Back

Differential Vulnerability of Stimulus-Locked and Persistent Gamma Oscillations: Implications in Schizophrenia

Chung, D. W.; Ermentrout, G. B.

2026-05-29 neuroscience
10.64898/2026.05.29.728634 bioRxiv
Show abstract

Working memory depends on gamma oscillations generated across sensory and prefrontal cortices. In sensory cortices such as primary visual cortex (V1), stimulus-locked gamma oscillations encode stimulus information, while in prefrontal cortex (PFC), persistent gamma oscillations maintain this information after the stimulus is removed. In schizophrenia (SZ), gamma power is reduced in both V1 and PFC, consistent with deficits in sensory encoding and working memory maintenance in the illness. These two regimes of gamma oscillations arise from a canonical microcircuit involving pyramidal neurons (PNs) and parvalbumin-expressing interneurons (PVIs). Yet, whether stimulus-locked and persistent gamma oscillations are similarly or differentially vulnerable to synaptic alterations within this circuit in SZ remains unknown. To investigate this question, we used a mean-field model of the PN-PVI circuit generating either stimulus-locked or persistent gamma oscillations. We then assessed the effects of three synaptic alterations found in SZ: lower excitatory drive to PVIs (E[->]I), lower inhibitory drive to PNs (I[->]E), and greater variability in E[->]I synaptic strength. Each alteration produced larger gamma power deficits in the persistent regime than in the stimulus-locked regime. When applied together, these alterations interacted synergistically to reduce gamma power in both regimes, with the persistent regime exhibiting a more pronounced deficit. Among the three parameters, E[->]I synaptic strength was the strongest contributor to the synergistic loss of gamma power. Two-dimensional bifurcation analyses further revealed that this differential vulnerability arises from a narrower margin of oscillatory stability in the persistent regime, where the parameter values producing maximum gamma power sit closer to the Hopf bifurcation boundary. Together, these findings identify the persistent regime as intrinsically more fragile than the stimulus-locked regime, with the implications for understanding regional patterns of synaptic pathology and cortical gamma oscillations with distinct dynamics in SZ. Author summaryWorking memory depends on stimulus-locked gamma oscillations in sensory cortices such as primary visual cortex (V1) for encoding stimulus information, and persistent gamma oscillations in prefrontal cortex (PFC) for maintaining this information after stimulus offset. In schizophrenia (SZ), gamma power is reduced in both V1 and PFC, and postmortem human brain studies suggest that the underlying synaptic alterations are more severe in V1 than in PFC. Our computational modeling results suggest that this regional pattern arises because persistent gamma oscillations are intrinsically more fragile than stimulus-locked gamma oscillations, so that smaller synaptic alterations are sufficient to disrupt gamma oscillations in PFC while larger alterations are required to produce comparable disruption in V1. Together, these findings give rise to a differential vulnerability model of cortical gamma oscillations in SZ, linking the regional patterns of synaptic pathology to the deficits in gamma oscillations observed across sensory and prefrontal cortices in the illness.

Matching journals

The top 9 journals account for 50% of the predicted probability mass.

1
The Journal of Neuroscience
928 papers in training set
Top 1%
9.9%
2
Schizophrenia Research
29 papers in training set
Top 0.1%
9.0%
3
Biological Psychiatry: Cognitive Neuroscience and Neuroimaging
62 papers in training set
Top 0.1%
8.2%
4
eneuro
389 papers in training set
Top 1%
6.3%
5
Biological Psychiatry
119 papers in training set
Top 0.7%
4.8%
6
NeuroImage: Clinical
132 papers in training set
Top 1%
4.1%
7
Neuroscience
88 papers in training set
Top 0.3%
3.5%
8
Translational Psychiatry
219 papers in training set
Top 2%
3.5%
9
Neuropsychologia
77 papers in training set
Top 0.3%
3.5%
50% of probability mass above
10
Schizophrenia Bulletin
29 papers in training set
Top 0.3%
2.7%
11
Scientific Reports
3102 papers in training set
Top 44%
2.7%
12
PLOS Computational Biology
1633 papers in training set
Top 15%
1.9%
13
eLife
5422 papers in training set
Top 39%
1.9%
14
Schizophrenia
19 papers in training set
Top 0.2%
1.7%
15
Cerebral Cortex
357 papers in training set
Top 0.9%
1.7%
16
Neurobiology of Disease
134 papers in training set
Top 3%
1.7%
17
Human Brain Mapping
295 papers in training set
Top 3%
1.6%
18
Proceedings of the National Academy of Sciences
2130 papers in training set
Top 37%
1.3%
19
Communications Biology
886 papers in training set
Top 13%
1.3%
20
Network Neuroscience
116 papers in training set
Top 0.8%
1.2%
21
Hippocampus
46 papers in training set
Top 0.3%
1.2%
22
Journal of Neurophysiology
263 papers in training set
Top 0.7%
0.9%
23
Biological Psychiatry Global Open Science
54 papers in training set
Top 1%
0.9%
24
NeuroImage
813 papers in training set
Top 6%
0.8%
25
Psychological Medicine
74 papers in training set
Top 2%
0.8%
26
Frontiers in Aging Neuroscience
67 papers in training set
Top 3%
0.7%
27
PLOS ONE
4510 papers in training set
Top 68%
0.7%
28
Molecular Psychiatry
242 papers in training set
Top 4%
0.7%
29
Neuropsychopharmacology
134 papers in training set
Top 3%
0.7%
30
Brain
154 papers in training set
Top 5%
0.7%