Abstract
Anxiety influences how the brain estimates and responds to uncertainty. These behavioural effects have been described within predictive coding and Bayesian inference frameworks, yet the associated neural correlates remain unclear. Recent work suggests that predictions in generative models of perception are represented in alpha-beta oscillations (8-30 Hz), while updates to predictions are driven by prediction errors weighted by precision (inverse variance; pwPE) and encoded in gamma oscillations (>30 Hz). We tested whether state anxiety alters the neural oscillatory activity associated with predictions and pwPE during learning. Healthy human participants performed a probabilistic reward-learning task in a volatile environment. In our previous work, we described learning behaviour in this task using a hierarchical Bayesian model, revealing more precise (biased) beliefs about the reward tendency in state anxiety, consistent with reduced learning in this group. The model provided trajectories of predictions and pwPEs for the current study, allowing us to assess their parametric effects on the time-frequency representations of EEG data. Using convolution modelling for oscillatory responses, we found that, relative to a control group, state anxiety increased alpha-beta activity in frontal and sensorimotor regions during processing pwPE, and in fronto-parietal regions during encoding predictions. No effects of state anxiety on gamma modulation were found. Our findings expand prior evidence on the oscillatory representations of predictions and pwPEs into the reward-learning domain. The results suggest that state anxiety modulates oscillatory correlates of pwPE and predictions in generative models, providing insights into a potential mechanism explaining biased belief updating and poorer reward learning.
Significance Statement Learning plays a central role in clinical and subclinical anxiety. This study tests whether a temporarily-induced state of anxiety in healthy human participants alters the neural oscillatory patterns associated with predicting and learning from rewards. We found that precision-weighted prediction errors were associated with increases in alpha-beta oscillations in our state anxious group. This finding suggested that anxiety states may inhibit encoding of relevant signals conveying the discrepancy between the predicted and observed reward. State anxiety also increased alpha-beta activity during processing predictions, indicating a stronger reliance on prior beliefs about the reward tendency. The results identify the alteration in alpha-beta oscillations as a candidate mechanism explaining misestimation of uncertainty and maladaptive learning in anxiety.
Competing Interest Statement
The authors have declared no competing interest.
Footnotes
Conflict of interest statement: The authors declare no competing financial interests.