Modulating hierarchical learning by high-definition transcranial alternating current stimulation at theta frequency.

Area of Science:

• Neuroscience
• Cognitive Science
• Computational Psychiatry

Background:

• The dorsolateral prefrontal cortex (DLPFC) is crucial for hierarchical learning.
• Previous electroencephalography (EEG) studies linked frontal theta oscillations (4-7 Hz) centered on the right DLPFC (rDLPFC) to low-level prediction errors.
• The causal role of these theta oscillations in hierarchical learning requires further investigation.

Purpose of the Study:

• To determine the causal effect of frontal theta oscillations on hierarchical learning.
• To investigate how modulating theta oscillations in the rDLPFC affects behavior and computational models of learning.

Main Methods:

• Utilized high-definition transcranial alternating current stimulation (HD-tACS) to deliver 6 Hz theta stimulation or sham stimulation over the rDLPFC.
• Participants performed a probabilistic reversal learning task in varying environmental volatilities.
• Employed a hierarchical Bayesian learning and decision model to analyze computational parameters.

Main Results:

• Theta HD-tACS significantly reduced behavioral accuracy in stable environments compared to sham stimulation.
• Computational modeling revealed that theta HD-tACS increased the low-level learning rate and estimation uncertainty.
• No significant changes were observed in the decision model's temperature parameter (inverse decision noise).

Conclusions:

• Theta frequency stimulation over the rDLPFC appears to modulate the low-level learning rate.
• Environmental stability influences the impact of theta stimulation on learning, suggesting adaptive learning mechanisms.
• These findings provide causal evidence for the role of theta oscillations in hierarchical learning processes.