JMIR Publications

ABSTRACT

Background:

Conversational agents (CAs) are a promising solution to support people to improve physical activity (PA) behaviours. However, there are a lack of CAs targeted at adolescents and which aim to provide support to overcome barriers to PA. This study reports the results of the co-design, development and evaluation of a prototype CA called Phyllis to support adolescents to overcome barriers to PA with the aim of improving PA behaviours. The study presents one of the first theory-driven conversational agent’s that utilise existing research, a theoretical framework and behaviour change model.

Objective:

The study used a mixed method approach to investigate the potential of a CA to support adolescents in overcoming barriers to PA and enhance their confidence and motivation to engage in PA. The hypothesis for the study is that a CA can support adolescents to overcome barriers to PA and be perceived by adolescents as being a tool to help them increase their confidence and motivation to participate in PA. The objectives of the study are: • Co-design a CA in collaboration with adolescents, incorporating Zhang's (2020) model and the theoretical framework proposed by Moore et al. (2023, under review), to assess its proof of concept. • Demonstrate the CA's ability to understand user input related to one of the 52 barriers to PA identified in the previous study (Moore, Vernon, et al., 2023). • Evaluate the usability and acceptability of the CA among adolescents. • Assess the perceived effectiveness of the solutions to barriers provided by the CA.

Methods:

The methodology involved co-design with eight adolescents to create a relational and persuasive conversational agent with a suitable persona and dialogue. The conversational agent was evaluated to determine acceptability, usability and effectiveness of the conversational agent with 46 adolescents participating in the study via an online survey.

Results:

Co-design participants were students aged 11 to 13, with a gender distribution of 56% female and 44% male, representing diverse ethnic backgrounds. Participants reported 37 specific barriers to PA and the most common barriers included 'lack of confidence,' 'fear of failure,' and a 'lack of motivation.' The CA's persona, named Phyllis, was co-designed with input from the students, reflecting their preferences for a friendly, understanding, and intelligent personality. Users engaged in 61 conversations with Phyllis, reported a positive user experience and 73% expressed a definite intention to use the fully functional CA in the future, with a Net Promoter Score indicating a high likelihood of recommendation. Phyllis also performed well recognising a range of different barriers to physical activity. The CA's persuasive capacity was evaluated in modules focusing on confidence and motivation, with a significant increase in students' agreement in feeling confident and motivated to engage in PA after interacting with Phyllis. Adolescents also expect to have a personalised experience and be able to personalise all aspects of the conversational agent.

Conclusions:

Results showed high acceptability and positive user experiences, indicating the CA's potential. Promising outcomes were observed in increasing confidence and motivation for PA. Further research and development are needed to create further interventions to address other barriers to PA and assess long-term behaviour change. Addressing concerns about bias and privacy is crucial to achieve acceptability in future. The CA's potential extends to healthcare systems and multi-modal support, providing valuable insights for designing digital health interventions including tackling global inactivity issues among adolescents.