Beyond simple conversation, the rabbit also needed to model correct English gently. Instead of directly pointing out mistakes, which can make learners self-conscious, I scripted it to rephrase sentences in correct grammar. If a student typed, “I goed to the store,” the rabbit wouldn’t say, “That’s wrong.” Instead, it would respond naturally with, “I went to the store,” as though it were just part of the conversation. This way, students would hear the correct structure without feeling like they were being corrected. Cultural sensitivity was another important factor. Since students come from diverse backgrounds, I didn’t want the rabbit engaging in discussions that could be controversial or uncomfortable. To address this, I added a script that filtered out certain topics like politics and religion. If a student brought up one of these topics, the rabbit would simply redirect the conversation with a friendly, “Let’s talk about something else!”

Once all the scripts were in place, I tested everything. I greeted the rabbit, and it responded correctly. I walked around to see if it would follow me, but for some reason, it didn’t. I also tested both correct and incorrect sentences to check if it would model proper grammar, but instead of rephrasing mistakes, it just kept saying, “That sounds interesting!” However, when I tested the sensitive topics filter, that part worked. It successfully redirected the conversation away from restricted topics (see screenshot). I also tried adding an animation script to make the rabbit move its mouth and body while talking, creating a more warm, patient, and engaging ESL tutor. But I ran into a problem: I couldn’t get it to animate. It didn’t have any pre-set animations in its inventory, and I wasn’t sure if I had permission to add my own.Ultimately, I wanted the rabbit to be an active part of the students’ learning experience, encouraging them to practice speaking, reinforcing correct grammar, and helping them feel comfortable using English. Unfortunately, some of my scripts didn’t work as expected. After nearly two or three hours of troubleshooting, I suspect that some scripts overlapped or conflicted with each other. Despite the setbacks, the process was a valuable learning experience, and I now have a clearer idea of what adjustments are needed to make the rabbit function as I originally envisioned.

This is the second version of the instructions, drawn after a good look at Helena's Totaller. The task, the target and personality are defined, but there is room for improvement, as suggested during one of the live sessions.

Dickens sounds very much as I expected and can be funny and warm. It doesn't correct one's language mistakes as much as I would have loved, but maybe it is wise so.

I feel very proud of this creation, but my students were not equally fascinated... Creating AI agents is not enough, you also need a good instructional design, as Jennifer Hamilton correctly pointed out.

Target Audience: Grade 10 students at an international school in Ho Chi Minh City, Vietnam

Link to the project: https://framevr.io/midsummidsum

Teacher: Michelle Kang

Project Description

This immersive virtual reality project uses Frame VR technology to bring to life Shakespeare's A Midsummer Night's Dream. Students will explore a mystical forest near Athens and interact with the four young lovers—Helena, Hermia, Lysander, and Demetrius speaking in modern American English, complying with CEFR (Common European Framework of Reference for Languages) B2-C1 descriptors.

Objectives

• Enhance understanding of the play's characters and plot.

• Improve engagement with Shakespearean literature through interactive technology.

• Develop critical thinking and analysis skills.

Virtual Environment

The Frame VR platform will host a 3D representation of the enchanted forest, complete with:

• Interactive avatars of the four lovers.

• Atmospheric elements like magical plants.

Student Activities

1. Virtual Exploration: Students will navigate the forest, encountering and conversing with the young lovers.

2. Character Interactions: Using provided dialogue samples and character descriptions, students will converse with the lovers, exploring their motivations and relationships.

3. Reflection Writing: After the VR experience, students will compose a reflective essay on their interactions and insights gained.

4. Class Discussion: Facilitated debates on character actions, themes of love and magic, and the play's relevance to modern relationships.

Resources

• Full text of "A Midsummer Night's Dream" (accessed via website).

• Concise YouTube video summarizing the play's plot.

• Uploaded character descriptions and sample dialogues.

• Ready Player Me (RPM) editing the four AI agents.

• Frame VR platform for hosting the virtual environment.

• CEFR Global Scale: https://www.coe.int/en/web/common-european-framework-reference-languages/table-1-cefr-3.3-common-reference-levels-global-scale.

Assessment

Students will be evaluated based on:

• Quality of interactions in the virtual environment.

• Depth of reflection in their written essays.

• Participation in class discussions.

• Understanding of key themes and character motivations.

This project leverages Frame VR's immersive capabilities to create a memorable and engaging learning experience, bringing Shakespeare's timeless comedy to life for modern students.

Cheers, Michelle