Social robots are the flavor of the year these days. If robots are to live with humans (which is not a foregone conclusion, IMO), they need to mesh with human psychology. This means they need to appear harmless and attractive, they need to understand and emit unconscious signals, and generally play nicely. It doesn’t matter what they do, so much as how they do it.
This has led to a variety of interesting research. Some pursue the goal of mimicking human behavior. Other approaches use non-human forms with intelligible behavior. There is a great range of possibilities, with more and less human-like appearance.
There are actually some really interesting questions here about the psychology of humans interacting with non-human machines, intelligent or not. It seems pretty clear that trying to faithfully imitate human forms and nuances isn’t necessary, nor is speech. (See perhaps the thoughtful work of Sense Thecla Schiphorst [2.3]).
This principle is clear in a new product, “Qoobo: A Tailed Cushion That Heals Your Heart”. While this has been described as a “robot”, it certainly lies at the edge of that term. It has only one behavior; waving the tail. No face. No dialog. Certainly no “useful” functions.
The Tokyo based inventor, Prof. Nobuhiro Sakata, (who apparently also created necomimi in 2011) believes that this is comforting. In fact, it is supposed to “heals your heart”, whatever that means exactly.
This is harmless, I guess, though vacuous.
But there are so many dubious aspects of this product, I can’t let it pass
First, they have tried to carefully reproduce the motion of a cat’s tail. It’s clear from the video that they haven’t succeeded in that effort, but in any case they seem to have no understanding of cats at all. Swishing the tail means the cat is agitated, not happy or friendly. A contented cat rubs and purrs, and does not swish the tail. If you pet a cat and its tail starts moving, it is unhappy and probably going to fight and/or run.
Second, setting aside the complete misunderstanding of natural cat behavior, the project claims that the responsive behavior of the tail enhances the human’s feelings. The crux of the case is that you “would project your emotions onto how the tail moves, and you could get a sense of healing from that”. Well, maybe so, though there is no evidence that this is actually true.
Third, the claimed benefits are nebulous and new agey. What exactly does “heals your heart” or a “sense of healing” mean? How ever these benefits may be defined, has Qoobo been shown to actually work as advertised? Furthermore, is it better than a placebo, such as a cushion without a tail, or a plush animal without animation? And how does it compare to alternatives such as a real cat or even to a virtual conversation via social media?
You might hope that the product would be proved safe and effective before it is sold, but that’s not how we do things these days. In fact, they are doing a kickstarter, and part of the work will be the unspecified pledge, “We will be conducting a proof of concept to ensure Qoobo is providing a sense of comfort to its users as intended.”
Qoobo is charming and cute and nice and all that. I really hate to criticize it. But I really think you should not make claims about supposed psychological or other benefits without legitimate evidence.
- Qoobo. Qoobo : A pillow with a wagging tail. 2017, https://www.kickstarter.com/projects/1477302345/qoobo?ref=484yf0.
- Thecla Schiphorst, soft(n): toward a somaesthetics of touch, in Proceedings of the 27th international conference extended abstracts on Human factors in computing systems. 2009, ACM: Boston, MA, USA. https://dl.acm.org/citation.cfm?doid=1520340.1520345
- Thecla Henrietta Helena Maria Schiphorst, THE VARIETIES OF USER EXPERIENCE: BRIDGING EMBODIED METHODOLOGIES FROM SOMATICS AND PERFORMANCE TO HUMAN COMPUTER INTERACTION, in Center for Advanced Inquiry in the Integrative Arts (CAiiA). 2009, University of Plymouth: Plymouth. https://www.academia.edu/207432/The_Varieties_of_User_Experience_Bridging_Embodied_Methodologies_from_Somatics_and_Performance_to_Human_Computer_Interaction