-
Notifications
You must be signed in to change notification settings - Fork 0
Summary about the two first studies
-
To identify the most appropriate way to hold a cubic-shaped device while being used in the context of text entry, in order to design the cubic layout.
-
To receive qualitative feedback from keycube implementation.
- 31 participants (15 women and 16 men)
- 23-53 years old with mean 31
- From two university campuses (Le Mans Université in France and Université TELUQ in Canada)
- A neutral cubic was used while connected via bluetooth to a laptop with a custom application. Each key corresponded to an alphanumeric code.
5 minutes free manipulation of the Keycube by each participants with 3 instructions
- To imagine using the prototype in the context of text entry for the whole experiment.
- To try as many different holding positions as possible
- To follow a think-aloud protocol while being video recorded After that theu had to follow a minimal set of different positions to try and had to choose their preferred positions to hold the device while simulating text entry, both in standing ansitting posture (with pictures). Finally, they had to fill a questionnaire. Each experiment lasted around 35 minutes.
Preferred holding position while in use
- Diagonal position
2. Front position
With two hands to use more fingers, forearms slightly bent and the touch screen facing down. Placing the little fingers under the bottom face or by increasing the pressing of the hand palms against the lateral edges of the device were adopted like two strategies to prevent the device from falling (only with the diagonal position).
A second user study was conducted in mean to design the cubic layout. Identify and measure the preferred and reachable key areas frome the fingers qhen the user hold a cubic-shaped device in the diagonal position were the goals.
- 22 participants (14 women 8 men) * 23-43 years old with mean 31
- Same campuses
- 13 participed in the previous study
- The same material was used in the study but with a new application to ease the mental representation of the device.
- A text area and symbolic hands accompanied the visual interface to guide the participants.
The blue and yellow faces were arbitrarily chosen to be external while the wi=hite and green faces were the internal (facing the user’s body), and the red face on top.
- An average of 48 minutes for each experiment lasted
- The alphanumeric codes of the keys were defined with the first letter of the surface and a number from 1 to 16 to retrieve the position of the keys among each surface.
- All the non-reported keys were considered unreachable by the participant. For the second part of the experiment, the instructor said an alphanumeric code to the participants, who in return reported the preferred finger to press the corresponding key. All keys of the prototype (N=80) WERE RANDIMLY GIVEN TO AVOID A POTENTIAL SUCCESSION BIAS FROM KEYS PLACED NEXT TO EACH other.
Dominant Fingers: Thumbs are the most versatile, covering the largest area of the cube (including the central red face and internal white/green faces).
Least Used Fingers: Little fingers were preferred for only one specific key each.
Agreement Score: Researchers used a "confidence score" (0.14 to 1.0) to measure how much participants agreed on which finger to use for a specific key. Most keys had a clear favorite, though three keys (b1, b4, b11) showed a split preference between two fingers.
Scoring System: Keys were rated from 0 (unreachable) to 3 (easily reachable).
Easiest Areas: The most accessible keys are those located near the fingertips, particularly around the middle fingers, which can be supported by the index and ring fingers.
Difficult Areas: Keys located near the edges in contact with the palms are the hardest to reach.
Some areas (like the top surface) have low "total" scores because only one finger (the thumb) can reach them, even though that finger reaches them easily. Other areas have low scores because they require effort from any finger that tries to touch them.
The study generated a large dataset of 19,514 distinct entries, including hand measurements and reachability scores. This data is released for scientific use and can be applied to the design of any cubic-shaped interface beyond just text entry.