Aim: To improve access to cognitive testing for older adults, the reliability and acceptability of a speech-based cognitive test administered by a social robot were investigated. Methods:The Japanese version of the Telephone Interview for Cognitive Status was administered by a social robot to participants recruited from retirement homes and adult daycare facilities. The robot's dialogue and gestures were preprogrammed, while the researcher controlled the timing of proceeding to the next question and scored participants' responses. We examined the internal consistency, alternate form reliability (experiment 1) and test-retest reliability (experiment 2) of the cognitive test. The acceptability of the cognitive test was also examined using a questionnaire in experiment 2.Results: A total of 66 individuals (mean age 81.2 AE 5.8 years) participated in experiment 1; the internal consistency (Cronbach's α) of the test was 0.691 and its alternate form reliability (measured by interclass correlation coefficient) was 0.728. A total of 40 of these individuals (mean age 82.0 AE 5.4 years) also participated in experiment 2, and the test-retest reliability was 0.818. According to the questionnaire responses, over half of the participants wanted (or very much wanted) to use the robot version of the test to measure the deterioration of their cognitive function. Conclusions:A robot-administered cognitive test might have satisfactory reliability and acceptability to community-dwelling older adults if those aspects of the test implemented by the researcher can also be successfully automated.
Personal communication robots are expected to assist daily living of elderly people. Aiming at developing computerized cognitive assessment systems, we collected human-robot spoken dialog of a cognitive impairment test scenario based on TICS (Telephone Interview for Cognitive Status) and COGNISTAT (Cognitive Status Examination) in Japanese. For the efficient acquisition of the spoken dialog corpus of this scenario, we implemented a WOz (Wizard of Oz) style spoken dialog system on a commercial personal robot, PaPeRo by NEC. By using this system, we collected 147 dialogs spoken by 48 elderlies whose ages varied about 75-85 and MMSE (Mini Mental State Examination) scores varied around 26.5. Each dialog took about 30 min, and contains around 100 human utterances. In order to evaluate feasibility of automatic assessment, we conducted speech recognition experiments with the speech corpus. In the recognition experiments of the elderly speech of answering to the temporal orientation test of asking today's month and date, the accuracies of discriminating correctness of the answers exceeds 80% by using speech dictation engines. This preliminary results show potential feasibilities for computerized cognitive assessment systems.
The purpose of this study was to reveal comprehensible instructions from an assistive robot for older adults, across cognitive levels and characteristics. Participants included 19 older adults with or without cognitive impairment. We administered cognitive tests assessing all major domains (e.g., memory and attention). Participants were required to listen to robot instructions carefully, and perform three activities of daily living (e.g., taking medicine) with three different types of instructions. In instruction pattern 1 (IP1), the robot informed seniors of the task in one sentence, while in instruction patterns 2 and 3 (IP2 and IP3), the steps of each activity were split into two and three sentences, respectively. Participants with lower cognitive level showed lower task performance with IP1, whereas almost all participants completed tasks with IP2 and IP3. Cognitive domains such as working memory significantly affected task performances. Participants with lower attention made mistakes in taking their medicine. The results imply that step-by-step instructions should be used for older people with lower levels of cognitive function, especially working memory, and repeated instructions may be required for lower attention. Types of instruction should be selected depending on cognitive characteristics.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.