Pervasive Health Workshop on Affective Interaction with Virtual Assistants within the Healthcare Context'

Research Article

Designing personalized prompts for a virtual assistant to support elderly care home residents

  • @INPROCEEDINGS{10.4108/eai.16-5-2016.2263874,
        author={Alexandra K\o{}nig and Aarti Malhotra and Jesse Hoey and Linda Francis},
        title={Designing personalized prompts for a virtual assistant to support elderly care home residents},
        proceedings={Pervasive Health Workshop on Affective Interaction with Virtual Assistants within the Healthcare Context'},
        publisher={ACM},
        proceedings_a={AIVAHC2016},
        year={2016},
        month={6},
        keywords={assistive technology; gerontechnology; clinical research; qualitative interviews; user feedback; affective computing intelligent interactive systems; affective interaction; virtual assistant; prompting; support in adl; elderly care; dementia},
        doi={10.4108/eai.16-5-2016.2263874}
    }
    
  • Alexandra König
    Aarti Malhotra
    Jesse Hoey
    Linda Francis
    Year: 2016
    Designing personalized prompts for a virtual assistant to support elderly care home residents
    AIVAHC2016
    EAI
    DOI: 10.4108/eai.16-5-2016.2263874
Alexandra König1,*, Aarti Malhotra1, Jesse Hoey1, Linda Francis2
  • 1: University of Waterloo, Waterloo, Canada
  • 2: Cleveland State University, USA
*Contact email: akonig@uwaterloo.ca

Abstract

In this paper, we present the first results of the ACT@HOME research project which aims to develop an an artificially intelligent virtual assistant (VA) to engage and help older adults with Alzheimer’s disease (AD) to complete activities of daily living (ADL) more independently. In order to define the most appropriate prompting style for each user profile, we performed 12 semi structured qualitative interviews with dyads of elderly care home residents and their family caregiver. During these interviews, we presented the virtual assistant and the different ‘static’ prompts to support people in the activity of hand washing. We gathered as much feedback and suggestions as possible coming directly from the end users about how to improve the provided prompts and thus, increase acceptability. The results are presented around three extracted themes: a) comments on current design of the virtual assistant, b) perceived usefulness, user adoption and c) suggestions for improvements. These should guide in the future developers of assistive technology to support elderly care home residents.