Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Piece of Theatre: Investigating How Teachers Design LLM Chatbots to Assist Adolescent Cyberbullying Education

Michael A. Hedderich, Natalie N. Bazarova, Wenting Zou, Ryun Shim, Xinda Ma, Qian Yang

TL;DR

This work investigates how teachers can design LLM-based chatbots to support adolescent cyberbullying education. By introducing Co-Pilot, a no-code builder and tester powered by LLM-Chains, the study gathers 13 middle school teachers’ perspectives to identify the levers needed to empower teachers as playwriters of classroom role-plays. It reveals that teachers seek both precise control and improvisational freedom to enable multi-participant scenarios that foster socio-emotional learning, empathy, and perspective-taking, while highlighting gaps requiring new design levers and collaborative workflows with students. The findings offer practical design guidance for teacher-centered chatbot tools and delineate research questions about control, collaboration, and classroom integration of AI-supported bystander interventions. The work argues that such tools can extend teachers’ instructional reach while preserving human-guided pedagogy and emphasizes future work to ensure safe, effective classroom deployment.

Abstract

Cyberbullying harms teenagers' mental health, and teaching them upstanding intervention is crucial. Wizard-of-Oz studies show chatbots can scale up personalized and interactive cyberbullying education, but implementing such chatbots is a challenging and delicate task. We created a no-code chatbot design tool for K-12 teachers. Using large language models and prompt chaining, our tool allows teachers to prototype bespoke dialogue flows and chatbot utterances. In offering this tool, we explore teachers' distinctive needs when designing chatbots to assist their teaching, and how chatbot design tools might better support them. Our findings reveal that teachers welcome the tool enthusiastically. Moreover, they see themselves as playwrights guiding both the students' and the chatbot's behaviors, while allowing for some improvisation. Their goal is to enable students to rehearse both desirable and undesirable reactions to cyberbullying in a safe environment. We discuss the design opportunities LLM-Chains offer for empowering teachers and the research opportunities this work opens up.

A Piece of Theatre: Investigating How Teachers Design LLM Chatbots to Assist Adolescent Cyberbullying Education

TL;DR

This work investigates how teachers can design LLM-based chatbots to support adolescent cyberbullying education. By introducing Co-Pilot, a no-code builder and tester powered by LLM-Chains, the study gathers 13 middle school teachers’ perspectives to identify the levers needed to empower teachers as playwriters of classroom role-plays. It reveals that teachers seek both precise control and improvisational freedom to enable multi-participant scenarios that foster socio-emotional learning, empathy, and perspective-taking, while highlighting gaps requiring new design levers and collaborative workflows with students. The findings offer practical design guidance for teacher-centered chatbot tools and delineate research questions about control, collaboration, and classroom integration of AI-supported bystander interventions. The work argues that such tools can extend teachers’ instructional reach while preserving human-guided pedagogy and emphasizes future work to ensure safe, effective classroom deployment.

Abstract

Cyberbullying harms teenagers' mental health, and teaching them upstanding intervention is crucial. Wizard-of-Oz studies show chatbots can scale up personalized and interactive cyberbullying education, but implementing such chatbots is a challenging and delicate task. We created a no-code chatbot design tool for K-12 teachers. Using large language models and prompt chaining, our tool allows teachers to prototype bespoke dialogue flows and chatbot utterances. In offering this tool, we explore teachers' distinctive needs when designing chatbots to assist their teaching, and how chatbot design tools might better support them. Our findings reveal that teachers welcome the tool enthusiastically. Moreover, they see themselves as playwrights guiding both the students' and the chatbot's behaviors, while allowing for some improvisation. Their goal is to enable students to rehearse both desirable and undesirable reactions to cyberbullying in a safe environment. We discuss the design opportunities LLM-Chains offer for empowering teachers and the research opportunities this work opens up.
Paper Structure (42 sections, 1 figure, 1 table)

This paper contains 42 sections, 1 figure, 1 table.

Table of Contents

  1. Introduction
  2. Related Work
  3. Teaching Adolescents about Cyberbullying and Bystander Intervention
  4. Teachers Creating Chatbots for Teaching
  5. Creating Controllable LLM Chatbots
  6. Method
  7. Designing a Chatbot Building and Testing Tool as a Probe
  8. Design goals.
  9. Design of the Chatbot Builder.
  10. Design of the Chatbot Tester.
  11. Co-Pilot Implementation.
  12. User Study Design.
  13. Participants
  14. Task
  15. Interview Protocol
  16. ...and 27 more sections

Figures (1)

  • Figure 1: Co-Pilot ChatbotBuilder interface showing the beginning of a dialogue flow for the cyberbullying scenario. Teachers define the possible behaviors of their students in each situation (green components) and the reaction the chatbot should give (yellow components). The teacher specifies example utterances for both types of components (chip elements).