In this paper, we propose a new interview method of eliciting needs for support completely and accurately from people with early-stage dementia for identifying their real needs, with the eventual objective of designing effective Assistive Technology. The interview procedure consists of the following steps: (1) entirely identifying tasks of a subject's everyday living, leisure and social activities (referred to, in all, as "social living activities"), (2) evaluating the subject's confusion levels (SCLs) about identified tasks based on the subject's self-rating, together with the caregiver's rating as a reference, and confusion level discrepancies (CLDs) are found, and (3) predicting the subject's support requirement levels (SRLs) about confusing tasks based on confusion levels followed by the subject's revision to thereby find SRL discrepancies (SRLDs). Subjects are asked to reconfirm SRL ratings associated with tasks having SRLDs and/or encouraged to raise the ratings of tasks having CLDs to accurately identify the subject's SRLs. Six subject-caregiver dyads were interviewed and 22 support needs, including 10 needs that were extracted by reconfirmation or encouragement, were elicited from four subjects. These elicited needs covered the entire social living scene from domestic affairs to social activities, showing that the proposed method was effective in systematically eliciting support needs from people with early-stage dementia. No support needs were elicited, however, from subjects with very little awareness of memory impairment, demonstrating one limitation of the proposed method.
All Science Journal Classification (ASJC) codes
- Computer Science(all)
- Electrical and Electronic Engineering