One Handed Bend Gestures

UX Design / UX Research / Development

Intro

This was a group project at Creative Interactions Lab in collaboration with Blackberry. We started off with an initial survey to understand the type of applications people use with one hand in their smartphones. We also learned about the possible scenarios where one-handed use cases were preferred. Followed by extensive literature review and brain-storming sessions, we short-listed eight bend gestures to evaluate. We designed and developed a functional prototype and conducted usability testing with 20+ participants.

Demo

Here’s the 30-second summary of our project available at ACM’s YouTube Channel:


User Survey

We started off with an online survey to find out the common use cases of one handed mobile interactions and the context of those use cases. Most of our 158 participants used their cellphone one-handed using their right hand, in context of mobile scenarios, to do single, short tasks.

Data analysis and Brainstorming

Based on the most frequent one-handed tasks and literature review, we then selected 8 deformable gestures, based on four locations (top left corner, top right corner, squeezing the central location vertically, or squeezing the bottom location vertically), and two directions (up and down).

Bend Locations


Prototype

We decided to identify users’ performance and comfort level for these gestures. We designed a prototype that provided the users with a set of tasks that required using those eight gestures.

All Gestures


Prototype GUI

Prototype GUI


Usability Testing

We ran an experiment to evaluate those 8 gestures performed with each hand, in the context of single tasks and repeated tasks.

User Study Session


Findings:

Our results show promise for one handed gestures, and warrant further exploration. The two best gestures were the top right up and the center squeeze up, which were faster, preferred, and more comfortable than the rest. We found no hand preference, which indicates that the gestures could be implemented to fit the needs of a wider range of the population, instead of favoring right-handers. We noticed that occlusion, one of touch’s weaknesses, is not a problem for deformable gestures. While the experiment did not explicitly measure changes in hand posture and grips, we observed that almost all participants had to re-grip to perform certain gestures.

Well-designed bend gestures would likely minimize the overall need for repositioning one’s hand(s) on the device. Hence, we believe that one-handed deformable gestures are an interesting, complementary interaction to touch, as they can alleviate some of one-handed touch’s issues.

User performance varied across different gestures for duration of interaction:

Duration


And spurious interaction:

Errors


User preference also varied for different gestures:

Likert Scale Data


This project was published in CHI 2015 – check it out here.