Combine the Cognitive Architecture ACT-R with user data
Project description
pyactcv - python interface
This library connects the cognitive architecture ACT-R with the programming language python to load user data into ACT-R's visicon.
The cognitive architecture ACT-R is able to monitor a human operator’s interactions with a system using the concept of model-tracing, a concept previously implemented within an ACT-R tutoring system [1]. This software library adapted the work of [2] to establish such a connection between the programming language python and ACT-R version 7.12. For exemplary usage of the library please see [3] and [4].
Installation
$ pip install pyactcv
or
$ pip install git+https://github.com/seblum/actcv
Usage
Take a look at the examples folder for an exemplary use case.
import pandas as pd
import actr
import pyactcv as cv
data = pd.read_csv('userData.csv', sep = ';', dtype = {'alarmactivecolumn' : float, 'alarmnumbercolumn' : float, 'timecolumn' : float})
header = list(data)
data = data.where((pd.notnull(data)), None)
frequency = 3000
duration = 3
starttime = 0
indexinput = 0
timebreak = 0.1
actcv = cv.ActCV(data, 'timecolumn' )
actcv.load_states()
actcv.schedule_visicon()
actcv.schedule_tone()
actr.run()
Files
-
actcv.py - Contains the class ActCV and methods to create the interface to load user data set into the visicon of ACT-R.
-
actr.py - Contains the dispatcher of ACT-R version 7.12., which is necessary to form a connection between python and ACT-R (see http://act-r.psy.cmu.edu/).
TODO
Possible additional feature to add:
- Add more dynamic read in for data
- Add selection of what to load ("visual", "audio")
- Add debugging support
Developing pyactcv
To install pyactcv along with the tools to develop and run tests please run the following in your virtualenv:
$ pip install -e .[dev]
Bibliography
[1] Fu, W.-T., Bothell, D., Douglass, S., Haimson, C., Sohn, M.-H., & Anderson, J. (2006). Toward a real-time model-based training system. Interacting with Computers, 18(6), 1215–1241.
[2] Halbruegge, M. (2013). Act-cv - bridging the gap between cognitive models and the outer world. In E. Brandenburg (Ed.), Grundlagen und Anwendungen der Mensch- Maschine- Interaktion: 10. Berliner Werkstatt Mensch- Maschine-Systeme (pp. 205–210). Berlin: TU Berlin.
[3] Klaproth, O. W., Halbruegge, M., Krol, L. R., Vernaleken, C., Zander, T. O. and Russwinkel, N. (2020). A Neuroadaptive Cognitive Model for Dealing With Uncertainty in Tracing Pilots’ Cognitive State. Topics in Cognitive Science, 12(3), p. 1012-1029.
[4] in review
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.