Exploring the impact of robot-delivered interpretation bias modification on depressed young adults in Hong Kong : a mixed-methods iterative approach

Abstract

Depression is a significant mental health issue affecting young adults in Hong Kong. Social robots offer a promising platform for delivering accessible and engaging interventions. This thesis explores the impact of robot modalities (text, audio, video) on user perceptions, experiences, and outcomes of an online imagery-enhanced elaborative interpretation bias modification (eiIBM) program for depressed young adults in Hong Kong. The research employs an iterative, scientific-centered approach across three studies. Study 1, a between-subjects experiment, examines differences in user perceptions, experiences, and outcomes between varied robot modalities and a no-robot control. Study 2, a within-subject interview, explores reasons for the differences or similarities found in Study 1 and elicits additional information on modality preferences. Study 3 improves the eiIBM program based on insights from Studies 1 and 2 and re-examines the effects of robot modalities. The findings contribute to theoretical models of technology acceptance for healthcare robotics (relational agents), inform design principles for effectively incorporating social robots into digital mental health interventions, and offer implications for developing accessible and research-centered robot-assisted therapies that promote cognitive resilience among vulnerable populations. This thesis advances knowledge on the nuances of how robot modalities shape user experiences and therapeutic outcomes, guiding the development of future AI-powered mental health solutions.