Recent blog post roundup and Sun letter on fetal learning

Recent posts for the ACS Chemical Biology Community have been on social taboos about discussing compensation, what to wear to work, and gender and scientific authorship.

The Baltimore Sun also published my letter in response to a pop science story they ran on fetal learning.  (The Sun's version appears to have fallen down the memory hole, but here's the original Reuters reports: it's very similar to the Sun story, except the Sun changed "babies in the womb" to "fetuses".)  The Sun has a pay wall, so I'm reproducing the letter here:

Your headline "Fetuses show signs of learning at 34 weeks" (July 27) was inconsistent with the content of the story accompanying it.

The author admitted that the findings were "not statistically significant." In layperson's terms, "not statistically significant" means that the evidence is not strong enough support a conclusion one way or the other.

Scientist Carl Sagan famously said that extraordinary claims require extraordinary evidence. To claim that fetuses show evidence of learning three weeks earlier than previously thought, the authors would have to show convincing evidence that the earlier work was mistaken. But because their findings were not statistically significant, they can offer no such evidence.

Science journalists are the final interpreter of science for the public. It is their duty not just to report recent findings but also — and more importantly — to critically evaluate them for their non-scientist readers.

The letter was edited somewhat: for brevity, I guess.  Here’s the original text:

Your recent headline "Fetuses show signs of learning at 34 weeks" was inconsistent with the content of the story.  The author admitted that the findings were "not statistically significant".  In layperson's terms, "not statistically significant" means "the evidence is not strong enough support a conclusion one way or another".  Carl Sagan famously said that extraordinary claims require extraordinary evidence.  To claim that fetuses show evidence of learning three weeks earlier than previously thought, the authors need to show convincing evidence that the previous work was mistaken: because their findings were not statistically significant, their findings are not such evidence.  The author was wrong to "cautiously" conclude, the reviewers were wrong not to insist on removing that conclusion, and the journal was wrong to publish that conclusion, but science journalists are the final interpreter of science for the public.  It is the sacred duty of a science journalist not just to report recent scientific findings, but also, more importantly, to critically evaluate them for their non-scientist readers.