The exact roles of the medial prefrontal cortex (mPFC) in conditional choice behavior are unknown and a visual contextual response selection task was used for examining the issue. Inactivation of the mPFC severely disrupted performance in the task. mPFC inactivations, however, did not disrupt the capability of perceptual discrimination for visual stimuli. Normal response selection was also observed when nonvisual cues were used as conditional stimuli. The results strongly suggest that the mPFC is not necessarily involved in the inhibition of response or flexible response selection in general, but is rather critical when response selection is required conditionally using visual context in the background."Contextual response selection" enables an organism to respond flexibly and adaptively in various situations when encountering similar objects (potentially associated with conflicting responses) across different contextual settings. It is implicated in the literature that the prefrontal cortex (PFC) is critically involved in making flexible responses across different contexts (Kesner and Ragozzino 2003;Haddon and Killcross 2006;Marquis et al. 2007;Lee and Solivan 2008;Horga et al. 2011). The PFC receives direct afferent fibers from the hippocampus (Jay and Witter 1991;Thierry et al. 2000), which is one of the critical structures for contextual memory tasks (Hirsh 1974;Good and Honey 1991;Kim and Fanselow 1992;Kim and Lee 2011). Presumably related to such "contextual connections," perturbations of PFC lead to severe impairment in contextual response selection tasks in which visual contextual cues play significant roles (Haddon and Killcross 2006;Jo et al. 2007;Marquis et al. 2007;Lee and Solivan 2008).In prior studies, however, behavioral tasks required subjects to learn complicated multiple relationships between noncontextual stimuli (e.g., object, light, tone, etc.) and visual contexts (e.g., room cues) when testing contextual response selection (Haddon and Killcross 2006;Marquis et al. 2007;Lee and Solivan 2008). In other cases, rats learned biconditional associations between two different objects and locations in a maze and flexibly responded to a context-relevant object for obtaining reward (Lee and Solivan 2008). What these studies have tested is whether rats could learn associations between multiple individual cues and discrete contexts and whether PFC was necessary in flexibly responding (or not responding) to particular cue-context paired associates. It is unclear in these behavioral paradigms, however, how much "response selection" component was tested and also what proportion of resulting performance deficits following PFC manipulations could be attributed to contextual response selection per se. In the current study, in order to establish a more straightforward relationship between context and response selection behavior, we removed the requirement of associating other sensory cues (e.g., light, tone, and objects) with context and just made a single object (i.e., sand-filled jar) allow two distinctively...

No abstract