Meta Spark

Tips for AR UX Researchers

Navigating the unique challenges
of UX research for AR

By: Dana Lee
December 16, 2021
Meta Spark

Tips for AR UX Researchers

Navigating the unique challenges of UX research for AR

By: Dana Lee
December 16, 2021

In our ongoing AR careers series, we’re inviting members from the Spark AR team to share their career journeys and experiences, and to provide their perspective and advice for others exploring career paths in AR. Our latest series contributor is Dana Lee from the Spark AR user experience team.

Hey everyone, I’m Dana, and I’m a user experience researcher focused on augmented reality. My colleague Stef Hutka recently wrote about her career journey becoming a user experience researcher, so I thought I’d dig further into the researcher role, and share some of the unique challenges of doing UX research in AR.

In this post I’ll cover topics like how we present emerging technology (like AR) in unbiased ways, how we evaluate social acceptability, parse out what people want, avoid novelty, manage hardware implications, and most important, how we advocate for the user. Within each of these topics, I’ll highlight the special considerations inherent to research in AR and provide some tips to address them. Let’s dig in!

Presenting emerging technology

Overall, people are still learning what augmented reality is — even if they’ve used a face effect on Instagram, the concept of using the rear-facing camera to see a world effect may be a bit confusing. As a researcher, how do you explain AR without being leading, yet make sure we’re talking about the same thing? This challenge of conducting unbiased research is not limited to AR. However, making the unfamiliar familiar comes up often with emerging technologies.

Prototypes and concepts are great tools to communicate potential ideas within AR without requiring too much explanation. Some people are more visual and better understand through concrete examples that they can explore, such as videos or click through prototypes. Depending on your research questions, the focus may not even be AR. In such cases, you can draw parallels from real life behaviors to explain how it works in AR. If this doesn’t quite capture your research task, you can explore analogous experiences. For example, trying on clothing in AR should be analogous to the real world experience.

Consider this as an opportunity to work with your team to gather assets for user research. For example, a designer or prototyper can create concepts for people to explore. Or, a data scientist can help parse large arrays of examples to help understand which are best to use in a study.

Evaluating social acceptability

Working in emerging products can yield confounding results — oftentimes people say they love a product but then in reality won’t use it. They may value the technology, but need to feel comfortable using it, especially around other people. To capture this potential issue, a great question to ask people is “if you saw someone else doing this, what would you think?” By taking a step back from the benefits of the product, this question gets people to think about the social world that we live in. Would you judge that person? Are there privacy concerns that pop up? Are there safety concerns? All these questions can help bring this future world into a more grounded reality.

Helping people imagine new technology adoption through the lens of social acceptability makes it easier to talk about, making it more likely the technology will get adopted in the first place.

In some past research, we’ve explored various scenarios and places in which people can use AR. People are generally excited about AR and it’s potential. It’s fun to imagine using it everywhere. However, when we ask if you saw someone holding their phone up and chasing something on say their commute or in the library, sentiments change. In the real world, we’ve observed people feeling self-conscious about body gestures related to AR. People want to be able to show off their tech savviness but within the confines of what feels acceptable and doesn’t call too much attention.

Parsing out what people need (and value)

If you are researching a topic with low awareness and low usage, some abstraction needs to happen. This is where the “art” of research comes in. Insights in AR require inferences (or reading between the lines) because people struggle to articulate what is unfamiliar to them. Henry Ford famously said, "If I had asked people what they wanted, they would have said faster horses." We face the same issue with AR.

It's difficult for people to know what exactly they want from augmented reality. One way to overcome this is by understanding people’s underlying needs, pain points and attitudes.

The Jobs To Be Done (JTBD) framework can help ground these insights and inform our work. It involves looking at the fundamental tasks that people “hire” services and products to do. This helps us think more concretely about what users actually need and are trying to do, regardless of the technology.

For instance, we asked people what they’d like to see in AR and they mentioned liking the idea of a calendar in AR. This makes sense because people want value from new technology and calendars help people know the day. Maybe people could even put the calendar into videos or stories to provide context. When it comes to mobile AR, users can already get calendar information from their phone’s calendar app. In this situation, we would want to dig deeper into why a calendar. Why do people cite a calendar? What does a calendar do for them? If they had a calendar, what would they do with it in mobile AR? What is their calendar app missing or doesn’t do a good job of? You might uncover that the root is really around wanting something to help them stay organized. Instead of a calendar in AR, which they already have on their phones, maybe there are other ways of presenting reminders in AR that’s more salient than within the calendar app.

Avoiding novelty

The core utility of Meta is social connection, largely achieved through engagement powered by entertainment. We have to take a hard look at what we are doing with the technology and ensure that our work is more than novelty but something that provides value, even if that value is something lighthearted, like having fun. When asking people what they value they will inevitably share a wide range of answers. One constant that usually shines through in our research is the desire for clear value and utility from new technology.

For example, imagine seeing virtual pizza raining down on the real Empire State building through your phone's camera. It’s fun and very New York. It’s a great image to take a picture of to show that you’re in the city but it lacks depth — there’s not much else that can be done beyond a single use. On the other hand, placing some AR characters in a scene empowers people to tell a story more effectively, in versatile new and different ways than before.

When evaluating AR experiences, we have to get beyond the novelty of the technology. And this starts by asking some really fundamental questions.

Are people able to tell different stories with this? Would people revisit this effect in the future? Would others seeing the effect want to try it out themselves? What “job” does this accomplish in the jobs to be done framework?

Managing hardware implications

My favorite part of doing research in AR is going out into the field and meeting people. Observing people using AR effects in the field is more representative of real life use — in parks, on commutes, in homes, etc. These are difficult situations to replicate in the lab. However, being in-field isn’t always an option. This has become a particular challenge in the pandemic.

Because remote interviews present more of a “lab-like" experience than one that might mimic real life, we’ve started running remote interviews and/or diary studies with follow up interviews. The diary studies help follow participants in a more natural environment; however, having two parts makes it a longer, more expensive study that requires more resources. One benefit of going remote has been the ability to recruit a more diverse group of users, since we can sample people outside of major cities.

Oftentimes, our participants are not familiar with wearable technology. This brings in an additional layer of human factor feedback to the research. Although the study may be focused on a software aspect, we can receive feedback influenced by the hardware and human factors instead. In these instances, we make sure to be clear about what we are asking about, and isolate questions specific to the user experience.

For example, while studying an AR product which leverages multiple form factors, we’ve observed how hardware and software findings can become intertwined. Tablets are extremely heavy when held up for an extended period of time, and this discomfort can bleed into the overall feedback about the user experience. When the study is in person, the struggles with the tablet weight could be observed. This gave extra context as to where the feedback was coming from. However, if this were a survey, user feedback forum, or diary study, it might be hard to realize that. It is important to look at the association between the feedback and the hardware. So when someone says "I'll never use this again!", this might be because of the form factor and not necessarily about the software.

Advocating for the user

In industry research, our role as user researchers is to translate our findings and help our product teams apply the learnings to our work. Our team doesn’t always have all the information to make the right decision, so we use research to help guide us.

Communicating people’s wants and desires is key here, and that can be challenging if non-existent technology is required to fulfil these needs. To help your team understand what is truly important, use today's technology as an anchor, make sure to remain conscious of current constraints when framing insights, and make sure to understand people’s current mental models.

All these together help us identify and close gaps with the technology we launch. After each launch, re-evaluate what worked and what didn’t to tackle these problems over time. This helps us build for the future today in a way that stays true to core user needs.

Our thanks to Dana for sharing her perspective today. If you would like to learn more about a career in AR UXR, please check out our career opportunities page.

Subscribe to the Meta Spark Blog