What Does a Meow Mean? In Search of Intuitively Understandable Communication by a Nonverbal Companion Robot
Vivienne Bihe Chi, Claudia B. Rébola, Bertram F. Malle · 2026
Read What Does a Meow Mean? In Search of Intuitively Understandable Communication by a Nonverbal Companion Robot by Vivienne Bihe Chi, Claudia B. Rébola, Bertram F. Malle (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-01251.txt.
Introduction
Vivienne Bihe Chi, Claudia B. Rébola, and Bertram F. Malle test whether older adults can understand nonverbal signals from a cat-like companion robot. Older adults living alone face challenges that robots may ease through reminders, activity prompts, or comfort. As part of developing a cat robot with limited assistive functions, the team designed auditory cat sounds and visual icons on a small display as communication signals.
Evaluation used a mixed-methods, user-centered approach: a pilot study, then a focus group with older adults that suggested revisions, then a large online experiment with adults over sixty-five. With both signal types present, intention inference was highly accurate. Auditory signals alone were less helpful, and removing them sometimes improved accuracy, except when the robot conveyed strong sentiments such as purring.
Catalog record
| CREATOR | Vivienne Bihe Chi, Claudia B. Rébola, Bertram F. Malle |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 36 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.01251
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
What Does a Meow Mean? In Search of Intuitively Understandable Communication by a Nonverbal Companion Robot Vivienne Bihe Chi, Claudia B. Rébola, Bertram F. Malle Abstract Older adults living alone have a number of challenges, and robots can help with some of them—by providing reminders, initiating activity, or offering comfort. As part of developing a cat robot with limited assistive functions, we designed a set of nonverbal communication signals, both auditory (cat sounds) and visual (icons on a small display). To evaluate these signals we used a mixed-methods, user-centered approach. After a pilot study, a focus group with older adults suggested revisions to the initial signal set. A large-sample online experiment then tested whether adults over the age of 65 could accurately infer the robot’s communicative intentions. When both visual and auditory signals were present, accuracy was high. When visual signals were absent, accuracy often decreased; when auditory signals were absent, accuracy sometimes increased. So the auditory signals were less helpful, except when the robot conveyed strong sentiments (e.g., purring while being petted). 1 Introduction Nearly three in ten older adults reside alone in the U.S. [ 19 ] and face challenges from falling to accidental prescription overdose to loneliness [ 15 ] . Robots may help alleviate these challenges, but older adults often encounter barriers when operating technology. Intended to simplify their lives, technology frequently causes anxiety and feelings of incompetence. Any assistive robot must have an intuitive and accessible design, enabling effortless interaction and facilitating adoption and long-term use. In such design, understandable communication is central. We have been developing a zoomorphic robot (built on the Ageless Innovation Joy For All animatronic cat [ 23 ] , Figure 1 ), designed with limited assistive functions but intended to be affordable for broad deployment. In developing the communicative signals that the robot emits, we needed to confirm that any human user can easily understand these signals. We report on the development and evaluation of intuitively understandable communication signals expressed by this nonverbal robot. We used a mixed-methods, user-centered approach, including a focus group study and a large-sample online experiment with older adults. Figure 1: Ageless Innovation’s Joy For All cat, with a smart collar added. 2 Related Work 2.1 Social Robots for Older Adults The proportion of older adults is increasing globally, with those aged 65 and older projected to represent an even larger share of society in the near future [ 6 ] . There are rising concerns about addressing their social and emotional needs, as many experience isolation and loneliness [ 3 ] . Social robots have emerged to offer companionship [ 4 ] and reduce loneliness by facilitating human-human interaction [ 28 , 32 ] . Social robots can serve a role in healthcare settings by providing mental stimulation, assisting with daily tasks, and supporting health monitoring [ 10 , 48 ] . For older adults with dementia, these robots can lower stress, help the person keep a schedule, and improve well-being [ 13 ] . Many challenges remain in the integration of social robots into older adults’ lives. Some healthy older adults feel uneasy about being led to believe these robots are real companions [ 32 , 4 ] . Further, privacy concerns about their advanced sensory capabilities and the handling of personal data are common, as are worries that robots could reduce opportunities for human-human interaction [ 4 , 29 , 18 ] . Addressing these concerns is essential if we want to ensure that social robots become meaningful and effective companions supporting older adults. 2.2 Animal Companion Robots Companion robots may be particularly important in assisting older adults. They don’t clean and iron and are often stationary, but they engage the user through touch, sound, and reactive movement, which creates a comforting, stress-reducing presence [ 36 , 21 ] . In addition to emotional comfort, companion robots can provide cognitive stimulation and even healthcare support. Some are designed to assist with memory exercises and daily routines [ 36 , 25 ] , others can support compliance with medication schedules or promote healthy behaviors [ 44 , 33 ] . Companion robots often mimic the appearance of familiar pets such as cats and dogs [ 36 , 41 ] , but they can also take on less conventional shapes, such as Paro the seal, PLEO the dinosaur, or Baby Whale [ 36 , 33 , 9 ] . Interaction complexity varies, from Paro’s subtle movements to Aibo’s dynamic behavior [ 5 ] . Older adults, especially those experiencing loneliness or cognitive decline, prefer familiar forms like robotic cats and dogs for their emotional comfort [ 21 , 46 ] . For older adults, robots’ interactive movement and responsiveness to simple commands are more important than technical sophistication [ 5 ] . 2.3 Universal Design & Accessibility Human-centered design aims to make companion robots functional and accessible for distinct user groups with different needs, abilities, and preferences. The principles of Universal Design, in particular, favor products that are usable by people of all abilities without requiring adaptations [ 37 , 51 ] . Thus, social robots should be accessible to older adults with sensory, cognitive, or mobility changes often associated with aging [ 17 ] . A fundamental principle of accessibility is the use of redundant communication signals by delivering information through multiple channels [ 14 ] , such as visual, auditory, and tactile cues [ 38 , 20 ] . This redundancy allows robots to convey critical messages to users with diverse abilities. 2.4 Nonverbal Communication Verbal communication abilities are ubiquitous in virtual agents and humanoid robots, but such abilities would be unusual, perhaps disturbing, when given to animal robots like cats and dogs. Pet companion robots are therefore best equipped with nonverbal communication [ 16 ] . In human-human interactions, nonverbal cues of smiling, nodding, and gestures are highly influential in social interactions [ 26 ] , from job interviews [ 45 ] to physician-patient communications [ 1 ] . However, a smiling, nodding cat robot may be just as uncanny as a talking one [ 30 ] . Some researchers have applied insights from ethology to design robot behaviors that resemble natural animal behavior [ 27 ] and are therefore easily readable by human users [ 12 ] . Indeed, older adults seem to expect and prefer zoomorphic robots with natural sounds and colors [ 11 ] . Selecting communication signals solely from natural origins, however, may limit a companion robot’s design. Natural and artificial signals can be combined, as in the case of Cozmo’s “emotion” expressions that include shape of eyes, head position, forklift movements, and sounds. Such complex expressions take time to unfold, however, and are potentially ambiguous [ 40 ] . Additional artificial signals may include differently colored lights [ 39 ] or icons on display screens that indicate mood [ 8 ] or that provide reminders [ 33 ] . Extensive work is available on sound design for robots [ 43 ] . Some have argued that all robotic sounds should strive to convey affective qualities [ 22 ] , and in practice, robotic sounds are often used to communicate emotions [ 52 , 50 ] . However,