 |
 |
 |
 |
 |
Forschung > HeiKA 2013
HeiKA 2013 Project: Combining Statistical and Computational Modeling Approaches for User State Prediction
Information from the application:
We propose a collaborative research project between the Section of Experimental and Theoretical Psychology (Prof. Joachim Funke) of Heidelberg University and the Cognitive Systems Lab (Prof. Tanja Schultz) at the Karlsruhe Institute of Technology. The aim of this collaboration is the development of a neuro-cognitive user model as the basis of assistive computer technology for complex mental tasks. The model will predict user states like confusion, confidence or mental load by integrating user interaction behavior and neural activity (measured by EEG) with a computational model implemented in the cognitive architecture ACT-R. As such it will contribute both to the rapidly developing field of integrated neuro-cognitive user modeling and address basic science questions related to the cognitive validity of neural EEG markers.
The collaboration brings together the specific expertise of two internationally established research groups in a field where the close cooperation between engineering and behavioral sciences is a critical success factor (KA: biosignal processing, machine learning and human-computer interaction; HD: cognitive modeling and complex problem solving research). The combination of anthropological and technical aspects in human-computer interaction fits the HEiKA research bridge “Nature, Technology and Society” (NTS) and also serves the HEIKA objective of achieving international excellence through combining complementary competencies from HD and KA.
The project builds on an initial collaboration effort of both groups that already resulted in first scientific outputs (Putze et al, 2012), but as of now lacks seed funding to develop it to a point where a full DFG research proposal becomes feasible. We therefore think that this project represents a good match for objectives and scope of HEiKA. The aims outlined in the detailed proposal define a research program of 1.5 years, including preparation and dissemination. We plan to execute two experiments with EEG recordings and one behavioral experiment.
We expect multiple lasting results of the project: 1) A comprehensive data corpus which can be further examined to target new research questions. 2) The computational model and EEG analysis protocols will enable future research in the area of user behavior prediction. 3) Research results as well as the developed model itself will be released to the scientific community in conference and journal publications of different communities 4) Finally, those publications will document the success and the benefits of a collaboration between both institutions and form a basis for a joint DFG research proposal, which will address the transfer of methodology and results to a more ecologically valid task setting and the implementation of adaptive assistive technology based on the user model developed.
Project Partner
Prof. Dr. Tanja Schultz, Cognitive Systems Lab, KIT Karlsruhe.
Project Assistants
Dr. Daniel Holt (Heidelberg), Dipl.-Ing. Felix Putze (Karlsruhe).
Funding and resources
A one year seed funding for this project, starting January 1, 2013, is given by HeiKA, the Heidelberg Karlsruhe Research Partnership.
Reference
Putze, F., Holt, D. V., Meyer, J., Borné, J., Schultz, T. & Funke, J. (2012). Combining cognitive modeling and EEG to predict user behavior in a search task. 11th International Conference on Cognitive Modeling (ICCM), Berlin, Germany.
|
 |
|
 |
|||||||||||
|
||||||||||||||
|
|
|
|
|
||||||||||
 |
|
 |
||||||||||||
|
||||||||||||||
 Top |
Zuletzt bearbeitet am 25.07.2013 von JF. |