TU Berlin

Kognitive Modellierung in dynamischen Mensch-Maschine-SystemenLinda Heimisch


Inhalt des Dokuments

zur Navigation

Linda Heimisch


Wissenschaftliche Mitarbeiterin



  • Kognitive Modellierung
  • Schnittstelle zwischen kognitiver Modellierung und EEG
  • Methoden der Kognitions- und Neurowissenschaften, Datenanalyse
  • Mathematische Modellierung



Bringing together Cognitive Models and EEG through Hidden Semi-Markov Models. An analysis of processing stages in a Mental Rotation task

Can we gain information from neuronal data that has the potential to refine computational cognitive models, make them more plausible, or improve their predictions about human performance in cognitive tasks? In order to give an answer to this question, one also must address two underlying questions: To what extent are neuronal data a good benchmark for the true nature of mental processes? And how well do neuronal data analysis and computational cognitive modelling go together, regarding both methodological/statistical aspects and theoretical/epistemological assumptions? In my thesis I will transform neuronal data obtained from an electroencephalography (below: EEG) study into a stochastic model that will allow to quantitatively compare these data to a cognitive model that is implemented in the cognitive architecture ACT-R. Thereby, I want to evaluate whether the assumptions that the ACT-R model makes about the number and the duration of cognitive stages involved in a Mental Rotation process match with the neuronal stages extracted from the EEG data - and, in case of mismatches, propose and discuss a way how the information obtained from the EEG data could be used to improve the ACT-R model. Particular emphasis will be put on the question if the assumptions that the ACT-R model makes about the contribution of a newly developed spatial module to the Mental Rotation process have a close match in the stochastic model derived from the EEG data. 



  • 2017 - heute   Humboldt-Universität zu Berlin
                          M.Sc. Mind and Brain - Track Brain

  • 2013 - 2017    Albert-Ludwigs-Universität Freiburg
                          B.A. Kognitionswissenschaft und Politikwissenschaft



  • 2019 - heute    Universität Potsdam
                           Studentische Hilfskraft im Sonderforschungsbereich 1294
                           Data Assimilation - Projekt B03
                           Parameter inference and model comparison in dynamical cognitive models

  • 2017 - 2018     Max-Planck-Institut für Bildungsforschung, Berlin
                           Studentische Hilfskraft am Harding-Zentrum für Risikokompetenz
                           Forschungsbereich: Adaptives Verhalten und Kognition


Combining ACT-R models with EEG data
Zitatschlüssel heimisch2021
Autor Heimisch, L.
Buchtitel Talk held at the 19th International Conference on Cognitive Modeling
Jahr 2021
Monat July
Link zur Originalpublikation Download Bibtex Eintrag



Schnellnavigation zur Seite über Nummerneingabe