CS 247B: Design for Behavior Change

Team 15: Mindful Movements

Amantina Rossi, Uma Phatak, Nadia Wan Rosli, Melody Fuentes, Rui Ying

 

Assumptions

Based on our intervention study results, we proposed three assumptions and designed experiments for each of them.

Assumption #1 Progress

We believe that: users are more likely to complete their mindful movement if they see their progress.

To verify that, we will: ask participants to do mindful movements three times during one day. We will tell them how much progress they’ve made via text-based affirmation and a visual.

And measure: people’s desire to continue with the movement by seeing if they continue to move with each text rather than quitting, as well as their motivation and stress levels in regard to mindful movement.

We are right if: People are more likely to want to continue to move due to the visual and textual representation of progress.

Assumption #2 Snooze

We believe that: users can benefit from being re-prompted to complete a movement if they are not able to complete it immediately the first time they receive a reminder.

To verify that, we will: send all participants an initial reminder to do a mindful movement and have them respond if they did the movement after receiving their reminder. If they do not respond within 15 minutes, we will send them another reminder every 15 minutes until they respond, up to 3 additional times (4 reminders in one hour).

And measure: rate of completion of the mindful movement as well as how long it took them after receiving the reminder to do the movement.

We are right if: the group of participants who receive follow-up reminders have a higher rate of completion of mindful movements.

Assumption #3 Timer 

We believe that: Users can benefit from extra guidance in addition to receiving a reminder in order to successfully complete a mindful movement via a timer. 

To verify that, we will: Send participants a 15-30 second timer with a reminder that they would use to guide them through the time it takes to complete a short, mindful movement. If they want to continue after, they are more than welcome to. 

And measure: Whether the participants are more likely to act on a mindful movement as a result, along with the amount of time they do the movement for, and how stressed the timer made them.

We are right if: The proportion of reminders sent versus actions completed goes up, showing that this little push of guidance makes participants more likely to follow through on their habit creation.

 

Experiments

According to the assumptions, we designed 3 separate experiments to gather insights.

Protocols

For detailed experiment protocols and materials, please refer to this document.

Participants

Assumption #1 Progress

3 new participants who have expressed interest in moving: 2 Stanford students and 1 post-grad with a remote job. They fit in with our personas.

Assumption #2 Snooze

3 participants: 1 new participant, 1 participant from the diary study only, 1 participant from the intervention study only. 2 students, 1 working adult (office/warehouse job).

Assumption #3 Timer

3 participants: 1 participant who was in the intervention study, working woman in a lab at a pharmaceutical company; 2 new participants, 1 UCI student, often working at a computer and 1 Stanford student, part-time secretary worker.

Results

Assumption #1 Progress

Sending texts about streaks

Sending texts about streaks

Survey responses:

• All participants completed at least ⅓ of the 3 movements.
• Streak visualization was at least neutrally useful to participants (3+ on a scale of 1-5).
• Showing progress was not stressful at all (3- on a scale of 1-5).

Participants’ feedback:

• Make it a daily streak instead of a streak that happens within one day (this is what we originally wanted to test but we didn’t have enough time)
  • They specifically said: “I think I could have been motivated more if it was more of a daily streak type thing instead of a number of times a day because it seems easier to achieve and more motivational to do something every day.”
• Include fun facts about movement to get me inspired!!

Assumption #2 Snooze

Note: Everyone ended up only stretching.

How to read table: Time of initial reminder → (Snooze, if applicable) → (Follow-up Reminder [R], if applicable) → Time of doing movement / Movement not completed

Participant 1: Daniel (student)

4:15 PM → 4:16 PM

5:30 PM → Snooze → 2nd R → Snooze → 3rd R → Snooze → 4th R → Movement not completed

9:00 PM → Snooze → 9:16 PM

Daniel’s feedback:

Daniel could not complete his movement in the 2nd row because he ended up unexpectedly driving out of town and was too busy. He wishes he either got more reminders from snoozing or had the option to reschedule the reminder in case the whole hour gets blocked out. He thinks if he got an additional reminder after he was not busy anymore, he would have done the movement (but because he didn’t, he did not do it at all for that round). He thinks the snooze function is a great idea and helpful for allowing more flexibility in a busy day.

Participant 2: Wan (working adult)

12:30 AM → 12:31 AM

1:30 AM → 1:31 AM

2:30 AM → 12:31 AM

11:00 AM → Snooze → 11:07 AM

12:30 PM → Snooze → 2nd R → 12:46 PM

Wan’s feedback:

Wan used the snooze function when he was driving and when he was in the middle of watching something. The first time, he did the movement before receiving a follow-up reminder because only 7 minutes passed, so the first reminder was still on his mind. The second time, he said the follow-up reminder helped him remember to do the movement because he got distracted by then. He finds reminders as well as the snooze function helpful and would not have otherwise done the movements if he did not receive the reminders.

Participant 3: Sydney (student)

12:00 PM → Snooze → 12:11 PM

2:00 PM → Snooze → 2nd R → Snooze → 3rd R → 2:40 PM

4:00 PM → Snooze → 2nd R → Snooze → 3rd R → Snooze → 4th R → 4:52 PM

6:00 PM → Snooze → 2nd R → 6:23 PM

8:00 PM → Snooze → 2nd R → 8:12 PM

Sydney’s feedback:

Sydney noted that after snoozing, she would sometimes try to do the movement in a timely enough manner so she wouldn’t have to receive the next follow-up reminder. In other words, she was sometimes motivated by the anticipation of a follow-up reminder and not wanting to trigger it. Most of the time, the reason she kept pressing snooze was either because she was busy or to procrastinate doing the movement. But knowing that she was going to get follow-up reminders motivated her to ultimately complete all the movements within the next hour. There were a few times when she also just forgot about doing the movement after snoozing and the follow-up reminders helped her remember. She thought knowing she would only get 3 follow-up reminders helped motivate her to do the movement within the next hour, but she still would have wanted the ability to snooze more in case she was still busy. She thought the snooze function was helpful or else she probably would not have done some of the scheduled movements.

Assumption #3 Timer

Test results:

• For the most part, all participants completed all three movements for all three days. The only exceptions was participant #2 missing one movement due to an impromptu meeting, and participant #3 receiving an alert while driving, and so they were not able to start the timer. To compensate, they did a breathing exercise while driving while counting in their head.
  • User 1: 9/9 movements (3/3, 3/3, 3/3)
  • User 2: 8/9 movements (3/3, ⅔, 3/3)
  • User 3: 8/9 movements with timer (⅔, 3/3, 3/3), 9/9 not accounting timer
• Day 1 I let users completely decide their own movements, Day 2 I proposed optional movements, Day 3 I let them decide whether they wanted to be provided with optional proposals or not. 
  • All 3 users choose to be provided with optional guidance on day 3.

Survey results:

• On a scale of 1-5, 
  • 4+ average on motivation level (1: no motivation – 5: extremely motivated)
  • 3+ average on engagement level (1: not engaged – 5: completely engaged)
  • 2 average stress level (1: no stress – 5: extremely stressed)
• User 1, who completed initial intervention study, preferred the timer.
• All 3 users preferred guidance.

Participant’s feedback:

User 1:

• Preferred timer over self-timed.
• “At a point where I would have stopped stretching on my own, seeing that I only had five seconds left kept me going for a little bit longer”.

User 2:

• “Sometimes benefitted, sometimes kept me from doing more. I would stop exactly as the timer stopped.”
• “Sometimes, especially while doing a strenuous stretch where it’s not as comfortable to keep close watch on my phone, I would be less engaged in the movement itself as I kept track of how much time I had left until I could go back to do what I was doing.”
  • “Same with breathing exercises, though it was easier to look at my phone, but I was not completely immersed in breathing, if that makes sense.”

User 3:

• “I enjoyed the very small added pressure of bringing a movement to completion. It motivated me to start just to see the clock hit zero.
  • “Once the concept of movement was brought back into my attention, I would continue to do it past the timer. Similarly, because my previous deep-focus activity was interrupted, I would remember to do small stretches at my desk without the reminder. Just remembering that movement was something that existed that I could do made it happen more.” 
• “Movements outside the notices felt less like an achievement, and so were less pressured, but also less satisfying to do. Though, I realized I did it more than usual before this activity.”

Synthesis

Based on the experiment results, we synthesized the findings into the following.

Assumption #1 Progress

We believe that: users are more likely to complete their mindful movement if they see their progress.

We observed: participants were neutrally motivated by the streak progress and were not stressed by keeping a streak.

From that we learned that: showing progress such as streaks are a nice way to keep users motivated. However, the frequency of streaks (daily streaks in a week vs streaks in one day) may have different effects on users’ stress level as well as motivation level.

Therefore, we will: not implement a streak progress system in our app, since the delicacy of its balance is hard to determine in a short time.

Assumption #2 Snooze

We believe that: users can benefit from being re-prompted to complete a movement if they are not able to complete it immediately the first time they receive a reminder.

We observed: that participants used the snooze feature when they were too busy to do the movement at the time of the initial reminder. They also most often did the movement after being re-prompted with another reminder, rather than on their own after choosing to snooze.

From that we learned that: The snooze option is useful to users if the reminder isn’t perfectly timed to their current schedule or task, and that including extra reminders may be necessary to ensure users don’t forget to do the movement after snoozing.

Therefore, we will: Implement a snooze button that will re-prompt the user in customizable intervals.

Assumption #3 Timer

We believe that: Users can benefit from extra guidance in addition to receiving a reminder in order to successfully complete a mindful movement via a timer.

We observed: that participants sometimes benefitted from the timer but sometimes didn’t, and only one participant preferred the timer over self timing. We also observed that users preferred guidance for the movements.

From that we learned that: the timer may take away from users’ ability to be mindful during their movement, but extra guidance in the form of instruction on the type of movement may be more effective.

Therefore, we will: add an option to add a timer but it won’t be the default. We will also provide suggestions for types of movement.