nimal models are important in investigating theorigin and the mechanisms underlying a human diseaseand designing new therapies, and have been widely usedin various areas of medical research.Animal models havenot been, however, very popular in psychiatric research.Reproducing psychiatric disorders in animals has oftenbeen considered difficult, if not impossible. Modelingschizophrenia is an example of a particularly difficult task,because it is a uniquely human disease, and its mostprominent symptomshallucinations, delusions, andthought disordercannot be reproduced in an animal.Recent new evidence about the neurobiology of this dis-ease has opened new possibilities of animal research. Inparticular, abnormalities in the neural circuitry involvingthe hippocampus, prefrontal cortex, and the dorsomedialthalamus have been reported recently, in addition to pre-viously recognized abnormal function of the dopaminer-gic system. Cytoarchitectural and molecular studies of thebrain, as well as neuropsychological studies showing thatschizophrenia symptoms emerge in young adulthood butsubtle motor and behavioral abnormalities are presentearly in life, suggest a neurodevelopmental origin of thedisease.To address a neurodevelopmental origin of schizophrenia,numerous studies modeling schizophrenia in animals havefocused on neonatal damage of restricted brain regions inrats1-11 and in monkeys.12-15 The main objective of many ofthese studies is to disrupt development of the hippocam-pus, a brain area consistently implicated in human schizo-phrenia,16-25 and thus disrupt development of the wide-B a s i c r e s e a r c h3 6 1Neonatal disconnection of the rat hippocampus: a neurodevelopmental model of schizophrenia Barbara K. Lipska, PhDAKeywords: animal model; schizophrenia; hippocampus; prefrontal cortex; neuro-development; neonate; locomotion; dopamineAuthor affiliations: Clinical Brain Disorders Branch, Intramural ResearchProgram, National Institute of Mental Health, NIH, IRP, Bethesda, Md, USAAddress for correspondence: Barbara K. Lipska, 10 Center Drive, Bldg 10,Rm 4N306, Bethesda, MD 20892-1385, USA(e-mail: lipskab@intra.nimh.nih.gov)In the context of our current knowledge about schizo-phrenia, heuristic models of psychiatric disorders may beused to test the plausibility of theories developed on thebasis of new emerging biological findings, explore mech-anisms of schizophrenia-like phenomena, and developpotential new treatments. In a series of studies, we haveshown that neonatal excitotoxic lesions of the rat ventralhippocampus (VH) may serve as a heuristic model. Themodel appears to mimic a spectrum of neurobiologicaland behavioral features of schizophrenia, including func-tional pathology in presumably critical brain regions inter-connected with the hippocampal formation and targetedby antipsychotic drugs (the striatum/nucleus accumbensand the prefrontal cortex), and leads in adolescence orearly adulthood to the emergence of abnormalities in anumber of dopamine-related behaviors. Moreover, ourdata show that even transient inactivation of the VH dur-ing a critical period of development, which produces sub-tle, if any, anatomical changes in the hippocampus, maybe sufficient to disrupt normal maturation of the pre-frontal cortex (and perhaps, other interconnected late-maturing regions) and trigger behavioral changes similarto those observed in animals with the permanent excito-toxic lesion. These results represent a potential new modelof aspects of schizophrenia without a gross anatomicallesion. Dialogues Clin Neurosci. 2002;4:361-367.