View in EDS HTML Full Text PDF Full Text

Causal discovery and counterfactual reasoning to optimize persuasive dialogue policies.

Saved in:
Bibliographic Details
Title: Causal discovery and counterfactual reasoning to optimize persuasive dialogue policies.
Authors: Zeng, Donghuo (AUTHOR), Legaspi, Roberto (AUTHOR), Sun, Yuewen (AUTHOR), Dong, Xinshuai (AUTHOR), Ikeda, Kazushi (AUTHOR), Spirtes, Peter (AUTHOR), Zhang, Kun (AUTHOR)
Source: Behaviour & Information Technology. May2026, Vol. 45 Issue 8, p1515-1529. 15p.
Subjects: Reinforcement (Psychology), Architecture, Persuasion (Rhetoric), Medical informatics, Causal models, Learning, Behavior, Problem solving, Causality (Physics), Thought & thinking, Algorithms
Abstract: Tailoring persuasive conversations to users leads to more effective persuasion. However, existing dialogue systems often struggle to adapt to dynamically evolving user states. This paper presents a novel method that leverages causal discovery and counterfactual reasoning for optimising system persuasion capability and outcomes. We employ the Greedy Relaxation of the Sparsest Permutation (GRaSP) algorithm to identify causal relationships between user and system utterance strategies, treating user strategies as states and system strategies as actions. GRaSP identifies user strategies as causal factors influencing system responses, which inform a Bidirectional Conditional Generative Adversarial Network (BiCoGAN) in generating counterfactual utterances for the system. Subsequently, we use the Dueling Double Deep Q-Network (D3QN) model to utilise counterfactual data to determine the best policy for selecting system utterances. Our experiments with the PersuasionForGood dataset show measurable improvements in persuasion outcomes using our approach compared to baseline methods. The observed increase in cumulative rewards and Q-values highlights the effectiveness of causal discovery in enhancing counterfactual reasoning and optimising reinforcement learning policies for online dialogue systems. [ABSTRACT FROM AUTHOR]
Copyright of Behaviour & Information Technology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database: Psychology and Behavioral Sciences Collection
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
Text:
  Availability: 1
Header DbId: pbh
DbLabel: Psychology and Behavioral Sciences Collection
An: 193710028
AccessLevel: 6
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Causal discovery and counterfactual reasoning to optimize persuasive dialogue policies.
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Zeng%2C+Donghuo%22">Zeng, Donghuo</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Legaspi%2C+Roberto%22">Legaspi, Roberto</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Sun%2C+Yuewen%22">Sun, Yuewen</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Dong%2C+Xinshuai%22">Dong, Xinshuai</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Ikeda%2C+Kazushi%22">Ikeda, Kazushi</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Spirtes%2C+Peter%22">Spirtes, Peter</searchLink> (AUTHOR)<br /><searchLink fieldCode="AR" term="%22Zhang%2C+Kun%22">Zhang, Kun</searchLink> (AUTHOR)
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="JN" term="%22Behaviour+%26+Information+Technology%22">Behaviour & Information Technology</searchLink>. May2026, Vol. 45 Issue 8, p1515-1529. 15p.
– Name: Subject
  Label: Subjects
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Reinforcement+%28Psychology%29%22">Reinforcement (Psychology)</searchLink><br /><searchLink fieldCode="DE" term="%22Architecture%22">Architecture</searchLink><br /><searchLink fieldCode="DE" term="%22Persuasion+%28Rhetoric%29%22">Persuasion (Rhetoric)</searchLink><br /><searchLink fieldCode="DE" term="%22Medical+informatics%22">Medical informatics</searchLink><br /><searchLink fieldCode="DE" term="%22Causal+models%22">Causal models</searchLink><br /><searchLink fieldCode="DE" term="%22Learning%22">Learning</searchLink><br /><searchLink fieldCode="DE" term="%22Behavior%22">Behavior</searchLink><br /><searchLink fieldCode="DE" term="%22Problem+solving%22">Problem solving</searchLink><br /><searchLink fieldCode="DE" term="%22Causality+%28Physics%29%22">Causality (Physics)</searchLink><br /><searchLink fieldCode="DE" term="%22Thought+%26+thinking%22">Thought & thinking</searchLink><br /><searchLink fieldCode="DE" term="%22Algorithms%22">Algorithms</searchLink>
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Tailoring persuasive conversations to users leads to more effective persuasion. However, existing dialogue systems often struggle to adapt to dynamically evolving user states. This paper presents a novel method that leverages causal discovery and counterfactual reasoning for optimising system persuasion capability and outcomes. We employ the Greedy Relaxation of the Sparsest Permutation (GRaSP) algorithm to identify causal relationships between user and system utterance strategies, treating user strategies as states and system strategies as actions. GRaSP identifies user strategies as causal factors influencing system responses, which inform a Bidirectional Conditional Generative Adversarial Network (BiCoGAN) in generating counterfactual utterances for the system. Subsequently, we use the Dueling Double Deep Q-Network (D3QN) model to utilise counterfactual data to determine the best policy for selecting system utterances. Our experiments with the PersuasionForGood dataset show measurable improvements in persuasion outcomes using our approach compared to baseline methods. The observed increase in cumulative rewards and Q-values highlights the effectiveness of causal discovery in enhancing counterfactual reasoning and optimising reinforcement learning policies for online dialogue systems. [ABSTRACT FROM AUTHOR]
– Name: AbstractSuppliedCopyright
  Label:
  Group: Ab
  Data: <i>Copyright of Behaviour & Information Technology is the property of Taylor & Francis Ltd and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract.</i> (Copyright applies to all Abstracts.)
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=pbh&AN=193710028
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1080/0144929X.2025.2478276
    Languages:
      – Code: eng
        Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 15
        StartPage: 1515
    Subjects:
      – SubjectFull: Reinforcement (Psychology)
        Type: general
      – SubjectFull: Architecture
        Type: general
      – SubjectFull: Persuasion (Rhetoric)
        Type: general
      – SubjectFull: Medical informatics
        Type: general
      – SubjectFull: Causal models
        Type: general
      – SubjectFull: Learning
        Type: general
      – SubjectFull: Behavior
        Type: general
      – SubjectFull: Problem solving
        Type: general
      – SubjectFull: Causality (Physics)
        Type: general
      – SubjectFull: Thought & thinking
        Type: general
      – SubjectFull: Algorithms
        Type: general
    Titles:
      – TitleFull: Causal discovery and counterfactual reasoning to optimize persuasive dialogue policies.
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Zeng, Donghuo
      – PersonEntity:
          Name:
            NameFull: Legaspi, Roberto
      – PersonEntity:
          Name:
            NameFull: Sun, Yuewen
      – PersonEntity:
          Name:
            NameFull: Dong, Xinshuai
      – PersonEntity:
          Name:
            NameFull: Ikeda, Kazushi
      – PersonEntity:
          Name:
            NameFull: Spirtes, Peter
      – PersonEntity:
          Name:
            NameFull: Zhang, Kun
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 15
              M: 05
              Text: May2026
              Type: published
              Y: 2026
          Identifiers:
            – Type: issn-print
              Value: 0144929X
          Numbering:
            – Type: volume
              Value: 45
            – Type: issue
              Value: 8
          Titles:
            – TitleFull: Behaviour & Information Technology
              Type: main
ResultId 1