Computer's human-likeness and its effects on willingness to disclose health-relevant information in healthy adults

Not Applicable

Completed

• Conditions: Self-disclosure of health behaviours; Self-disclosure of emotional events; Mental Health - Studies of normal psychology, cognitive function and behaviour

• Registration Number: ACTRN12622000676718
• Lead Sponsor: The University of Auckland

Brief Summary

Overall, this study found differences in health information disclosure to a virtual human, a chatbot, and an online questionnaire. First, with respect to the disclosure of sensitive information: (1) The digital human elicited the highest socially desirable responses to the perceived loneliness questions, and the online questionnaire group elicited the lowest, with the chatbot falling somewhere in between. Socially desirable responses in disclosing sexual risk behaviours did not differ between groups; (2) The proportion of participants who ever declined to answer a sensitive item was significantly higher in the digital human group compared to that in the chatbot group, with the online questionnaire falling somewhere in between; (3) The disclosure amounts of recent interpersonal conflicts and sadness did not differ by group, yet the disclosure amount of a recent stressful event was greater in the digital human group than the chatbot or online questionnaire group. Second, with respect to the disclosure of non-sensitive information: (1) Socially desirable responses to any of the non-sensitive close-ended items did not differ between groups; (2) The proportion of participants who ever declined to answer a non-sensitive item did not differ by group; (3) The disclosure amounts of all three positive emotional experiences (i.e., a recent happy event, happiness routines and grateful things in life) were greater in the digital human group than that in the online questionnaire group, with the chatbot group falling in between. Perceived anthropomorphism did not mediate the observed between-group differences in the self-disclosure outcomes. Perceived anthropomorphism towards the chatbot was higher than that towards the digital human or the online questionnaire. The thematic analysis identified that participants in each group had mixed experiences with their allocated method’s human likeness or the lack thereof. Several meaningful patterns arose: (1) Participants in all groups reported feeling more comfortable answering sensitive questions to a digital method than real humans due to lower judgement concerns. Yet, some participants in the digital human groups experienced social evaluative pressure when sharing sensitive information. This experience (i.e., social evaluative pressure) was not identified for those in the chatbot or online questionnaire groups; (2) Participants in both digital human and chatbot groups shared they felt a sense of rapport with the digital interviewers, whereas this aspect of the user experience was not identified for those in the online questionnaire group; (3) Participants in all groups expressed a strong desire for interpersonal connections when disclosing emotional experiences, especially positive emotional experiences that did not expose them to vulnerabilities or negative judgements; (4) The digital human induced some unintended consequences that were not desirable in the context of assessing health information. As well as social evaluative pressure, these consequences included causing disappointment by over-raising users’ expectations towards the technology, the uncanny valley effect, and a sense of “talking to no one”.

Detailed Description

Not available

Study Timeline

• Study Start: 2022-05-10
• Study Completion: 2022-09-30
• Last Update Posted: 2 September 2024

Recruitment & Eligibility

• Status: Completed
• Sex: All
• Target Recruitment: 160

Inclusion Criteria

The participants will be adults between 18 to 35 years old with English fluency (i.e., can speak, read, and write in fluent English).

Exclusion Criteria

Participants will be excluded from the study if they have hearing difficulties or vision loss (As these participants may need special assistance with using the computer or hearing the researcher. Due to the limited resource, this study is not equipped to provide such assistance).

Study & Design

• Study Type: Interventional
• Study Design: Not specified