Vol.:(0123456789) 1 3 Journal of Ambient Intelligence and Humanized Computing https://doi.org/10.1007/s12652-019-01423-x ORIGINAL RESEARCH Elder user’s attitude toward assistive virtual agents: the role of voice and gender Anna Esposito 1  · Terry Amorese 1  · Marialucia Cuciniello 1  · Maria Teresa Riviello 1  · Antonietta M. Esposito 2  · Alda Troncone 1  · Maria Inés Torres 3  · Stephan Schlögl 4  · Gennaro Cordasco 1 Received: 19 November 2018 / Accepted: 2 August 2019 © Springer-Verlag GmbH Germany, part of Springer Nature 2019 Abstract This paper reports on experiments devoted to explore the role of specifc attributes of humanoid virtual agents that may infuence elderly users’ perception and attitude, determining their acceptance and adoption as assistive devices. In particular, it investigates elderly preference on agents’ gender and the role of the agents’ ability to use voice during the interaction. To this aim two diferent groups of seniors were involved in the experiments. The frst group evaluated talking virtual agents, the second one the same virtual agents, but silenced. The data shows that elderly users, independently from their gender, prefer to interact with female agents, especially when they are able to talk to them, revealing the role played by the voice. Furthermore, it was found a signifcant efect of the elderly level of experience with technology: when interacting with agents with voice, elderly users with high technological experience were less interested and considered the proposed agents less attractive and appealing, while just the opposite occurred when interacting with silenced agents. 1 Introduction The continuous and rapid development in information and communication technology (ICT) produces innovations in several areas, many of these increasingly involves devices and applications that aims at promoting health and wellbe- ing for elderly people (Mansouri et al. 2017; Bonaccorsi et al. 2016; Pollack 2005). Intelligent technology for sup- porting eldercare, indeed, has become crucial in the recent decades considering the ageing population is growing sig- nifcantly all over the world (United Nations report 2017), resulting in dramatic shortage of qualifed care personnel (https://eldercareworkforce.org/workforce-short age/) and increase of expenditures for national health care institutions (Rowe et al. 2016). Therefore, the challenge is to develop new technological solutions to support elderly individuals to stay healthy and active, enabling them to remain at home and live independently to the maximum possible extent (Chris- tophorou et al. 2016). Among these solutions, complex autonomous interfaces, such as virtual agents, are developed to support elderly in their everyday life, promoting health behaviors, such as fol- lowing medical treatments, e.g. reminding to take medicines or do exercises, assisting them in their day to day tasks, as well as encouraging them to create and/or maintaining social contacts (Katrin and Michael 2014; Yaghoubzadeh et al. 2013; Ring et al. 2013). In defining and developing assistive virtual agents account has to be taken of the specifc needs and require- ments of the elderly users, in order to ensure accessibility, usability and usefulness, that are primary factors infuencing adoption of technology by older individuals (Heerink et al. 2010). The presence of these factors, however, does not pre- dict the efective use of assistive technologies among elderly (Goher et al. 2017; Broadbent et al. 2012). Some character- istics of assistive virtual agents’ design (e.g. appearance, attributes, etc.) infuence users’ perception and attitude and consequently afect users’ behavior, in terms of adoption and correct use of technology (Díaz-Boladeras et al. 2013). Hence, it is of interest to explore which are virtual agents’ features that move elder users’ preference when interacting * Anna Esposito iiass.annaesp@tin.it; anna.esposito@unicampania.it 1 Department of Psychology and IIASS, Università degli Studi della Campania “L. Vanvitelli”, Caserta, Italy 2 Istituto Nazionale di Geofsica e Vulcanologia, Sez. di Napoli Osservatorio Vesuviano, Naples, Italy 3 Speech Interactive Research Group, Universidad del País Vasco UPV/EHU, Bilbao, Spain 4 Department of Management, Communication and IT, MCI Management Center Innsbruck, Innsbruck, Austria