Sir

Sandra Titus and her colleagues examine a problem that many scientists would prefer to ignore in their Commentary 'Repairing research integrity', (Nature 453, 980-982; 2008). Their survey of extramural scientists funded by the US National Institutes of Health indicates that many cases of scientific misconduct may not be reported to their institutions and that even fewer come to the attention of the government funding agency. They point out that institutional leaders have a strong financial incentive to silence the would—be whistle—blower, because a confirmed case of research misconduct can hurt the institution's reputation and impede the flow of its funding.

But in calling for more tutoring and stricter supervision of junior scientists, Titus and colleagues drastically oversimplify the nature of the problem. They assume that most scientific leaders behave with integrity in their own work, but somehow fail to pass this trait to those whose research they supervise.

On the basis of our own discussions with biomedical scientists at the predoctoral, postdoctoral and faculty level, we hold a different view. The academic and financial rewards of calculated, cautious dishonesty on the part of some scientific leaders are, we believe, all too apparent to the junior scientists they supervise. No amount of tutoring, stricter supervision or courses in research ethics will fix this problem.

We, the writers of this Correspondence, are the authors of a report written 21 years ago on misconduct in biomedical research (Nature 325, 207–214; 1987).

See also: Integrity: Croatia's standards unusual in much of Europe Integrity: misconduct by a few damages credibility for many Integrity: how to measure breaches effectively Titus et al. reply