Collaborators: [Michael
Dorr]
Goal: To make gaze guidance on a PC monitor smoother (more comfortable).
When one writes a text on your PC monitor, there are many visual
distractions occurring, such as suddenly appearing advertisement in your
browser, email notifications and pop-up dialogs informing you about on-going
processes. Many of these distractions will capture your gaze – even if it is
for a fraction of a second only. But finding back to the cursor position in
your text editor is not always straight forward – it often results in a short
visual search, which one is hardly aware of. It would be better if gaze was
recaptured after such a distraction. Although the editor cursor is blinking, in
many situations this ‘blinking’ is not enough, because once you leave the
editor window, your gaze is too peripheral to sense the blinking cursor. In
this project, we attempt to make the cursor size gaze-contingent, that is, the
farther away your gaze moves away your cursor – the more eccentric it is -, the
larger the cursor becomes and that should therefore facilitate your way back to
the position of text entry. We call it the gaze-recapturing editor cursor
(GREC). The GREC is not only beneficial when the writer was distracted by
notifications, but also for instance when switching between windows and copying
and pasting text.
Method: The goal was to engage the subject into a task which required frequent
scrolling within the windows and switching between windows in order to exploit
the use of the gaze-contingent cursor as much as possible, given the limited
duration of a typical evaluation. Subjects were asked to create hyperlinks
between specific words in a text placed in a word document and specific web
addresses. At the end of the test, subjects were asked to evaluate the use of
the gaze-contingent cursor. The specific instructions were as follows (see file
task instructions).
Results: We are in the process of testing.
Discussion: The broad success of the gaze-recapturing editor cursor ultimately
depends on the success of providing a PC monitor with a cheap, unobtrusive
eye-tracker. The technology exists in principal, e.g. [link].
Such eye-trackers do not necessarily need to be very accurate. We think an
accuracy of 1 degree is sufficient for a GREC.
The project is a spin-off from our research on a
gaze-guidance system:
[link to project website] the website of the 5-lab European
collaboration
[pdf] manuscript full of ideas for human-computer
interaction
[talk] slides about
this issue