Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

PITCH: Productivity and Mental Well-being Coaching through Daily Conversational Interaction

Adnan Abbas, Sang Won Lee

TL;DR

The paper tackles the challenge of improving productivity and mental well-being through conversational AI-assisted planning. It introduces PITCH, an LLM-powered agent that externalizes daily tasks in the morning and prompts reflection in the evening, augmented by a rotation-based and context-aware prompting strategy to sustain engagement. The authors describe a Slack-based prototype, a within-subjects field study comparing static and rotation-enabled interventions, and mixed-method evaluation plans to assess productivity and well-being impacts. If effective, this approach could provide scalable, personalized daily coaching that enhances time management and healthier work-life boundaries in real-world settings.

Abstract

Efficient task planning is essential for productivity and mental well-being, yet individuals often struggle to create realistic plans and reflect upon their productivity. Leveraging the advancement in artificial intelligence (AI), conversational agents have emerged as a promising tool for enhancing productivity. Our work focuses on externalizing plans through conversation, aiming to solidify intentions and foster focused action, thereby positively impacting their productivity and mental well-being. We share our plan of designing a conversational agent to offer insightful questions and reflective prompts for increasing plan adherence by leveraging the social interactivity of natural conversations. Previous studies have shown the effectiveness of such agents, but many interventions remain static, leading to decreased user engagement over time. To address this limitation, we propose a novel rotation and context-aware prompting strategy, providing users with varied interventions daily. Our system, PITCH, utilizes large language models (LLMs) to facilitate externalization and reflection on daily plans. Through this study, we investigate the impact of externalizing tasks with conversational agents on productivity and mental well-being, and the effectiveness of a rotation strategy in maintaining user engagement.

PITCH: Productivity and Mental Well-being Coaching through Daily Conversational Interaction

TL;DR

The paper tackles the challenge of improving productivity and mental well-being through conversational AI-assisted planning. It introduces PITCH, an LLM-powered agent that externalizes daily tasks in the morning and prompts reflection in the evening, augmented by a rotation-based and context-aware prompting strategy to sustain engagement. The authors describe a Slack-based prototype, a within-subjects field study comparing static and rotation-enabled interventions, and mixed-method evaluation plans to assess productivity and well-being impacts. If effective, this approach could provide scalable, personalized daily coaching that enhances time management and healthier work-life boundaries in real-world settings.

Abstract

Efficient task planning is essential for productivity and mental well-being, yet individuals often struggle to create realistic plans and reflect upon their productivity. Leveraging the advancement in artificial intelligence (AI), conversational agents have emerged as a promising tool for enhancing productivity. Our work focuses on externalizing plans through conversation, aiming to solidify intentions and foster focused action, thereby positively impacting their productivity and mental well-being. We share our plan of designing a conversational agent to offer insightful questions and reflective prompts for increasing plan adherence by leveraging the social interactivity of natural conversations. Previous studies have shown the effectiveness of such agents, but many interventions remain static, leading to decreased user engagement over time. To address this limitation, we propose a novel rotation and context-aware prompting strategy, providing users with varied interventions daily. Our system, PITCH, utilizes large language models (LLMs) to facilitate externalization and reflection on daily plans. Through this study, we investigate the impact of externalizing tasks with conversational agents on productivity and mental well-being, and the effectiveness of a rotation strategy in maintaining user engagement.
Paper Structure (10 sections, 2 figures, 1 table)

This paper contains 10 sections, 2 figures, 1 table.

Table of Contents

  1. Introduction
  2. PITCH - Conversational Agent for Reflecting on Task Management
  3. Design Goals
  4. Help users to externalize their plans through conversation at the beginning of their day
  5. Facilitate personalized reflection on users’ externalized plan and detachment from work through context-aware prompting
  6. Maintain user's interest by introducing randomness in externalization inquiry and reflection prompts
  7. ONGOING EFFORTS
  8. Development
  9. User study
  10. Evaluation

Figures (2)

  • Figure 1: Scenario 1 - Morning and Evening example conversation
  • Figure 2: Scenario 2 - Morning and Evening example conversation