (4.08–4.26 on a 1–5 scale, with 1 corresponding to “very unlikely” and 5 to “very likely”). This cluster included the most popular hypothesis for the subcutaneous fat layer (energy reserve especially for the developing brain), the descended larynx (required by articulate speech), bipedalism (use of tools and weapons), speech (social pressure for elaborate communication), and the big brain (complex social organization).

This combination might be the most popular overall scenario for the origin of these traits, but the next most popular 2–3 explanations for bipedalism (freeing hands for foraging, better view over tall grass), large brain (required by either language or collaborative hunting), and speech (required by either collaborative hunting or transmitting cultural tradition; triggered by the descended larynx) also received high average credibility scores (3.53–3.96). Their proximity in ordination space indicated that they were found credible by the same respondents, which makes it difficult to identify a single most popular overall scenario. The hypotheses explaining hairlessness were not found convincing by the respondents, as even the two most popular ones (avoidance of overheating when hunting, avoidance of ectoparasites) had average credibility scores of only 3.48 and 3.17, respectively.

Eleven of the twelve most popular hypotheses were based on inherent drivers of evolution, that is, proposing that morphological traits emerged in response to selection pressure either from a novel behavior or from a pre‐existing morphological trait. Hypotheses based on selection pressure from a new kind of external environment were less popular even within the dryland group, and the credibility scores of all the hypotheses in the water‐related group were low to intermediate (2.26–2.99). The hypotheses proposing that encephalization was triggered by improved nutrition also received intermediate popularity scores, whether achieved by cooking or by increased consumption of fish or meat (all three with credibility scores in the range 2.61–2.77). The four least popular hypotheses of all (credibility scores 1.95–2.20) were based on inherent drivers operating on dry land.

The ordination results suggest that the respondents viewed the water‐related hypotheses as an ensemble whose overall credibility they assessed independently of how they scored the credibilities of the other hypotheses. This impression is strengthened when viewing the ordination of the respondents (the gray cloud in Figure  2 a) in more detail (Figure  3 ). The main gradient among the respondents follows the average credibility score they gave for the water‐related hypotheses (Figure  2 a), and this is almost perpendicular to the (less clear) gradient of average credibility scores given for the twelve most popular hypotheses (Figure  3 b).

The positions of the survey respondents in the space of the principal coordinates analysis shown in Figure  2 a. The ordination is the same in each panel, but colors illustrate different kinds of information related to each respondent. The colored crosses indicate the mean position of the respondents belonging to the respective subgroup. (a) Average credibility score given to the hypotheses in the water‐related group (the smaller cloud of points in Figure  2 a). (b) Average score given to the 12 most popular hypotheses in Figure  2 a. (c) Number of scientific publications authored or co‐authored (crosses of all three categories overlap). (d) Field of expertise. (e) Familiarity with hypotheses on human evolution. (f) Experience in teaching human evolution

The respondents’ position in the ordination did not seem to be related with how much scientific experience they had in general, as measured with the total number of scientific publications they had authored (Figure  3 c), but it was related with how much they knew about human evolution. Those having more background information on this specific topic (by self‐assessment, by main field of expertise being paleoanthropology or anthropology, or by having taught university courses on the topic) appeared to be more often plotted in the upper part of the ordination than respondents representing other backgrounds (Figure  3 d–f).

The visual impressions were confirmed by statistical analyses. These were carried out separately for five different subgroupings of the hypotheses. Three of these were chosen because they formed clear groups in the ordination of Figure  2 a (the dryland hypotheses, the water‐related hypotheses, the 12 most popular dryland hypotheses). The dryland hypotheses were also split into those based on environmental adaptation and those evoking behavioral drivers.

The largest effect by far on the responses was that of the field or expertise, with (paleo)anthropologists being more critical overall than representatives of any other expertise group (Table  2 ). The difference was especially large for the water‐related hypotheses: The average credibility score given by (paleo)anthropologists to this group of hypotheses (2.10 on the 1–5 scale) was much lower than the average score given by human biologists (3.02), with biologists (2.70), and others (2.67) being intermediate. For the dryland hypotheses, the difference between (paleo)anthropologists (2.97) and human biologists (3.22) was only 0.25 (vs. 0.92 in the case of the water‐related hypotheses), and the differences in the scores given by biologists, human biologists, and others were not statistically significant.

Results of Tukey's HSD test between different subgroups of respondents (line starting with Test result ~) and their average credibility scores (standard deviation in parentheses) for different groups of hypotheses: the most popular 12 hypotheses; the dryland hypotheses (the larger hypothesis group in Figure 2a); the water‐related hypotheses (the smaller hypothesis group in Figure 2a); dryland hypotheses based on behavioural demands; dryland hypotheses based on adaptation to the external environment

The results obtained with respondent subgroups based on total number of authored peer reviewed publications and total number of authored popular science publications are not shown, because they were not associated with significantly different ( p  < .05) means in any comparisons.

*** p  < .001; ** p  < .01; * p  < .05.

Overall scientific experience (as measured with the number of scientific publications authored) had no effect on the scores given to either the dryland or the water‐related hypotheses (Table  2 ). However, the more knowledge the respondents had on human evolution specifically (self‐assessed familiarity with the hypotheses, number of scientific publications on human evolution or experience in teaching human evolution), the lower the scores they gave to the water‐related hypotheses. Among biologists, those who knew more about human evolution were more critical than the less knowledgeable ones, and (paleo)anthropologists were more critical than human biologists with the same self‐assessed knowledge level.

When the dryland hypotheses were split into two groups depending on whether they were based on behavioral arguments or environmental adaptation, both groups obtained rather similar results. The main difference was that the behavioral hypotheses received somewhat higher average credibility scores, which reflects the fact that 10 of the 12 most popular hypotheses were based on behavior (on the other hand, so were the four least popular hypotheses).

To visualize the differences in opinion among the (paleo)anthropologists and representatives of other fields, we repeated the ordination of the hypotheses for each of the four respondent groups separately. In accordance with the fact that most respondents were biologists, the ordination based on the biologists’ data only (Figure  2 c) was very similar to the ordination based on all respondents (Figure  2 a). The ordination based on (paleo)anthropologists’ views (Figure  2 b) differed especially in relation to the hypotheses for bipedalism: Hypotheses that explained bipedalism by foraging, tool use, or carrying were very far removed from the main cloud and toward the opposite side than the water‐related hypotheses. In addition, the average credibility scores given to the water‐related hypotheses were among the lowest of any hypotheses. This contrasted with the situation in the ordination based on human biologists’ data (Figure  2 d), in which the water‐based hypotheses had intermediate credibility scores.

The hypotheses differed clearly from each other in the frequencies of different credibility scores, but there were some similarities in the overall pattern among those six traits for which three or more hypotheses were evaluated (Figure  4 ). None of the hypotheses received the “very likely” score from more than 46% of the respondents, but most traits had at least one hypothesis that was considered “very likely” by more than 23% and likely (either “very likely” or “moderately likely”) by 72%–90%. Many of the intermediately popular hypotheses divided the respondents rather evenly between those who found them likely and those who found them unlikely (the latter referring to the scores “very unlikely” and “moderately unlikely” combined).

Credibility scores given by survey respondents to hypotheses that aim to explain the evolutionary origin of specific human traits. The hypotheses are sorted in order of decreasing popularity as estimated by the percentage of respondents who scored them likely (i.e., either “very likely” or “moderately likely”). Descriptions of the hypotheses as they were given in the survey are shown in Table  1

A causal relationship between articulate speech and descended larynx was accepted by most respondents, but there was no consensus on the direction of the causality. That the larynx descended because this was required by articulate speech was found likely by 84% and very likely by 43%. At the same time, that the evolution of speech was triggered by the descended larynx was found likely by 61% and very likely by 18%. In fact, 36% of the respondents scored both directions as equally likely.

Traits in the category “other” had only one explanatory hypothesis each in the survey, and this was water‐related. All of these hypotheses received many more “very unlikely” than “very likely” scores. However, four hypotheses (that baby swimming, profuse sweating, diving ability, and magnitude of diving reflex evolved as adaptations to a semi‐aquatic way of life) received so many “moderately likely” scores that the percentage of respondents who found them likely was slightly larger than the percentage who found them unlikely (Figure  4 ).

Details on how the hypotheses were scored by respondents representing different fields of expertise are shown in Figure  5 . In accordance with the statistical test results, most hypotheses received rather similar scores from respondents of all fields of expertise. However, (paleo)anthropologists were clearly more critical than representatives of the other fields in relation to several hypotheses, including: that nakedness evolved to avoid ectoparasites, that the big brain evolved because warfare caused pressure for higher intelligence, and that any traits evolved as adaptations to swimming or diving.

Frequencies of credibility scores given to hypotheses aiming to explain different traits (columns) by respondents of different fields of expertise (rows). In each panel, the answers are, from left to right, “very likely,” moderately likely,” “no opinion,” “moderately unlikely,” and “very unlikely.” Hypotheses that have been included in the aquatic ape hypothesis are shown in shades of blue and green. Those dryland hypotheses for which the opinions of anthropologists and other expertise groups clearly diverged are shown in magenta. The other hypotheses are in shades of brown, with darker colors given to hypotheses that received higher average credibility scores in the survey

There was a lot of variation among the traits in how many of the proposed explanations the respondents found convincing (Figure  6 ). For any one trait, 33%–64% of the respondents did not find any of the proposed hypotheses “very likely,” while 19%–38% found exactly one and 8%–45% more than one. Ten respondents (0.8%) explained that they did not score any of the hypotheses as likely, because they do not believe that humans have evolved at all (most of them explicitly referred to special creation by God).

The number of hypotheses (colors) proposed to explain each human trait (rows) that each respondent found very likely (left panel) or likely (either very likely or moderately likely; right panel). The total number of hypotheses included in the survey is shown after the name of each trait

The survey asked respondents’ opinions on twenty critical arguments that have been presented against the aquatic ape hypothesis. For most arguments, the modal response was “no opinion,” especially among those 43% of the respondents who had never heard of AAH before. Nevertheless, some arguments were clearly more frequently agreed with than others (Figure  7 and Table  3 ). The most widely accepted critique was that not all aquatic mammals have naked skin, so hairlessness cannot be considered an aquatic adaptation. In the other extreme, less than 3% of the respondents fully agreed and less than 12% mostly agreed with the critique that AAH is unscientific or not worthy of attention for the reasons given; in most cases, the number of respondents who strongly disagreed with these critiques was larger than the number who mostly or fully agreed.

The degree to which respondents representing different expertise fields agree with critique presented against the aquatic ape hypothesis. The full description of each point of critique can be found in Table  3

Points of critique presented against the aquatic ape hypothesis (AAH). The abbreviations are used in Figure  7 , and the full text is copied verbatim from the survey

4. DISCUSSION

The main results of our survey can be summarized as follows: (1) There was no general agreement among the respondents on why any of the uniquely human traits have evolved: None of the proposed hypotheses was universally either accepted or rejected. (2) For any individual trait, the percentage of respondents who found none of the hypotheses “very likely” was between >30% (bipedalism) and >65% (nakedness). (3) In general, opinions on the credibility of the hypotheses were independent of a person's background (gender, age, field of expertise, degree of scientific experience), but (paleo)anthropologists were clearly more critical than representatives of other fields. (4) The hypotheses that mention adaptation to swimming or diving as an explanatory factor were found much less credible by (paleo)anthropologists and slightly more credible by human biologists than by biologists and representatives of other fields. (5) Most respondents were critical about the aquatic ape hypothesis (AAH), but only a small minority considered it to be unscientific.

Of course, all conclusions based on the survey data must be considered tentative only, because the response rate was very low, and it is possible that the results are biased. Members of some subgroup might have been more likely to respond than members of some other subgroup, and the average credibility scores given to the different hypotheses by the respondents may not be representative of the opinions of all scientists in the background population. However, it is unlikely that a lack of general agreement on the drivers of trait evolution or such a clear difference in opinion between (paleo)anthropologists and others could have emerged just as a result of biased sampling.

Our results did not reveal a set of explanations that would collectively provide a coherent and popular scenario for the origin of all (or even many) human traits. Indeed, some of the hypotheses that had almost equal and rather high average credibility scores explained the same trait, whereas for other traits, no hypothesis emerged as particularly popular. Against this background, it is interesting that almost half of the respondents fully or mostly agreed with the statement that the aquatic ape hypothesis “is not needed, because all human traits can be explained by terrestrial scenarios”.

The lack of agreement on why humans evolved the traits we have today is very obvious in our results: No hypothesis was universally accepted, and for most traits, there were several almost equally popular alternative hypotheses rather than one that would generally be considered superior to the others. None of the hypotheses received the score “very likely” from more than half of the respondents or obtained an average credibility score higher than 4.26 (of 5). For hairlessness, the most popular hypothesis was thought to be “very likely” by only 16% of the respondents, and its average credibility score (3.48) was closer to 3 (which is the limit between being considered more likely than unlikely) than to 4 (moderately likely). In addition, for only two of the traits (subcutaneous fat layer and descended larynx), the most popular hypothesis was found at least moderately likely by almost all respondents at the same time as the next most popular hypothesis was found clearly less likely. This may partly reflect the fact that fewer alternative hypotheses have been proposed for these traits than for many of the others included in the survey.

Importantly, lack of agreement did not reflect just ignorance on the topic among nonspecialists, because the responses were, in general, very similar between anthropologists and respondents representing other fields of science. In fact, anthropologists were even more skeptical about all hypotheses than representatives of the other fields were. In other words, outsiders were slightly more convinced that the proposed hypotheses are plausible than those who work in the field. Maybe anthropologists (especially paleoanthropologists) are more systematically trained to be wary of just‐so‐stories (explanations of past events and processes backed up by little or no evidence) than students in nearby fields are. It is also possible that outsiders are somewhat less likely to question hypotheses proposed within an unfamiliar field. This could be because they do not feel qualified to do so, or because they have not heard of the debates that draw attention to the weaknesses of the hypotheses.

Our results conform with the widespread belief that professionals in the field of human evolution are more critical toward the aquatic ape hypothesis (AAH) than outsiders are (Langdon, 1997 ; Bender et al., 2012 ; see also nonscientific sources such as Hawks, 2005 ; Moore, 2012 and Wikipedia: Aquatic Ape Hypothesis: Talk). However, this did not seem to be due to overall scientific ignorance, because how respondents assessed the credibility of the hypotheses proposing adaptation to swimming or diving was independent of both their overall scientific experience level and how they assessed the credibility of the other hypotheses. Interestingly, those whose main field of expertise is human biology had the most positive attitudes toward the water‐related hypotheses, giving them an average credibility score that was as much as 0.9 units higher (on a 1–5 scale) than the average score given by (paleo)anthropologists.

The difference in average opinion between (paleo)anthropologists and other scientists can be interpreted in two opposite ways. On the one hand, those who know the field of human evolution best may be best positioned to make a justified evaluation of the validity of the alternative hypotheses. On the other hand, prior knowledge may induce one to reject unconventional hypotheses offhand merely because they challenge the established paradigms of a field (Bender et al., 2012 ; Klayman, 1995 ). Obviously, the two interpretations lead to opposite conclusions on whether or not the critical attitude of the (paleo)anthropologists can be taken as evidence that AAH is flawed. In our survey, a vast majority of the respondents who had an opinion on the issue disagreed with the statement that AAH can be ignored because its main proponents are not professionals in the field of human evolution. This was the case both overall and within each field of expertise separately, although the proportion of respondents who agreed with the statement was higher among (paleo)anthropologists than among representatives of the other fields.

In this context, it is also interesting that the respondents’ assessment of the credibility of the water‐related hypotheses did not depend on the number of scientific papers they had authored. This indicates that established scientists are no more likely to reject or accept these hypotheses than junior scientists are—unless their scientific experience relates directly to the field of human evolution. A vast majority of the respondents disagreed with the critique that AAH is unscientific. Of course, this does not mean that they would consider the explanations proposed by AAH to be correct, and indeed, all the hypotheses related to AAH received relatively low credibility scores (although not as low as the least popular dryland hypotheses).

If, for the sake of argument, we accept the most popular explanation for each trait to be the correct one, a scenario of evolution by internal drive emerges: The large brain evolved because complex social organization required higher intelligence, the subcutaneous fat layer evolved to serve as an energy reserve for the developing brain, articulate speech evolved because there was social pressure for elaborate communication, the larynx descended because this was required by articulate speech, bipedalism evolved to make the use of tools and weapons easier, and nakedness evolved to avoid overheating when hunting. For most traits, the next most popular explanation was not far behind in popularity. Most of these were also based on inherent drivers, but sometimes in the opposite temporal sequence (e.g., articulate speech was triggered by the descended larynx; large brain evolved because it was required by articulate speech). We found this result disturbing, because the overwhelming popularity of hypotheses based on inherent drivers gives the impression that human evolution is generally thought to have been goal‐directed. This would be in conflict with the current understanding (explained in every evolutionary biology textbook) that evolution has no foresight.

Overall, the survey revealed no general agreement among the respondents: None of the proposed hypotheses on why specific uniquely human traits have evolved was universally either accepted or rejected. Nevertheless, identifying and quantifying what is not generally known and agreed upon can be useful in itself, as it may help to focus future research on answering the most important open questions. Clearly, there is still a long way to go before the question “why are humans so different from other primates” has been answered in a comprehensive and generally satisfactory way.

DATA ACCESSIBILITY

Conflict of interest.

None declared.

AUTHOR CONTRIBUTIONS

HT designed and conducted the survey and led the writing. All authors discussed the results and planned the data analyses together. The R code used to analyze the data and draw the figures was written by MT with contributions from JT.

Supporting information

Acknowledgments.

We thank Carlos Peña for writing the code to extract respondents’ email addresses from the Internet; Mirkka Jones, Kalle Ruokolainen, and Timo Vuorisalo for comments that helped to improve the survey questions; and Jouko Tuomisto for comments on the manuscript.

Tuomisto H, Tuomisto M, Tuomisto JT. How scientists perceive the evolutionary origin of human traits: Results of a survey study . Ecol Evol . 2018; 8 :3518–3533. https://doi.org/10.1002/ece3.3887 [ PMC free article ] [ PubMed ] [ Google Scholar ]

• Barber, N. (1995). The evolutionary psychology of physical attractiveness: Sexual selection and human morphology . Ethology and Sociobiology , 16 , 395–424. https://doi.org/10.1016/0162-3095(95)00068-2 [ Google Scholar ]
• Bartholomew, G. A. , & Birdsell, J. B. (1953). Ecology and the Protohominids* . The American Anthropologist , 55 , 481–498. [ Google Scholar ]
• Bender, R. , Tobias, P. W. , & Bender, N. (2012). The savannah hypotheses: Origin, reception and impact on paleoanthropology . History & Philosophy of the Life Sciences , 34 , 147–184. [ PubMed ] [ Google Scholar ]
• Carrier, D. R. , Kapoor, A. K. , Kimura, T. , Nickels, M. K. , Scott, E. C. , So, J. K. , & Trinkaus, E. (1984). The energetic paradox of human running and hominid evolution [and comments and reply] . Current Anthropology , 25 , 483–495. https://doi.org/10.1086/203165 [ Google Scholar ]
• Crompton, R. H. , Sellers, W. I. , & Thorpe, S. K. S. (2010). Arboreality, terrestriality and bipedalism . Philosophical Transactions of the Royal Society B: Biological Sciences , 365 , 3301–3314. https://doi.org/10.1098/rstb.2010.0035 [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Cunnane, S. C. , & Crawford, M. A. (2014). Energetic and nutritional constraints on infant brain development: Implications for brain expansion during human evolution . Journal of Human Evolution , 77 , 88–98. https://doi.org/10.1016/j.jhevol.2014.05.001 [ PubMed ] [ Google Scholar ]
• Dart, R. (1925). Australopithecus africanus : The man‐ape of South Africa . Nature , 115 , 195–199. https://doi.org/10.1038/115195a0 [ Google Scholar ]
• Domínguez‐Rodrigo, M. (2014). Is the “Savanna Hypothesis” a dead concept for explaining the emergence of the earliest hominins? Current Anthropology , 55 , 59–81. https://doi.org/10.1086/674530 [ Google Scholar ]
• Gee, H. (2013). The accidental species: Misunderstandings of human evolution . Chicago, IL: University of Chicago Press; https://doi.org/10.7208/chicago/9780226044989.001.0001 [ Google Scholar ]
• Giles, J. (2011). Naked love: The evolution of human hairlessness . Biology Theory , 5 , 326–336. [ Google Scholar ]
• Hardy, A. (1960). Was man more aquatic in the past? The New Scientist , 7 , 642–645. [ Google Scholar ]
• Hawks, J. (2005). Why anthropologists don't accept the Aquatic Ape Theory . Retrieved from http://johnhawks.net/weblog/topics/pseudoscience/aquatic_ape_theory.html [accessed 5 September 2017]
• Hewes, G. W. (1961). Food transport and the origin of hominid bipedalism . The American Anthropologist , 63 , 687–710. https://doi.org/10.1525/aa.1961.63.4.02a00020 [ Google Scholar ]
• Hunt, K. D. (1994). The evolution of human bipedality: Ecology and functional morphology . Journal of Human Evolution , 26 , 183–202. https://doi.org/10.1006/jhev.1994.1011 [ Google Scholar ]
• Hunt, K. D. (1996). The postural feeding hypothesis: An ecological model for the evolution of bipedalism . South African Journal of Science , 92 , 77–90. [ Google Scholar ]
• Isler, K. , & Van Schaik, C. P. (2014). How humans evolved large brains: Comparative evidence . Evolutionary Anthropology: Issues, News, and Reviews , 23 , 65–75. https://doi.org/10.1002/evan.21403 [ PubMed ] [ Google Scholar ]
• Jolly, C. J. (1970). The seed‐eaters: A new model of hominid differentiation based on a baboon analogy . Man , 5 , 5–26. https://doi.org/10.2307/2798801 [ Google Scholar ]
• Kingdon, J. (2003). Lowly origin: Where, when, and why our ancestors first stood up . Princeton, NJ: Princeton University Press. [ Google Scholar ]
• Kingston, J. D. (2007). Shifting adaptive landscapes: Progress and challenges in reconstructing early hominid environments . The American Journal of Physical Anthropology , 134 , 20–58. https://doi.org/10.1002/(ISSN)1096-8644 [ PubMed ] [ Google Scholar ]
• Klayman, J. (1995). Varieties of confirmation bias In Busemeyer J., Hastie R., & Medin D. L. (Eds.), Psychology of learning and motivation (pp. 385–418). Cambridge, MA: Academic Press. [ Google Scholar ]
• Kovarovic, K. , & Andrews, P. (2007). Bovid postcranial ecomorphological survey of the Laetoli paleoenvironment . Journal of Human Evolution , 52 , 663–680. https://doi.org/10.1016/j.jhevol.2007.01.001 [ PubMed ] [ Google Scholar ]
• Kuliukas, A. (2002). Wading for food the driving force of the evolution of bipedalism? Nutrition and Health , 16 , 267–289. https://doi.org/10.1177/026010600201600402 [ PubMed ] [ Google Scholar ]
• Langdon, J. H. (1997). Umbrella hypotheses and parsimony in human evolution: A critique of the Aquatic Ape Hypothesis . Journal of Human Evolution , 33 , 479–494. https://doi.org/10.1006/jhev.1997.0146 [ PubMed ] [ Google Scholar ]
• Leakey, R. , & Lewin, R. (1977). Origins: The emergence and evolution of our species and its possible future . Boston, MA: E. P. Dutton. [ Google Scholar ]
• Lovejoy, C. O. (1981). The origin of man . Science , 211 , 341–350. https://doi.org/10.1126/science.211.4480.341 [ PubMed ] [ Google Scholar ]
• Maslin, M. A. , & Christensen, B. (2007). Tectonics, orbital forcing, global climate change, and human evolution in Africa: Introduction to the African paleoclimate special volume . Journal of Human Evolution , 53 , 443–464. https://doi.org/10.1016/j.jhevol.2007.06.005 [ PubMed ] [ Google Scholar ]
• Moore, J. (2012). Aquatic ape theory: Sink or swim? Retrieved from www.aquaticape.org [accessed 5 September 2017]
• Morgan, E. (1982). The aquatic ape: A theory of human evolution . London, UK: Souvenir Press. [ Google Scholar ]
• Morgan, E. (1990). The scars of evolution: What our bodies tell us about human origins . London, UK: Souvenir Press. [ Google Scholar ]
• Morgan, E. (1997). The aquatic ape hypothesis . London, UK: Souvenir Press. [ Google Scholar ]
• Niemitz, C. (2010). The evolution of the upright posture and gait—a review and a new synthesis . Naturwissenschaften , 97 , 241–263. https://doi.org/10.1007/s00114-009-0637-3 [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Oksanen, J. , Blanchet, F. G. , Kindt, R. , Legendre, P. , Minchin, P. R. , O'Hara, R. B. , … Wagner, H. (2015). Vegan: Community ecology package . R package version 2.3‐2. http://CRAN.R-project.org/package=vegan
• Pontzer, H. , Raichlen, D. A. , & Sockol, M. D. (2009). The metabolic cost of walking in humans, chimpanzees, and early hominins . Journal of Human Evolution , 56 , 43–54. https://doi.org/10.1016/j.jhevol.2008.09.001 [ PubMed ] [ Google Scholar ]
• Potts, R. (1998a). Environmental hypotheses of hominin evolution . The American Journal of Physical Anthropology , 107 , 93–136. https://doi.org/10.1002/(ISSN)1096-8644 [ PubMed ] [ Google Scholar ]
• Potts, R. (1998b). Variability selection in hominid evolution . Evolutionary Anthropology: Issues, News, and Reviews , 7 , 81–96. https://doi.org/10.1002/(ISSN)1520-6505 [ Google Scholar ]
• Ravey, M. (1978). Bipedalism: An early warning system for Miocene Hominoids . Science , 199 , 372 https://doi.org/10.1126/science.199.4327.372 [ PubMed ] [ Google Scholar ]
• Rodman, P. S. , & McHenry, H. M. (1980). Bioenergetics and the origin of hominid bipedalism . The American Journal of Physical Anthropology , 52 , 103–106. https://doi.org/10.1002/(ISSN)1096-8644 [ PubMed ] [ Google Scholar ]
• Stout, D. , & Chaminade, T. (2012). Stone tools, language and the brain in human evolution . Philosophical Transactions of the Royal Society B: Biological Sciences , 367 , 75–87. https://doi.org/10.1098/rstb.2011.0099 [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Sutou, S. (2012). Hairless mutation: A driving force of humanization from a human–ape common ancestor by enforcing upright walking while holding a baby with both hands . Genes to Cells , 17 , 264–272. https://doi.org/10.1111/j.1365-2443.2012.01592.x [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Tanner, N. M. (1981). On becoming human . Cambridge, UK: CUP Archive. [ Google Scholar ]
• Thorpe, S. K. S. , Holder, R. L. , & Crompton, R. H. (2007). Origin of human bipedalism as an adaptation for locomotion on flexible branches . Science , 316 , 1328–1331. https://doi.org/10.1126/science.1140799 [ PubMed ] [ Google Scholar ]
• Vaneechoutte, M. , Kuliukas, A. , & Verhaegen, M. (2011). Was man more aquatic in the past? Fifty years after Alister hardy – Waterside hypotheses of human evolution . Sharjah, UAE: Bentham Science Publishers. [ Google Scholar ]
• Verhaegen, M. , Puech, P.‐F. , & Munro, S. (2002). Aquarboreal ancestors? Trends in Ecology & Evolution , 17 , 212–217. https://doi.org/10.1016/S0169-5347(02)02490-4 [ Google Scholar ]
• Washburn, S. L. (1960). Tools and human evolution . Scientific American , 203 , 62–75. https://doi.org/10.1038/scientificamerican0960-62 [ PubMed ] [ Google Scholar ]
• Watson, J. C. , Payne, R. C. , Chamberlain, A. T. , Jones, R. K. , & Sellers, W. I. (2008). The energetic costs of load‐carrying and the evolution of bipedalism . Journal of Human Evolution , 54 , 675–683. https://doi.org/10.1016/j.jhevol.2007.10.004 [ PubMed ] [ Google Scholar ]
• Wells, J. C. K. (2006). The evolution of human fatness and susceptibility to obesity: An ethological approach . Biological Reviews , 81 , 183–205. https://doi.org/10.1017/S1464793105006974 [ PubMed ] [ Google Scholar ]
• Wheeler, P. E. (1984). The evolution of bipedality and loss of functional body hair in hominids . Journal of Human Evolution , 13 , 91–98. https://doi.org/10.1016/S0047-2484(84)80079-2 [ Google Scholar ]
• Wheeler, P. E. (1991). The influence of bipedalism on the energy and water budgets of early hominids . Journal of Human Evolution , 21 , 117–136. https://doi.org/10.1016/0047-2484(91)90003-E [ Google Scholar ]

Numbers, Facts and Trends Shaping Your World

Read our research on:

Full Topic List

Regions & Countries

Publications

• Our Methods
• Short Reads
• Tools & Resources

Read Our Research On:

Biotechnology Research Viewed With Caution Globally, but Most Support Gene Editing for Babies To Treat Disease

Majorities say scientific research on gene editing is a misuse – rather than an appropriate use – of technology. But public acceptance of gene editing for babies depends on how it will be used, and views often differ by age and religion.

How Many Creationists Are There in America?

A new survey shows the number can vary considerably depending how you ask questions about evolution

For Darwin Day, 6 facts about the evolution debate

Tuesday is the 210th anniversary of Charles Darwin’s birth. Roughly eight-in-ten U.S. adults say humans have evolved over time.

How highly religious Americans view evolution depends on how they’re asked about it

Evolution remains a contentious issue. When asked about it, highly religious Americans’ responses can vary depending on how the question is asked.

The Evolution of Pew Research Center’s Survey Questions About the Origins and Development of Life on Earth

Measuring public opinion on evolution has never been an easy task for survey researchers.

Darwin in America

Almost 160 years after Charles Darwin publicized his groundbreaking theory on the development of life, Americans are still arguing about evolution. In spite of the fact that evolutionary theory is accepted by all but a small number of scientists, it continues to be rejected by many Americans.

The Scientific and Ethical Elements of Human Enhancement

Human enhancement may be just around the corner. How do Americans view these emerging technologies that may one day enhance our human capabilities?

U.S. Public Wary of Biomedical Technologies to ‘Enhance’ Human Abilities

Americans are more worried than enthusiastic about using gene editing, brain chip implants and synthetic blood to change human capabilities

Video: Are science and religion in conflict with each other?

A majority of the public says science and religion often conflict, but fewer say science conflicts with their own beliefs. And highly religious Americans are less likely than others to see conflict between faith and science.

Appendix A: About the Survey

The bulk of the analysis in this report stems from a Pew Research Center survey conducted by telephone with a national sample of adults (18 years of age or older) living in all 50 U.S. states and the District of Columbia. The results are based on 2,002 interviews (801 respondents were interviewed on a landline […]

REFINE YOUR SELECTION

• Cary Funk (22)
• Lee Rainie (15)
• David Masci (4)
• Brian Kennedy (2)
• Alec Tyson (1)
• Courtney Johnson (1)
• Elizabeth Podrebarac Sciupac (1)
• Michael Lipka (1)
• Tom Rosentiel (1)

Research Teams

• Internet and Technology (23)
• Religion (23)
• Science (21)
• Methods (1)

901 E St. NW, Suite 300 Washington, DC 20004 USA (+1) 202-419-4300 | Main (+1) 202-857-8562 | Fax (+1) 202-419-4372 |  Media Inquiries

Research Topics

• Email Newsletters

ABOUT PEW RESEARCH CENTER  Pew Research Center is a nonpartisan, nonadvocacy fact tank that informs the public about the issues, attitudes and trends shaping the world. It does not take policy positions. The Center conducts public opinion polling, demographic research, computational social science research and other data-driven research. Pew Research Center is a subsidiary of The Pew Charitable Trusts , its primary funder.

© 2024 Pew Research Center

• Conferences
• Last Updated: September 13, 2024
• In AI Mysteries

Top Machine Learning Research Papers

• by Dr. Nivash Jeevanandam

Advances in machine learning and deep learning research are reshaping our technology. Machine learning and deep learning have accomplished various astounding feats, and key research articles have resulted in technical advances used by billions of people. The research in this sector is advancing at a breakneck pace and assisting you to keep up. Here is a collection of the most important scientific study papers in machine learning.

Rebooting ACGAN: Auxiliary Classifier GANs with Stable Training

The authors of this work examined why ACGAN training becomes unstable as the number of classes in the dataset grows. The researchers revealed that the unstable training occurs due to a gradient explosion problem caused by the unboundedness of the input feature vectors and the classifier’s poor classification capabilities during the early training stage. The researchers presented the Data-to-Data Cross-Entropy loss (D2D-CE) and the Rebooted Auxiliary Classifier Generative Adversarial Network to alleviate the instability and reinforce ACGAN (ReACGAN). Additionally, extensive tests of ReACGAN demonstrate that it is resistant to hyperparameter selection and is compatible with a variety of architectures and differentiable augmentations.

This article is ranked #1 on CIFAR-10 for Conditional Image Generation.

For the research paper, read here .

For code, see here .

Dense Unsupervised Learning for Video Segmentation

The authors presented a straightforward and computationally fast unsupervised strategy for learning dense spacetime representations from unlabeled films in this study. The approach demonstrates rapid convergence of training and a high degree of data efficiency. Furthermore, the researchers obtain VOS accuracy superior to previous results despite employing a fraction of the previously necessary training data. The researchers acknowledge that the research findings may be utilised maliciously, such as for unlawful surveillance, and that they are excited to investigate how this skill might be used to better learn a broader spectrum of invariances by exploiting larger temporal windows in movies with complex (ego-)motion, which is more prone to disocclusions.

This study is ranked #1 on DAVIS 2017 for Unsupervised Video Object Segmentation (val).

Temporally-Consistent Surface Reconstruction using Metrically-Consistent Atlases

The authors offer an atlas-based technique for producing unsupervised temporally consistent surface reconstructions by requiring a point on the canonical shape representation to translate to metrically consistent 3D locations on the reconstructed surfaces. Finally, the researchers envisage a plethora of potential applications for the method. For example, by substituting an image-based loss for the Chamfer distance, one may apply the method to RGB video sequences, which the researchers feel will spur development in video-based 3D reconstruction.

This article is ranked #1 on ANIM in the category of Surface Reconstruction. 

EdgeFlow: Achieving Practical Interactive Segmentation with Edge-Guided Flow

The researchers propose a revolutionary interactive architecture called EdgeFlow that uses user interaction data without resorting to post-processing or iterative optimisation. The suggested technique achieves state-of-the-art performance on common benchmarks due to its coarse-to-fine network design. Additionally, the researchers create an effective interactive segmentation tool that enables the user to improve the segmentation result through flexible options incrementally.

This paper is ranked #1 on Interactive Segmentation on PASCAL VOC

Learning Transferable Visual Models From Natural Language Supervision

The authors of this work examined whether it is possible to transfer the success of task-agnostic web-scale pre-training in natural language processing to another domain. The findings indicate that adopting this formula resulted in the emergence of similar behaviours in the field of computer vision, and the authors examine the social ramifications of this line of research. CLIP models learn to accomplish a range of tasks during pre-training to optimise their training objective. Using natural language prompting, CLIP can then use this task learning to enable zero-shot transfer to many existing datasets. When applied at a large scale, this technique can compete with task-specific supervised models, while there is still much space for improvement.

This research is ranked #1 on Zero-Shot Transfer Image Classification on SUN

CoAtNet: Marrying Convolution and Attention for All Data Sizes

The researchers in this article conduct a thorough examination of the features of convolutions and transformers, resulting in a principled approach for combining them into a new family of models dubbed CoAtNet. Extensive experiments demonstrate that CoAtNet combines the advantages of ConvNets and Transformers, achieving state-of-the-art performance across a range of data sizes and compute budgets. Take note that this article is currently concentrating on ImageNet classification for model construction. However, the researchers believe their approach is relevant to a broader range of applications, such as object detection and semantic segmentation.

This paper is ranked #1 on Image Classification on ImageNet (using extra training data).

SwinIR: Image Restoration Using Swin Transformer

The authors of this article suggest the SwinIR image restoration model, which is based on the Swin Transformer . The model comprises three modules: shallow feature extraction, deep feature extraction, and human-recognition reconstruction. For deep feature extraction, the researchers employ a stack of residual Swin Transformer blocks (RSTB), each formed of Swin Transformer layers, a convolution layer, and a residual connection.

This research article is ranked #1 on Image Super-Resolution on Manga109 – 4x upscaling.

Artificial Replay: A Meta-Algorithm for Harnessing Historical Data in Bandits

Ways to incorporate historical data are still unclear: initialising reward estimates with historical samples can suffer from bogus and imbalanced data coverage, leading to computational and storage issues—particularly in continuous action spaces. The paper addresses the obstacles by proposing ‘Artificial Replay’, an algorithm to incorporate historical data into any arbitrary base bandit algorithm. 

Read the full paper here . 

Bootstrapped Meta-Learning

Author(s) – Sean R. Sinclair et al.

The paper proposes an algorithm in which the meta-learner teaches itself to overcome the meta-optimisation challenge. The algorithm focuses on meta-learning with gradients, which guarantees performance improvements. Furthermore, the paper also looks at how bootstrapping opens up possibilities. 

Read the full paper here .

LaMDA: Language Models for Dialog Applications

Author(s) – Sebastian Flennerhag et al.

The research describes the LaMDA system which caused chaos in AI this summer when a former Google engineer claimed that it had shown signs of sentience. LaMDA is a family of large language models for dialogue applications based on Transformer architecture. The interesting feature of the model is its fine-tuning with human-annotated data and the possibility of consulting external sources. This is a very interesting model family, which we might encounter in many applications we use daily. 

Competition-Level Code Generation with AlphaCode

Author(s) – Yujia Li et al.

Systems can help programmers become more productive. The following research addresses the problems with incorporating innovations in AI into these systems. AlphaCode is a system that creates solutions for problems that require deeper reasoning. 

Privacy for Free: How does Dataset Condensation Help Privacy?

Author(s) – Tian Dong et al.

The paper focuses on Privacy Preserving Machine Learning, specifically deducting the leakage of sensitive data in machine learning. It puts forth one of the first propositions of using dataset condensation techniques to preserve the data efficiency during model training and furnish membership privacy.

Why do tree-based models still outperform deep learning on tabular data?

Author(s) – Léo Grinsztajn, Edouard Oyallon and Gaël Varoquaux

The research answers why deep learning models still find it hard to compete on tabular data compared to tree-based models. It is shown that MLP-like architectures are more sensitive to uninformative features in data compared to their tree-based counterparts. 

Multi-Objective Bayesian Optimisation over High-Dimensional Search Spaces 

Author(s) – Samuel Daulton et al.

The paper proposes ‘MORBO’, a scalable method for multiple-objective BO as it performs better than that of high-dimensional search spaces. MORBO significantly improves the sample efficiency and, where existing BO algorithms fail, MORBO provides improved sample efficiencies over the current approach. 

A Path Towards Autonomous Machine Intelligence Version 0.9.2

Author(s) – Yann LeCun

The research offers a vision about how to progress towards general AI. The study combines several concepts: a configurable predictive world model, behaviour driven through intrinsic motivation, and hierarchical joint embedding architectures trained with self-supervised

learning. 

TranAD: Deep Transformer Networks for Anomaly Detection in Multivariate Time Series Data

Author(s) –  Shreshth Tuli, Giuliano Casale and Nicholas R. Jennings

This is a specialised paper applying transformer architecture to the problem of unsupervised anomaly detection in multivariate time series. Many architectures which were successful in other fields are, at some point, also being applied to time series. The research shows improved performance on some known data sets. 

Differentially Private Bias-Term only Fine-tuning of Foundation Models

Author(s) – Zhiqi Bu et al. 

In the paper, researchers study the problem of differentially private (DP) fine-tuning of large pre-trained models—a recent privacy-preserving approach suitable for solving downstream tasks with sensitive data. Existing work has demonstrated that high accuracy is possible under strong privacy constraints yet requires significant computational overhead or modifications to the network architecture.

ALBERT: A Lite BERT

Usually, increasing model size when pretraining natural language representations often result in improved performance on downstream tasks, but the training times become longer. To address these problems, the authors in their work presented two parameter-reduction techniques to lower memory consumption and increase the training speed of BERT. The authors also used a self-supervised loss that focuses on modelling inter-sentence coherence and consistently helped downstream tasks with multi-sentence inputs. According to results, this model established new state-of-the-art results on the GLUE, RACE, and squad benchmarks while having fewer parameters compared to BERT-large. 

Check the paper here .

Beyond Accuracy: Behavioral Testing of NLP Models with CheckList

Microsoft Research, along with the University of Washington and the University of California, in this paper, introduced a model-agnostic and task agnostic methodology for testing NLP models known as CheckList. This is also the winner of the best paper award at the ACL conference this year. It included a matrix of general linguistic capabilities and test types that facilitate comprehensive test ideation, as well as a software tool to generate a large and diverse number of test cases quickly. 

Linformer is a Transformer architecture for tackling the self-attention bottleneck in Transformers. It reduces self-attention to an O(n) operation in both space- and time complexity. It is a new self-attention mechanism which allows the researchers to compute the contextual mapping in linear time and memory complexity with respect to the sequence length. 

Read more about the paper here .

Plug and Play Language Models

Plug and Play Language Models ( PPLM ) are a combination of pre-trained language models with one or more simple attribute classifiers. This, in turn, assists in text generation without any further training. According to the authors, model samples demonstrated control over sentiment styles, and extensive automated and human-annotated evaluations showed attribute alignment and fluency. 

Reformer 

The researchers at Google, in this paper , introduced Reformer. This work showcased that the architecture of a Transformer can be executed efficiently on long sequences and with small memory. The authors believe that the ability to handle long sequences opens the way for the use of the Reformer on many generative tasks. In addition to generating very long coherent text, the Reformer can bring the power of Transformer models to other domains like time-series forecasting, music, image and video generation. 

An Image is Worth 16X16 Words

The irony here is that one of the popular language models, Transformers have been made to do computer vision tasks. In this paper , the authors claimed that the vision transformer could go toe-to-toe with the state-of-the-art models on image recognition benchmarks, reaching accuracies as high as 88.36% on ImageNet and 94.55% on CIFAR-100. For this, the vision transformer receives input as a one-dimensional sequence of token embeddings. The image is then reshaped into a sequence of flattened 2D patches. The transformers in this work use constant widths through all of its layers.

Unsupervised Learning of Probably Symmetric Deformable 3D Objects

Winner of the CVPR best paper award, in this work, the authors proposed a method to learn 3D deformable object categories from raw single-view images, without external supervision. This method uses an autoencoder that factored each input image into depth, albedo, viewpoint and illumination. The authors showcased that reasoning about illumination can be used to exploit the underlying object symmetry even if the appearance is not symmetric due to shading.

Generative Pretraining from Pixels

In this paper, OpenAI researchers examined whether similar models can learn useful representations for images. For this, the researchers trained a sequence Transformer to auto-regressively predict pixels, without incorporating knowledge of the 2D input structure. Despite training on low-resolution ImageNet without labels, the researchers found that a GPT-2 scale model learns strong image representations as measured by linear probing, fine-tuning, and low-data classification. On CIFAR-10, it achieved 96.3% accuracy with a linear probe, outperforming a supervised Wide ResNet, and 99.0% accuracy with full fine-tuning and matching the top supervised pre-trained models. An even larger model, trained on a mixture of ImageNet and web images, is competitive with self-supervised benchmarks on ImageNet, achieving 72.0% top-1 accuracy on a linear probe of their features.

Deep Reinforcement Learning and its Neuroscientific Implications

In this paper, the authors provided a high-level introduction to deep RL , discussed some of its initial applications to neuroscience, and surveyed its wider implications for research on brain and behaviour and concluded with a list of opportunities for next-stage research. Although DeepRL seems to be promising, the authors wrote that it is still a work in progress and its implications in neuroscience should be looked at as a great opportunity. For instance, deep RL provides an agent-based framework for studying the way that reward shapes representation, and how representation, in turn, shapes learning and decision making — two issues which together span a large swath of what is most central to neuroscience. 

Dopamine-based Reinforcement Learning

Why humans doing certain things are often linked to dopamine , a hormone that acts as the reward system (think: the likes on your Instagram page). So, keeping this fact in hindsight, DeepMind with the help of Harvard labs, analysed dopamine cells in mice and recorded how the mice received rewards while they learned a task. They then checked these recordings for consistency in the activity of the dopamine neurons with standard temporal difference algorithms. This paper proposed an account of dopamine-based reinforcement learning inspired by recent artificial intelligence research on distributional reinforcement learning. The authors hypothesised that the brain represents possible future rewards not as a single mean but as a probability distribution, effectively representing multiple future outcomes simultaneously and in parallel. 

Lottery Tickets In Reinforcement Learning & NLP

In this paper, the authors bridged natural language processing (NLP) and reinforcement learning (RL). They examined both recurrent LSTM models and large-scale Transformer models for NLP and discrete-action space tasks for RL. The results suggested that the lottery ticket hypothesis is not restricted to supervised learning of natural images, but rather represents a broader phenomenon in deep neural networks.

What Can Learned Intrinsic Rewards Capture

In this paper, the authors explored if the reward function itself can be a good locus of learned knowledge. They proposed a scalable framework for learning useful intrinsic reward functions across multiple lifetimes of experience and showed that it is feasible to learn and capture knowledge about long-term exploration and exploitation into a reward function. 

AutoML- Zero

The progress of AutoML has largely focused on the architecture of neural networks, where it has relied on sophisticated expert-designed layers as building blocks, or similarly restrictive search spaces. In this paper , the authors showed that AutoML could go further with AutoML Zero, that automatically discovers complete machine learning algorithms just using basic mathematical operations as building blocks. The researchers demonstrated this by introducing a novel framework that significantly reduced human bias through a generic search space.

Rethinking Batch Normalization for Meta-Learning

Batch normalization is an essential component of meta-learning pipelines. However, there are several challenges. So, in this paper, the authors evaluated a range of approaches to batch normalization for meta-learning scenarios and developed a novel approach — TaskNorm. Experiments demonstrated that the choice of batch normalization has a dramatic effect on both classification accuracy and training time for both gradient-based and gradient-free meta-learning approaches. The TaskNorm has been found to be consistently improving the performance.

Meta-Learning without Memorisation

Meta-learning algorithms need meta-training tasks to be mutually exclusive, such that no single model can solve all of the tasks at once. In this paper, the authors designed a meta-regularisation objective using information theory that successfully uses data from non-mutually-exclusive tasks to efficiently adapt to novel tasks.

Understanding the Effectiveness of MAML

Model Agnostic Meta-Learning (MAML) consists of optimisation loops, from which the inner loop can efficiently learn new tasks. In this paper, the authors demonstrated that feature reuse is the dominant factor and led to ANIL (Almost No Inner Loop) algorithm — a simplification of MAML where the inner loop is removed for all but the (task-specific) head of the underlying neural network. 

Your Classifier is Secretly an Energy-Based Model

This paper proposed attempts to reinterpret a standard discriminative classifier as an energy-based model. In this setting, wrote the authors, the standard class probabilities can be easily computed. They demonstrated that energy-based training of the joint distribution improves calibration, robustness, handout-of-distribution detection while also enabling the proposed model to generate samples rivalling the quality of recent GAN approaches. This work improves upon the recently proposed techniques for scaling up the training of energy-based models. It has also been the first to achieve performance rivalling the state-of-the-art in both generative and discriminative learning within one hybrid model.

Reverse-Engineering Deep ReLU Networks

This paper investigated the commonly assumed notion that neural networks cannot be recovered from its outputs, as they depend on its parameters in a highly nonlinear way. The authors claimed that by observing only its output, one could identify the architecture, weights, and biases of an unknown deep ReLU network. By dissecting the set of region boundaries into components associated with particular neurons, the researchers showed that it is possible to recover the weights of neurons and their arrangement within the network.

Cricket Analytics and Predictor

Authors: Suyash Mahajan,  Salma Shaikh, Jash Vora, Gunjan Kandhari,  Rutuja Pawar,

Abstract:   The paper embark on predicting the outcomes of Indian Premier League (IPL) cricket match using a supervised learning approach from a team composition perspective. The study suggests that the relative team strength between the competing teams forms a distinctive feature for predicting the winner. Modeling the team strength boils down to modeling individual player‘s batting and bowling performances, forming the basis of our approach.

Research Methodology: In this paper, two methodologies have been used. MySQL database is used for storing data whereas Java for the GUI. The algorithm used is Clustering Algorithm for prediction. The steps followed are as

• Begin with a decision on the value of k being the number of clusters.
• Put any initial partition that classifies the data into k clusters.
• Take every sample in the sequence; compute its distance from centroid of each of the clusters. If sample is not in the cluster with the closest centroid currently, switch this sample to that cluster and update the centroid of the cluster accepting the new sample and the cluster losing the sample.

For the research paper, read here

2.Real Time Sleep / Drowsiness Detection – Project Report

Author : Roshan Tavhare

Institute : University of Mumbai

Abstract : The main idea behind this project is to develop a nonintrusive system which can detect fatigue of any human and can issue a timely warning. Drivers who do not take regular breaks when driving long distances run a high risk of becoming drowsy a state which they often fail to recognize early enough.

Research Methodology : A training set of labeled facial landmarks on an image. These images are manually labeled, specifying specific (x, y) -coordinates of regions surrounding each facial structure.

• Priors, more specifically, the probability on distance between pairs of input pixels. The pre-trained facial landmark detector inside the dlib library is used to estimate the location of 68 (x, y)-coordinates that map to facial structures on the face.

A Study of Various Text Augmentation Techniques for Relation Classification in Free Text

Authors: Chinmaya Mishra Praveen Kumar and Reddy Kumar Moda,  Syed Saqib Bukhari and Andreas Dengel

Institute: German Research Center for Artificial Intelligence (DFKI), Kaiserslautern, Germany

Abstract: In this paper, the researchers explore various text data augmentation techniques in text space and word embedding space. They studied the effect of various augmented datasets on the efficiency of different deep learning models for relation classification in text.

Research Methodology: The researchers implemented five text data augmentation techniques (Similar word, synonyms, interpolation, extrapolation and random noise method)  and explored the ways in which we could preserve the grammatical and the contextual structures of the sentences while generating new sentences automatically using data augmentation techniques.

Smart Health Monitoring and Management Using Internet of Things, Artificial Intelligence with Cloud Based Processing

Author : Prateek Kaushik

Institute : G D Goenka University, Gurugram

Abstract : This research paper described a personalised smart health monitoring device using wireless sensors and the latest technology.

Research Methodology: Machine learning and Deep Learning techniques are discussed which works as a catalyst to improve the  performance of any health monitor system such supervised machine learning algorithms, unsupervised machine learning algorithms, auto-encoder, convolutional neural network and restricted boltzmann machine .

Internet of Things with BIG DATA Analytics -A Survey

Author : A.Pavithra,  C.Anandhakumar and V.Nithin Meenashisundharam

Institute : Sree Saraswathi Thyagaraja College,

Abstract : This article we discuss about Big data on IoT and how it is interrelated to each other along with the necessity of implementing Big data with IoT and its benefits, job market

Research Methodology : Machine learning, Deep Learning, and Artificial Intelligence are key technologies that are used to provide value-added applications along with IoT and big data in addition to being used in a stand-alone mod.

Single Headed Attention RNN: Stop Thinking With Your Head 

Author: Stephen Merity

In this work of art, the Harvard grad author, Stephen “Smerity” Merity, investigated the current state of NLP, the models being used and other alternate approaches. In this process, he tears down the conventional methods from top to bottom, including etymology.

The author also voices the need for a Moore’s Law for machine learning that encourages a minicomputer future while also announcing his plans on rebuilding the codebase from the ground up both as an educational tool for others and as a strong platform for future work in academia and industry.

EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks

Authors: Mingxing Tan and Quoc V. Le 

In this work, the authors propose a compound scaling method that tells when to increase or decrease depth, height and resolution of a certain network.

Convolutional Neural Networks(CNNs) are at the heart of many machine vision applications. 

EfficientNets are believed to superpass state-of-the-art accuracy with up to 10x better efficiency (smaller and faster).

Deep Double Descent By OpenAI

Authors: Mikhail Belkin, Daniel Hsu, Siyuan Ma, Soumik Mandal

In this paper , an attempt has been made to reconcile classical understanding and modern practice within a unified performance curve. 

The “double descent” curve overtakes the classic U-shaped bias-variance trade-off curve by showing how increasing model capacity beyond the point of interpolation results in improved performance. 

The Lottery Ticket Hypothesis

Authors: Jonathan Frankle, Michael Carbin

Neural network pruning techniques can reduce the parameter counts of trained networks by over 90%, decreasing storage requirements and improving computational performance of inference without compromising accuracy. 

The authors find that a standard pruning technique naturally uncovers subnetworks whose initializations made them capable of training effectively. Based on these results, they introduce the “lottery ticket hypothesis:”

On The Measure Of Intelligence 

Authors: Francois Chollet

This work summarizes and critically assesses the definitions of intelligence and evaluation approaches, while making apparent the historical conceptions of intelligence that have implicitly guided them.

The author, also the creator of keras, introduces a formal definition of intelligence based on Algorithmic Information Theory and using this definition, he also proposes a set of guidelines for what a general AI benchmark should look like. 

Zero-Shot Word Sense Disambiguation Using Sense Definition Embeddings via IISc Bangalore & CMU

Authors: Sawan Kumar, Sharmistha Jat, Karan Saxena and Partha Talukdar

Word Sense Disambiguation (WSD) is a longstanding  but open problem in Natural Language Processing (NLP).  Current supervised WSD methods treat senses as discrete labels  and also resort to predicting the Most-Frequent-Sense (MFS) for words unseen  during training.

The researchers from IISc Bangalore in collaboration with Carnegie Mellon University propose  Extended WSD Incorporating Sense Embeddings (EWISE), a supervised model to perform WSD  by predicting over a continuous sense embedding space as opposed to a discrete label space.

Deep Equilibrium Models 

Authors: Shaojie Bai, J. Zico Kolter and Vladlen Koltun 

Motivated by the observation that the hidden layers of many existing deep sequence models converge towards some fixed point, the researchers at Carnegie Mellon University present a new approach to modeling sequential data through deep equilibrium model (DEQ) models. 

Using this approach, training and prediction in these networks require only constant memory, regardless of the effective “depth” of the network.

IMAGENET-Trained CNNs are Biased Towards Texture

Authors: Robert G, Patricia R, Claudio M, Matthias Bethge, Felix A. W and Wieland B

Convolutional Neural Networks (CNNs) are commonly thought to recognise objects by learning increasingly complex representations of object shapes. The authors in this paper , evaluate CNNs and human observers on images with a texture-shape cue conflict. They show that ImageNet-trained CNNs are strongly biased towards recognising textures rather than shapes, which is in stark contrast to human behavioural evidence.

A Geometric Perspective on Optimal Representations for Reinforcement Learning 

Authors: Marc G. B , Will D , Robert D , Adrien A T , Pablo S C , Nicolas Le R , Dale S, Tor L, Clare L

The authors propose a new perspective on representation learning in reinforcement learning

based on geometric properties of the space of value functions. This work shows that adversarial value functions exhibit interesting structure, and are good auxiliary tasks when learning a representation of an environment. The authors believe this work to open up the possibility of automatically generating auxiliary tasks in deep reinforcement learning.

Weight Agnostic Neural Networks 

Authors: Adam Gaier & David Ha

In this work , the authors explore whether neural network architectures alone, without learning any weight parameters, can encode solutions for a given task. In this paper, they propose a search method for neural network architectures that can already perform a task without any explicit weight training. 

Stand-Alone Self-Attention in Vision Models 

Authors: Prajit Ramachandran, Niki P, Ashish Vaswani,Irwan Bello Anselm Levskaya, Jonathon S

In this work, the Google researchers verified that content-based interactions can serve the vision models . The proposed stand-alone local self-attention layer achieves competitive predictive performance on ImageNet classification and COCO object detection tasks while requiring fewer parameters and floating-point operations than the corresponding convolution baselines. Results show that attention is especially effective in the later parts of the network. 

High-Fidelity Image Generation With Fewer Labels 

Authors: Mario Lucic, Michael Tschannen, Marvin Ritter, Xiaohua Z, Olivier B and Sylvain Gelly 

Modern-day models can produce high quality, close to reality when fed with a vast quantity of labelled data. To solve this large data dependency, researchers from Google released this work , to demonstrate how one can benefit from recent work on self- and semi-supervised learning to outperform the state of the art on both unsupervised ImageNet synthesis, as well as in the conditional setting.

The proposed approach is able to match the sample quality of the current state-of-the-art conditional model BigGAN on ImageNet using only 10% of the labels and outperform it using 20% of the labels.

ALBERT: A Lite BERT for Self-Supervised Learning of Language Representations

Authors: Zhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin G, Piyush Sharma and Radu S

The authors present two parameter-reduction techniques to lower memory consumption and increase the training speed of BERT and to address the challenges posed by increasing model size and GPU/TPU memory limitations, longer training times, and unexpected model degradation

As a result, this proposed model establishes new state-of-the-art results on the GLUE, RACE, and SQuAD benchmarks while having fewer parameters compared to BERT-large.

GauGANs-Semantic Image Synthesis with Spatially-Adaptive Normalization 

Author: Taesung Park, Ming-Yu Liu, Ting-Chun Wang and Jun-Yan Zhu

Nvidia in collaboration with UC Berkeley and MIT proposed a model which has a spatially-adaptive normalization layer for synthesizing photorealistic images given an input semantic layout.

This model retained visual fidelity and alignment with challenging input layouts while allowing the user to control both semantic and style.

📣 Want to advertise in AIM? Book here

Subscribe to The Belamy: Our Weekly Newsletter

Biggest ai stories, delivered to your inbox every week..

Discover how Cypher 2024 expands to the USA, bridging AI innovation gaps and tackling the challenges of enterprise AI adoption

© Analytics India Magazine Pvt Ltd & AIM Media House LLC 2024

• Terms of use
• Privacy Policy

Subscribe to Our Newsletter

The Belamy, our weekly Newsletter is a rage. Just enter your email below.

Subscribe to Our Youtube channel