Most people know MRI as a way of taking pictures of the brain. Far fewer know that the same scanner can also measure the chemical composition of small brain regions — without contrast injection, without radiation, and in a single sitting. That technique is magnetic resonance spectroscopy (MRS), and over the past two decades it has become an important tool for studying the chemistry of schizophrenia.
MR spectroscopy lets researchers non-invasively measure concentrations of brain metabolites — including glutamate, GABA, glutathione, and N-acetylaspartate — and has produced some of the most cited recent evidence implicating glutamate dysfunction in schizophrenia.
How MRS works
MR spectroscopy uses the same nuclear magnetic resonance physics as MRI but reads out the signal differently. Each chemical in the brain has a slightly different magnetic environment, producing a distinct frequency in the recorded signal. By analysing those frequencies, researchers can identify and quantify several metabolites within a defined voxel — typically a cube a few centimetres on a side.
The most commonly measured metabolites in schizophrenia research include:
- Glutamate (Glu) and glutamate+glutamine (Glx) — the brain's main excitatory neurotransmitter system
- GABA — the main inhibitory neurotransmitter
- N-acetylaspartate (NAA) — often interpreted as a marker of neuronal integrity
- Creatine and phosphocreatine — markers of energy metabolism
- Glutathione (GSH) — a major brain antioxidant
- Choline-containing compounds — markers of cell membrane turnover
MRS at higher field strengths (3T, 7T) gives better separation of overlapping peaks, which has been particularly important for distinguishing glutamate from glutamine and for measuring GABA reliably.
What MRS has shown in schizophrenia
Glutamate
The largest and most replicated MRS finding in schizophrenia concerns glutamate. Multiple meta-analyses, including the influential 2016 paper by Merritt and colleagues in JAMA Psychiatry, have shown:
- Elevated glutamate in the medial frontal cortex and basal ganglia in clinical high-risk and first-episode patients
- A pattern of glutamate change that may shift with illness stage and treatment
- Higher glutamate in some studies of treatment-resistant schizophrenia
These findings have helped anchor the glutamate hypothesis of schizophrenia in direct human evidence, complementing pharmacological models based on NMDA antagonists like ketamine. See our piece on the glutamate hypothesis.
GABA
GABA MRS has produced more variable findings. Some studies show reduced GABA in prefrontal regions in schizophrenia, consistent with post-mortem findings of altered parvalbumin-positive interneurons. Other studies have not replicated this. The technical demands of GABA MRS — particularly editing techniques like MEGA-PRESS — partly explain the variability.
NAA
Reduced NAA, often in prefrontal and hippocampal regions, has been one of the longest-standing MRS findings in schizophrenia. NAA is sometimes interpreted as a measure of neuronal integrity, though it is also affected by mitochondrial function and brain energetics. Reduced NAA has been observed in chronic schizophrenia and to a lesser extent in first-episode patients, suggesting both early and progressive components.
Glutathione
Glutathione is the brain's most abundant antioxidant. Several MRS studies have shown reduced cortical glutathione in schizophrenia, supporting models that include oxidative stress as a contributor. This has fed interest in N-acetylcysteine (NAC), a glutathione precursor, as an adjunctive treatment. See our pieces on NAC in schizophrenia.
Energy metabolism
Phosphorus MRS (31P-MRS) and proton MRS measures of creatine and lactate have suggested altered brain energy metabolism in schizophrenia. This has informed renewed interest in metabolic interventions including the ketogenic diet.
Limitations and caveats
MRS findings come with important caveats:
- Spatial limitation. A single MRS voxel covers a few cubic centimetres. Whole-brain MRS is harder and lower resolution. Researchers usually choose voxel locations based on hypothesis, which can introduce bias.
- Quantification challenges. Absolute metabolite concentrations are sensitive to the calibration method. Many studies report ratios (e.g., Glu/Cr) which depend on whether the denominator is itself stable.
- Overlapping peaks. Glutamate and glutamine peaks overlap at lower field strengths, leading many studies to report combined Glx rather than separate Glu and Gln.
- Medication effects. Antipsychotics may alter measured metabolites in ways not yet fully understood.
- Population variability. Sex, age, smoking, and other factors influence metabolite levels and need to be controlled.
From MRS to drug development
One of the practical impacts of MRS findings has been to motivate clinical trials of glutamate-modulating agents in schizophrenia. The 2024 FDA approval of xanomeline-trospium (Cobenfy), a muscarinic-targeted antipsychotic, was not directly based on MRS data, but the broader push beyond dopamine that the approval reflects is informed by MRS evidence of glutamate dysfunction.
Several clinical trials have used MRS as a pharmacodynamic biomarker — measuring how a candidate drug changes brain glutamate or GABA — to test target engagement before committing to large efficacy studies. This is one of the more promising applications of MRS for drug development.
Clinical use
MRS is not used to diagnose schizophrenia. There are some narrow clinical applications — for example, measuring NAA in suspected mitochondrial disorders or differentiating tumour types — but psychiatric clinical use is essentially non-existent. The findings remain a research tool.
MRS measures total tissue concentrations within a voxel. It does not directly measure synaptic neurotransmitter release, the rate of metabolic flux, or function. Interpretations linking MRS measurements to specific functional changes require care.
Where the field is going
Several directions are active:
- Higher-field MRS at 7T and beyond, allowing better separation of metabolites
- 2D MRS techniques that resolve overlapping peaks
- Functional MRS — detecting metabolite changes in response to a task or stimulus
- Multimodal studies combining MRS with PET, fMRI, and EEG
- Larger longitudinal datasets to track metabolite changes across illness stages
For more imaging perspectives, see our pieces on fMRI, PET imaging, and brain volume changes.
This article is for educational purposes only and is not medical advice, diagnosis, or treatment. Always consult a qualified mental health professional. If you or someone you know is in crisis, call or text 988 in the US, or your local emergency number.