Learning of irrelevant stimulus-response associations modulates cognitive control

Area of Science:

• Cognitive Neuroscience
• Computational Neuroscience

Background:

• Conflict tasks like the Simon task show effects modulated by trial proportions.
• Two mechanisms proposed: control learning and irrelevant stimulus-response (S-R) learning.
• Neural basis of S-R learning modulation on conflict effects is poorly understood.

Purpose of the Study:

• Investigate neural substrates of conflict effects modulated by irrelevant S-R learning.
• Utilize a model-based fMRI approach to quantify S-R learning.
• Examine the relationship between learning prediction errors and cognitive control.

Main Methods:

• Employed a model-based functional Magnetic Resonance Imaging (fMRI) study.
• Dynamically altered congruent/incongruent trial proportions in a Simon task.
• Quantitatively modeled irrelevant S-R association learning and analyzed prediction errors.

Main Results:

• Behavioral data revealed unsigned prediction errors correlated with reaction times.
• fMRI results indicated fronto-parietal and cingulo-opercular networks modulated by unsigned prediction errors.
• Neural modulation occurred irrespective of trial congruency, linking prediction errors to reactive control.

Conclusions:

• Irrelevant stimulus-response learning modulates reactive cognitive control.
• This provides a novel mechanism for cognitive control modulation, distinct from control learning.
• Highlights the role of learning in dynamic cognitive control adjustments.