What might my child be seeking when an AI conversation feels supportive or friend-like?
Anne J Maheux; Debra Boeldt; Samir Akre-Bhide; Jessica Flannery; Allison Paige; Giavanna Villella; Kaitlyn Burnell; Eva H Telzer; Scott H Kollins · 2026-07-27 · Journal of Medical Internet Research
Parent-reported ages 1–17; most users 10–17; ages below 10 especially uncertainCC BY 4.0Contemporary generative AI chatbots and companions; observational use, not an interventionChild-facing evidence
Which ways of asking about feelings help my teenager feel comfortable rather than interrogated?
Audrey Mariamo; Caroline Elizabeth Temcheff; Pierre-Majorique Léger; Sylvain Senecal; Marianne Alexandra Lau; Andre Kushniruk; Stefan Morana; Alfred Brendel; Rebecca Krukowski; Monet Slinowsky · 2021-03-18 · JMIR Human Factors
14–17 yearsCC BY 4.0Hypothetical chatbot interface and prewritten questions; no live GenAIChild-facing evidence
When a robot says it likes the same things as my child, how might that shape what my child shares?
Mike E U Ligthart; Mark A Neerincx; Koen V Hindriks · 2022-09-15 · Frontiers in Robotics and AI
8–10 years in study 1; 8–11 in study 2CC BY (version not stated)Wizard-of-Oz pilot plus autonomous speech-recognition social robot with constrained dialogueChild-facing evidence
What would make digital self-compassion support feel relevant to my teen’s real daily challenges?
Anna Boggiss; Nathan Consedine; Sarah Hopkins; Connor Silvester; Craig Jefferies; Paul Hofman; Anna Serlachius; Yu Kuei Lin; Joanna McClintock; Santosh Pandey; Sun Young Park · 2023-05-05 · JMIR Diabetes
12–16 yearsCC BY 4.0Rule-based lessons plus explicitly reported limited AI recognition of emotions, management status and risk words; demonstrated prototypeChild-facing evidence
Eligible 6–12 years; analyzed mean 10.79 years (SD 1.12)CC BY 4.0Autonomous social robot with basic NLP and preprogrammed adaptive activitiesChild-facing evidence
When a support chatbot reports that users feel better, how can I judge what caused the change?
Dana Vertsberger; Navot Naor; Mirène Winsberg; Khaled El Emam; Bradley Malin; Sebastian Fudickar; Zhenwei Dai · 2022-11-29 · JMIR AI
Advertised to 14–18-year-olds; actual ages and gender were not collected or verifiedCC BY 4.0Automated ACT conversational service with human distress handoff; model architecture unspecifiedChild-facing evidence
Does a child looking longer at a robot mean the child is learning to understand people’s emotions?
Maria J. Pinto-Bernal; Sergio D. Sierra M.; Marcela Munera; Diego Casas; Adriana Villa-Moreno; Anselmo Frizera-Neto; Martin F. Stoelen; Tony Belpaeme; Carlos A. Cifuentes · 2023-03-02 · Frontiers in Neurorobotics
Validation eligibility 5–10 years; design work involved children 3–9CC BY 4.0Remotely controlled social robot with automated behavior measurementChild-facing evidence
How can I tell whether a comforting robot changed my child’s pain, mood or both?
Fang-Yu Hsu; Yun-Hsuan Lee; Chih-Yuan Yang; Sue-hsien Chen; Shih-Ming Chu; Angela Shin-Yu Lien; Matthew Balcarras · 2026-08-07 · Journal of Medical Internet Research
4–12 years; mean 7.35CC BY 4.0Wizard-of-Oz socially assistive robot plus automated facial-expression measurementChild-facing evidence
If my child enjoys a school robot buddy, what would show that the support helps beyond enjoyment?
Anne Zijp; Jiska J Aardoom; Olivier A Blanson Henkemans; Sylvia van der Pal; Eline Vlasblom; Anke Versluis; Amaryllis Mavragani · 2025-08-29 · JMIR Formative Research
8–12 years; mean 9.9CC BY 4.0Structured robot-delivered psychological training with tablet responses; no GenAI evaluatedChild-facing evidence
Approximately 3–6 years (recruited 3 years 3 months–6 years 1 month)CC BY 4.0Fully teleoperated NAO with matched recorded speech and programmed actionsChild-facing evidence
Can a shared activity with a robot open a small opportunity to include a child who is new to the group?
Sylvaine Tuncer; Sarah Gillet; Iolanda Leite · 2022-03-03 · Frontiers in Robotics and AI
School children; mean 10.46 years (SD 0.67)CC BY (version not stated)Programmed/adaptive game-mediating social robot; qualitative analysis, not GenAIChild-facing evidence
Adolescents; ASD mean 15.91 years, comparison mean 15.73; full range not stated in methodsCC BY 4.0Teleoperated android and simple humanoid with scripted emotional-disclosure interviewsChild-facing evidence
What would my child and I want a hospital support robot to do before we trusted it near a difficult procedure?
Fareha Nishat; Summer Hudson; Prabdeep Panesar; Samina Ali; Sasha Litwin; Frauke Zeller; Patricia Candelaria; Mary Ellen Foster; Jennifer Stinson · 2023-08-24 · Journal of Clinical and Translational Science
Children 5–11 years; caregivers were adultsCC BY 4.0Proposed adaptive AI-enhanced social robot; qualitative design study, no live AI interventionDirect parenting evidence