03/04/2025
Are dogs really using buttons to talk? Or is it fake news?
Answers are starting to come in.
I’ve been skeptical about the “talking dog” trend, where dogs use soundboard buttons to “speak” to their guardians. But new research (2024) is challenging my skepticism, at least partially, and shedding light on what’s really happening when our dogs press these buttons.
Two 2024 peer-reviewed studies explored whether dogs genuinely understand the words they’re pressing, and whether their button-press combinations are intentional or just random.
The first study (Bastos et al., 2024) tackled comprehension. Researchers conducted two complementary experiments: an in-person study with 30 dogs and a citizen science study with 29 dogs, all trained by their guardians.
Dogs responded appropriately to ‘play’ and ‘outside’ button presses significantly more often than expected by chance, even when pressed by strangers without additional contextual cues. This wasn’t random; dogs were significantly more likely to respond appropriately to familiar words, which is evidence that at least basic word comprehension seems to be genuinely occurring.
The second study (Rossano et al., 2024) analyzed over 56,000 multi-button presses from more than 150 dogs. The goal was to investigate whether these multi-button combinations were intentional, and results showed dogs weren’t randomly pressing buttons or just imitating their humans.
Certain button sequences appeared far more often than chance would predict. Dogs frequently pressed meaningful combinations like their name + “want,” or “play” + “outside,” at rates significantly higher than chance, suggesting deliberate, purposeful communication. However, researchers are still working to figure out whether dogs might be using memorized button “chunks” reinforced by positive outcomes, rather than creatively building novel sentences.
Here’s the important part: Yes, dogs are intentionally pressing buttons and associating specific words with specific outcomes. But the “sentences” in these studies are simple and mostly about immediate desires (like play, food, or going outside) and not complex grammar or abstract thoughts. We don’t have evidence they’re using buttons to convey complex ideas and language - yet!
So why does this matter?
First, it supports that dogs are incredibly capable associative learners. Like us, they grasp words as symbols connected to specific real-world results.
Second, it underscores the individuality of canine intelligence. Some dogs thrive with buttons, others don’t, which is a great reminder of how training and enrichment must be tailored to each dog’s unique cognitive style.
Personally, we’ve taught button use to our dogs and noticed clear signs they associate their buttons with specific outcomes. But let’s remember to stay grounded: pressing buttons alone isn’t proof of deep linguistic thought. It’s evidence that dogs can communicate their basic thoughts in a new, intriguing way.
Hopefully, future research will test if dogs use buttons to communicate information humans don’t already know, how well they understand novel combinations, and whether these alternative communication devices genuinely enrich dogs’ lives.
Meanwhile, let’s appreciate that our dogs are already excellent communicators, in dog language. The buttons just add another fascinating layer!
What are your thoughts?
Have you tried buttons with your dog? What’s your experience?
Studies referenced:
Bastos, A. P. M., et al. (2024). How do soundboard-trained dogs respond to human button presses? An investigation into word comprehension. https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0307189
Rossano, F., et al. (2024). Soundboard-trained dogs produce non-accidental, non-random and non-imitative two-button combinations. https://www.nature.com/articles/s41598-024-79517-6