Maya Sneller
Research Mentor(s): Lauren Girouard
Mentor Department: Psychology
Authors: Maya Sneller, Lauren Girouard-Hallam
Session: Session 3 (11:00am – 11:50am)
Presentation Type: Poster 5
Abstract
As artificial intelligence (AI) chatbots like ChatGPT gain visibility in mainstream media and become a common part of internet usage, questions are being raised about their role in children’s online experiences and how much children trust the information these chatbots provide. This study seeks to explore how children perceive and place trust in AI chatbots. When data collection begins, this experiment will involve 216 children aged 7 to 14, who will evaluate the reliability of responses to health-related science questions. These answers, delivered either by a human or an AI chatbot, will be shown through a PowerPoint presentation during an online Zoom session with the experimenter. The children will also analyze specific types of mistakes—such as fabricated answers, logical errors, and typos—made by both humans (including adults and doctors) and AI conversation bots. Additionally, the study will investigate whether children’s verbal abilities and their familiarity with the internet impact their trust in chatbots and their ability to detect errors. This study aims to deepen understanding of children’s trust in and comprehension of AI chatbots. It will also examine how cognitive skills and prior online experience shape their views, providing valuable insights into the role of AI chatbots in children’s digital lives.
