Experimental models of hepatic encephalopathy (HE) are limited by difficulties in objectively monitoring neuronal function. There are few models that examine a well-defined neuronal pathway and lack the confounding effects of anesthetics. Motor-evoked potentials (MEPs) assess the function of the motor tract, which has been shown to be impaired in patients with cirrhosis. MEPs were elicited by cranial stimulation (central) and compound motor action potential by sciatic nerve stimulation (peripheral) in several models of HE in the rat. The experiments were performed using subcutaneous electrodes without anesthetics. Brain water content was assessed by gravimetry, brain metabolites were measured by magnetic resonance spectroscopy, and amino acids in microdialysates from the frontal cortex were analyzed by high-performance liquid chromatography. Abnormalities of MEP were observed in acute liver failure (ALF) induced by hepatic devascularization in relation to the progression of neurological manifestations. Similar disturbances were seen in rats with portocaval anastomosis after the administration of blood or lipopolysaccharide, but were absent in rats with biliary duct ligation. Hypothermia (≤35°C) and mannitol prevented the development of brain edema in acute liver failure, but only hypothermia avoided the decrease in the amplitude of MEP. Disturbances of MEP caused by the administration of blood into the gastrointestinal tract in rats with portocaval anastomosis were associated with an increase in ammonia, glutamine, and glutamate in brain microdialysate. Conclusion: Assessment of MEP in awake rats is a valid method to monitor HE in models of ALF and precipitated HE. This method shows the lack of efficacy of mannitol, a therapy that decreases brain edema, and relates disturbances of the function of the motor tract to ammonia and its metabolites. Copyright © 2010 American Association for the Study of Liver Diseases.