Biologically constrained action selection improves cognitive control in a model of the Stroop task.

Area of Science:

• Cognitive Psychology
• Computational Neuroscience

Background:

• The Stroop task is crucial for studying stimulus conflict and response selection.
• Existing models, like Cohen et al.'s, do not fully account for all Stroop task data.
• The diffusion model of choice behavior is insufficient for explaining response selection in this task.

Purpose of the Study:

• To improve computational models of the Stroop task by addressing limitations in current response selection mechanisms.
• To integrate biologically plausible mechanisms, specifically the basal ganglia/thalamo-cortical complex, into cognitive models.
• To enhance the explanatory power of cognitive models for stimulus conflict and response selection.

Main Methods:

• Adapted the Cohen et al. model of the Stroop task.
• Replaced the diffusion model's response selection mechanism with one based on the basal ganglia/thalamo-cortical complex.
• Evaluated the adapted model's ability to match core human Stroop task data and accommodate cognitive control mechanisms.

Main Results:

• The adapted model shows improved accuracy in matching core human Stroop task data.
• The new mechanism allows for principled integration of additional cognitive control mechanisms.
• The study critiques the diffusion model's suitability for response selection.

Conclusions:

• Biologically constrained models of action selection are vital for advancing cognitive models.
• The basal ganglia/thalamo-cortical complex offers a more accurate basis for response selection mechanisms in cognitive models.
• This research provides a more refined understanding of stimulus conflict and response selection in the Stroop task.