A MEG Study of Visual Repetition Priming in Schizophrenia: Evidence for Impaired High-Frequency Oscillations and Event-Related Fields in Thalamo-Occipital Cortices

Andreas Sauer, Tineke Grent-‘t-Jong, Michael Wibral, Michael Grube, Wolf Singer, Peter J. UhlhaasFrontiers in Psychiatry 2020 · DOI

TL;DR

This study used MEG to investigate how the brain processes repeated visual stimuli in individuals with schizophrenia. While patients performed similarly to controls on a behavioral task, their brain activity showed significant deficits in processing novel stimuli and in adapting to repeated stimuli, particularly in high-frequency brain waves and early electrical responses. This suggests a fundamental impairment in how the brains of individuals with schizophrenia modify neural circuits based on experience.

What Problem Does This Paper Address?

Schizophrenia is characterized by significant cognitive deficits that predict functional outcomes and are not easily treated with medication. While these deficits are often attributed to higher-level cognitive functions, there's growing evidence for impairments in sensory processing and the brain's ability to adapt to new information. A key mechanism for learning and memory is synaptic plasticity, which involves experience-dependent changes in neural connections. This paper investigates whether schizophrenia patients have difficulties with "repetition suppression" and "repetition enhancement"—neural processes that reflect how the brain adapts its response when a stimulus is presented multiple times. Understanding these fundamental neural adaptation mechanisms could shed light on the underlying causes of cognitive impairments in schizophrenia.

Methods

The researchers employed magnetoencephalography (MEG), a non-invasive neuroimaging technique that measures magnetic fields produced by electrical activity in the brain, offering excellent temporal resolution. They used a visual priming paradigm where participants viewed images. The core of the experiment involved presenting images in two conditions:

  1. NOLAG: Three identical images presented consecutively.
  2. LAG: Three identical images presented with 5-15 different intervening images. This condition was crucial to distinguish repetition effects from simple habituation.

MEG data were analyzed for both:

  • Event-Related Fields (ERFs): These are time-locked electrical responses to stimuli, providing information about the timing and location of neural activity.
  • Spectral Power: This analysis examined the strength of brain oscillations (brain waves) across a wide frequency range (1-200 Hz), focusing on gamma-band (high-frequency) and beta-band (mid-frequency) oscillations, which are thought to be critical for neural processing and plasticity.

Regions of interest (ROIs) were defined in the occipital cortex (involved in visual processing) and thalamus (a relay center for sensory information). Virtual channels were created to analyze data from these specific brain areas.

Datasets

Not discussed in the paper. The study used data collected from participants in their experiment.

Key Insights and Ideas

  • Dissociation between Behavior and Brain Activity: The most striking finding is that while patients with schizophrenia performed behaviorally on par with controls, their brain activity revealed significant differences. This highlights that behavioral performance alone might not capture the full extent of neural dysfunction in schizophrenia.
  • Impaired Neural Adaptation: The study strongly suggests that individuals with schizophrenia have a fundamental deficit in how their brains adapt to repeated sensory input. This isn't just about recognizing something familiar; it's about the brain's ability to efficiently update its processing based on that familiarity.
  • Role of High-Frequency Oscillations: The research points to a critical role for gamma-band oscillations in processing novel stimuli and for beta-band oscillations in repetition enhancement. Deficits in these specific frequency bands in schizophrenia patients suggest a disruption in the neural mechanisms underlying learning and prediction.
  • Thalamo-Occipital Involvement: The observed deficits were not confined to visual processing areas (occipital cortex) but also extended to the thalamus, indicating that impairments in sensory relay and integration might contribute to these issues.

Key Results

  • Behavioral Priming: No significant difference was found between schizophrenia patients and healthy controls in behavioral priming (i.e., faster reaction times or improved accuracy for repeated stimuli).
  • Novel Stimuli Response: Schizophrenia patients showed significantly reduced gamma-band oscillatory responses to novel stimuli in occipital regions.
  • Repetition Suppression (Gamma-band): Patients exhibited significantly reduced repetition suppression of gamma-band activity in occipital cortex for both consecutive (NOLAG) and interleaved (LAG) repetitions.
  • Repetition Enhancement (Beta-band): Schizophrenia patients showed significantly reduced repetition enhancement of beta-band power in occipital cortex for both NOLAG and LAG conditions.
  • Event-Related Fields (ERFs):
    • Healthy controls showed significant repetition suppression of the early C1m component (30-80 ms) in occipital and, to a lesser extent, thalamic regions. This suppression was largely absent in schizophrenia patients.
  • Controls also showed repetition suppression of the late positive component (LPC) in the NOLAG condition, a finding that was not statistically significant in patients, though a trend was observed.

Limitations and Future Work

  • Small Sample Size: The authors acknowledge that the relatively small sample size (n=14 per group) might limit the statistical power to detect correlations between behavioral and neuroimaging measures, as well as between neuroimaging data and psychopathology.
  • Eye Movements: The study did not systematically assess the contribution of eye movements to the observed differences.
  • MEG Limitations for Deep Structures: While the study successfully used MEG to investigate thalamic signals, the authors note that deep brain structures can be challenging to study with MEG due to signal decay. However, the significant effect sizes observed suggest the feasibility of this approach.

Future Directions: * Investigate repetition suppression in at-risk populations and larger cohorts of schizophrenia patients to better understand its role in the disorder's development and progression. * Explore whether deficits in repetition suppression are linked to aberrant connectivity between cortical areas. * Further research into the neural mechanisms underlying these deficits, potentially using more advanced computational models or different neuroimaging techniques.

Summary updated

Abstract

<b>Background:</b> Cognitive dysfunctions represent a core feature of schizophrenia and a predictor for clinical outcomes. One possible mechanism for cognitive impairments could involve an impairment in the experience-dependent modifications of cortical networks. <b>Methods:</b> To address this issue, we employed magnetoencephalography (MEG) during a visual priming paradigm in a sample of chronic patients with schizophrenia (<i>n</i> = 14), and in a group of healthy controls (<i>n</i> = 14). We obtained MEG-recordings during the presentation of visual stimuli that were presented three times either consecutively or with intervening stimuli. MEG-data were analyzed for event-related fields as well as spectral power in the 1-200 Hz range to examine repetition suppression and repetition enhancement. We defined regions of interest in occipital and thalamic regions and obtained virtual-channel data. <b>Results:</b> Behavioral priming did not differ between groups. However, patients with schizophrenia showed prominently reduced oscillatory response to novel stimuli in the gamma-frequency band as well as significantly reduced repetition suppression of gamma-band activity and reduced repetition enhancement of beta-band power in occipital cortex to both consecutive repetitions as well as repetitions with intervening stimuli. Moreover, schizophrenia patients were characterized by a significant deficit in suppression of the C1m component in occipital cortex and thalamus as well as of the late positive component (LPC) in occipital cortex. <b>Conclusions:</b> These data provide novel evidence for impaired repetition suppression in cortical and subcortical circuits in schizophrenia. Although behavioral priming was preserved, patients with schizophrenia showed deficits in repetition suppression as well as repetition enhancement in thalamic and occipital regions, suggesting that experience-dependent modification of neural circuits is impaired in the disorder.
BibTeX
@inproceedings{sauer2020meg,
  author = {Andreas Sauer and Tineke Grent-‘t-Jong and Michael Wibral and Michael Grube and Wolf Singer and Peter J. Uhlhaas},
  title = {{A MEG Study of Visual Repetition Priming in Schizophrenia: Evidence for Impaired High-Frequency Oscillations and Event-Related Fields in Thalamo-Occipital Cortices}},
  booktitle = {Frontiers in Psychiatry},
  year = {2020},
  doi = {10.3389/fpsyt.2020.561973},
}
Figure