Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Before and After ChatGPT: Revisiting AI-Based Dialogue Systems for Emotional Support

Daeun Lee, Dongje Yoo, Migyeong Yang, Jihyun An, Christine B. Cha, Jinyoung Han

Abstract

Mental health remains a major public health concern, while access to timely psychological support is often limited. AI-based dialogue systems have emerged as promising tools to address these barriers, and recent advances in large language models (LLMs) have significantly transformed this research area. However, a systematic understanding of this technological transition is still limited. This study reviews the technological evolution of AI-driven dialogue systems for mental health, focusing on the shift from task-specific deep learning models to LLM-based approaches. We conducted a bibliometric analysis and qualitative trend review of studies published between 2020 and May 2024 using Web of Science, Scopus, and the ACM Digital Library. The qualitative analysis compared research conducted before and after the widespread adoption of LLMs. Pre-LLM research was represented by highly cited studies and work based on the ESConv dataset, while post-LLM research included highly cited dialogue systems built on LLMs. A total of 146 studies met the inclusion criteria, showing a steady growth in publications over time. Before the widespread use of LLMs, empathetic response generation mainly relied on task-specific deep learning models. Highly cited and ESConv-based studies commonly focused on multi-task learning and the integration of external knowledge. In contrast, recent LLM-based dialogue systems demonstrate improved linguistic flexibility and generalization for emotional support. However, these systems also raise concerns related to reliability and safety in mental health applications. This review highlights the technological transition of AI-based dialogue systems for mental health in the LLM era. By identifying current research trends and limitations, the findings provide guidance for developing more effective and reliable AI-driven counseling systems.

Before and After ChatGPT: Revisiting AI-Based Dialogue Systems for Emotional Support

Abstract

Mental health remains a major public health concern, while access to timely psychological support is often limited. AI-based dialogue systems have emerged as promising tools to address these barriers, and recent advances in large language models (LLMs) have significantly transformed this research area. However, a systematic understanding of this technological transition is still limited. This study reviews the technological evolution of AI-driven dialogue systems for mental health, focusing on the shift from task-specific deep learning models to LLM-based approaches. We conducted a bibliometric analysis and qualitative trend review of studies published between 2020 and May 2024 using Web of Science, Scopus, and the ACM Digital Library. The qualitative analysis compared research conducted before and after the widespread adoption of LLMs. Pre-LLM research was represented by highly cited studies and work based on the ESConv dataset, while post-LLM research included highly cited dialogue systems built on LLMs. A total of 146 studies met the inclusion criteria, showing a steady growth in publications over time. Before the widespread use of LLMs, empathetic response generation mainly relied on task-specific deep learning models. Highly cited and ESConv-based studies commonly focused on multi-task learning and the integration of external knowledge. In contrast, recent LLM-based dialogue systems demonstrate improved linguistic flexibility and generalization for emotional support. However, these systems also raise concerns related to reliability and safety in mental health applications. This review highlights the technological transition of AI-based dialogue systems for mental health in the LLM era. By identifying current research trends and limitations, the findings provide guidance for developing more effective and reliable AI-driven counseling systems.
Paper Structure (21 sections, 5 figures, 10 tables)

This paper contains 21 sections, 5 figures, 10 tables.

Table of Contents

  1. Introduction
  2. Method
  3. Data Sources and Search Strategy
  4. Eligibility Criteria
  5. Study Screening and Selection
  6. Analysis Methodology
  7. Result
  8. Overall Publication Trend
  9. Pre-LLM Approaches: Task-Specific Deep Learning Models
  10. Overview of Highly Cited Top 10 Publications
  11. Overview of Publications using Emotional Support Conversation (ESConv) Dataset
  12. Post-LLM Approaches: Large Language Model–Based Dialogue Systems
  13. Discussion
  14. Implications & Recommendations
  15. The Lack of Training Dataset for Counseling System Development
  16. ...and 6 more sections

Figures (5)

  • Figure 1: Search Query Categories with Results
  • Figure 2: PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flowchart moher2010preferred
  • Figure 3: Keyword Co-occurrence Network Graph from 2020 to May 2024
  • Figure 4: Yearly Keyword Co-occurrence Network Graphs (2020–2024)
  • Figure 5: We present figures from the original paper liu2021towards to illustrate three stages of the ESConv framework and the corresponding eight counseling strategies. According to liu2021towards, this framework comprises three stages, each with specific support strategies. The exploration stage aims to help individuals identify underlying issues; the comforting stage focuses on providing empathy and understanding; and the action stage involves offering practical information or suggestions. Typically, the emotional support process follows a sequential order from 1. Exploration → 2. Comforting → 3. Action, as indicated by black arrows, though it can also be adjusted to suit the conversation's needs, as represented by dashed gray arrows.