Why Charlie Doesn't Have a Yes or No Button
One of the most common questions I get is why I haven't given Charlie yes/no or "all done" buttons. The short answer is that he already communicates those things clearly. The longer answer is where the science gets interesting.
Charlie tells me "no" or "I'm finished" by turning away. It is unambiguous, he does it every time, and it involves his whole body. Adding a button for a message he already sends fluently would not expand what he can say. In communication terms it is redundant, a second channel carrying information the first channel already delivers reliably. In human AAC (augmentative and alternative communication) design, the guiding principle is to prioritize vocabulary that widens a person's expressive range, not to duplicate a signal they already produce well. The same logic applies here.
But there is a deeper reason, and it comes from one of the most important cautionary tales in animal cognition: Clever Hans. Hans was a horse in the early 1900s who appeared to do arithmetic by tapping his hoof. Careful investigation by Oskar Pfungst showed Hans was not doing math at all. He was reading tiny, involuntary changes in the posture and expression of the humans around him, cues they did not know they were giving, and stopping when he sensed the expected answer had arrived. Hans was brilliant. Just not at arithmetic. He was brilliant at reading people.
This is exactly why yes/no buttons are risky in a way Charlie's other buttons are not. When he presses "Neck" or "Booty," the answer lives in him. I do not know in advance what is bothering him, so I cannot cue it. But a yes/no button is different. When you ask an animal "do you want this?", you usually already have an expectation, and that is precisely the situation where unconscious cueing (sometimes called the ideomotor effect or expectancy effect) creeps in. A "yes" button would not give me cleaner information. It would give me a channel where my own hopes could leak into his answer.
His turn-away is more trustworthy for another reason too. It is a costly, whole-body signal, hard to produce by accident and hard to misread. A button press is small and easy. If I gave him a "no" button, I would eventually face presses that contradict his body language, and I would have to decide which to believe. I would have manufactured an ambiguity that does not currently exist.
There is a principle in comparative psychology called Morgan's Canon: do not explain a behavior as complex reasoning when a simpler explanation will do. Good communication research means building systems that resist over-interpretation, not ones that invite it. Choosing not to add yes/no buttons is one of those choices. It keeps Charlie's "no" honest, in his body, on his terms, and free of my influence.
The novelty of a button press is easy to celebrate. The harder and more important work is making sure each button actually means something.
