Forschung

•   Instruction-based Learning

Volitional control of action allows for the flexible selection of the one behavior that appears most appropriate in the current situation. To behave “appropriately” poses several challenges to an organism, depending on its learning history. If the goal is known (e.g. to have a coffee), but the means to achieve this goal in the current situation are unknown, it is crucial to figure out what the “correct” action is and to avoid errors. This can be sovlved via trial-and-error learning which relies on evaluating retrospectively whether an action performed under certain stimulus conditions yielded desirable outcomes. By contrast, exploiting our highly developed communication skills humans can use explicit instructions as an acquisition shortcut specifying prospectively how to yield intended outcomes under the appropriate stimulus conditions (e.g., by consulting the operating instructions of the coffee maker or by observing someone else demonstrating the operation of the machine). The central aim of this project is to advance our surprisingly limited knowledge of the neurocognitive mechanism underlying the instruction-based acquisition of novel goal-directed behaviors. This research is supported by a DFG-grant of Hannes Ruge and Uta Wolfensteller (SFB940 project A2).

• Action Control

The question how (and why) people act in a goal-oriented manner is one of the core questions in psychology. We explore this question focusing on fundamental associational learning mechanisms as well as competition resolution mechanisms.

Learning Mechanisms

A fundamental prerequisite of goal-oriented behavior is to correctly recognize associations between events, for instance, between a certain action (pressing a green button) in a certain context (the coffee maker in the cafeteria) and a certain subsequent event (coffee). What is remarkable is that people are extremely fast in learning such associations between events. By means of functional magnetic resonance imaging (fMRI), one main goal of the project is to elucidate the brain mechanisms that underlie the rapid learning of such associations and are hence fundamental to goal-directed behaviour. What is special about learning the type of associations that ultimately enable goal-oriented behaviour, compared to learning associations allowing stimulus-driven behaviour, or event anticipation lacking implications for one’s own behaviour? Different variations of the experimental paradigm will provide insight not only regarding whether (i.e., under which circumstances) and how rapid learning is implemented in the brain, but also whether and how learnt associations are translated into assumptions and expectations in the context of goal-directed behavior. This research is supported by a DFG-grant of Hannes Ruge and Uta Wolfensteller (RU 1539/2-1 and SFB940 project A2).

Competition Resolution

Another crucial aspect of volitional control pertains to the fact, that in virtually any situation we can choose among a bunch of different actions, each of which has successfully served to yield different goals in the past (just imagine the numerous goals you could pursue by different uses of a kitchen knife). There is rarely one single correct action in a given situation and the goal to simply “take the correct action” is obviously underspecified unless a specific goal is determined. Thus, in this case, volitional control means selecting the action that has previously served to achieve specifically the currently pursued goal. Importantly, action selection is typically not only challenged due to competing goals or competing action-goal associations, but also due to directly stimulus-induced action tendencies that may override any goal-directed choices. It is therefore crucial to examine goal-directed action control in direct contrast to stimulus-bound control and to consider possible variables that may shift the balance between these competing influences on choice behavior. This research is supported by a DFG-grant of Hannes Ruge and Uta Wolfensteller (SFB940 project A2).

•   Preparatory control in task switching

Human behavior is flexible in that it is organized according to the anticipated future consequences of action (i.e., goals) rather than merely reacting upon current sensory inputs (see project ‘action control’). Another variety of behavioral flexibility is that goals can be adjusted and switched momentarily. This ability has been extensively examined within the task switching paradigm. Using fMRI and EEG methodologies, we have been interested in characterizing the multi-facetted nature of preparatory processes in anticipation of implementing a changed task goal.

• Fmri Design & Analysis

FMRI-based mental chronometry

Mental processes naturally evolve across time both serially and in parallel. An important endeavor tightly linked to the classical notion of mental chronometry therefore is to determine the flow of information processing that transforms sensory inputs into high-order mental representations and overt behavior. Due to the slow evolution of event-related hemodynamic responses fMRI-based mental chronometry has been notoriously difficult. This project attempts to recover temporal process information by extracting BOLD response timing parameters associated with the experimentally controlled temporal structure of successive events. Currently, this approach is used to evaluate the impact of random vs. blocked foreperiods in forced-choice and task switching paradigms. Furthermore we explore the challenges and potential benefits of sub-second whole brain acquisition of functional MRI data for fMRI-based mental chronometry.

Utilization and development of advanced analysis methods.

• Large-scale pattern classification.
• Global functional network analysis.
• EEG-informed FMRI-based mental chronometry using simultanenous EEG-FMRI acquistion. 

This research is supported by a DFG-grant of Hannes Ruge and Michael Marxen (SFB940 project Z2).