Summary: New research finds that elevated glutamate-glutamine (Glx) levels in the anterior insular cortex (AIC) are linked to greater sensitivity to mistakes and to higher symptoms of anxiety and depression. Using functional magnetic resonance spectroscopy (fMRS) together with reinforcement-learning tasks, investigators showed that higher resting AIC Glx predicted a general internalizing symptom score, and that this relationship was explained by increased error sensitivity.
During reward learning, AIC Glx decreased modestly—consistent with rapid metabolic engagement—while Glx in the medial prefrontal cortex (mPFC) did not change. The findings suggest that excess glutamatergic signaling in the AIC amplifies perceived errors, which may drive the rumination and worry characteristic of anxiety and depression.
Key Facts
- AIC Glutamate: Higher resting glutamate-glutamine (Glx) in the anterior insula correlates with a transdiagnostic dimension of anxiety and depression.
- Error Sensitivity: Elevated AIC Glx is associated with a tendency to overweight prediction errors during learning, a cognitive bias linked to internalizing symptoms.
- Dynamic Changes: AIC Glx shows a modest decline during gain-based learning, indicating acute metabolic demand; mPFC Glx remains stable.
Source: Neuroscience News
New research indicates that an overactive anterior insular cortex (AIC)—a central hub for integrating bodily signals, emotion, and error monitoring—may contribute to anxiety and depression by increasing sensitivity to mistakes and negative feedback.
By combining functional magnetic resonance spectroscopy with computational models of decision-making, the study links elevated excitatory tone in the AIC to shared internalizing symptoms and clarifies how this neurochemical state affects learning from feedback.
The Brain’s Hub for Feelings and Fears
The anterior insula lies deep within the brain and plays a pivotal role in sensing internal bodily states, detecting errors, and integrating emotions into decisions. Prior imaging work has repeatedly shown higher insular activity in people with anxiety and depression, especially when they face uncertainty or receive negative feedback. The mechanisms behind that heightened activity, however, have been unclear.
Glutamate is the brain’s principal excitatory neurotransmitter and is essential for learning, plasticity, and emotional regulation. Dysregulated glutamatergic signaling has been implicated in psychiatric disorders, with excessive glutamate in frontal-insular regions linked to heightened stress responses and emotional reactivity. This raises the possibility that the AIC’s glutamatergic tone could explain why some people overreact to mistakes and become trapped in worry and rumination.
To investigate this idea, researchers recruited 56 healthy young adults for fMRS imaging and behavioral testing. Participants completed questionnaires assessing anxiety and depression symptoms, which were combined into a single general psychopathology factor (G-score) representing shared internalizing tendencies.
Linking Brain Chemistry to Errors and Emotions
Inside the MRI, participants performed a two-armed bandit decision task in which choices produced gains or losses. The task was designed to reveal how strongly each person weighted prediction errors—the difference between expected and actual outcomes—that drive learning. At the same time, single-voxel fMRS measured Glx in the AIC and in a comparison region, the medial prefrontal cortex (mPFC), both at rest and during the task.
Results showed that higher resting AIC Glx correlated with greater sensitivity to prediction errors for both gains and losses, and with higher G-scores. Crucially, error sensitivity fully mediated the relationship between AIC Glx and the G-score, indicating that increased excitatory signaling in the insula leads to exaggerated error weighting, which then contributes to anxiety-depression symptoms.
These associations were specific to the AIC: mPFC Glx did not relate to error sensitivity or to symptom scores. In other words, an elevated glutamatergic state in the anterior insula appears to bias cognition toward overestimating mistakes and negative feedback, feeding into maladaptive thought patterns.
Dynamic Changes During Learning
The study also captured rapid neurochemical changes: during the gain-learning block, AIC Glx dropped by a small but significant amount (approximately 2.2%), and remained lower afterward. This task-evoked reduction did not occur during loss-learning nor in the mPFC, suggesting a region- and context-specific metabolic response. Despite these transient dips, baseline differences persisted: participants with higher resting AIC Glx still showed greater error sensitivity and higher G-scores after the task, implying that acute metabolic fluctuations are layered on top of a more stable elevated excitatory tone.
Why Error Sensitivity Matters
Error sensitivity—the tendency to overweight mistakes or negative feedback—is adaptive when balanced, because it helps correct behavior. When it becomes excessive, however, it can drive rumination, self-criticism, avoidance, and other features of anxiety and depression. The findings suggest that an overactive glutamatergic system in the AIC magnifies error signals so that each mistake feels disproportionately salient or threatening, promoting a cycle of worry and low mood.
Given the anterior insula’s role in integrating bodily and emotional information, heightened excitatory activity there could “turn up” perceived error signals more readily than in other regions. The mPFC, while important for mood regulation, did not show the same neurochemical associations, underscoring the AIC’s specific contribution to error-related affective processing.
Toward Better Treatments
These results point to potential clinical strategies. Interventions that reduce insular glutamatergic overactivity—whether pharmacological agents that modulate glutamate, neuromodulation approaches, or targeted behavioral therapies—might lessen maladaptive error sensitivity and improve symptoms. For example, drugs that normalize glutamate signaling or neurofeedback training that helps patients lower AIC activity could reduce error overestimation and associated worry.
The study also highlights the value of combining computational models of learning with advanced neuroimaging and psychometric analysis. Measuring how people learn from feedback, how their brain chemistry shifts during tasks, and how these patterns map onto general psychopathology can reveal specific pathways linking brain chemistry to emotion and behavior.
Limitations and Future Directions
The study has limitations. The sample consisted of young, healthy adults, so findings should be replicated in larger and more clinically diverse groups. Because the data are cross-sectional, causality cannot be established: long-term anxiety or depression could also modify insular glutamate over time. Additionally, the loss block always preceded the gain block, which may have affected task-related Glx dynamics; future studies should counterbalance block order and examine longitudinal interventions that target AIC glutamate.
Despite these caveats, the evidence supports targeting insular glutamatergic signaling as a promising approach for alleviating internalizing symptoms, especially in individuals prone to excessive sensitivity to mistakes and negative feedback.
Conclusion
Elevated glutamatergic tone in the anterior insula appears to amplify how strongly we perceive and react to errors, contributing to the worry and rumination common in anxiety and depression. By identifying this neurochemical pathway, the study clarifies how brain chemistry, cognition, and emotion interact and points toward interventions that could rebalance an overactive error-monitoring system.
As the authors note: quieting overexcited error signals in the insula may help quiet the mind itself.
About this depression, neuroscience, and anxiety research news
Author: Neuroscience News Communications
Source: Neuroscience News
Contact: Neuroscience News Communications – Neuroscience News
Image: The image is credited to Neuroscience News
Original Research: Open access.
“Anterior insular cortex glutamate-glutamine (Glx) levels predict general psychopathology via heightened error sensitivity” by Bumseok Jeong et al., Frontiers in Neuroscience. DOI: 10.3389/fnins.2025.1592015
Abstract
Anterior insular cortex glutamate-glutamine (Glx) levels predict general psychopathology via heightened error sensitivity
Introduction: The anterior insular cortex (AIC) integrates interoceptive, cognitive-emotional, and error-monitoring signals, and is consistently hyperactive in anxiety and depression. Although elevated glutamate + glutamine (Glx) in fronto-insular regions has been linked to stress reactivity, it remained unclear whether AIC Glx relates to a transdiagnostic general psychopathology factor (G-score) or to a tendency to overweight prediction errors during learning.
To address this, the researchers combined functional MRS (fMRS) with reinforcement-learning modeling to test whether (i) baseline AIC Glx predicts a G-score derived from bifactor analysis of PHQ-9, GAD-7, and STAI-X1, and (ii) task-evoked Glx changes track individual differences in error sensitivity during gain- and loss-based learning.
Methods: Fifty-six healthy adults (mean age 22 ± 2 years, 16 women) completed questionnaires and performed a two-armed bandit task (40 loss then 40 gain trials) while single-voxel semi-LASER spectra were acquired from AIC and medial prefrontal cortex (mPFC) at rest and during each block. Six Rescorla-Wagner model variants were fitted to choices; the best model (lowest LOOIC) included error sensitivity, decision temperature, and value decay. Glx (CRLB < 20%) was quantified with LCModel and analyzed with repeated-measures ANOVA and Bonferroni-corrected correlations; mediation was assessed using Baron-Kenny steps (α = 0.05).
Results: Baseline AIC Glx correlated with the G-score (r = 0.39, p = 0.004) and with error sensitivity for gains and losses (r ≈ 0.41–0.44, p ≤ 0.005); mPFC Glx showed no such relations. AIC Glx fell during gain learning (−2.21%, p = 0.034) and remained low post-task, whereas mPFC Glx was unchanged. Error sensitivity fully mediated the AIC Glx–G-score link; associations were specific to Glx, not other metabolites.
Discussion: Higher excitatory tone in the AIC appears to increase prediction-error weighting, which in turn amplifies a shared anxiety-depression dimension. Dynamic Glx reductions during reward learning suggest acute metabolic demand superimposed on a trait-like baseline that biases cognition. Targeting insular glutamatergic function—pharmacologically or via neuromodulation—may therefore mitigate maladaptive error processing underlying internalizing psychopathology.