Problems & Role · Highlights · Exploration · Design · Evaluation · Introspection · Impact

Since early in my career, I got fascinated with the use of 'entertaining' mechanics in 'serious' software. The most common example would be integrating achievements and leaderboards into fitness apps. I wanted to explore the topic further. For my research project I chose the University of Melbourne, as it was the best Australian university for HCI research at the time:

Problems

Preliminary market research and academic literature review helped me identify the following problems with the existing 'gamified' solutions for fitness and running in particular:

• most solutions focus on external rewards, such as points and trophies
• most mixed reality systems require expensive hardware and are inconvenient for running
• running experience in the existing literature is seen as monolith and discussed mostly in terms of skill (amateur or pro) and frequency of exercise
• existing academic literature and commercial solutions mostly focus on the potential health benefits of running rather than enriching the experience

'Zombies, Run!' is an exception to all of the above. Instead of relying on achievement badges, it relies on a fictional story. With only a simple app and a pair of headphones, 'Zombies, Run!' story creates a mixed reality experience for the runner: with the app, they run not just in their regular environment, but in a post-apocalyptic world full of zombies.

The popularity and commercial success of 'Zombies, Run!' suggested that such 'mixed reality stories' (MRS) have potential, yet there was limited understanding of the format. Thus, the research question for the project became:

Role

While it's hard to overvalue the amount of support and guidance I received from my supervisors and peers, ultimately, as a PhD candidate, I was solely responsible for

• planning and time-management
• designing and conducting the studies
• recruiting participants and receiving approvals from the ethics committee
• presenting the findings
• bringing the project to completion

Highlights

Problems & Role · Highlights · Exploration · Design · Evaluation · Introspection · Impact

Runner Journey

One of the first findings was a more detailed understanding of the runner journey. This helped to identify some specific gaps in the current runner experience and centre the project on a particular part of the journey: the trajectory of the workout experience.

Jobs to Be Done

The research also helped to identify some non-obvious jobs to be done by mixed reality stories, for instance distractions and a sense of companionship for people too self-conscious to run together with someone else. Distraction in particular turned out to be an important direction for this project.

Runner Personas

One of the largest findings was identifying the differences between runner personas and their relationships with mixed reality stories. Interestingly, running experience did not play a big role in regards to attitudes towards MRS.

• Phase I revealed that whether runners are bored or engaged by the workout is more important in the context of MRS than their running experience
• Phase II identified three types of connection between physical and virtual in MRS: blending, ambiguity, contradiction.
• Phase III introduced a new axis for the direction of engagement, inward or outward. This provided a deeper understanding of the relationships between personas and types of connection in MRS.

Overall, the main contribution of the project was defining a framework for understanding and designing better mixed reality experiences for running.

Process

I broke down the overall project into three distinct phases, each building up towards answering the overall question. Given the nature of the question, I tended to adopt qualitative research methods in the process. Each phase followed a regular design thinking process, just on a larger scale:

I Exploration

Problems & Role · Highlights · Exploration · Design · Evaluation · Introspection · Impact

The aim of the first phase was to explore…

The main study during the first phase was a study with 14 runners and people who want to start running regularly. I asked each of them to use 'Zombies, Run!' during their workouts for at least three weeks. I interviewed them before and after they started using the app. Three participants did not finish the study (hopefully, not due to being caught by the zombies).

As participants were running on their own, at a convenient time and location, it was difficult to capture their thoughts during the workout. To address the issue, I asked the runners to record a short audio-diary entry after each run — that helped to capture participants' thoughts still fresh on their minds.

During the first phase, I also joined two Melbourne running clubs myself, to better understand runners, and also find potential participants.

Summary of Phase I Methods

Data Analysis & Findings

I extensively used DevonThink for data analysis. I chose the app for its flexibility with tagging and colour-labeling, and its ability to work with any files, including audio and video clips.

The main finding during that phase was a deeper understanding of a running experience journey.

The journey trajectory showed that the running experience is not monolith and that different parts of the journey require different design approaches. The focus of my work was on the trajectory of the workout experience. Here I found that the attitude people have towards running plays a larger part in how they experience an MRS than their running experience, age, gender or even how much they generally enjoy horror stories or survival games.

The results also demonstrated how runners switch between running and story mindsets by using voluntary engagement to fine-tune the hybrid experience of MRS.

There was one mission which was specifically for night, so I make sure I ran that at night.
Runner 5

II Design

Problems & Role · Highlights · Exploration · Design · Evaluation · Introspection · Impact

In the second phase, I switched perspectives to better understand how storytellers and designers can create new MRS.

I conducted workshops with 6 creative writersand a producer to get insights into their regular process and speculate how it can be adapted for the creation of MRS. The workshop structure was inspired by storytelling workshops I attended in Melbourne.

Two writers and the producer agreed to complete a sample MRS script, all three were interviewed as a group after they completed their scripts. I also developed a framework for analysing the scripts themselves. In addition, I analysed public interviews of creators of 'Zombies, Run!' and other professional storytellers.

Summary of Phase II Methods

Data Analysis & Findings

During this phase, in addition to traditional data analysis, I also integrated script analysis into the process:

The second phase revealed three approaches writers can use to connect or disconnect an MRS to and from the physical environment. Another main finding was a demonstration of how creative writers repurpose familiar tools to create mixed reality experiences.

I was really playing with the idea of having different sounds and giving almost like an alternative reality to the runner. [...] I suppose, this is coming from my radio background where I feel like you can create this whole theatrical experience using only sound…
Writer 7

III Evaluation

Problems & Role · Highlights · Exploration · Design · Evaluation · Introspection · Impact

The aim of the third phase was to evaluate and extend previous findings and bring the thesis to conclusion. The final research sub-question was:

This phase introduced a new set of runners to the mixed reality stories created during the second phase. 35 runners were asked to pick a story and (literally) run with it at the Melbourne University campus. Right after the workout, they were interviewed about the experience. Some participants chose to run in pairs, and there was even a group of three runners participating simultaneously. Some participants came back to experience a different MRS and were interviewed again about the differences in their experiences.

Listening back to previous interviews made me realise that I can shorten interview time by repackaging some of the typical questions into a questionnaire. This allowed me to make each interview 10-15 minutes shorter, without sacrificing the amount of data I was collecting.

Summary of Phase III Methods

Data Analysis & Findings

This phase required me to compare data not only across three different MRS, but also across all three phases. Thus, I made a giant spreadsheet to have an overview of all the potential intersections. The spreadsheet also allowed me to quickly zoom out for a bird's eye view of the whole project and to zoom in for a detailed view.

As expected, this phase deepened the findings from the previous phases. Namely, it introduced another important dimension to the running experience journey: the direction of runners' engagement. The study also explored different combinations of runner types with different kinds of MRS stories. The results provided more insight into how runners use voluntary engagement to control their experience.

That’s the power of imagination […] Everything near me became story. If I saw a car – that was the car of the story. If I saw a woman – the woman became the woman who was speaking to my earphones [...] Listening to [a] story is different to listening to music.
Runner 22

Introspection

Problems & Role · Highlights · Exploration · Design · Evaluation · Introspection · Impact

There were many challenges I encountered over the years of working on the project. To highlight just a few:

Ethics committee

For each study I conducted, I had to receive approval from the departmental human ethics advisory group. Each application provides a detailed description of a study along with potential risks. While the risks for the main study of phase three were considered minimal, the ethics application spanned across 12 pages… and was denied:

[Runners] will have their attention split between the listening to audio tracks and watching where they are running. In Princess’s park, the participants could run into each other, other runners, other people in the park, children in the play grounds, footy players practicing (and they won’t be happy) or traffic as the park is surrounded by two busy roads and traffic in and out of the footy ground.
Ethics review

Instead of searching for an alternative running track, I took the study in a different direction and suggested locating it at the Melbourne University campus. This location had several research advantages over any running track and allowed for more interesting connections with the stories. I wouldn't even consider it if I got the approval in the first place. Lesson learned:

Recruiting

During the final phase, I also struggled with recruiting. Due to the nature of the study, I needed more participants than in the two previous phases combined. As it turned out, simply putting available time slots on the recruitment website more than tripled the response rate.

Structure

The most difficult challenge, however, was finding the proper structure. It took me many months to organise disjointed findings into a coherent thesis.

I was only able to reach my goal when I truly internalised a different mindset for studying UX; a mindset that centres meaning – which technology brings to people's lives – over function. And that was the main lesson I learned:

Impact

Problems & Role · Highlights · Exploration · Design · Evaluation · Introspection · Impact

This project investigated how can the user experience of running with mixed reality stories be supported. In turn, a complex multi-faceted answer to this question strives to provide contributions towards several areas of HCI and UX research and practice.

First and foremost, the thesis defined the format of a 'mixed reality story' and demonstrated how it's different from audiobooks, podcasts and existing 'gamified' solutions for running. The project also provided a detailed description of how mixed reality experiences can be created without expensive hardware or complex software.

Moreover, the project demonstrated in detail how people use voluntary engagement to control their experience and switch between contexts on the run, which provides an important insight into how MRS work and how their design might be improved.

In addition, the findings challenge the 'monolith' view of the running experience, previously defined mainly by the skill or frequency of workouts. The project demonstrated how some runners actively look for distractions while others seek more of a meditative experience and want to focus on their runs. There is also an important distinction between runners focusing inwards, on their thoughts, and runners focusing outwards, on the environment around them. These distinctions were not previously described in the relevant literature, yet play a crucial role in how runners experience different types of MRS.

These findings allowed anticipation of some future trends in academic research and the industry. For instance, in the thesis I speculated how other activities, such as walking, could benefit from an audio-centric mixed reality story. About a year after the publication of the thesis, Apple has done exactly that with 'Time to walk'.

The work is now written about in a way that demonstrates expertise in the theoretical, methodological and practical aspects of the work. I think that this thesis is a brilliant piece of work, for which the student should be hugely proud.
Dr Conor Linehan, thesis examiner

For the Record, a service for logging board game sessions ›