Introduction
In the pursuit of a scientific understanding of human vision, precision is paramount—not only in experimental design and computational modeling but also in the formal documentation of scientific contributions. This article serves as a formal corrigendum to the 2023 publication, "Let’s move forward: Image-computable models and a common model evaluation scheme are prerequisites for a scientific understanding of human vision" (DiCarlo et al., 2023).
The Nature of the Correction
The primary purpose of this corrigendum is to rectify a typographical error in the acknowledgments section of the original paper. Specifically, the name of collaborator Michael J. Lee was incorrectly rendered as "Micheal Lee."
While such an error does not alter the scientific findings, data analysis, or the core arguments presented regarding image-computable models, its correction is vital for the integrity of the academic record. Proper attribution ensures that all individuals who contributed to the intellectual framework of the work are correctly recognized and searchable within citation databases.
Context: The "Image-Computable" Paradigm
The original article by DiCarlo et al. (2023) posits a significant shift in vision science. It argues that the field must move beyond conceptual descriptions and adopt "image-computable models." These are functional computational architectures, such as Deep Neural Networks (DNNs), that can process any arbitrary image and generate predictions comparable to biological vision.
Furthermore, the authors advocate for a "common model evaluation scheme." This involves standardized benchmarks (like Brain-Score) to evaluate how closely artificial models mimic human neural and behavioral responses. The corrigendum, while administrative, underscores the rigorous standards of accuracy that the authors believe should define the field.
Conclusion
The authors have expressed regret for the oversight. By issuing this correction, the academic record for "Let’s move forward" is now accurate, ensuring that Michael J. Lee is properly credited for his contributions to this influential work in computational neuroscience.
Summary updated