Billi the cat’s fans hang on her every word.
She’s a cute tabby with bright green eyes and an expressive face, but what sets her apart from the thousands of felines on Instagram isn’t her looks — it’s the fact that she communicates with her human using a talking board.
In her videos, the 13-year-old domestic shorthair pads over to a setup on the floor and presses buttons that trigger audio clips of words and short phrases: “Dog,” “Food,” “Tummy,” “All done” and many others.
“I started with a word that I’d really not recommend that you start with, which is ‘food,’ because it becomes very motivating for them. And Billi loves food,” Billi’s human, Kendra Baker, told Salon.
Starting with food “kind of backfired on me,” Baker admitted, “but it definitely got the ball rolling.”
Baker enrolled Billi in an informal online research group called TheyCanTalk comprised of pet owners who try to teach their animals to communicate using talking boards. Ninety five percent of the animals involved in the program are dogs, but TheyCanTalk founder Leo Trottier, a cognitive scientist, told Salon he was “pleasantly surprised” when people began signing their cats up as “learners.”
He’s not the only professional keeping tabs on the progress of felines using talking boards.
“I’m very intrigued by the cats that are using the boards, because there’s really a dearth in cat cognition studies, particularly those that happen in the home,” cognitive scientist Gabriella Smith said. “Cats are really kind of overlooked in the companion animal cognition world. I’ve been a big fan of Billi, and my animal cognition scientist brain just lights up because I see these behaviors that I know from my own cat — but now I’m able to look at it from a cognition lens.”
We had fun with the idea here on the blog exactly one year ago when we wrote about Kristiina Wilson, an animal behaviorist who battled the boredom of lockdown by constructing a DIY talking board for her cat, who is naturally communicative and “very clear about his needs and wants” even without the board.
We imagined Buddy taking to a talking board with gusto, sparking an arms race in which he rapidly educates himself, expands his talking board, hooks it up to the internet and builds himself a prosthetic opposable thumb as part of his evil plans to take over the world.
Watching clips of Baker’s cat, Billi, I really want to believe she’s learning rudimentary language, and that I’m seeing a cat pause thoughtfully after her human poses a question, thinking over her answer before deliberately pressing a button of her choosing. I want to believe our cats can process words and simple phrases, even if syntax is beyond them. I’d like to believe my cat, like Billi, would tap out a plea for me to stay at home and hang out instead of running out to do errands.
But I don’t.
There are a few things that stretch credibility as far as the videos go, starting with the fact that Billi’s communication array has more than 50 buttons and many of them represent abstract concepts like “before,” “want,” and “later” as well as mental states or reactions like “mad” and “ouch.”
At times Baker reminds me of Penny Patterson, the animal psychologist famous for teaching Koko the gorilla to “use” sign language. There are plenty of videos of Koko throughout the years on Youtube and one constant, as different visitors, celebrities and kittens interact with her, is Patterson’s trinary role as surrogate mother, scientist and interpreter.
Koko would sign a seemingly random assortment of words, and Patterson would explain to her befuddled guests that “play nipple eat” meant Koko wanted to play, examine their nipples, and then have lunch. (Koko was obsessed with nipples for unknown reasons, to the point where media reports described her obsession as a “fetish.” It even resulted in a lawsuit in which two female employees claimed Patterson threatened to fire them if they “did not indulge Koko’s nipple fetish” and expose their breasts to the gorilla. The parties eventually settled for an undisclosed sum.)
Koko’s communications were filtered through Patterson, and her antics — whether pretending to blow her nose, feeding her kitten or signing — were done at the behest, and urging of, Patterson.
Likewise, Baker praises Billi like any loving cat caretaker would do, but does a lot of interpreting. Billi hits the button for “bird” after sundown and Baker compliments the tabby on her ability to peer into the dark outside. Then Billi pads back over to her board and presses a button for “before,” and Baker decides her cat isn’t saying she sees a bird out there now, but that she saw one earlier in the day.
When Billi presses the buttons for “mom” and “settle,” Baker interprets it as a request to lay down and cuddle. But the buttons are right next to each other, and Billi doesn’t so much press them as she stands on them the way cats do when they rest their front paws on a surface.
When Baker prompts Billi to look for her toy mouse, she presses the buttons for “where,” “mousey” and “hmmm?” Billi looks around, scratches herself for a few seconds and looks around some more before Baker picks up the toy and gives it to her. Then Baker tries again, pressing the buttons in the same order.
Billi idles, glances at Baker and finally pads over to her toy. Baker sees it as confirmation that her cat understood the words, was able to string them together, understood the sequence was a request for information, and responded appropriately.
It’s a fun informal experiment, it’s neat to see a cat using a talking board, but I think there’s a significant burden of proof for anyone claiming it qualifies as science. Even the seemingly straightforward words and phrases — “food,” “catnip” and “love you” — are more likely conditioning than understanding.
Is Billi really saying she loves Baker when she presses the button for “love you,” or has she learned that pressing that particular button always rewards acknowledgment and attention? Does Billi know “love” corresponds to the feelings of affection she has for Baker? Does Billi understand that “food” means food, or does she know pressing that button results in Baker giving her a tasty snack?
Those of us in the audience aren’t watching in real time. Videos of Billi, like videos of “talking dogs” like Bunny the sheepadoodle, are heavily edited. Long stretches of inactivity are clipped or condensed. While almost all online influencers (Bunny has a million followers on Instagram alone) make heavy use of edits to satisfy the short attention spans of their viewers, communication has a crucial temporal context. There’s a chasm of difference between someone answering a question with “no” within a second, and an animal pressing a “no” button after three minutes have passed.
The videos also constitute editing by omission. Instagrammers who upload videos of their pets using talking boards are choosing the best ones, those that confirm their belief that their cats and dogs are genuinely learning human language. We don’t see the footage of pets banging on buttons randomly or activating them accidentally as they walk over the talking boards.
That’s not to say Billi doesn’t experience the emotions associated with her talking board buttons. Behaviorism fell out of favor in 1959 with the advent of the cognitive revolution, and experiments in recent years have removed any lingering doubt behaviorist diehards may have had about animals.
Animals think and feel
Our furry friends experience the full range of primary and secondary emotions. They feel pain, depression, excitement and joy just as acutely as humans do, and well-designed experiments — like the neurologist Gregory Berns training dogs to sit still in fMRI machines so their brains can be scanned — have confirmed animals have internal thought processes and rich cognitive lives. (Like all good science, Berns’ work is repeatable.)
The question isn’t whether animals think and feel, it’s whether we’re expecting them to do something they aren’t meant to do — and anthropomorphizing them in the process — by pushing them to learn human language.
They already meet us halfway, and in the case of cats in particular, more than halfway: The little ones communicate by scent and body language, but they’re smart enough to realize we humans are hopeless at learning those subtle languages, so when they communicate with us, they vocalize as we do.
We know that even great apes, our closest relatives in the animal kingdom, and dogs — who were the first animals to be domesticated and are born with the ability to parse human facial expressions — can’t get a handle on human language.
With Koko’s passing in 2018, the sun has set on the age of primate language experiments partly because funding has dried up, partly because the undertaking involves becoming a parent in a very real sense to an animal that can live for half a century or more, and mostly because it turns out gorillas, chimpanzees and bonobos weren’t really learning language at all.
Herbert Terrace, the scientist behind the infamous “Nim Chimpsky” experiment of the 1970s, set out to prove behaviorist B.F. Skinner right — and linguist Noam Chomsky wrong — and ended up admitting failure in a book that explains how he came to realize ape language experiments were unintentional exercises in the Clever Hans Phenomenon.
“Careful examination of videotapes revealed that the human trainers had inadvertently been cuing the chimp in advance of his signing,” the Wall Street Journal wrote of the Chimsky experiment. “To Mr. Terrace’s great credit, he acknowledged his error, and then went on to discover similar mistakes had been made in most—perhaps all—prior and subsequent claims of apes acquiring human-type language. He has, to some extent, been actively engaged as an animal-language myth-buster ever since.”
The horse who changed science
If a horse can pick up on unintentional body language cues, and fool tens of thousands of people into believing it can answer questions and perform calculations as a result, then it’s a certainty that primates and cats — our closest genetic relatives and domesticated animals who live with us as members of our family, respectively — can easily pick up on cues, especially when there are rewards involved.
“Clever Hans was hailed as the first and most famous ‘thinking’ animal,” the authors of a 2013 paper on the phenomenon wrote. “Except a few skeptics, the majority of biologists, psychologists, and medical doctors, experts of all kind, and laymen were rather convinced by this example that animals are able to think in a human way and to express human ideas in non-verbal human language. In 1904, the German board of education even set up a commission to determine if the claims made about Hans were genuine. After an extended period—a year and a half of study—they concluded that there was no hoax involved.”
Clever Hans fooled lay people, scientists and animal behaviorists alike until finally, “by the meticulous examination of Professor Oscar Pfungst, a biologist and psychologist,” Pfungst realized Hans couldn’t answer correctly if the person asking him the question did not know the answer either.
In other words, Hans was reading the answers off the faces and via the body language of the people around him. In the end it turned out Hans wasn’t a math genius or a connoisseur of the arts, but he was clever. Hans showed the world just how closely attuned animals can be to human behavior.
The fact that a horse, and not a primate or a dog, revealed such an ability reflects the thousands of years horses have been taking subtle cues from their trainers and riders. An experienced horse can determine a rider’s intent by the slightest tension on the reins or shift in weight.
In fact, it was the Clever Hans Phenomenon that led to innovations like double blind studies and forced scientists to really think about how they design their studies, as even the most careful and well intentioned researcher can unconsciously convey information to study participants. When we’re close to our subjects, as pet caretakers are, it’s almost a guarantee that we’re giving our pets cues without awareness of what we’re doing.
Animals like cats and dogs may not think like people do — and it’s a mistake to expect that of them — but that doesn’t mean they’re not experts at gleaning information from our body language, facial expressions and tone of voice. In fact, that’s their natural way of communicating information.
Even though I can’t bring myself to believe Billi, or any other cat, is learning to use human language, there’s value because failure tells us just as much as success does. Any attempt to better understand cats is a worthy pursuit.
Regardless, it’s clear Billi likes pressing the buttons on her talking board, and the entire exercise is a stimulating game for her. Cats may not have the ability to use human language, but they do like to play, and they like anything that results in interaction and attention.
Maybe that’s the most important takeaway.
“I really believe that the majority of house cats are bored and depressed,” Baker said. “We don’t give them any stimulation . . . and if this pandemic has taught us anything, it’s that staying inside your house all day is terrible. So you know, anything that we can do for them that gives them a better life, I’m for it.”
All photos credit Wikimedia Commons and Pexels unless otherwise noted.
7 thoughts on “Can Cats Talk? These Researchers Think They Can”
You may want to check out this article from The New Yorker. It did mention someday this developing area of technology may make inter-species communication possible.
Thanks for the link, Lavinia. While reading it, I found myself wondering how neuro-atypical people fit in to this voxelized model of mental space, the potential impact of qualia on shared concepts, and the ways this tech could change things. As the author put it, the mind reels at the possibilities, good and bad.
I think we’ve all had frustrating experiences in which we lack the vocabulary to describe concepts and ideas. I feel that way with concepts in linguistics and music, as someone who doesn’t have a formal education in either. The possibility of shortcuts mediated by computer is interesting. On the other hand, the potential applications can be terrifying. Imagine if this technology matures and a mental inventory becomes a requirement for sentencing in criminal cases, or becomes a routine part of investigative police work. We’ve already dispensed with so many elements of privacy in an always-online, over-sharing world of social media, with at least two generations of kids (so far) for whom sharing unedited thoughts with everyone is a normal part of every day existence.
The idea that our minds, our thoughts, would no longer be private is disturbing.
But it’s also disturbing to think that people may be judged on thoughts rather than actions, a bit like a Minority Report/science fiction vision of “justice.” (Or potentially more frightening: Gauging loyalty in an authoritarian society based on thought patterns. Chairman Xi demands pure thoughts! Juche demands that your relatives’ thoughts be “correct” for favorable songbun!)
Getting back to the neuro-atypical, one thing that distinguishes “normal” people from those with autism is the brain’s ability to filter and to “split and lump” concepts and experiences. I think anyone who has ever experimented with psychedelics knows what it’s like to have those filters temporarily disabled or muted.
I imagine animal minds filter experiences and concepts just like human minds do, but in ways that are probably as foreign as neuro-atypical mind. We know, for example, that cats are much less concerned with quantities than they are with size, which makes sense from an evolutionary perspective, as solitary hunters care more about whether potential prey can feasibly be taken down, and whether it’s enough to fill an empty stomach, vs the human idea of “I’m really hungry so I’ll eat two apples instead of one.”
This tech may prove once and for all whether we’re completely wasting our time trying to teach animals to understand basic human language, or whether certain animals can learn words as ideas rather than prompts and associations. I still think we’re asking too much and putting the burden on animals, rather than putting the burden on us to better understand them.
And that will conclude my book-size comment. 🙂
LikeLiked by 2 people
A lot of the New Yorker articles have been both fascinating and terrifying of late, especially this one. And you are right, the burden needs to be put on the humans to better understand other forms of life.
LikeLiked by 1 person
Buddy having access to a communication board would certainly result in some illicit messaging to the catnip cartel!
LikeLiked by 2 people
The idea of Bud having a line to the outside world is terrifying. I’d probably go broke within a week as packages with thousands of pounds of turkey pile up at my front door. He’d probably order a bull horn as well, to amplify his already-loud meows.
LikeLiked by 1 person
Years ago, I had this adorable little cat I named Bunny. I will never ever forget this: one day I asked – probably for the 10th time that day – ‘Who’s the very cutest kitty in the whole world?’ (Her. Obviously.) And she replied, in a little kitty voice: “I… am.’ This wasn’t a meow. It was two separate sounds that sounded a whole lot like the correct answer.