henhao
Sign in
Scientific Communication & EducationbioRxiv
Preprint

It runs in the family: New extinct species from charismatic clades are more likely to be published in high-impact journals than those from non-charismatic clades

Bryan Michael GeeiD

PDFDOICode
In-depth full-text digest available · Jump to digest ↓

Abstract

The cultural landscape of paleontology is dominated by dinosaurs and a few other charismatic clades that are more likely to be depicted in media and to be recognized by the public. There is a prevailing belief among researchers that these biases extend to scholarly publishing - that certain clades enjoy a 'taxonomic advantage' that renders them more likely to be published in high-impact journals like Nature and Science. Given the importance of publishing in high-impact journals for career advancement, quantitative testing of this hypothesis can inform on potential biases in this area. I examined articles describing new species that were published between 2000 and mid-2026 in one of thirteen journals (n = 4,248 articles, 8,331 new species), ranging from high-profile, multidisciplinary titles like Nature to lower-profile, specialist titles like the Journal of Paleontology. Comparisons of the taxonomic composition of new species between journals revealed a positive correlation between a journal's prestige (approximated by its impact factor) and the relative proportion of new species that belong to charismatic clades. Avemetatarsalians and mammaliamorphs represent more than 50% of new species in Nature and Science but less than 15% of new species in the control group of paleontological journals. These results provide support for the perception that there is a bias between extinct taxa as it relates to publication in high-impact journals. Given the potential for researchers to accrue cumulative advantage through publishing in prestigious journals, bias in favor of charismatic clades creates the potential for opportunities and knowledge generation to become increasingly skewed and consolidated towards these clades, at the expense of less charismatic clades and the people studying them. This threatens the discipline's ability to continue to construct a holistic understanding of the full history of life that can be disseminated to the public. Further attention should be directed towards examining potential biases in paleontological publications as the primary currency in academia.

02 /Specification & Classification

Click any tag to browse matching papers
Citations
1
OpenAlex Topics

In-depth summary

✦ Multi-agent verified full-text synthesis

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (13 Oct 2026) · CC BY 4.0

Among 8,331 new extinct species named in thirteen journals since 2000, avemetatarsalians and mammaliamorphs exceed 50% of species in Nature and Science but are under 15% in paleontology journals.

Papers in high-impact journals such as Nature and Science weigh heavily in hiring and funding, and paleontologists widely assume that dinosaurs and other charismatic clades reach these venues more easily, an idea the author says had never been tested quantitatively. Earlier bibliometric work in paleontology mostly asked who does the research; this study asks which clades get published where, using papers that name new species.

Key results

  1. ShownNew species from charismatic clades are heavily over-represented in Nature and Science compared with the paleontology control journals. Avemetatarsalians plus mammaliamorphs make up more than half of the species named in each of the two journals.

    Avemetatarsalians (avians included): 23.9% of new species in Nature, 15.7% in Science, under 5% in the control journals. Mammaliamorphs (primates included): 31.7% in Nature, 40% in Science, under 11% in the controls. For Nature and Science pooled: avemetatarsalians 21.8%, mammaliamorphs 33.9%. · There are clear disparities in taxonomic…; High-JIF journals: Nature and Science; Abstract

  2. SuggestedAcross the thirteen journals, the share of new species that are avemetatarsalians or mammaliamorphs rises with journal impact factor (JIF). How well the line fits depends strongly on the vertebrate-only Journal of Vertebrate Paleontology.

    Linear regression of log10 JIF against the percent of new species in Avemetatarsalia or Mammaliamorpha: r^2 = 0.32 with all journals and 0.65 without the Journal of Vertebrate Paleontology (JIF 1.9). Journal JIFs run from 1.3 (Journal of Paleontology) to 56.1 (Nature). · There are clear disparities in taxonomic…; Journal scope of examined publications; Figure legends

  3. ShownVertebrates gain their extra share in high-JIF journals mostly at the expense of invertebrates. Science named no new molluscs or lophophorates at all in the sampled period.

    Control journals vs Nature/Science: lophophorates 9.1% vs 0.7%, molluscs 14.8% vs 2.5%, non-chordate deuterostomes 6.1% vs 1.8%, non-insect pan-arthropods 15.7% vs 5.1%, 'Other' 9.2% vs 3.3%. Lophophorates and molluscs are 8.5% and 13.8% of the full sample, yet Science published none since 2000. · Control journals; There are clear disparities in taxonomic…

  4. ShownInsects make up a much larger share of new species in medium-JIF journals than in Nature and Science. This comes mostly from Cretaceous Myanmar amber species in Current Biology, Gondwana Research and Nature Communications.

    Insects are 23.8% of new species in the medium-JIF group vs 5.8% in Nature/Science. Each of the three journals has over 28% insects, and Cretaceous Myanmar insects are the most common combination in all three. Insects are 80% of new species from Myanmar; Myanmar species peaked in 2018 and have recently fallen to around 25% of that peak or less. · High-JIF journals: Nature and Science; Clade-specific patterns

  5. ShownAbout 1 in 9 genera with a new species in these journals is missing from the Paleobiology Database (PBDB). Non-vertebrates and 'Other' have higher missing rates than vertebrates; avians, non-avian avemetatarsalians and primates have very low rates, though non-primate mammaliamorphs have a fairly high one.

    After manual checks, 965 species (11.58%) in 823 genera (12.63%) were truly missing. Of these, 897 (92.95%) were published before 2026 and 416 before 2016. Avians, non-avian avemetatarsalians and primates had very low missing rates. Control journals account for over 88% of articles with at least one missing genus. · Retrieval of taxonomic rank information; Taxonomic biases are also present in databases

Methods, data and reproducibility

Ran the same Boolean full-text queries for new-species designators (e.g., 'sp. nov.', 'n. sp.') in each journal's own search interface for 2000 to 2026, and recorded each named species by hand with its age and provenance. Higher taxonomy came from the Paleobiology Database API, grouped into 20 clade categories. Category shares were compared across high-, medium- and low-JIF journal groups and against log10 JIF.

  • Research articles naming new extinct species in Nature, Science, six medium-JIF journals and five paleontology control journals, 2000 to mid-2026 · multiple species (fossil taxa; 99.35% classified under Eukaryota) · n = 4,248 articles; 8,331 species; 6,515 genera (6,027 articles examined)
  • Paleobiology Database (PBDB) higher-rank taxonomy for each genus, retrieved via API on September 8, 2026, with missing genera filled in by hand
  • OpenAlex author names and ORCIDs for Nature and Science naming papers (repeat-authorship analysis) · n = 863 of 1,083 unique author names with linked ORCIDs
  • Crossref citation metadata for all naming papers (used to generate the reference list)
  • 2025 Journal Impact Factors for the thirteen sampled journals
Methods & tools
Paleobiology Database (PBDB) API, OpenAlex API, pyalex, Crossref API, Google Sheets, Python, pandas, numpy, scipy, matplotlib, requests, pypandoc-binary
Code
https://github.com/bryanmgee/new-taxa-in-fancy-journals
Software & environment
Python scripts using csv, datetime, glob, json, matplotlib, numpy, os, pandas, pyalex, pypandoc-binary, re, requests, rich, scipy and time; no package versions are given.
Reproducibility gaps
The dataset, the scripts for this version and the supplemental appendices (including the search queries) are said to be on Zenodo, but no Zenodo DOI or URL appears in the text. The publisher search interfaces are closed-source and may return different results later.
Limitations (6)
  • The study shows a correlation, not a cause. It cannot separate editorial bias (including desk rejection) from specialists of less charismatic clades choosing not to submit, because journals do not release data on rejected submissions. · stated by the authors
  • Only the taxonomy of the study system is examined. Other influences on acceptance, such as methods, exceptional preservation (e.g., the Jehol and Chengjiang biotas) and author reputation, are not modelled. Around 10% of Nature/Science naming-paper authors with ORCIDs appear on two or more such papers. · stated by the authors
  • Charisma is not measured. It is defined from general observations of extant taxa: birds, dinosaurs and mammals above other vertebrates, and vertebrates above non-vertebrates. The 20 categories are not all monophyletic or of equal rank. · stated by the authors
  • Small samples in some journals (Science Advances in particular) and papers that name several species at once can skew the proportions. Change over time across the quarter-century window is not analysed because of confounds. · stated by the authors
  • Results depend on closed-source publisher search engines that can change silently, and on PBDB taxonomy that is incomplete and was not systematically checked for later reclassification. The control group contains only English-language, hybrid open-access journals from western societies. · stated by the authors
  • The JIF correlation rests on thirteen journal-level points and reports only r^2, with no significance test or uncertainty. Dropping one journal moves it from 0.32 to 0.65. · noted in this summary

Who should read:Read this if you study publication or taxonomic bias in paleontology or research assessment, or want a journal-level dataset of new extinct species with clade, age, provenance and PBDB coverage.

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISFischer '23Ottaviani '23Bielby '25Jaric '18Stoudt '21Nojiri '26
62% semantic matchEcology
S. E. Fischer, A. M. Lindsay, J. G. Otten+2 more·bioRxiv·Oct 26, 2023