Tortella Fabio (Azienda Ospedaliera Universitaria Integrata di Verona, Verona, Italy)
Gianola Silvia (Unit of Clinical Epidemiology, IRCCS Istituto Ortopedico Galeazzi, Milano, Italy)
Turolla Andrea (Department of Biomedical and Neuromotor Sciences-DIBINEM, Alma Mater Studiorum Università di Bologna, Bologna, Italy)
Greta Castellini (Unit of Clinical Epidemiology, IRCCS Istituto Ortopedico Galeazzi, Milano, Italy)
Pillastrini Paolo (Department of Biomedical and Neuromotor Sciences, University of Bologna, Bologna, Italy)
Palese Alvisa (Department of Medicine, University of Udine, Udine, Italy)
Landuzzi Maria Gabriella (Department of Human Sciences, University of Verona, Verona, Italy)
Corradi Federica (University of Verona, School of Speech Therapy, Verona, Italy)
Rossettini Giacomo (Department of Physiotherapy, Faculty of Sport Sciences, Universidad Europea de Madrid, Calle Tajo s/n, Villaviciosa de Odón 28670, Spain)
Artificial Intelligence (AI) chatbots, like ChatGPT, Microsoft Bing, and Google Bard, emulate human interaction through multimodal communication. In physiotherapy, these chatbots assist in clinical reasoning and treatment decisions by suggesting symptoms to investigate, tests to perform, and more. In education, they enhance learning experiences and student engagement through intelligent feedback and tutoring. Understanding physiotherapy students’ knowledge and perceptions of AI chatbots is essential for their effective implementation in education and training. This study aims to investigate the knowledge, perceptions, use and barriers of AI chatbots among Italian physiotherapy students.
An online survey was distributed to physiotherapy students from 9 universities across Italy, by using Survey Monkey, from 21 February 2024, collecting responses through a self-assessment questionnaire of 24 questions on chatbot knowledge, use, perceptions, and barriers. The questionnaire was constructed in accordance with the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) and STrengthening the Reporting of OBservational Studies in Epidemiology guidelines (STROBE). Descriptive and frequency statistics analyzed the responses. This study is still ongoing.
Currently, a total of 552 students participated, averaging 22.4 ± 4 years old, with n=231 (40.7%) females. Of these, n=459 (94.4%) had heard of AI chatbots, primarily through social media, and n=256 (52.7%) had used them for academic purposes. Common uses included learning support (n=179, 36.8%), text summarization (n=81, 16.67%), translation (n=62, 12.76%), and problem-solving (n=62, 12.76%). Only 5% (n=24) used chatbots during internships. Students recognized limitations in handling complex tasks (n=294, 60.5%) and potential inaccuracies (n=282, 58%). The perceived benefits included time savings (n=299, 63.3%) and accessibility (n=278, 58.9%), while barriers included impairment of critical thinking (n=256, 55%) and misinformation risks (n=282, 60.6%).
The results of this study are preliminary as it is still ongoing. Most Italian physiotherapy students are aware of and use AI chatbots for academic purposes, recognizing their benefits and limitations. Rather than banning their use, developing guidelines and providing training for both faculty and students are recommended to ensure the proper application of chatbots in the university setting.
