Summer Sprint 2: Another Prototype Pitstop!

The team explores modality moments with body storming and paper prototyping.

Written by the MHCI x 99P Labs Capstone Team
Edited by 99P Labs

The MHCI x 99P Labs Capstone Team is part of the Master of Human-Computer Interaction (MHCI) program at Carnegie Mellon University.

After hitting the ground running with our first summer sprint, our team was ready to continue on full speed ahead to sprint 2! From where we left off, the team was able to synthesize findings from our sprint 1 prototypes focused on passenger seating preferences, and we are currently in the process of finishing up sprint 2. For the second sprint, we explored the multimodal experience, and had a blast body storming and paper prototyping at our local farmer’s market! Let’s see what all we’ve learned from our various prototype solutions.

So, What Did We Learn From Sprint 1?

1. To what degree should this seating arrangement be movable and customizable, versus fixed?
2. What is the optimal seating arrangement for traveler safety and comfort?
3. How is this preference impacted by the presence of other passengers, both known and unknown?

For our prototypes, we tested the physical layout of a people mover in both a physical and app based format, and the primary takeaways we gathered from testing were:

1. Passengers indicated that they initially don’t care about customized seating, but this changes with factors such as trip duration and whether or not they’re traveling with friends. Passengers’ hesitation to customize may be related to their mental models of ride-shares and their inability to customize in those settings.
2. If seating is customizable, passengers prefer to make those selections during the pre-trip stage, to avoid awkward passenger dynamics and ensure efficiency of boarding.
3. Passengers want to face forward because of physical reasons like motion sickness, but also due to a desire for a sense of control while riding in an autonomous vehicle.

With these findings in mind, our future design explorations should explore the following questions:

• How might we design built-in interfaces that instill more confidence for passengers?
• How might we design a system that allows for customization in an intuitive way that doesn’t contradict existing mental models?

How Do We Move Forward? Models, Models, and More Models!

To begin sprint 2, our team began modeling to better understand and align on the ecosystem we wanted to design. To start, we began to create a service ecosystem model to align on the expectations of what an attendant should do — or, framed in the user’s perspective, what are the passengers’ needs and expectations throughout the different phases of using this future service? In other words, before we progress towards ideation of our product, we must be able to know how we can support passengers during each phase of transit.

From this model, we are looking into the “usage” quadrant. We hypothesize that the usage quadrant is the area with the most opportunity for interactions between our anticipated system and users in autonomous transit.

In addition to an ecosystem model, we also created an intersystem model that highlights all the interconnected domains that our solution may touch upon.

Lastly, to align on what exactly is a people mover, we constructed a physical model of what this vehicle might actually look like. Constructing a physical model helped the team solidify our mental models of the vehicle form and all that it is and isn’t capable of.

Kick-Start Sprint 2

After we aligned on the form and seating arrangement of our people mover, we kicked started Sprint 2 with a conversation about what we want to learn. Following Sprint 1, the main focus of this sprint was to flesh out the system that will be implemented in our people mover to support passengers’ needs. Specifically, how might we design a recognizable and seamless system that a user can feel confident in using while in autonomous transit?

We decided on the following learning objectives as guiding questions for this sprint:

1. What are the associated mental models of a technological ecosystem that supports user needs?
2. What expectations does a user have for a system rather than a singular, isolated attendant?
3. Where does the system start and where does it end?

With these learning objectives in mind, we moved forward with brainstorming activities and collaborative sketching sessions for possible solutions. During the course of exploration and ideation, we shifted the prototyping direction to test modalities in multiple in-vehicle scenarios. This allowed us to try understanding how passengers navigate available modalities of information in unexpected scenarios, when there is not a human present to help them. We also wanted to pinpoint gaps in information that may arise between modalities such as screen vs. voice for future designs.

Body Storming and Paper Prototyping at Farmer’s Market

On the sunny morning of Saturday June 4, we gathered at Bloomfield Farmer’s Market to set up our paper prototype for testing. We invited passersby to sit in our “vehicle” and interact with both a built-in screen and an audio system in two different scenarios. In the first scenario, the passenger suddenly doesn’t feel well and wants to get out of the vehicle sooner than planned. And the second scenario is being inside an autonomous vehicle and the vehicle suddenly stops moving. Through Bodystorming and Think-alouds, the participants provided tons of valuable information and feedback about passenger expectations, interface information hierarchy, communication preference, etc, which we later discussed and categorized in our synthesis session.

What Did We Find Out?

After affinity mapping and synthesis discussion, our team came up with the following key takeaways from Sprint 2 user testing:

1. In an autonomous shared vehicle, multi-modal reinforcement of information needs to happen simultaneously.
2. The level of detail of information should strike the balance between instilling trust and being confusing or overwhelming.
3. Urgency of the situation influences modality preference.
4. Passengers want to provide feedback through an effective communication channel.

Next Steps

As we concluded this sprint with rich findings and concrete takeaways, we wanted to carry them forward and further explore the following directions in the future:

• Appropriate level of specificity and road visibility increases passenger confidence
• Intentional communication will reduce passenger cognitive load and improve rider experience

Stay tuned for more prototyping and user testing stories from our next exciting journey!

Follow 99P Labs here on Medium and on our Linkedin for future updates on this project and other student research!