[ACT-R-users] Re: ACT-R-users Digest, Vol 7, Issue 6

[ben.willems at faa.gov]

I understand that a lot can happen during such long fixations, but would 
ACT-R be able to predict such long fixations?  I have attached a snippet 
of about 150 seconds for one of our air traffic controllers for one 
aircraft while it is entering his airspace.  Controllers in this air 
traffic control simulation dealt with 35(!) aircraft at any point in time. 
 When separting events related to a single aircraft there are clear 
patterns of task related activity.  In te example blow a controller enters 
a command to move the aircraft label, the takes responsibility for the 
aircraft.  This is perception, cognition, and motor action in the wild! We 
have tried to create "traces" if you like of observable events.  Question 
now is how we can cluster these events to form subtasks, subtasks to for 
tasks, etc.  Of course I can take an arbitrary interval between events and 
suggest that anything beyond that interval must be the start of a new 
sub-task, but I had hoped to use something that has better underpinnings 
than just Benny coming up with a number.

Anyway, thanks for all the feedback.  I hope to be able to use a model 
like ACT-R/PM in the future to see if we can explain some of the things 
that we see in our human data.

Ben




Ben Willems
Engineering Research Psychologist
William J. Hughes Technical Center
NAS Human Factors Group (ACB-220)
Building 28
Atlantic City International Airport, NJ 08405
USA
Phone:  609-485-4191
Fax:  609-485-6218
E-mail: Ben.Willems at faa.gov
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/act-r-users/attachments/20060216/a21fc6dc/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: image/gif
Size: 32461 bytes
Desc: not available
URL: <http://mailman.srv.cs.cmu.edu/pipermail/act-r-users/attachments/20060216/a21fc6dc/attachment.gif>