A recent study published in iScience by Rakshita Deshmukh and Arjun Ramakrishnan from the Indian Institute of Technology Kanpur explores how high trait anxiety affects the brain's ability to distinguish between genuine environmental changes and random noise. Their research shows that anxious individuals often misinterpret random fluctuations in rewards as indications that the rules have changed. Remarkably, the study also indicates that a single night of deep, slow-wave sleep, known as N3 sleep, can help recalibrate this misunderstanding, aligning anxious learners' strategies closer to optimal performance.
Read More
The researchers designed a three-option probabilistic reversal-learning task, framed as a fishing game, where participants selected among three islands to maximize their catch. This task allowed for independent manipulation of stochasticity, or the variability of outcomes under stable conditions, and volatility, the frequency of changes in reward contingencies. In low-stochasticity environments, the best option yielded rewards on 80 percent of trials, compared to only 60 percent in high-stochasticity conditions, where outcomes were less reliable. The volatility was adjusted by changing which option was best every 25 trials in slow conditions and every 10 trials in fast conditions.
The first experiment involved fifty participants who completed six blocks of trials across all combinations of stochasticity and volatility. The study assessed trait anxiety using the State-Trait Anxiety Inventory. Results revealed that individuals with high trait anxiety (a score of 45 or above) earned considerably fewer rewards than their low-anxiety counterparts, specifically in environments that combined stable yet noisy conditions.
The performance gap was attributed to the elevated learning rates observed in high trait anxious individuals, not due to a generalized sensitivity to volatility but rather to an inability to appropriately discount random noise during stable environments. Higher learning rates under these conditions led to reduced reward accumulation for anxious participants, suggesting they were misattributing random fluctuations as meaningful changes.
In the second experiment, forty participants performed the task before and after a night of sleep in the laboratory, monitored with a wearable EEG device. Findings indicated that greater N3 sleep correlated with reductions in state anxiety, and intriguingly, high trait anxious individuals showed more significant reductions in anxiety with increased N3 sleep. Increased N3 sleep also linked to higher post-sleep reward earnings and better identification of the most rewarding option in anxious participants, particularly in stable environments.
This study highlights a key finding: N3 sleep may play a corrective role in miscalibrated anxiety, reinforcing optimal learning strategies while reducing the adverse effects of trait anxiety on decision-making. Although the research suggests promising implications for enhancing learning processes in anxious individuals, the authors note that further studies are needed to address constraints such as the sample population and potential confounding factors.