>If ANY points stray outside of either area (during lock-on or selection) then I discard the whole thing and the user can start again.
Oh I think I see it now.
I was thinking of the exiting of a magnification box.
I just read this part of the documentation:
> Point lock-on radius (pixels):
>
> Specific to 'Point selection source' of 'Fixations', this setting dictates the size of the initial 'lock-on' area.
> Your attention must remain completely inside this area for the lock-on time before a fixation begins.
> A larger radius results in a larger lock-on area (the area is a circle) making it easier to keep your attention within the bounds and begin a fixation, but reduces the accuracy of your selection.
> A smaller radius results in a smaller lock-on area making it more challenging to keep your attention within the bounds and begin a fixation, but increasing the accuracy of your selection
>
> Point fixation radius (pixels):
>
> Specific to 'Point selection source' of 'Fixations', this setting dictates the size of the fixation area (after the initial lock-on).
> Your attention must remain completely inside this area for the fixation to complete.
> A larger radius results in a larger fixation area (the area is a circle) making it easier to keep your attention within the bounds and complete the fixation, but more difficult to deliberately break the fixation by directing your attention elsewhere.
> A smaller radius results in a smaller fixation area making it more challenging to keep your attention within the bounds and complete the fixation, but easer to deliberately break the fixation by directing your attention elsewhere.
That's kind of like the invisible "air brush" circle I was thinking of.
I'm guessing that it's like letting us choose the brush size.
Choose the circle radius that will enclose and consider the gaze points to be deemed adjacent and close, and thus worthy of an acceptance that the user is showing interest.
Interest is either for lock-on to begin fixation, or the fixation itself.
I didn't know that radius was tweakable, and I didn't know that that there were 2 radiuses, with the fixation circle being larger than the beginning lock on radius (beginning lock on radius is smaller to prevent false starts, and the creation of ready-to-proceed-and-fixate circles that might need to be escaped?).
I thought that fixation was set by Eye Tribe: viewtopic.php?f=8&t=277&p=1163&hilit=fixation&sid=8492dfa2cd3960590de4306fc0f01064#p1163
Looking at MastaLomaster’s comment:
> In particular, your "fixation radius" is called gDISPERSION_LIMIT in my program;
the "fixation period" is called FIXATION_LIMIT.
>Thank you for sharing this, but I consider the fixation detection as solved in my program.
I guess that it’s up to the programmer to let some parameters be loose.
E.g. I don't recall ever seeing a setting for a lock on fixation before the main fixation in a program before.
I haven't delved deeply in to various eye-tracking programs, but I usually just see fixation time".
Here, it's "lock radius", "lock time", "fixation radius", "fixation time".
(And that's excluding adding magnification options).
---
In regards to thinking that the “If ANY points stray outside of either area” had to do with magnification, there are indeed magnification settings too:
> Magnify source (percentage of screen): Size of the area which will be captured around the selected point for magnification.
>
> Magnify destination (percentage of screen): Size of the area in the middle of the screen which will display the magnified area.
We can change the size of the magnification box, and how much area around the fixation is captured to the box?
Sweet.Statistics: Posted by skataJeffKang — 31 Jul 2015, 07:28
]]>
