[ACT-R-users] ACT-R for simulation purposes

[Ana Maria Navarro Cerda]

Dear all,
In the framework of a project of the 7th Framework Programme, called VERITAS, we are aiming to construct virtual user models that represent the capabilities and disabilities of both elderly and disabled users. For developing these virtual models, two parallel researches are being performed: one focused on the existing suitable physical models, and the other focused on cognitive models. These cognitive models should allow representing cognitive disabilities and psychological states such as stress and fatigue. The virtual models will be use in a simulation where they will perform a number of tasks in different domains: automotive, workspace, infotainment, etc. The purpose of this simulation is to see whether this virtual model representing a human being with a disability, can perform the proposed tasks in a simulation.(ie: a person with upper-limb impairment won't be able to perform a range of tasks with the hands or a person with a severe memory impairment cannot retain new memories and therefore cannot memorize a telephone number).
After an initial research, it seems that ACT-R is one of the most promising cognitive architectures, since it allows us to "moderate/overlay" the models in order to represent a varied number of users (in our case it can be used to modified the performance of user's memory and attention that decreases for elderly and some cognitive disabilities).
As I mentioned before, in our project, these "virtual models" are the input of a simulation. This simulation has as input these virtual models, but also the simulation models ( based on a composition of tasks previously defined in XML).
So, if I'm not mistaken, if you want to simulate a person performing a task (ie: driving) you have to "write in ACT-R" all the information regarding the declarative and procedural memory. In this example model the productions that model how people drive and the facts (chunks) of driving.
However, there are some issues that remain unclear to me:

·         These simulation models  we are developing (task models) are not fixed, we have a repository of tasks and the developers can "compose" the final simulation model as a sequence of tasks, allowing to have an infinite number of simulation of models. So we would need that Act-R allowed certain modularity and reuse of the code for the models. Is this feasible? Is there any modification of ACT-R allowing this?

·         Another question is concerning the simulation: if there are external events happening in the simulation (ie: a car in front of us when driving),  should these "events"  also be part of the declarative memory? If the answer is yes, do we have APIs that allow us to integrate this "events" into chunks in an automatic way?

·         Does it exist t any open repository of ACT-R tasks models that we could use?


I would really appreciate any advise you could give me. This is a new world for me, and it's difficult to go into it without getting lost.
Best regards and thanks in advance,
Ana Navarro

Ana Navarro Cerdá
R&D Department
ITACA-TSB
[cid:image001.gif at 01CBD285.3F517F20] Health & Wellbeing Technologies
Univ. Politécnica de Valencia
Edificio G8 - Camino de Vera s/n
46022 Valencia

tel:          +34 96 387 76 06
fax:         +34 96 387 72 79
e-mail: annacer at itaca.upv.es<mailto:annacer at itaca.upv.es>
http://www.tsb.upv.es









-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/act-r-users/attachments/20110222/5e0a004a/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.gif
Type: image/gif
Size: 598 bytes
Desc: image001.gif
URL: <http://mailman.srv.cs.cmu.edu/pipermail/act-r-users/attachments/20110222/5e0a004a/attachment.gif>