Donders Institute for Brain, Cognition and Behaviour
Zoek in de site...

Thesis defense Marrit Zuure (Donders series 532)

15 December 2021

Promotors: Prof. dr. P.H. E. Tiesinga, Dr. M. X Cohen

The Origins of Response Conflict Theta: An Adventure in Source Separation

Response conflict pertains to resolving a choice between two concurrently active motor responses: one that is automatic, and another that is goal-relevant. In humans, response conflict is accompanied by a short-lived burst of theta-frequency (4-8 Hz) brain waves. The generating mechanisms of these brain waves - which are presumably the same mechanisms that implement the processing of response conflict - remain poorly understood.

This thesis aims to explore the origin of response conflict theta.

Previously, conflict theta was taken to be a signal with a single origin, varying in magnitude with the strength of conflict. We challenged the idea of conflict theta as a singular phenomenon. By applying advanced source separation methods to a unique data set consisting of combined EEG and MEG, we were able to decompose the conflict theta signal into activity from multiple sources. This finding was robust to alternate explanations. Source activity was unique and uncorrelated, suggesting that sources engaged in independent conflict-related computations. We next developed a novel extension of the source separation method used and validated it in simulated as well as real-world data. Finally, we used said extension to validate the multiple-thetas findings in LFP data recorded from rats. Results in rats were not conclusive; this discrepancy and others were discussed and integrated into a theoretical framework informing possible future work.

The findings in this thesis suggest that response conflict theta in humans, and possibly other species, is driven by multiple sources in the brain. Activity from these sources mixes together, but can be separated using advanced analysis methods. We hypothesize that individual theta sources represent local networks of neurons that perform independent conflict-processing computations. In addition to contributing to the body of response conflict literature, this thesis highlights the importance of signal de-mixing in order to investigate the functional underpinnings of cognitive processes.