Paper Reading #15 - Enhanced Area Cursors: Reducing Fine Pointing Demands for People with Motor Impairments

Comments:


Reference:
Enhanced Area Cursors: Reducing Fine Pointing Demands for People with Motor Impairments
Leah Findlater, Alex Jansen, Kristen Shinohara, Morgan Dixon, Peter Kamb, Joshua Rakita, Jacob O. Wobbrock
UIST '10

Summary:
This paper explores ways to improve mouse-based interfaces for individuals with motor imparements.  The researchers developed four enhanced area cursors, two of which use magnification while the others use goal crossing.  The two crossing methods used are click-and-cross and cross-and-cross.  The click-and-cross cursor works as follows.  The user moves a cursor with a circular area over the desired target, then clicks to activate the area trigger.  The cross-and-cross method employs a point cursor inside of an area cursor.  A "trigger arc" is used on the area cursor, and it always follows the point cursor opposite the direction of movement.  When the user wishes to activate the cross-and-cross cursor, they reverse movement direction and cross the trigger arc, activating the area cursor.  For both methods, when the area cursor is activated the targets within are assigned to various crossing arcs.  The user selects a target by crossing the appropriate arc and stopping the mouse for at least 300ms.

The two magnification-based cursors are the motor-magnifier and the visual-motor-magnifier.  For both cursors, the user moves and area cursor over the target area and clicks once to enter the selection phase.  The motor-magnifier decreases the mouse gain, effectively magnifying the motor space, though there are no changes to the visual representation of the targets.  The visual-motor-magnifier works in the opposite way by increasing the targets covered by the area cursor without changing the mouse gain. 

For the experiment, 24 participants were used (half of whom had motor impairments).  All participants were regular computer users.  Participants were exposed to the four cursors described above, with the addition of a  point cursor and a bubble cursor.  They had to operate in environments where targets were 4, 8, or 16 pixels wide.  In addition, they had to navigate environments with no, half-target width, or full target width spacing.  Target clutter was also varied, with 250 and 1000 distracters to see performance in sparse vs. cluttered areas.

Most area cursors did not degrade with decreasing target size, with VMM, CLC, and bubble performing the best for small targets.  In addition, most enhanced area cursors performed well with decreased spacing.  Crossing cursors did not perform well with cluttored spaces.  CLC, VMM, and bubble cursors all had fewer submovements than their competitors, and submovement reduction proved to be greatest with the smallest target size.  Seven of the impared participants preferred the VMM cursor, while the CRC and CLC cursors had 3 and 2 supporters respectively.  The point cursor was the least preferred (9 of 12), while MM, CRC, and CLC received one response each.  The able bodied participants all performed better with the point and bubble cursors, and no data indicated if any of the enhanced area cursors improved accuracy.

Discussion:

I have seen research before that dealt with mouse cursors, but this is the first I've seen to specifically focus on motor-impared users.  That being said, while reading this paper I kept thinking that while the described cursors did not sound appealing to me, they would sound fantastic to someone with limited movement.  It was interesting to see multiple solutions to this problem, and it makes me wonder what other methods are being researched.  A computer is a powerful tool, and I like that work is being done to make them more accessible to all members of the public.