Ivan: Co-founder of Retraction Watch and Executive Director of the Center for Scientific Integrity. He is a distinguished journalist in residence at NYU and the Editor-in-Chief of the TRANS Center.
Stephanie Kinan: Vice President of the International Society of Managing and Technical Editors (ISMTE), Committee on Publication Ethics (COPE) Council member, and Assistant Director of Publications at the American Society for Gastrointestinal Endoscopy (ASGE).
Event Sponsors:
Gold Sponsor:River Valley Technologies.
Bronze Sponsors:Scholastica and Silverchair.
Historical Evolution of Research Integrity and the Science Sleuth
Definition of a Science Sleuth: Individuals who identify issues in scientific literature and take action to rectify them. Their activities often result in correcting the scientific record through corrections, retractions, or letters to the editor.
Historical Timeline of Integrity Landmarks:
2004: Mike Rossner of Rockefeller University Press "sounded the alarm" regarding image manipulation, specifically focusing on western blot figures. This represents the beginning of public discourse on image-based misconduct.
2010:Retraction Watch was launched by Adam Marcus and Ivan to provide transparency regarding the retraction process.
2011: Yana Christopher was hired by Embo Press as the first-ever Image Integrity Analyst, identifying fifteen years ago that journals required in-house integrity specialists.
2012: The American Society for Biochemistry and Molecular Biology (ASBMB) hired a Manager of Publication Ethics to address image manipulation and respond to sleuth comments.
2012:PubPeer was created as a public platform for researchers and sleuths to leave comments on papers, facilitating community-driven peer review and error identification.
2016: Elizabeth Bick published a sentinel study (initially as a preprint) examining 20,000 papers. She found that 2% of these papers contained images with features suggestive of deliberate manipulation.
2021: The Problematic Paper Screener was established by computer scientists to detect papers produced by software like SciGen and MathGen, as well as identifying "tortured phrases."
Quantitative Trends and Statistical Rise in Retractions
Database Statistics: The Retraction Watch database, now owned by Crossref but still operated/maintained by the Center for Scientific Integrity, currently contains more than 65,000 retractions.
Frequency Growth:
Twenty to twenty-five years ago, the retraction rate was approximately 50001 papers.
Currently, the rate has increased to approximately 5001 papers.
Case Study: The 2023 Spike:
There was a massive increase in retractions in 2023, largely attributed to Hindawi.
Wiley acquired Hindawi in 2021 and discovered massive integrity failures in late 2022.
This led to approximately 11,600 retractions over an eighteen-month period.
Primary Drivers: While misconduct may be increasing, the dominant driver of the rise in retractions is the increased scrutiny of literature by sleuths and specialized tools.
The Impact of Misconduct, AI, and Image Manipulation
Reasons for Retraction: Approximately two-thirds of retractions are due to misconduct. This includes:
Fabricating or faking data.
Plagiarism.
Fudging data (subtle manipulation).
Artificial Intelligence (AI) and Large Language Models (LLMs):
LLMs are currently impacting the literature through "hallucinated references" (fake citations) and "AI swaps" where systems are overwhelmed by generated content.
Tortured Phrases: These occur when authors use AI to evade plagiarism software. Examples include:
"Breast cancer" becoming "bosom trouble."
"Renal failure" becoming "kidney disappointment."
Efficiency Issues: Sleuths frequently describe the current process for correcting the record as "slow, opaque, and inconsistent," with investigations often taking months or years.
COPE: Standardizing Publication Ethics and Global Guidance
Organizational Overview: The Committee on Publication Ethics (COPE) has existed for about 30 years. It is an international community with approximately 15,000 members across nearly 100 countries.
Mission: To provide guidance and standards for ethical issues in scholarly publishing.
Core Resources for Journal Offices:
Flow charts: Step-by-step visual guides for handling specific ethical breaches.
E-learning modules and webinars.
Guidelines and discussion documents.
Key Tenets: COPE emphasizes transparency, consistency, and the responsibility of journals to investigate all credible concerns brought to their attention.
Practical Implementation of Ethical Investigations in Journal Offices
Internal Infrastructure:
Journals should create clear guidelines based on their specific workflow for responding to ethical breaches.
A designated contact person (e.g., Managing Editor, Editor-in-Chief, or Research Integrity Officer) should be appointed to maintain consistency.
Policies must be established for internal misconduct (reviewers, staff, editors) and for when to outsource investigations to institutions or legal counsel.
Prioritization: Offices should prioritize investigations for recent papers, high-impact articles, or articles receiving frequent reports to stop the spread of unreliable information.
Protocol for Managing Complainants and Confidentiality
Professional Conduct:
Acknowledge receipt of complaints promptly.
Maintain a neutral, factual tone; avoid personal exchanges or back-and-forth arguments.
If behavior becomes abusive or threatening, cease communication and escalate to authorities.
Confidentiality: Protect the identity of complainants unless explicit permission is granted to identify them. Retraction or correction notices should remain anonymous regarding the whistleblower unless otherwise agreed upon.
Closure: Notify the complainant of the final outcome, but do not engage in debate if they disagree with the decision.
Guidelines for Issuing Retractions, Corrections, and Expressions of Concern
Retractions:
Purpose: Solely to correct the record and identify unreliable work; it is not a punishment for the author.
Appropriate use: When findings are unreliable due to misconduct (fabrication, plagiarism) or honest error (methodological failure).
Inappropriate use: Minor errors that don't affect science, authorship disputes that don't affect reliability, or failures to replicate findings.
Expressions of Concern (EOC):
Issued when an investigation is ongoing but inconclusive, or when there is insufficient evidence for an immediate retraction but the reader should be warned of potential issues.
Corrections/Errata: Used for minor errors that do not compromise the overall scientific integrity or conclusions of the paper.
Questions and Discussion
Resource Allocation: Ivan argued that publishers must staff and resource integrity teams as seriously as they do pre-publication peer review. This prevents cases from "dragging on" or being missed due to high volume.
Metadata and Discoverability:
Retractions and EOCs must have their own Digital Object Identifiers (DOIs) to be discoverable in the metadata.
The National Information Standards Organization (NISO) has a working group on corrections and retractions meant to standardize these metadata tags.
HTML-only notes or "publisher notes" without DOIs are ineffective because they are not tracked by external systems.
Apathetic Editors and AI: If an Editor-in-Chief refuses to act on an ethical breach (like an AI-written letter), Stephanie suggests pointing to COPE standards or contractual obligations that require adherence to ethical protocols as a means of leverage.
Legal Threats and the Chilling Effect:
Legal threats (e.g., suing for libel) are common and create a "chilling effect" that prevents journals from retracting papers.
Case Example: The American Diabetes Association (ADA) successfully fought a federal court case to issue an expression of concern, citing the First Amendment.
Advice: Use legal teams and escalate threats to publishers immediately; do not attempt to handle legal intimidation at the editorial assistant or managing editor level.