Smell circuitry, stalled stem cells and more reader feedback

Odor confusion

People just aren’t very good at putting names to smells. It turns out that the odor-to-language interface in the brain may be less efficient than the sophisticated neural circuitry that processes visual and auditory information, as Rachel Ehrenberg reported in “Brain regions link odors to words” (SN: 12/13/14, p. 8).

“Imagine my surprise when I read your short article on how the brain links words to odors,” wrote Tim Geho. “It states that ‘people have a hard time identifying odors.’ Just the very day before, I read … that instead of the long-held belief that people can distinguish about 10,000 scents, it is now thought that the average person can detect at least a trillion different smells. These two articles seem to reach vastly different conclusions.”

The two studies were examining separate things, explains Ehrenberg. The “trillion different smells” study (SN: 12/27/14, p. 25) concluded that humans can discriminate between lots of odor pairs — basically, it showed that people have a good sense of smell. The brain regions study, however, looked at people’s ability to name odors. The authors were trying to figure out why humans are so bad at describing odors in words, even though they’re good at telling smells apart. The research suggests that neural information about odors is still pretty crude when it reaches the brain’s language network, so people depend more on context to describe what they smell.

Stem cells stalled out

Two papers detailing a simple method for making stem cells were retracted last year when other researchers were unable to replicate the results. Tina Hesman Saey summed up the controversy in “Easy stem cells a no go” (SN: 12/27/14, p. 25) as part of Science News’ Top 25 stories of the year.

Some readers felt that Nature, the journal that published the now-retracted papers, deserved a share of the blame. “It was reported that the peer reviewers did not recommend publication because of ‘grave concerns over the work,’ but it was published anyway. Nature is now trying to spin the controversy as something that could not have been spotted prior to publication despite its own peer reviewers being on record as having rejected it,” wrote commenter wooter. “Perhaps the editors at Nature need to reevaluate their motivation regarding the publication of papers that promote something that appears to be too good to be true, because it may just turn out to be the case.”

Commenter Physics Police, however, came to the journal’s defense. “Let’s be careful in how we criticize Nature. They clearly blew this one. But just because some reviewers had some negative feedback doesn’t mean a paper should go unpublished. We don’t know the full story about how this paper was reviewed. For a hypothetical example, some ‘grave concerns’ might go away with a revision that includes missing data tables. Remember that fraud and scientific misconduct are often invisible to the reader.”

Support for SN

Long-time reader Paul Ebel wrote in with some kind words for Science News. “I have been a subscriber of SN for more than 20 years and have bought subscriptions for my grandchildren and for my business partners. I think this magazine is one of the most important ways to vet news about science,” he said. “Thanks for all that you do. In particular, thank Ms. Emerson for her notes on the second page. I always read every word she writes, and that sets the stage for the evening of enjoying the magazine.”


The magnitude 5.7 earthquake that rattled Oklahoma in 2011 is attributed to the Wiltzetta Fault in “Carbon quakes” (SN: 1/24/15, p. 14). The correct spelling of the fault’s name is Wilzetta.

On Page 15 of the same feature, the wrong issue date is provided for an earlier Science News story. That story can be found at SN: 9/8/12, p. 20.

More Stories from Science News on Science & Society