Link to recording of research discussion 2025-01-15 @ FRIB room 1200
TL/DR: Goodstein’s Framework for Scientific Fraud: “In my experience three factors are nearly always present whenever fraud occurs in science….
- The scientist involved is under career pressure,
- [they] think they know how the experiment would come out if it were performed properly, and
- The research is being conducted in a field where precise reproducibility is not expected.
It is by no means true that fraud occurs whenever these three factors are present. The factors are quite common and fraud is thought to be rare in science.”
- The circumstances that forces a scientist to succumb to temptation because of a lack sufficient self-discipline and self-confidence – generative AI erodes self-discipline…
- The rationalization a scientist provides to themselves which reframes the unethical behavior as simply a “convenient short-cut” that they or someone else will confirm the results later – a form of self-deception – generative AI provides a convenient short-cut…
- The perceived safety net that allows them plausible deniability in the event that they or someone else cannot confirm the results later – a form of self-preservation – generative AI makes it so much easier to perceive a safety net…
Action Items For The Community (Point 1):
- Be curious, not judgmental: when a case of scientific fraud becomes a big news item, have a thoughtful conversation within your research groups about how the offender could have slowly and systematically rationalized their behavior that have led up to that point – use yourself as an example – in other words ”imagine how it could have happened here.”
- Forgive the smaller mistakes but use them as teachable moments – people are not unethical, behavior is! Forgiveness does not mean no consequences; it means limited corrective-based consequences that are not permanent.
- Simple (but not easy): change the incentive structure of merit and reward in science – first small step: send your ”reproducibility studies” and “negative or null results” to the new APS journal Open Science (launched last month) – this will hopefully bring more balance to what type of research is valued and shift how scientific impact is quantified and measured.
Action Items For The Individual (Points 2 and 3):
JTS Addendum to Goodstein Framework: in every detailed study of scientific fraud, it was discovered later that the offender had a long track record of ever-increasing unethical behavior. They all started small, sometimes unintentionally (Quiz Show), finding out that there were seemingly no negative consequences until their behavior escalated into epic career-ending proportions.
Think of acting ethically (or unethically) is like “training a muscle” which requires deliberate practice:
- Every once in a while, imagine credible circumstances that you could find yourself in where you might not have enough self-discipline and self-confidence to resist temptation
- Routinely remind yourself of the corrosive effect of making seemingly small and ”consequence-free” unethical choices that could very well grow in time
- Periodically self-reflect with brutal honesty on your choices through the lens of ethics and document:
- your ethical successes as examples of what went right – this helps builds up your pattern of good habits and self-identity as a scientist – think of yourself as a scientist who is highly ethical (a defining trait) as opposed to highly intelligent/creative/good at math (merely correlated traits)
- your ethical failures and how to recognize these types of challenging choices and failure modes faster in the future – this helps build up your self-discipline and self-confidence
The incentive structure of Science
is not inherently conducive to ethical behavior,
so one must proactively choose to act ethically,
which is The One Defining Trait of the Platonic Ideal Scientist.
To do so, one must maintain constant vigilance and
build and grow a pattern of good habits,
self-discipline, and
self-confidence without self-deception.