Collaborating digital twins for health coaching

Sana Salman; Deborah Richards

doi:10.1007/978-3-031-70415-4_20

Collaborating digital twins for health coaching

Sana Salman, Deborah Richards^*

^*Corresponding author for this work

School of Computing

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

Abstract

Embodied conversational agents (ECAs) are digital characters that behave like humans and utter humanlike dialogues. They are digital twins of a human coach and are cost effective and reachable as compared to human coaches who typically are costly and have long wait times for appointment. To support a healthy life-style, multiple health experts may be needed and our multi-agent digital twins give access to all the coaches in the same session. To provide motivation and encourage following the advice given, these coaches use relational cues in their dialogues including empowerment, working alliance and affirmation cues which are found in actual human-human coaching sessions. Our digital twins simulate three coaches who are experts in diet, physical activity and cognitive health which collaborate with each other, and the human, to provide a holistic coaching experience. This paper reports on our use of Generative AI to modify neutral dialogues with these relational cues. We recommend that both automated and human validation be undertaken in the context of health advice.

Original language	English
Title of host publication	Advances in Practical Applications of Agents, Multi-Agent Systems, and Digital Twins
Subtitle of host publication	The PAAMS Collection : 22nd International Conference, PAAMS 2024, proceedings
Editors	Philippe Mathieu, Fernando De la Prieta
Place of Publication	Cham
Publisher	Springer, Springer Nature
Pages	228-239
Number of pages	12
ISBN (Electronic)	9783031704154
ISBN (Print)	9783031704147
DOIs	https://doi.org/10.1007/978-3-031-70415-4_20
Publication status	Published - 2025
Event	22nd International Conference on Practical Applications of Agents and Multi-Agent Systems, PAAMS 2024 - Salamanca, Spain Duration: 26 Jun 2024 → 28 Jun 2024

Publication series

Name	Lecture Notes in Artificial Intelligence
Publisher	Springer
Volume	15157
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	22nd International Conference on Practical Applications of Agents and Multi-Agent Systems, PAAMS 2024
Country/Territory	Spain
City	Salamanca
Period	26/06/24 → 28/06/24

Keywords

ECA
relational cue
digital twin
Generative AI

Access to Document

10.1007/978-3-031-70415-4_20

Cite this

Salman, S., & Richards, D. (2025). Collaborating digital twins for health coaching. In P. Mathieu, & F. De la Prieta (Eds.), Advances in Practical Applications of Agents, Multi-Agent Systems, and Digital Twins: The PAAMS Collection : 22nd International Conference, PAAMS 2024, proceedings (pp. 228-239). (Lecture Notes in Artificial Intelligence; Vol. 15157). Springer, Springer Nature. https://doi.org/10.1007/978-3-031-70415-4_20

Salman, Sana ; Richards, Deborah. / Collaborating digital twins for health coaching. Advances in Practical Applications of Agents, Multi-Agent Systems, and Digital Twins: The PAAMS Collection : 22nd International Conference, PAAMS 2024, proceedings. editor / Philippe Mathieu ; Fernando De la Prieta. Cham : Springer, Springer Nature, 2025. pp. 228-239 (Lecture Notes in Artificial Intelligence).

@inproceedings{7945f1c6f8344287a8b5c51516854b87,

title = "Collaborating digital twins for health coaching",

abstract = "Embodied conversational agents (ECAs) are digital characters that behave like humans and utter humanlike dialogues. They are digital twins of a human coach and are cost effective and reachable as compared to human coaches who typically are costly and have long wait times for appointment. To support a healthy life-style, multiple health experts may be needed and our multi-agent digital twins give access to all the coaches in the same session. To provide motivation and encourage following the advice given, these coaches use relational cues in their dialogues including empowerment, working alliance and affirmation cues which are found in actual human-human coaching sessions. Our digital twins simulate three coaches who are experts in diet, physical activity and cognitive health which collaborate with each other, and the human, to provide a holistic coaching experience. This paper reports on our use of Generative AI to modify neutral dialogues with these relational cues. We recommend that both automated and human validation be undertaken in the context of health advice.",

keywords = "ECA, relational cue, digital twin, Generative AI",

author = "Sana Salman and Deborah Richards",

year = "2025",

doi = "10.1007/978-3-031-70415-4\_20",

language = "English",

isbn = "9783031704147",

series = "Lecture Notes in Artificial Intelligence",

publisher = "Springer, Springer Nature",

pages = "228--239",

editor = "Philippe Mathieu and \{De la Prieta\}, Fernando",

booktitle = "Advances in Practical Applications of Agents, Multi-Agent Systems, and Digital Twins",

address = "United States",

note = "22nd International Conference on Practical Applications of Agents and Multi-Agent Systems, PAAMS 2024 ; Conference date: 26-06-2024 Through 28-06-2024",

}

Salman, S & Richards, D 2025, Collaborating digital twins for health coaching. in P Mathieu & F De la Prieta (eds), Advances in Practical Applications of Agents, Multi-Agent Systems, and Digital Twins: The PAAMS Collection : 22nd International Conference, PAAMS 2024, proceedings. Lecture Notes in Artificial Intelligence, vol. 15157, Springer, Springer Nature, Cham, pp. 228-239, 22nd International Conference on Practical Applications of Agents and Multi-Agent Systems, PAAMS 2024, Salamanca, Spain, 26/06/24. https://doi.org/10.1007/978-3-031-70415-4_20

Collaborating digital twins for health coaching. / Salman, Sana ; Richards, Deborah.
Advances in Practical Applications of Agents, Multi-Agent Systems, and Digital Twins: The PAAMS Collection : 22nd International Conference, PAAMS 2024, proceedings. ed. / Philippe Mathieu; Fernando De la Prieta. Cham: Springer, Springer Nature, 2025. p. 228-239 (Lecture Notes in Artificial Intelligence; Vol. 15157).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - Collaborating digital twins for health coaching

AU - Salman, Sana

AU - Richards, Deborah

PY - 2025

Y1 - 2025

N2 - Embodied conversational agents (ECAs) are digital characters that behave like humans and utter humanlike dialogues. They are digital twins of a human coach and are cost effective and reachable as compared to human coaches who typically are costly and have long wait times for appointment. To support a healthy life-style, multiple health experts may be needed and our multi-agent digital twins give access to all the coaches in the same session. To provide motivation and encourage following the advice given, these coaches use relational cues in their dialogues including empowerment, working alliance and affirmation cues which are found in actual human-human coaching sessions. Our digital twins simulate three coaches who are experts in diet, physical activity and cognitive health which collaborate with each other, and the human, to provide a holistic coaching experience. This paper reports on our use of Generative AI to modify neutral dialogues with these relational cues. We recommend that both automated and human validation be undertaken in the context of health advice.

AB - Embodied conversational agents (ECAs) are digital characters that behave like humans and utter humanlike dialogues. They are digital twins of a human coach and are cost effective and reachable as compared to human coaches who typically are costly and have long wait times for appointment. To support a healthy life-style, multiple health experts may be needed and our multi-agent digital twins give access to all the coaches in the same session. To provide motivation and encourage following the advice given, these coaches use relational cues in their dialogues including empowerment, working alliance and affirmation cues which are found in actual human-human coaching sessions. Our digital twins simulate three coaches who are experts in diet, physical activity and cognitive health which collaborate with each other, and the human, to provide a holistic coaching experience. This paper reports on our use of Generative AI to modify neutral dialogues with these relational cues. We recommend that both automated and human validation be undertaken in the context of health advice.

KW - ECA

KW - relational cue

KW - digital twin

KW - Generative AI

UR - https://www.scopus.com/pages/publications/85210323104

U2 - 10.1007/978-3-031-70415-4_20

DO - 10.1007/978-3-031-70415-4_20

M3 - Conference proceeding contribution

AN - SCOPUS:85210323104

SN - 9783031704147

T3 - Lecture Notes in Artificial Intelligence

SP - 228

EP - 239

BT - Advances in Practical Applications of Agents, Multi-Agent Systems, and Digital Twins

A2 - Mathieu, Philippe

A2 - De la Prieta, Fernando

PB - Springer, Springer Nature

CY - Cham

T2 - 22nd International Conference on Practical Applications of Agents and Multi-Agent Systems, PAAMS 2024

Y2 - 26 June 2024 through 28 June 2024

ER -

Salman S , Richards D. Collaborating digital twins for health coaching. In Mathieu P, De la Prieta F, editors, Advances in Practical Applications of Agents, Multi-Agent Systems, and Digital Twins: The PAAMS Collection : 22nd International Conference, PAAMS 2024, proceedings. Cham: Springer, Springer Nature. 2025. p. 228-239. (Lecture Notes in Artificial Intelligence). doi: 10.1007/978-3-031-70415-4_20

Collaborating digital twins for health coaching

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Using large language models to embed relational cues in the dialogue of collaborating digital twins

Cite this