Erratum: Fish can infer social rank by observation alone

Logan Grosenick, Tricia S. Clement, Russell D. FernaldNature 2007 · DOI · OpenAlex

Abstract

This paper reviews a collection of formal amendments published in Nature (March 2007), focusing on the correction of statistical significance in fish social intelligence research, undisclosed financial interests in medical studies, and technical errors in paleontology and physics. These corrections underscore the importance of post-publication scrutiny in maintaining scientific rigor and the integrity of the global research record.

Introduction

The scientific record is a living document, subject to refinement and correction even after the rigorous process of peer review. In March 2007, the journal Nature published a series of formal amendments—comprising corrigenda, errata, and an addendum—to rectify errors in previously published research. These corrections span diverse fields, including behavioral ecology, immunology, paleontology, and condensed-matter physics, illustrating the multifaceted nature of post-publication oversight.

Statistical Re-evaluation in Behavioral Ecology

A primary focus of these amendments is the study titled "Fish can infer social rank by observation alone." In the original report, the authors cited a statistical significance level of $p < 0.01$ for their "first-choice" results regarding social rank inference. Upon further review, this value was officially adjusted to $p < 0.05$.

While the findings remain statistically significant by the standard threshold of 0.05, this adjustment is critical for the scientific community. It represents a shift in the confidence interval from 99% to 95%, a nuance that is vital for researchers attempting to replicate the study or meta-analyze cognitive abilities in non-human species. The authors noted that while the print version required this correction, the online PDF and HTML versions had already been updated to reflect the accurate value.

Ethical Disclosures and Transparency

Two separate corrigenda addressed failures to disclose competing financial interests, highlighting the evolving standards of transparency in research. In papers concerning malaria vaccines and the role of the GREPTIN protein in osteoarthritis, the authors had not initially declared pending patent applications.

The journal clarified that patent applications—even those held by non-profit or academic institutions—constitute a competing financial interest that must be formally declared. These amendments reinforce the necessity of absolute transparency to maintain public and professional trust in biomedical research and to ensure that potential biases are visible to the reader.

Technical Accuracy and Nomenclature

Technical errors, though often typographical, can have significant implications for data interpretation. In a study on chronic polyarthritis, the units on a figure axis (Figure 3d) were corrected from micrograms ($\mu$g) to picograms (pg). This represents a million-fold difference in scale, which is essential for accurate quantitative comparison in immunological studies.

Furthermore, the correction of the Mesozoic gliding mammal's name from Volaticotherium antiquus to Volaticotherium antiquum serves as a reminder of the ongoing relevance of classical linguistic rules in biological nomenclature. Following the International Code of Zoological Nomenclature, the species name suffix must agree with the gender of the generic name (in this case, neuter). Such corrections ensure that taxonomic records remain consistent and linguistically accurate.

Experimental Clarifications in Physics

Finally, an addendum to a paper on "artificial spin ice" provided necessary procedural details that were missing from the original publication. The authors provided a detailed step-by-step description of the magnetic field treatment, explaining how the field's polarity was switched in conjunction with magnitude adjustments while rotating the sample. These details are essential for the reproducibility of complex physics experiments and the further development of magnetic materials research.

Conclusion

The publication of these amendments serves as a testament to the self-correcting nature of science. Whether addressing statistical confidence, financial transparency, or typographical errors in units and nomenclature, the formal process of correction ensures that the scientific record remains a reliable foundation for future discovery. These notices finalize the record of past work, ensuring that subsequent researchers build upon accurate and fully disclosed data.

Summary updated

BibTeX
@inproceedings{grosenick2007erratum,
  author = {Logan Grosenick and Tricia S. Clement and Russell D. Fernald},
  title = {{Erratum: Fish can infer social rank by observation alone}},
  booktitle = {Nature},
  year = {2007},
  doi = {10.1038/nature05646},
}

Loading related content...

Figure