ACT-R multitasking

[Dario Salvucci]

A question about ACT-R and multitasking...

We have recently begun an attempt to model driving in ACT-R/PM, and
immediately the issue arose of how to handle multiple tasks.  There
seem to be at least two (related) problems:

1. Interleaving tasks.  Since ACT-R is serial, we require some
scheduling that allocates pieces of time to each task.  One
straightforward approach might use several top-level productions that
compete through conflict resolution and push task subgoals onto the
stack.  Each subgoal would do some incremental piece of the task and
return control to the higher-level productions.  However, this raises
a problem of time allocation: the noisy process may happen to allocate
large chunks of time to a single process without others getting their
share.  In addition, there is a question of whether the subtasks
should dictate when they complete or whether the scheduler should
interrupt subtasks when needed.

2. Prioritizing tasks.  In tasks such as driving, certain tasks aren't
as important as others (e.g., steering vs. changing radio stations).
How can the model allocate time according to task priorities?  Again
with the straightforward approach above, we could set conflict
resolution parameters to values that represent task priorities.
However, often these priorities are dynamic and time-related -- for
instance, after a steering correction, steering may not be important
for another one or two seconds, during which time a lower-priority
task could be executed.

I'm aware of some work on interleaving for short cognitive tasks
(e.g., Byrne's PRP work) and some other work on large-scale ACT-R
models of dynamic tasks, but I'm unclear as to how much has been done
for multitasking/scheduling of prioritized tasks.  Might anyone have
any ideas, papers, or experiences related to these issues?

Thanks,
Dario


PS: The EMMA model of eye movements and visual attention introduced at
ICCM 2000 is now available at < http://www.cbr.com/~dario/EMMA >.  The
model is in its alpha version and is probably best suited for serious
ACT-R/PM junkies. :) We expect to have a beta version that is fully
integrated into ACT-R/PM by the time the ACT-R workshop rolls around.
Feel free to email me for more information.


---------------------------------------------------------------------------
Dario Salvucci                             Email: dario at cbr.com
Cambridge Basic Research                   Phone: (617) 374-9669
Four Cambridge Center                      Fax:   (617) 374-9697
Cambridge, MA 02142  USA                   WWW:   http://www.cbr.com/~dario