Modelling prospective memory and resilient situated communications via Wizard of Oz

Many of these services necessitate that the robots can communicate and interact with users. However, to interact with robots fluently can be a challenge, for reasons such as an inappropriate mental model of the robot [8]. For example, based on the wide range of types of sensors of the robot (e.g., camera, radar, Infra-red detector, microphone, speaker, etc.), users may assume they have multimodal communication capabilities and even expect that the robot can remember them and recall details of their previous interactions[7]. Furthermore, speech may have a variety of accents, dialects, grammatical faults, and disfluencies, making interaction more difficult. This is one of the reasons why much research on humanrobot interaction(HRI) has focused on short-term interactions and relied on Wizard of Oz or"constrained rule-based" methods to get around these issues[2, 3]. However, communication failures may also happen in these kinds of sets up, especially with elderly participants. This abstract presents a scenario for human-robot action in a home setting involving an older adult and a robot. The scenario is designed to explore the envisioned modelling of memory for communication with a SAR. The scenario will enable the gathering of data on failures of speech technology and human-robot communication involving shared memory that may occur during daily activities such as a music-listening activity.