Previous studies have utilized a lesion model of cortical injury that produces transient behavioral impairments to investigate the recovery of function process. To better understand the recovery process, it would be beneficial to use a lesion model that produces more severe, enduring, behavioral impairments. The purpose of experiment 1 was to validate whether large lesions of the sensorimotor cortex (SMC), which included the rostral forelimb and caudal forelimb regions, produced enduring behavioral deficits. Rats were given large unilateral electrolytic lesions of the SMC, administered either the N-methyl-D-aspartate (NMDA) antagonist, MK-801 or saline 16 h after injury, and tested on a battery of behavioral tests. Enduring behavioral deficits were observed, for at least 6 months, on two tests of forelimb placing while transient deficits were observed on the foot-fault and somatosensory neutralization tests. Administration of MK-801 facilitated recovery on the somatosensory neutralization test; however, it did not induce recovery on either forelimb placing test. A second experiment was performed to determine if earlier administration of MK-801, the NMDA antagonist magnesium chloride (MgCl2), or the antioxidant N-tert-butyl-alpha -phenylnitrone (PBN) could induce behavioral recovery in this chronic model. Treatment with these drugs induced behavioral recovery on the forelimb placing tests, whereas, the saline-treated rats did not show any signs of behavioral recovery for at least 3 months. Anatomical analysis of the striatum showed that MK-801 and MgCl2 but not PEN reduced the extent of lesion-induced striatal atrophy. These results suggest that administration of MK-801, MgCl2, or PEN shortly after cortical injury can induce recovery of function when recovery is otherwise not expected in un-treated rats. (C) 2000 Elsevier Science Inc.