Grant: Travel Awards - Institute Community Support

The expanding aging population and the impending demand on healthcare and economic resources has led to a significant increase in research pertaining to the development of advanced technologies to support healthy aging. Notably, the interfaces used in these technologies increasingly involve virtual communicative agents (e.g., GoogleHome, smart home interfaces, etc.). However, the design focus has mainly been on appearance and acoustics rather than the cognitive processes that underlie successful communication. For example, the ability to take the perspective of a communicative partner is essential for interpreting words such as "my", "left", and "that". Although there are multiple perspective cues available in human-human communication (e.g., gaze cues, gestures, shared knowledge), these are often absent or limited in human-computer interaction. Thus, being able to follow a virtual agent's instructions may involve extra demands on perspective taking, or may require a change to the kinds of language used. The proposed research will involve a series of studies examining older adults' ability to take the perspective of a virtual agent while following spoken instructions to complete everyday tasks (e.g., preparing a sandwich). Eye movement measures will be used to track the time-course and efficiency of listeners' real-time comprehension. Drawing from psycholinguistic literature, we will manipulate factors that have shown to aid or challenge comprehension in human-human interaction such as word choice, sentence structure, speech rate, pauses, and feedback. This methodology allows for a fine grained analysis of communicative perspective taking, enabling improvements in human-computer interaction that would enhance older adults' experience with such assistive tools. My research results will therefore inform existing efforts for designing assistive tools that can enhance the physical and mental well-being of older adults.