Powerpoint template for scientific posters Swarthmore College

1
GridMouse Using mobile phones to navigate large
public displays Anna Ostberg, Jon Campbell, Brian
Evans Advisor Bill GriswoldCalit2, University
of California San Diego, La Jolla, California
Introduction The increasing role of interactive
public displays is important because they allow
users to employ technology in place of such
things as chalkboards and bulletin boards. The
ability to use computation in conjunction with a
display can allow for easier manipulation of
information and an enriched user
experience. However, in the implementation of
large public displays, the technology used for
interaction can pose a difficult problem. People
will want to be able to control and manipulate
things on the display, but we do not want users
to be tied to a keyboard and mouse, or for them
to have to gain access to special equipment in
order to use the public displays. Our solution
is to use the number-pad on a mobile phone to
select grids on the display, allowing the user to
navigate the display by zooming in. When users
press a number on the phone, the corresponding
grid on the display is highlighted and they can
choose to zoom in, creating a new and smaller
grid. Users can also use the track-stick on the
phone for fine-grain navigation.
Conclusions Interestingly, even though GridMouse
was much slower than the gyroscopic mouse, the
subjective ratings of the participants do not
indicate a great difference. The subjective
results indicate that speed is not always the
most important factor in evaluating these
devices. Four users preferred the gyroscopic
mouse (Users 1, 2, 4, 5) while two preferred
GridMouse (Users 3, 6). Beyond these basic
preferences, the average ratings for the ease of
learning to use each device were equal. This
discrepancy between speed and user satisfaction
is indicative of the fact that the speed of the
gyroscopic mouse may be hindered by the
difficulty of selecting small targets. Several
users commented on the sensitivity of the mouse
and the overall difficulty of using it. Also,
based on observations of the subjects, they all
struggled with getting the mouse to hover over
smaller targets and frequently overshot the
target multiple times. The problem of accuracy
when selecting small targets is almost completely
eliminated in GridMouse because the grids make it
easy to see how the mouse will move.
Additionally, because the track-stick moves
slowly and predictably, subjects found that it
was much easier to select a target. Another
important issue was the ergonomics of the mobile
phone. The buttons on the phone are quite small
and close together, and several users made the
mistake of pressing the wrong button. It was
particularly common for them to press 7 instead
of the key when they tried to click on a
target. This issue could potentially be resolved
by moving the left click key to a more easily
accessible button, or by using a different mobile
phone with larger buttons.

Results The user study was measured both
objectively and subjectively. Our objective
measure was the time-stamps of every click that
the user made. For the subjective portion,
participants were also asked to fill out a short
questionnaire where they rated the two devices
and evaluated their preferences.
Following their completion of the clicking task,
subjects completed the questionnaire. On this
questionnaire, they rated the ease of learning
how to use each device, as well as the ease of
actually using each device.

Question Average Ease of learning how to
use the gyroscopic mouse 4.33 Ease of
learning how to use GridMouse 4.33 Ease of
using gyroscopic mouse 3.83 Ease of using
GridMouse 3.50
Another important result of the study was seeing
what kinds of styles and techniques the subjects
developed. This was observed while they completed
the tasks. On the gyroscopic mouse, subjects
coped with the sensitivity and awkward button
positioning by holding the mouse with two
hands. While using GridMouse, on the other hand,
subjects developed a number of distinct
interaction styles. One such style was to never
zoom in. These users chose to use the numbers on
the phone to navigate between the 9 grids that
are accessible at the largest zoom. Once they had
selected the correct grid, they simply used the
track-stick to move in the direction of the
target. This eliminated the need to zoom-out
every time and subjects who used the technique
said that it felt faster. In another style,
subjects did use the zooming functionality, but
if the next target was outside the current grid
but somewhere near the current mouse location,
they got as close as they could using the grid
and then used the track-stick (Figure 4).
Figure 2. A screenshot of the task that subjects
performed on each device.
The time-stamps are an important part of the
study because they indicate how quickly the user
is able to complete the task on each device.
Figure 3 indicates the total duration of the
tasks
Materials and methods Subjects were asked to
complete navigation tasks on a large public
display (four 52 widescreen displays). Subjects
used the GridMouse software running on an iMate
SP5m mobile phone and a gyroscopic mouse
(Gyration GC15M). The mobile phone transmitted
key presses to the computer through WiFi. Each
trial consisted of the exact same distribution of
60 dots but was inverted across the x-axis and
the order was reversed to account for any
possible learning effects. Circular targets were
used because they eliminated any differences that
might result from the angle of approach.
Figure 3. The task completion times for each
subject on each device. The dashed red line is
the average times. The results from the first
participant are not included because the task was
not the same length as for users 2 through 6.
The average speeds (minutesseconds.milliseconds)
for the two devices are Gyroscopic Mouse
235.925 GridMouse 936.148 On average,
GridMouse takes 3.695 times as long.
Figure 4. How a user might navigate to the blue
target without having to zoom out. The user moves
to zone 3 (orange arrow) to get as close to the
target as possible, and then uses the track-stick
(green arrow).
Note Due to time constraints, this preliminary
study consisted of only six subjects, and thus is
not considered statistically significant.
Figure 1. Gyroscopic mouse (left), iMate mobile
phone (right)
Related work Douglas, S.A., Mithal, A.K.. 1994.
The effect of reducing homing time on the speed
of a finger-controlled isometric pointing device.
In Proceedings of the SIGCHI conference on Human
factors in computing systems celebrating
interdependence (Boston, Massachusetts, USA,
April 24 28, 1994). SIGCHI 94. ACM Press, New
York, NY, 411-416. Nielsen, J., Levy, J. 1994.
Measuring usability preference vs. performance.
In Communications of the ACM v.37 n.4 (April
1994). ACM Press, New York, NY, 66-75.
Future work During the school year, I will be
working on completing an actual user study with
15-20 subjects and writing a paper. For further
information Please contact aostberg_at_ucsd.edu.
Sas, C,. Dix, A. 2008. Designing and evaluating
mobile phone-based interaction with public
displays. In CHI 08 extended abstracts on Human
factors in computing systems (Florence, Italy,
April 05 10, 2008). CHI 08. ACM Press, New
York, NY, 3941-3944. Wobbrock, J.O., Myers, B.A.
2008. Trackball Text Entry for People with Motor
Impairments. In CHI 06 Proceedings (Montreal,
Quebec, Canada, April 22 27, 2006). CHI 06.
ACM Press, New York, NY, 479-488.