This project investigates how interactive narrative in virtual reality can be used to foster greater empathy towards the LGBTQ+ community.

Design Intro

Genre

Science fiction

Story Summary

A young gay protagonist’s journey of retrieving his lost memory and remember who he really is.

World Setting

A fictional dystopian future world ruled by an oppressive regime, where only heterosexuality is allowed. By using brainwashing, the regime programmed most non-heterosexual people’s memory to be make them “normal”.

User Role

Vincent, a 19-year-old gay teenager, got his memory programmed.

User Goal

To retrieve their lost memory through a journey of reliving their past.

Story & Interactions

The user will pick one of four collectors from the podium at the center to start the journey, each collector corresponds to one space.

The prison will have four doors, each leading to a different memory space.

When the user approaches the door, it will open automatically.

Once the user passes through the door, they will arrive at the new space.

How to keep users motivated to explore?

The Memory Collector is essential for the user experience, as it collects all lost memories.

The Memory Collector will flash to indicate to the user that they are near a memory.

To create a more engaging experience, users must explore the environment to locate the memory.

All the characters in the memories will be presented as particles.

When the user is close enough to the location of the memories, it will appear as illuminated particles.

The more dense the particles, the more significant the memory is to the narrative.

When a memory is collected, the particles from the memory are absorbed into the prop, and the light indicates to the user how many memories they have collected.

Once they have collected all the memories, the collector will be fully illuminated.

What is the end of experience?

After each memory collection from a space, the user can return to the prison of memory and place the collector back on the podium at the center.

Once the user has collected all the memories from four spaces and returned the collectors, the prison will unfold and the user will return to reality.

Product Design

Design direction

Considering the fact that the current AI technology is still primitive which can only realize limited semantic recognition, combined with insights from the survey, we decided to design an online HR interviewer and an HR robot.

Part 1. Online AI HR interviewer prototype

The AI HR interviewer is mainly used for an online interview during the campus recruitment months. With the AI HR interviewer, they don't need to bother interview prospective employees one by one that doubles their effort in the past.

Finally, when the interview is finished. I designed a simplified version of the MBTI psychological test to match interviewees with the best positions with a sense of humor, thus reducing the psychological burden for interviewee after the interview.

The NLP technology I used was still very basic which mainly analyzes the language by searching the keywords. In such a case, I intentionally lower users' exception by designing a "humble" personality for the AI robot.

Part 2. Cardboard AI HR robot develop

The AI HR robot is designed for campus recruitment events, in which students usually want to learn a lot about the job position and company information and also send their CVs. With this robot, HRs don't need to answer questions especially those been repetitively asked.

Using JD's recycled cardboard, I designed a robot that can be easily assembled and disassembled.

The robot's face has five different states, including confusion, sleepy, smiling, talking, and blinking, and would be displayed depending on the stage of the conversation. (e.g. When the robot doesn't talk to users for some time, the sleepy face would be shown)

Users can chat with the robot not only by talking but also by touching on the screen which is designed as the robot's face and also a user interface through that users can easily review conversation history that is presented as conversation bubbles.

When the conversation ends, users can download the conversation history to their mobile phone for subsequent review by scanning the QR code on the screen.

Design process

Designing the conversation flow

In order to realize an intelligent conversation experience in the case of limited NLP technology, I designed the entire dialogue flow. The functional dialogue mainly includes core, skip, bridge and chatter.

The core section contains topics that are frequently asked by users at campus recruitment events.
The skip and bridge sections are used to jump between different functional sections through keyword recognition.
The chatter is the section that makes the conversation more interesting.

Skip example:

User interface iteration

In an effort to make the user interface and robot appearance looks consistent with the overall style of the cardboard, I used a pixelated user interface style.

before:

after:

Design patterns

Voice user interface design principles:

Dialogue bubbles contains text that is synchronized with speech should not be too long, and one bubble per sentence is the best.
The GUI should indicate whether the user speaks or not by using different animation, to inform the user that the device is recording well.
When the robot is in different states (such as sleeping, talking, error, etc.), the robot's visual appearance should change accordingly.
When the robot trying to inform a large amount of information and selections to the user, the touch interaction can be combined to make the user's selection easier.

The Prison of Memory

Design Intro

Story & Interactions

Gameplay Demo (very very early stage...)

Product Design

Design direction

Part 1. Online AI HR interviewer prototype

Part 2. Cardboard AI HR robot develop

Design process

Designing the conversation flow

User interface iteration

Design patterns

Voice user interface design principles:

Dialogue bubbles contains text that is synchronized with speech should not be too long, and one bubble per sentence is the best.

The GUI should indicate whether the user speaks or not by using different animation, to inform the user that the device is recording well.

When the robot is in different states (such as sleeping, talking, error, etc.), the robot's visual appearance should change accordingly.

When the robot trying to inform a large amount of information and selections to the user, the touch interaction can be combined to make the user's selection easier.