The present study reports the first comprehensive map of brain networks underlying latent inhibition learning and the first application of structural equation modeling to cytochrome oxidase data. In latent inhibition, repeated exposure to a stimulus results in a latent form of learning that inhibits subsequent associations with that stimulus. As neuronal energy demand to form learned associations changes, so does the induction of the respiratory enzyme cytochrome oxidase. Therefore, cytochrome oxidase can be used as an endpoint metabolic marker of the effects of experience on regional brain metabolic capacity. Quantitative cytochrome oxidase histochemistry was used to map brain regions in mice trained on a tone-footshock fear conditioning paradigm with either tone preexposure (latent inhibition), conditioning only (acquisition), conditioning followed by tone alone (extinction), or no handling or conditioning (naïve). The ventral cochlear nucleus, medial geniculate, CA1 hippocampus, and perirhinal cortex showed modified metabolic capacity due to latent inhibition. Structural equation modeling was used to determine the causal influences in an anatomical network of these regions and others thought to mediate latent inhibition, including the accumbens and entorhinal cortex. An uncoupling of ascending influences between auditory regions was observed in latent inhibition. There was also a reduced influence on the accumbens from the perirhinal cortex in both latent inhibition and extinction. The results suggest a specific network with a neural mechanism of latent inhibition that appears to involve sensory gating, as evidenced by modifications in metabolic capacity and effective connectivity between auditory regions and reduced perirhinal cortex influence on the accumbens. Keywords latent inhibition; metabolic mapping; structural equation modeling; cytochrome oxidase; sensory gating; perirhinal cortex; learning and memory In latent inhibition (LI), preexposure to a stimulus inhibits subsequent associations with that stimulus (Lubow and Moore, 1959). LI has been demonstrated in several behavioral paradigms in a variety of species, including humans, and is disrupted in schizophrenics (Lubow, 1989). Lesion and pharmacological studies have explored the neural mechanisms underlying LI and have identified several key brain regions. Conditioned drink suppression and eye blink conditioning studies demonstrated that lesions to the entorhinal cortex disrupt LI (Coutureau et al., 1999;Shohamy et al., 2000;Coutureau et al., 2002). Other studies using a conditioned Correspondence should be addressed to: Prof. F. Gonzalez-Lima, University of Texas at Austin, Department of Psychology, 1 University Station A8000, Austin, TX 78712-0187, USA. Phone (512) Fax (512) 471-4728, email: gonzalez-lima@mail.utexas.edu. Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will under...