And let me be clear about my bias here: I want dogs to catch yawns, as any good animal lover would.
In 2008, a paper declared, "Dogs catch human yawns." But since then, things haven't been looking good for the Contagious Yawners (i.e. the good guys). A 2009 study found no evidence of dogs catching yawns from people yawning on video. And a 2011 study found no evidence of dogs catching yawns from their owners or strangers.
"Our results provide no support for empathy-based, emotionally connected yawning contagion in dogs and casts doubt on the recently documented phenomenon of cross-species contagious yawning," the researchers in the latter study concluded. "We interpret our findings as showing that if dogs are seen yawning contagiously then the contagion must be explained on less cognitively stringent grounds than empathy."
To be honest, I had pretty much given up hope that dogs were the empathic beasts I want them to be, finely tuned to humans' every emotional quiver. But today, good news arrived!
A newly released study suggests that as dogs mature, they do, in fact, start to catch yawns from humans. Cute little puppies can't do the trick, but Old Yeller can. The authors of the new research, a Swedish team from Lund University, acknowledge the controversy over contagious yawning and draw out its import to empathy research at length. They tried to correct for the various sources of error in previous studies and then tested 35 dogs to see if they'd yawn when a human did. Mostly, this test involves variations on having a human do nothing, yawn, or make a mouth opening motion they call a 'gape,' which won't make other humans yawn.
This is a yawn (A) and gape (B) example from an earlier paper.
They found two key things. One, that it didn't matter if the dog knew the person yawning. And two, the age of the dog did matter. They hypothesize that some kind of mental machinery comes into play somewhere around seven months old, and after that, the dogs were much more likely to catch yawns from humans. (You can check out the whole paper here.)
I bring this paper to your attention for two reasons. One, hooray, perhaps our dogs really do sense our emotional states! And two, science does not progress along a straight line. Any time I look into any crevice of science, you find disputes that take years and years to get resolved. Individual experiments and measurable quantities (like yawns per dog) become tied up in larger ideas about how the world does or should work. It's messy and if you follow anything for long enough, you'll see the field meandering like the Mississippi flowing down to the Gulf. The editors of one journal might publish several papers supporting a contention, while a rival group publishes takedowns. At the ground level, that's how knowledge gets made.
We'd like the story to be tidy. Hell, me personally, I would like the story to be settled: Dogs either catch our yawns and are custom-bred empaths or they don't and aren't. But the world is messier than that, even when you strip away as much complexity as possible. A model experiment, as long as it includes living beings, can be too complex a system to understand easily.