I wondered if maybe the style of horse training where we say "Its very good if the horse does the thing we ask, and very bad if they do not" gets people into the habit of thinking in a very good/bad yes/no green/red dicotomy.
And I think this kind of teaches people to put horses responses into very big bucket - its a yes if the horse did what you asked, and its a no if the horse didnt do what you asked. And thats it. YES, or NO.
And then when those same people start trying to allow their horses more choice and more freedom, they are still in the habit of only ever hearing YES, or NO, from their horses.
So their horse might be trying to communicate "hey do you think there will be a tractor coming out of that driveway again soon cause last time there was a tractor and we had to move off the path to let them past" but the human who can only recognise YES or NO sees the horse stop and look and hears NO.
Or the human asks the horse to trot but the horse trots forward in a kind of awkward and tense way - the horse might be trying to communicate "the different way you tried to girth me up today is causing the back of my saddle to move too much and its bouncing awkwardly on my back" but the human who can only recognise YES or NO feels the horse trot when asked and hears YES.
Trying to look for the longer sentence of communication instead of chucking everything in YES/NO buckets does seem to be benefitting me and my horse. It seems to be more useful model, a better reflection of reality, and is helping me notice patterns in and offer support to my pony that I might have otherwise missed.
But I am also cautious and suspicious of my own idea here. I could see how I could accidently end up telling myself a story in my head of what my pony was thinking - a story which may or may not be reflective of her actual experiences. I'm trying to protect us against this potential pitfall where I can by making hypotheisises and then testing them in the moment.