Learning Cognitive Flexibility: Neural Substrates of Adapting Switch-Readiness to Time-varying Demands.
Anthony W Sali1, Christina Bejjani2, Tobias Egner2
1Wake Forest University, Winston-Salem, NC.
Journal of Cognitive Neuroscience
|November 27, 2023
Summary
Cognitive flexibility, or task-switching ability, adapts to changing environments through reinforcement learning. The frontoparietal cortex tracks task-switch likelihood, optimizing performance in dynamic settings.
Area of Science:
- Cognitive Neuroscience
- Neuroeconomics
Background:
- Cognitive flexibility, the ability to switch between tasks, is crucial for navigating dynamic environments.
- Adaptation in cognitive flexibility is influenced by the statistical regularities of the environment, particularly the frequency of task switching.
- The neural mechanisms underlying these moment-to-moment adjustments in cognitive flexibility remain incompletely understood.
Purpose of the Study:
- To investigate the neural basis of adaptive cognitive flexibility in response to changing task-switching demands.
- To explore how reinforcement learning mechanisms contribute to behavioral and neural adjustments in task-switching costs.
- To identify brain regions involved in tracking the likelihood of upcoming task switches.
Main Methods:
- Utilized a cued task-switching paradigm with manipulated switch probabilities across blocks.
- Employed functional magnetic resonance imaging (fMRI) to capture brain activity during task performance.
- Applied reinforcement learning models to behavioral reaction time (RT) data to derive trial-wise prediction errors.
Main Results:
- Behavioral switch costs decreased as the probability of switching increased, demonstrating adaptive behavioral adjustments.
- Neural switch costs were identified in the lateral and medial frontoparietal cortex.
- Activity in the lateral frontal and parietal cortex correlated positively with unsigned switch prediction error, indicating a tracking of switch likelihood.
Conclusions:
- The frontoparietal cortex plays a key role in mediating cognitive flexibility by adapting to varying task-switching demands.
- Reinforcement learning signals, specifically unsigned prediction errors related to task switches, are encoded in the frontoparietal cortex.
- These findings highlight the brain's capacity to dynamically adjust cognitive control based on environmental statistics to optimize task performance.
