TABLE 1.

Summary of evidence for excitatory-inhibitory(E-I) imbalance in schizophrenia.

	Excitatory-inhibitory(E-I) balance
	Glutamatergic dysfunction	GABAergic dysfunction
Cerebrospinal fluid (CSF) and Post-mortem studies	Findings of CSF glutamate levels are inconsistent. Enhanced levels of glutamate receptor antagonist kynurenic acid (KYNA) is consistent.	Lower CSF and plasma GABA levels in schizophrenic patients were first reported. The GABA levels increased with age, duration of illness, and with therapy of long-term neuroleptic.
Microcircuits	Increased glutamatergic activity in prefrontal, precommissural dorsal-caudate, basal ganglia, thalamus, and medial temporal lobe, decreased glutamatergic activity in the anterior cingulate cortex were reported.	Increased prefrontal, anterior cingulate cortex, and parieto-occipital cortex GABAergic activity; reduced GABA levels in ACC showed correlation with illness severity were reported.
Animal models	Genetic models: DISC1, RELN, CLU3 and ATX, PGC-1α–/–, 22q11.2 1.5 Mb deletion [Df(16)A ± ], NL2 R215H knock-in mouse model of schizophrenia suggested alterations of E-I balance.
Pharmacological studies	Dysfunction of NMDAR could cause schizophrenia-like behaviors in animal models and humans. Antagonists of NMDAR could produce positive, negative and cognitive symptoms in schizophrenia. D-serine and riluzole may be effective in the treatment of negative symptoms.	Dysfunction of ionotropic GABA type A receptors represents a core pathophysiological mechanism underlying cognitive dysfunction in schizophrenia. Lorazepam could exacerbate working memory deficits in schizophrenia. However, flumazenil could ameliorate working memory deficits in schizophrenics.