Between Us

Project Info

Exhibition piece
2 week project

My Role

Design Engineer
Design Lead
Team of 5 designers

Tools

Unity
Meta SDK, Meta Quest 3
Custom AI pipeline (Deepgram, Llama)
Between Us is an interactive spatial experience that encourages in-situ and post-conversation emotional awareness for conflict resolution between intimate relationships.
By using scrapped-together voice-to-emotional classification pipelines, we aimed to create a new kind of empathy interface, one that allowed users to experience the very emotional space they were themselves taking up, giving form and motion to emotional vulnerability and avoidance in real time.

In designing the experience, the system had to be able to...

Reflect emotion type and intensity...

Positive emotions trigger green or blue hues...
and negative emotions trigger red or yellow hues.

Reflect emotional avoidance and vulnerability...

Vulnerability invites emotional charge...
While avoidance repels it.

And trace each conversation into a "learning" constellation.

A conversational artefact that could be used after the fact, to learn and encourage retrospection throughout distinct conversations.
A balanced conversation started by one person
A one-sided negative/positive conversation
An entirely one-sided emotional outburst

Debuting our experience at our prototyping exhibition...

It was an incredible experience setting our studio space up into an exhibition, hosting guest critics, professors and friends to try out our conversational experience (and I have to say, I'm thankful there weren't many intimate conflicts to mediate!).

In all seriousness, it was also a chance to receive feedback on how we could improve such an embodied and involved experience.

Moving Forward

We moved and produced a concept quickly, and developed something raw that many folks resonated with, in visualizing and tracing emotion over time. Folks found the conversation artefact especially intriguing.

If we had more time and more resources (even just another headset), we would have loved create a more robust input pipeline: detecting vocal tone, color, silence, as well as body language.

We'd also love to talk to experts in the couples therapy and conflict resolution spaces to learn more about what couples struggle with, which techniques work, and ultimately try the tool out on more couples as well as other intimate relationships.