Presentation
DH10 - Healthcare Chatbots: Effects of Information Quantity and Adaptability on User Trust and Usability
SessionPoster Session 1
DescriptionIntroduction
Healthcare chatbots offer an accessible and quick alternative to individuals who are unable to seek health services due to cost, location, or stigma. However, despite the growing use of healthcare chatbots, it is still not clear how best to present information to avoid information overload while maximizing trust and usability. The goal of this study is to explore the benefits and limitations of an adaptable approach to information presentation in healthcare chatbots. The research questions (RQs) in this study are:
1) RQ1: How does the amount of information presented within a healthcare chatbot influence user trust and usability?
2) RQ2: How does the ability to control the amount of information (adaptability) presented within a healthcare chatbot affect user trust and usability?
Methods
Participants will be asked to read a sample medical scenario and interact with different healthcare chatbot prototypes. These prototypes will be modeled after existing chatbots and will contain either low, high, or adaptable medical information. For the adaptable prototypes, participants will have the option to click buttons to see more information if they choose. Users’ eye movements will be tracked to understand where their attention was focused and for how long. Participants’ interaction with the chatbot will also be recorded and they will be asked to complete a knowledge check after each trial. At the completion of the experiment, measures of trust, usability, and workload will also be collected.
Results and key takeaway points
The findings will contribute to the human factors’ literature on information overload and how that relates to the design of interfaces that support healthcare diagnoses. In particular, the findings will fill a gap in relation to the effects of information overload on user trust and usability in healthcare chatbots, and whether adopting an adaptable approach can mitigate the effects of information overload.
Healthcare chatbots offer an accessible and quick alternative to individuals who are unable to seek health services due to cost, location, or stigma. However, despite the growing use of healthcare chatbots, it is still not clear how best to present information to avoid information overload while maximizing trust and usability. The goal of this study is to explore the benefits and limitations of an adaptable approach to information presentation in healthcare chatbots. The research questions (RQs) in this study are:
1) RQ1: How does the amount of information presented within a healthcare chatbot influence user trust and usability?
2) RQ2: How does the ability to control the amount of information (adaptability) presented within a healthcare chatbot affect user trust and usability?
Methods
Participants will be asked to read a sample medical scenario and interact with different healthcare chatbot prototypes. These prototypes will be modeled after existing chatbots and will contain either low, high, or adaptable medical information. For the adaptable prototypes, participants will have the option to click buttons to see more information if they choose. Users’ eye movements will be tracked to understand where their attention was focused and for how long. Participants’ interaction with the chatbot will also be recorded and they will be asked to complete a knowledge check after each trial. At the completion of the experiment, measures of trust, usability, and workload will also be collected.
Results and key takeaway points
The findings will contribute to the human factors’ literature on information overload and how that relates to the design of interfaces that support healthcare diagnoses. In particular, the findings will fill a gap in relation to the effects of information overload on user trust and usability in healthcare chatbots, and whether adopting an adaptable approach can mitigate the effects of information overload.
Event Type
Poster Presentation
TimeMonday, March 314:45pm - 6:15pm EDT
LocationFrontenac Foyer
Digital Health (DH)
Simulation and Education (SE)
Hospital Environments (HE)
Medical and Drug Delivery Devices (MDD)
Patient Safety and Research Initiatives (PS)